Download PDF
If the PDF viewer below does not load, use the download button above or the link to view on BHL.
If the PDF does not load in the viewer above, use the download button or the link to view this item on the Biodiversity Heritage Library website.
Full Text (OCR)
Show OCR Text (479,802 characters)
L. - /-C.7TC
s&aa-
NATIONAL ANTARCTIC EXPEDITION
1901-1904
// ^
V, /
Moll,
NATURAL HISTORY
IVISZIZ
VOL. I.
GEOLOGY
(FIELD-GEOLOGY : PETROGRAPHY)
Diviiion of Ma3W®
LONDON :
PRINTED BY ORDER OF THE TRUSTEES OF
THE BRITISH MUSEUM.
1907
{All Rights Reserved)
'smithso/v^
3 0 1986
Sold by Longmans and Co., 39 Paternoster Row, E.C. ; Bernard Quaritch, 15 Piccadilly, W.
Dulau and Co., 37 Soho Square, W. ;
AND AT
The British Museum (Natural PIistory), Cromwell Road, London, S.W.
PREFACE.
When, iii 1901, the Expedition of the S.S. ‘Discovery,’ under Captain Scott, R.N.,
was sent to the Antarctic Regions, the Trustees of the British Museum gave their
assistance to this national enterprise by allowing the cases eontainiug the natural
history specimens which might be obtained by the Expedition to be sent to the
Natural History Museum for unpacking and sorting. They further undertook to
publish a detailed report on the collections so obtained, under the superintendence
of the Director of the Natural History Departments.
Some of the most important collections have been dealt with by naturalists who
were members of the Expedition. Thus, the Mammals and Birds are described by
Dr. Edward A. Wilson, the Isopoda and Pycnogonida by Mr. T. V. Hodgson, and
the Rocks (in relation to Field Geology) by Mr. II. T. Ferrar. Other groups have
been dealt with by members of the staff of the Natural History Departments of
the British Museum : Mr. Boulenger describes the Fishes ; Mr. E. A. Smith, the
Gastropoda, Lamellibranchia, and Brachiopoda ; Mr. Jeffrey Bell, the Echinoderma ;
Dr. Caiman, the Crustacea Decapoda, and the Cumacea ; Mr. Kirkpatrick, the non-
caleareous Sponges ; whilst Dr. G. T. Prior has prepared a petrographical description
of the Rock-specimens.
It has been necessary to obtain the assistance of other specialists in order to deal
with the rest of the collections. So far as the latter group of contributors is
concerned, the following is a list of the subject-matters, together with the name of the
naturalist who has undertaken the work in each case : —
Embryos of Seals
Anatomy of Emperor Penguin
Tunicata ....
Cephalodiscus .
Cephalopoda
Nudibranohs and Pteropods
POLYZOA ....
Eggs and Young of Asterias
Amphipoda
ScHIZOPODA
NeBALIzE ....
OSTRACODA
Copepoda ....
Dr. Marrett Tims.
Mr. W. P. Pycraft.
Prof. IIerdman.
Dr. Ridewood.
Dr. IIoyle.
Sir Charles Eliot, K.C.M.G.
Mr. H. W. Burrows.
Prof. MaoBride.
Mr. A. 0. Walker.
Mr. Holt.
Dr. J. Thiele.
Prof. Brady.
Dr. Wolfenden.
b 2
PREFACE.
Cirri pedia ....
. Prof. Gruvel.
Myzostoma ....
. Prof. v. Graff.
Acari .....
. Dr. Trouessart.
COLLEMBOLA ....
. Prof. Carpenter.
POLYCH/ETA ....
Prof. Ehlers.
Gephyria ....
Mr. A. E. Shipley.
Ch.etognatha ....
Dr. Fowler.
Nemertines ....
. Prof. IIubrecht.
Free Platyhelminthes
. Mr. F. F. Laidlaw.
Cestoda .....
Mr. A. E. Shipley.
Nematoda ....
Dr. v. Linstow.
ZOANTHARIA ....
Mr. Clubb.
Alcyonaria and Pennatulida
. Prof. Hickson.
IIydromedus.e ....
. Mr. E. T. Brown.
Calcareous Sponges .
Mr. Frewen Jenkin.
PiADIOLARIA ....
Mr. Lewis II. Gough.
Mosses .....
M. Jules Cardot.
Lichens .....
Mr. Darbishire.
Alga; (Marine) ....
Mrs. Gepp.
Alga: (Fresh- water) .
. Dr. Fritsch.
Alga; (Calcareous) .
Dr. Foslie.
Phytoplankton ....
. Dr. Lewis II. Gough.
The work of securing the assistance of
these specialists and of distributing
the
collections has been performed by Mr. Jeffrey
Bell, of the Zoological Department,
who
has also acted as sub-editor of the Zoological and Botanical portions of the reports.
The Keeper of Minerals, Mr. Fletcher, has superintended the reports in the subjects
belonging to his department.
The Director desires to acknowledge the ability and energy which have been
brought to bear on the preparation of the Zoological reports by Mr. Jeffrey Bell.
Owing to his care, the reports have been got ready by the various contributors and
published within a reasonable time after the return of the ‘ Discovery ’ from the
Antarctic Regions. Neither trouble nor expense has been spared in order to
render the illustration and presentation of the Natural History of the Expedition
worthy of the generous efforts both of Captain Scott and his fellow-explorers and
of those who provided the funds for that enterprise.
E. Ray Lank ester.
October, 1906.
PREFACE TO VOLUME I.
The mineral-specimens collected during the ‘ Discovery ’ Antarctic Expedition being
virtually all of them rock-specimens, their importance depends, not merely on their
own characters, but on the mutual relations of the masses which they represent ; in
these circumstances, a Report descriptive of the specimens themselves can be of little
scientific value unless preceded by an account of the rock-masses of which they have
formed part.
Mr. II. T. Ferrar, Geologist to the Expedition, had lived in the region and
collected nearly all the specimens, and was obviously the one to be invited to
prepare a monograph of the Field-geology. Fortunately he was able to accept the
invitation, and to submit the manuscript of his Report before leaving England to take
up an appointment on the Geological Survey of Egypt.
The scientific description of the specimens was entrusted to Dr. G. T. Prior,
Assistant in the Mineral Department, who had already examined and described
the mineral-specimens collected during the ‘ Ross ’ and the ‘ Southern Cross ’
Antarctic Expeditions.
The points regarded by the authors as deserving special attention are con-
veniently indicated in the respective Summaries (pp. 98, 139).
The elaborate Index to the volume has been made by Dr. Prior.
It has been my duty, as Keeper of the Mineral Department, to supervise the
preparation and publication of these Reports, but the scientific part of the work has
been done entirely by the respective authors.
Mineral Department,
British Museum (Natural History),
May 1, 1907.
L. Fletcher.
TABLE OF CONTENTS
I.— REPORT ON THE FIELD-GEOLOGY OF THE REGION EXPLORED DURING THE
CHAPTER
I.
‘DISCOVERY’ ANTARCTIC EXPEDITION, 1901-4.
•
By H. T. Ferrar, M.A., F.G.S., Geologist to the Expedition.
Islands, chiefly off the coast of South Victoria Land
PAGE
1
II.
The Ross Archipelago ........
.
8
III.
The Mainland of South Victoria Land .....
17
IV.
The Gneissic Rocks and Crystalline Limestone
25
V.
The Granites ..........
32
VI.
The Beacon Sandstone Formation ......
39
Appendix to Chapter VI. — Report on the Plant-remains from
Sandstone. By E. A.. Newell Arber, M.A., F.L.S., F.G.S. .
the Beacon
48
VII.
The Dolerites ..........
49
VIII.
The Sea-ice and the Shore-ice .......
55
IX.
The Land-ice ..........
63
X.
The Land-ice — continued ........
76
XI.
Denudation
87
Appendix to the Report on the Field-geology. — Notes relative to Macquarie
and Auckland Islands, outside the Antarctic Circle ....
95
Summary .............
98
II.— REPORT ON THE ROCK-SPECIMENS COLLECTED DURING THE
‘DISCOVERY’ ANTARCTIC EXPEDITION, 1901-4.
By G. T. Prior, M.A., D.Sc., F.G.S., Assistant in the Mineral Department, British Museum.
PAGE
INTRODUCTORY ............... 101
CHAPTER .
I. Volcanic Rocks 102
Basalts 102
Ivenytes ............. 110
Phonolitic Trachytes and Phonolites 113
Chemical Relations of the Volcanic Rocks 119
II. The Basement-rocks of South Victoria Land 124
Crystalline Limestone and Gneiss . . . . . . . .124
Granites and Diorites .......... 125
III. Dyke-Rocks (Lamprophyres, *etc.) ......... 129
Camptonites ............ 129
Kersantites ............ 130
Banakites . . . . . . . . . . . . .131
IV. The Beacon Sandstone and other Sedimentary Rocks 134
V. The Dolerites . . . . . . . . . . . . .136
Summary ............... 139
III.— INDEX TO THE VOLUME.
ILLUSTRATIONS IN THE TEXT.
Fig. 1. — Sturge Island, Balleny Islands, showing transition from “ Piedmont-aground”
to “ Piedmont-afloat
Fig. 2. — Two of the Possession Islands. The taller one shows the junction of two
types of rock ............
Fig. 3. — East side of Coulman Island, showing the horizontal structure of the
ROCKS, AND THE “ PlEDMONT-AGROUND ” WHICH SURROUNDS THE ISLAND .
Fig. 4. — Castle Rock and Mount Erebus .........
Fig. 5. — Cape Crozier and Mount Terror
Fig. 6. — Cape Ad are Peninsula, from Ross Sea ........
Fig. 7. — A volcanic cone on the mainland ; the summit of Cape Jones. The
‘Discovery’ in a gulf in the Lady Newnes “Piedmont-afloat”
Fig. 8. — Mount Nansen, the tabular mountain south of Cape Washington .
Fig. 9. — Mount Huggins and the Royal Society Range
Fig. 10. — King Edward VII Land
Fig. 11. — The crystalline limestone on the hill Jl5 south side of the Blue G-lacier .
Fig. 12. — The crystalline limestone on the north side of the Blue Glacier at G4
Fig. 13. — The gneiss at the east end of the Lower Kukri Hills, near the hill H
Fig. 14. — Looking up the Ferrar Glacier, Northern Foothills on the left, Cathedral
Rocks near the centre, and the Kukri Hills on the right ....
Fig. 15. — Dolerite upon granite on the north side of Granite Harbour
Fig. 16. — Hollowed granite-boulder in the Snow Valley near the Royal Society Range
Fig. 17. — Dolerite-cliff standing back from the edge of the granite of Cathedral
Rocks
Fig. 18. — The horizontal upper surface of the granite on the south side of the
Kukri Hills
Fig. 19. — Dolerite-sill in the Beacon Sandstone near Finger Mountain
Fig. 20. — The Inland Forts. Sandstone capped by dolerite
Fig. 21. — Impression in sandstone at West Groin. Copy of sketch made in the field .
Fig. 22.— -Finger Mountain. Wedge of sandstone in the dolerite .....
Fig. 23. — Terra Cotta Mountains, showing dykes of dolerite
Fig. 24. — Depot Nunatak, from the east .........
Fig. 25. — Columnar dolerite of Depot Nunatak ........
Fig. 26. — Columnar dolerite at the foot of Knob Head. The large boulder on the
SKY-LINE IS OF GRANITE ...........
Fig. 27. — The dark band in the Kukri Hills on the right shows the dolerite-sheet
RESTING UPON THE EVEN SURFACE OF THE GRANITE ......
Fig. 28. — The South Arm, with tabular features exhibited on the left, and Knob
Head on the right ............
PAGE
3
4
5
9
10
17
19
21
23
24
26
27
28
29
32
34
35
37
40
42
43
45
47
49
50
52
53
54
VOL. I.
C
X
ILLUSTRATIONS IN THE TEXT.
PAGE
Fig. 29. Crystals of ice which have grown upon a fishing line several fathoms
BELOW THE LOWER (OR FREEZING) SURFACE OF THE SEA-ICE ..... 55
Fig. 30.— The pack-ice, seen from the crow’s nest of the ship 56
Fig. 31.— The ‘Discovery’ brought to a standstill by pack-ice 57
Fig. 32.— Water-holes in sea-ice at Cape Armitage and Hut Point in January, 1904 . 58
Fig. 33.— Sea-ice breaking away from the Winter Quarters in 1902 .... 59
Fig. 34. — The Relief-ships forcing their way through the barrier of floe-ice in 1904 60
Fig. 35. — The ice-foot at Hut Point 61
Fig. 36. — Shore-ice wrapping the land near the foot of Castle Rock . . . .61
Fig. 37.— Ice-foot and pack-ice in Wood Bay at foot of Mount Melbourne ... 65
Fig. 38.— The Ross Piedmont from the side of Mount Terror, showing the cliff-edge
AND FLAT UPPER SURFACE ........... 68
Fig. 39. — Three ice-tongues falling into North Fork ....... 72
Fig. 40. — An iceberg, over 150 feet high, tilted through nearly 90° .... 74
Fig. 41. — Uplift of morainic material in the ice at the foot of Knob Head . . 76
Fig. 42. — The dark band of ice-without-grain, below normal glacier-ice, at the foot
of Knob Head ............ 77
Fig. 43. — Moraine on the Ferrar Glacier ......... 78
Fig. 44. — Glacier-table formed by a layer of gravel ....... 79
Fig. 45. — Moraines on floating ice at the head of McMurdo Sound .... 80
Fig. 46. — Moraine-cone of ice-scratched stones, on which the Balanus shells were
found, on the floating glacier-ice in the bay between White Island and
Black Island ............. 80
Fig. 47. — Moraines supported by ice, on the west side of McMurdo Sound ... 81
Fig. 48.— Undulating surface of hard “ marbled ” snow ...... 84
Fig. 49. — The two lower men are standing upon the upper surface of sea-ice
DEPRESSED BY SNOW BELOW WATER-LEVEL ........ 85
Fig. 50. — Hollowed granite-boulder with incrustation of calcium carbonate, near
Descent Pass ............. 87
Fig. 51.— Saline Pond in moraines on west side of McMurdo Sound .... 88
Fig. 52. — Water separating mud from gravel in the moraines on Minna Bluff . . 89
Fig. 53. — Water-channel on floating ice in McMurdo Sound . . . ... .90
Fig. 54. — Seaward edge of the glacier-ice floating in McMurdo Sound ... 90
Fig. 55.— Fractured dome in the floating glacier-ice, near the spot where sodium
sulphate crystals were found, two miles from the north end of White Island 91
Fig. 56. — A Glacier descending from the top of Coulman Island into the sea . . 93
Fig. 57. — The strand and the steep coast-line of the east side of Macquarie Island . 95
Fig. 58. — The south side of Ross Harbour, Auckland Islands, showing submerged valleys 97
Fig. 59. — Pseudomorph after hornblende, in basalt (691) from Castle Rock, showing
inclusions of apatite. (Magnification, 10 diam.) ...... 103
Fig. 60. — Inclusions in augite of nodule from Turtle Back Island. (Magnification,
100 diam.) . . . . . . . . . . . . 108
Fig. 61.— Olivine with magnetite-inclusions, from Castle Rock. (Magnification, 60 diam.) 108
ILLUSTRATIONS IN THE TEXT.— PLATES.
xi
' PAGE
Fig.* 62. — Kenyte with porphyritic anorthoclase, boulder from Turtle Back Island . 110
Fig. 63.— Spherule with magnetite, in glassy kenyte, from Turtle Back Island. (Magnifi-
cation, 200 cliam.) . . . . . . . . . . . .111
Fig. 64. — Leucite-crystals in base of glassy hornblende-trachyte (264) from Observa-
tion Hill. The prismatic crystals are mostly augite : the long one at the
top is hornblende. (Magnification, 150 diam.) 118
Fig. 65. — Dendritic magnetite in glassy base of trachyte from Observation Hill.
(Magnification, 200 diam.) . . . . . . . . . . 118
Fig. 66.— Hornblende-inclusions in the trachyte of Observation Hill. (Magnification,
150 diam.) ............. 118
Fig. 67.— Graphical representation of the chemical composition of the olivine-basalt
(656) from near the Gap .......... 122
Fig. 68. — Graphical representation of the chemical composition of the leucite-kenyte
(818) from Cape Boyds ........... 122
Fig. 69. — Crystalline limestone with chondrodite, from Southern Foothills. (Natural size) 124
Fig. 70. — Micropegmatite surrounding felspar in augen-gneiss (727) from the Kukri
Hills. (Magnification, 25 diam.) . . . . . . . . .125
Fig. 71. — Quartz-grains in Beacon Sandstone (679) from Inland Forts. The dotted
LINES SHOW THE ORIGINAL ROUNDED OUTLINES OF THE GRAINS. (Magnification,
20 diam.) ............. 134
Fig. 72. — Micropegmatite in dolerite (662) from Depot Nunatak. (Magnification, 100 diam.) 136
PLATES.
{At end of volume.)
Plate I. — A Panorama of Mount Terror from the south-east.
Plate II. — A Panorama of Winter Quarters, showing Harbour Heights, Crater Hill and
Observation Hill.
Plate III. — The north end of the Roybal Society Range, and the south side of the Kukri
Hills. Looking up Ferrar Glacier from Descent Pass.
Plate IY. — View down the East Fork of the Ferrar Glacier, showing the low granite-
hills BETWEEN G3 AND Gs AND THE GNEISS-EXPOSURE AT THE FOOT OF CATHEDRAL
Rocks.
Plate V. — A Panorama of the south side of the Ferrar Glacier as seen from a point
ABOVE THE SOLITARY ROCKS.
Plate YI. — The overflow of the Koettlitz Glacier into a tributary valley containing
AN ICE-SLAB.
Plate YII. — Geological Sections.
Section I. — From east to west across the Royal Society Range, from McMurdo
Sound to the Inland-ice.
Section II. — From south to north, across East Fork, the Kukri Hills and
North Fork.
Section III. — From west to east along the Kukri Hills.
Xll
PLATES.— MAPS.
Plate VIII. — Fig. 1. — Olivine-basalt (656) from cliff between Gap and Horseshoe Bay (p. 104).
Fig. 2. — Gabbro-i.ike nodule (415) in limburgite, from Winter Quarters (p. 108).
Fig. 3. — Leucite-kenyte (818) from Cape Royds (p. 111).
Fig. 4. — Phonolitic trachyte (248) from Mount Terror (p. 115).
Fig. 5. — Phonolite (530) from Black Island (p. 11G).
Fig. G. — Phonolitic hornblende-trachyte (277) from Observation Hill (p. 117).
Plate IX. — Fig. 1. — Augen-gneiss beloiv D4, Kukri Hills (p. 125).
Fig. 2.— Diorite (715) from Cathedral Rocks (p. 127).
Fig. 3. — Diorite to essexite (572) from the Blue Glacier (p. 128).
Fig. 4. — Kersantite (579) from Northern Foothills (p. 130).
Fig. 5. — Camptonite (839) from Southern Foothills (p. 129).
Fig. G. — Dyke-rock (714) related to banakite, from the Northern Foothills
(p. 131).
Plate X.— Fig. 1.— Dolerite (662) from Depot Nunatak (p. 136).
Fig. 2. — Dolerite (69G), 2 ft. from junction with sandstone, Dry Valleys (p. 138).
Fig. 3. — Dolerite (687), 6 in. from junction with sandstone, Inland Forts (p. 138).
Fig. 4. — Dolerite (695), 2 in. from junction with sandstone, Dry Valleys (p. 138).
Fig. 5.— Junction of dolerite and sandstone (669) at B, (p. 138).
Fig. 6. — Dolerite (154) with granitic patches, Granite Harbour (p. 139).
MAPS.
(In 2Joclcet at end of volume.)
Chart of the Antarctic Ocean between Lat. 66° S. and 83° S., and Long. 150° E. and 150° W.
Map of the District near the ‘ Discovery ’ Winter Quarters.
BE PORT ON THE FIELD-GEOLOGY
OF THE REGION EXPLORED DURING THE
‘DISCOVERY’ ANTARCTIC EXPEDITION, 1901-4.
By II. T. Ferrar, M.A., F.G.S., Geologist to the Expedition.
Chapter I.
ISLANDS, CHIEFLY OFF THE COAST OF SOUTH VICTOPJA LAND.
The part of South Victoria Land known to us consists of a great range, or series
of mountain-ranges, stretching in an almost straight line from latitude 71° S. to
lat. 82° S., a distance of about 800 miles. Some of the mountains rise to a height
of 13,000 feet, and it is remarkable that there is no extensive area of land lower
than 4000 feet. Off this bold coast-line is a shallow sea (Ross Sea), with occasional
islands arranged along a line roughly parallel to the coast and close in under it.
The earliest specimens brought back from the Ross Quadrant were those obtained
by Captain Balleny in the year 1839 from the Balleny Islands. Shortly afterwards
the ‘ Erebus ’ and ‘ Terror ’ Expedition under Sir James Clarke Ross brought back rock-
specimens from other outlying islands, and until the year 1895 no additional specimens
of Antarctic rocks were obtained from this area. It was also known that (l) the
Balleny Islands are volcanic, one of them possessing an active volcano ; (2) South
Victoria Land consists of a great range of mountains probably volcanic,* and with at
least one volcano still active. The specimens include scoriae and olivine-basalt from
Young Island, one of the Balleny group, f basalts, palagonite-tuffs, and granites
from the largest of the Possession Islands, and basalt from Franklin Island, one
of the isolated islands off the coast, j
A French expedition contemporary with that of Ross also obtained granites § from
* Ross, ‘ Voyage in the Southern and Antarctic Regions, 1839-43,’ 1847, vol. ii, p. 415.
t 1 The Antarctic Manual’ (Roy. Geogr. Soe.), 1901, p. 341.
t Prior, Mineralogical Magazine, 1899, vol. xii, p. 91.
§ ‘ The Antarctic Manual ’ (Roy. Geogr. Soc.), 1901, p. 449.
VOL. i.
B
2
H. T. FERRAR.
low rocky islets lying off the coast of Adelie Land, and these strongly suggested the
existence of a continental mass of land.
The fact that blocks of gneiss and granite, probably dropped from icebergs, were
dredged up in high southern latitudes during the ‘ Challenger ’ expedition was also
regarded as evidence of the existence of a continent. Fragments of mica-schists, sand-
stones, limestones and shales, were also dredged up at the same time.* * * § This fact was
sufficient to render it extremely probable that sooner or later fossiliferous sedimentary
rocks would be discovered.
In the year 1895 Mr. Borchgrevink obtained schistose and granitic rocks from
Cape Adare.
The ‘ Southern Cross ’ collection described by Dr. Prior f includes various
plutonic and volcanic rocks as well as siliceous slates, the latter being apparently
the first sedimentary rocks found in situ in South Victoria Land. The slates are
noted as occurring at the head of Robertson Bay. These slates are directly coyered
by the basalts of Cape Adare on the east, and they have been followed northward
along the coast for some five miles.
The islands may be conveniently considered in the order of increasing latitude,
commencing with the Balleny group near the Antarctic circle.
Balleny Islands.
This group consists of five islands lying between longitudes 161° E. and 1G5° E.,
and latitudes 66° S. and G8° S., that is to say, about the Antarctic circle. They were
discovered by Captain Balleny in 1839. He brought back specimens from Young
Island, and reported the presence of an active volcano on Buckle Island, a report
afterwards confirmed by the ‘ Southern Cross ’ Expedition. J
Rowe Island, the most northerly of the group, was very distant from the
‘ Discovery’s ’ track. Balleny remarks that it is low and offers no remarkable
feature.
Young Island, one of the largest, is roughly 10 miles long and 5 miles broad, and,
according to Balleny, is the highest. It rises to an estimated height of 12,000 feet. It
is girt by a high cliff and has the form of a terraced cone. The rock-specimens
collected here in 1839 were the first obtained from what is now known as the Ross
Quadrant. They “ prove to be scoriae and basalt with crystals of olivine.” §
Borradaile Island is about 500 feet high and 2 miles long, and, like the others, is
bounded by vertical cliffs.
* See also 1 Nature,’ 1898, vol. lvii, p. 420.
t Prior, Rep. ‘ Southern Cross ’ Collections (British Museum), 1902, p. 325.
t ‘The Antarctic Manual ’ (Roy. Geogr. Soc.), 1901, pp. 499, 500.
§ ‘The Antarctic Manual’ (Roy. Geogr. Soc.), 1901, p. 341.
BALLENY ISLANDS.— SCOTT ISLANDS.
Buckle Island, which bears the active volcano, is probably 15 miles long. The
surrounding cliff varies from 100 to 1000 feet in height, while above it the land rises
as a dome to a height of about 4000 feet. The volcano is situated on the north end,
which is otherwise low and flat.
Sturge Island (Fig. 1) is about 20 miles long and 7 broad, and rises to a height
of over 10,000 feet. It is claviform and smooth in outline, and appears as an elongated
dome surrounded by a rock-cliff which varies from 1000 to 3000 feet in height. At
the north end are seen roughly parallel irregular lines, dipping at an angle of about
15° to the west. Approximately parallel to them are numerous conspicuous lenticular
bands of light-yellow colour. The soundings in this area are uniform over great
Fig. 1. — Stuege Island, Balleny Islands, showing tbansition prom “ Piedmont-agbodnd ” to “ Piedmont-afloat.”
distances, ranging from 260 to 270 fathoms. The sea-bottom is covered with a mixture
of rock-flour (902) * and ice-scratched stones (866),* including fragments of dolerite and
micaceous schists (see pp. 135 and 139).
Scott Islands.
The Scott Islands were discovered by Captain Colbeck in December, 1902,f and
are flat-topped rocky islets, which rise 300 feet sheer out of the shallow sea. The
smaller is practically an isolated pillar (Haggitt’s Pillar), while the larger is about two
miles across. Mr. J. D. Morrison landed on the larger island and collected specimens
of a trachytic rock (see p. 114) regarding which he gave the following information : — -
“No. 1 was taken from the south-east side, lat. 67° 24 '5' S., long. 179° 55*5' W.,
* The numbers refer to the author’s List of Specimens,
t Colbeck, Geog. Journ., 1905, vol. xxv, plate, p. 402.
B 2
4
H. T. FERRAR.
being broken off the highest accessible point, about 12 feet above water-level. The
strata seemed about 2 feet thick, dipping to the south-east side at an angle of 45°,
striking S.W. and N.E.”
The Possession Islands.
This group consists of two large and five small islands, close under the highest
peaks of South Victoria Land and about 5 miles off shore and a little north of the 72nd
parallel of latitude (Fig. 2). They were discovered by Sir James Clarke Ross in 1841,
when a landing was made and rock-specimens were collected. They include basalts,
palagouite-tuff, phonolite, and fragments of granite, but the latter were probably not
found in situ * Mr. C. E. Borchgrevink and Captain Jenssen both landed here in
Fig. 2. — Two op the Possession Islands.
The tallee one shows the junction of two types op kock.
1895, and collected rock-material. The specimens brought back by the former have
been described by Messrs. T. W. E. David, W. F. Smeeth and J. A. Schofield. f
In December, 1902, Mr. Morrison, landing from the relief ship ‘ Morning,’
collected rock-specimens. He obtained only two types of rock in situ : one (No. 2) is a
palagonite-tuff and the other (No. 3) a grey hornblende-basalt. He gave the following
information relative to them : — No. 2 was taken from the south-west shore of Posses-
sion Island, 18 feet above water-level, lat. 71° 5G' S., long. 171° 10' E. There are no
signs of stratification, but there is a very distinct vertical line-of- parting between the
I'ocks forming Nos. 2 and 3. No. 3 was taken from a piece of ice floating close in
shore on the S.S.E. side. The cliffs are about 150 feet high and overhanging.”
From the view seen from the deck of the 1 Discovery,’ it would appear that the
higher part is composed of palagonite-tuff, and the south side, ending in a bold clift 300
* Prior, Mineralogical Magazine, 1899, vol. xii, p. 75.
f Journ. Roy. Soc., New South Wales, 1895, vol. xxix, pp. 461-492,
THE POSSESSION ISLANDS.— COULMAX ISLAND.
5
feet high, of basalt. This island, the largest of the group, is 2 miles long in a north-
and-south direction. The landing place is at the northern end, which is low, flat and
prominently terraced. The next largest island may be a mile across : like the first it
has vertical sides. Of the other islands, three are flat-topped, about a quarter of a
mile in diameter, and 50 feet high. They appear to be roughly rectangular in shape,
and have bare vertical sides. A fourth forms an isolated pillar appearing to be made
up of vertical columns, while a fifth is less than a quarter of a mile across and over
100 feet high. This last shows an uneven junction of two rocks at about 50 feet
above the sea (Fig. 2). The distribution of snow would seem to show that the rocks
are tuff and basalt, the tuft’ being uppermost. The junction is irregular, but on the
whole slopes from west to east, away from the high land.
Fig. 3. — East side of Coulman Island, showing the hoeizontal structoke of the
ROCKS, AND THE “ PIEDMONT- AGROUND ” WHICH SURROUNDS THE ISLAND.
Coulman Island.
This island is situated in latitude 73 1 S., longitude 170° E. It was discovered by
Sir James Clarke Ross, but the first rock-specimens were brought back by the
‘ Southern Cross ’ expedition ; they were determined to be hornl flende-basalt and
basalt-agglomerate,* but no details were given as to the distribution of the rocks. Here
also the land is characterised by bare rock-cliffs more than 1000 feet high, which fall
sheer away to the sea. The island has an even outline. Its top is nearly flat at Cape
Wadworth, the north end, but at the south end rises to nearly 3000 feet. It is about
20 miles long, and averages 7 miles in breadth. Cape Anne, the south end, terminates
in a bare cliff over 2000 feet high. It shows chrome-yellow patches in several places,
Prior, Rep. ‘ Southern Cross’ Collections (British Museum), 1902, p. 322.
II. T. FERRAR.
6
and, by analogy with the yellow lenticular patch of basalt-agglomerate (82) at Cape
Wadworth, we may infer that the south end is also partly of the same rock. As the
yellow patches lie almost horizontally it is highly probable that the island consists
of alternating sheets of basalt and basalt-agglomerate (Fig. 3).
In addition to bedding-planes visible near Cape Wadworth, there are dykes
running vertically up the cliffs. Among the specimens collected by the ‘ Discovery ’
from this Cape, there are basalt-scoria (80), basalt-agglomerate (82), and a basalt (81)
obtained from a dyke standing out on the cliff-face. Such dykes do not reach the
top of the cliff but, after extending some way up a steep slope, end off at the base of
the agglomerate. Only a quarter of an hour could be allowed on shore, but as the
ship steamed into Lady Newnes Bay quite similar structure-lines were seen on the
west side of the island and on the mainland (Cape Jones), areas which were possibly
at one time continuous. The distribution of snow on the west side of the island points
to gentle folding of the rocks about an east-and-west axis, but soundings of over
150 fathoms in the channel show no continuation of this fold towards the west.
Franklin Island.
Franklin Ishmd is situated in latitude 76° 8' S. and longitude 168° 12' E., and
until the ‘ Southern Cross ’ Expedition this was the most southerly land from which
rock -specimens had been obtained. The island was discovered by Sir James Clarke
Ross, who save its length as 12 miles and its breadth as G miles. He described its north
side* as a line of dark precipitous cliffs between 500 and GOO feet high, exposing
several longitudinal broad white bands and two or three bands of a red-ochre colour.
The specimens he collected arc all basalts of one type,f while in the ‘ Southern Cross ’
collection there is a specimen of magma-basalt (limburgite) remarkable for the number
and large size of the olivine-enstatite nodules. J
From Mr. J. D. Morrison of the ‘ Morning ’ five specimens of similar magma-
basalts with olivine-enstatite nodules were received ; with them was the following
note : — “ Nos. 4 and 5 were taken from Franklin Island, from a belt of rock about
30 feet thick running horizontally along one side about 300 feet above sea-level.
Height of hill about 700 feet; very difficult to ascend, as the slope is composed of
small stones lying at an angle of about 45°. Nos. 6 and 7 were broken from a large
boulder lying at the foot of the hill. The beach is about half a mile broad and a
mile long, almost flat and about 10 feet above sea-level. Large boulders and heaps of
shingle are scattered over the beach, which is on the south-west corner of the
island.”
* Ross, ‘ Voyage in the Southern and Antarctic Regions, 1839-43,’ 1847, vol. i, p. 215.
t Prior, Mineralogieal Magazine, 1899, vol. xii, p. 79.
I Prior, Rep. ‘ Southern Cross ’ Collections (British Museum), 1902, p. 328.
BEAUFORT ISLAND.
Beaufort Island.
This island lies in latitude 77° S., longitude 167° E., and about 12 miles off Cape
Bird, which is the north extremity of Mount Erebus. Sir James Clarke Boss described
it as small and conical ; * we saw it from many points of view and estimate its length
to be 5 miles, its breadth 2 miles, and height 1000 feet. It has a rugged outline,
with a very steep snow-covered slope on the west side and a bare precipitous cliff on
the east side. Its summit is a narrow ridge running north-and-south. No specimens
have been obtained from this island.
Table of Distances.
A brief table which shows roughly the distances between some of the volcanoes
and islands will not be out of place here.
Mount Erebus to Cape Horn .......
. 3000 miles
55
55
to Mount Haddington (Swedish Expedition) .
. 2500
55
55
55
to Mount Gauss (German Expedition)
. 1000
55
55
55
to Tongariro (an active volcano in New Zealand)
. 2000
51
55
55
to Buckle Island Volcano .....
. 720
55
55
55
to King Edward VII Land .....
500
55
55
to Cape Adare .......
420
51
55
55
to Mount Longstaff ......
380
55
55
55
to Possession Islands ......
375
55
55
55
to Coulman Island ......
200
55
9)
55
to Mount Melbourne ......
200
55
51
55
to Franklin Island ......
90
55
55
55
to Mount Discovery ......
70
55
55
55
to Mount Terror ......
25
55
* Ross, 1 Voyage in the Southern and Antarctic Regions, 1839-43,’ 1S47, vol. i, p. 217.
8
Chapter II.
THE ROSS ARCHIPELAGO.
This group of islands includes practically all the land within 50 miles of the
‘Discovery’s’ winter quarters, and is the most extensive area not directly joined to
the mainland. The group is important because it has long been a centre of volcanic
activity, which continues even to the present day.
Ross Island.
Ross Island is practically made up of the volcanic cones, Mounts Erebus and
Terror, Cape Bird (Mount Bird, as it may be called), and another convex dome, Mount
Terra Nova, lying between Mount Erebus and Mount Terror. This island therefore
consists of four distinct volcanoes, and of these the greatest, Mount Erebus, is still
active. This mass of ejected material lies between latitudes 77° 9' and 77° 49' 8., and
longitudes 1G6° 8' and 1G9° 10' E. It forms an island having roughly the shape of
an equilateral triangle with a side of 50 miles.
Soundings in the waters around this island are unfortunately incomplete, but the
few that we have would seem to show that the depth is greatest close to the shore and
decreases gradually outwards. Whether or not this anomalous deepening is due to
overweighting of the crust by so many huge volcanic piles close together is not clear,
but the occurrence is suggestive.
Mount Erebus (Fig. 4) is 12,922 feet high, and was active when seen by Sir James
C. Ross in 1841, “ emitting flame and smoke in great profusion.” * During our two
years’ stay in Winter Quarters at its base the snow was always white and continuous to
the summit, and only steam was ever seen to be erupted. On three sides the mountain
rises directly from sea -level, and has flowing convex curves, which give it a very
massive and undenuded aspect. Three stages in its history appear to be recorded in
its contours. Of these the first was by far the most violent, and produced a cone with
crater about 8 miles in diameter. The walls of this still stand, and encircle it as a
girdle about G000 feet above sea-level. In profile, on the north side this ring appears
as a strong outstanding crag, and is separated from the mountain-side by a deep notch,
while on the south side there is only a mere shoulder to interrupt the regular convex
curve. The second stage is rendered evident by the existence of the lip of a later
crater at a height of nearly 11,000 feet. Old lava-streams from it, swept bare of snow,
can now be seen. The latest stage is recorded by the present small cone, which has
been built up asymetrically within the second, and from this steam now issues. Dr.
Ross, 1 Voyage in the Southern and Antarctic Regions, 1839-43,’ 1847, vol. i, p. 216.
MOUNT EREBUS.
9
Wilson has recorded five or six other steam-jets issuing from the north-east side, but
from the ship it was unusual to distinguish more than two. Mount Erebus bears a
very striking resemblance to Mount Etna as shown in von Waltershausen’s picture,*
and is much more dome-like than is suggested by the published pictures of the better-
known active volcanoes.
Owing to the difficulty of access, few rock-specimens could be obtained from
Mount Erebus itself. Specimens, however, were got from the following localities :
(a) The Turk's Head, a bare cliff which rises sharply from the sea to a height of about
300 feet on the south-west side of the mountain. Mr. Hodgson tells me that the tuffs
O
which build up this headland are exposed for a length of about 200 yards along the
shore, are bedded, and dip to the north-west at an angle of nearly 40°. (b) Cape
Royds , a promontory on the west side of Mount Erebus, having an area of about
3 square miles. This area is bare of snow, and consists of dyke-outpourings of basalt
with lenticular crystals
of felspar (818) (leucite-
kenyte, see p. 111). The
Cape is rectangular in
shape and displays many
outstanding dykes which
rise to heights of 200 and
300 feet, but are now
being rapidly disintegrated.
A similar, but vesicular,
rock (lava) (820) forms a
small knoll, 1500 feet up
this side of Mount Erebus,
but, as the rest of the
surface was covered with snow, no relations between these rocks could be made
out. (c) Cape Barne, the bare rocks which lie about 3 miles south of Cape Royds
and are separated from the latter by a shallow bay. The Cape consists of
black vesicular basalt-lava (813) which dips to the west away from Mount Erebus.
The extreme end of the Cape is a pinnacle rising 200 feet sheer from the sea,
and is separated from the main mass of the Cape by a scree which prevents the
junction of the vesicular rock and the basalt-agglomerate (815) of the pinnacle being
seen, (d) The Skuary, an area of bare land, between Cape Barne and the Turk’s Head.
This is about 2 square miles in extent, and except along the shore, where rock in situ
is visible, is covered by moraines. The moraines include fine tuffs (808) and a compact,
grey alkaline-basalt or kenyte (812) containing parallel lenticular crystals of felspar.
Below them, and extending to the shore, vesicular glassy basalt- rock (811) of the same
character is seen in situ. This last is over 100 feet thick, and appears to consist of
* Scrope, ‘Volcanos,’ 2nd edit., 1862, p. 190, fig. 43.
Fig. 4. — Castle Rock and Mount Erebus.
YOL. I.
c
10
n. T. FERRAR.
successive lava-flows laid horizontally one upon another. These rocks are similar in
character to the lavas with lenticular felspars of Cape Royds, and are also related to
the rocks of the neighbouring Dellbridge Islands, to be considered later.
Mount Bird is a flattened dome over 3000 feet high and, like Mount Terra Nova,
which attains a height of about 7000 feet, is an undenuded volcanic cone. No
specimens have been obtained from either of these mountains.
Mount Terror (Plate I) is a quiescent volcano 10,750 feet high. It forms the
eastern part of Ross Island, and, though not quite so high as Mount Erebus,
covers almost, as great an area. Its base is circular and has a diameter of perhaps
20 miles ; its surface is almost completely snow-covered. On the south side the
covering is so thick that no parasitic vents, if present, could be distinguished. On
the north side, especially above Cape Crozier, is the largest area of rock-exposure,
and here all eminences which have been examined were of the nature of subsidiary
surface lava-flows. Some
of these are quite con-
spicuous, and from them
many specimens have been
collected.
The ‘Southern
Cross ’ Expedition * ob-
tained hornblende-basalts
from a bare rock-cliff
10 miles or so to the
west of Cape Crozier, and
from the red and yellow
“blazes” that occur in
this cliff it would seem
that basalt-agglomerate is also present. From the cones on the east side of the
mountain above the Great Ice Barrier of Ross, Dr. Wilson collected basalt-scoriae
(824) and limburgite (825) and proved the cones to be due to subsidiary eruptions.
From the bare rock-cliffs (Crozier Cliffs), against which the ice-sheet abuts, Dr. Wilson
also collected rock, in situ. His specimens are of two kinds: (1) columnar basalt (830
and 848), which forms the mass of the cliffs and reaches a height of 800 feet above
sea-level; and (2) a yellow trachytic rock (831), occurring in irregular lenticles in
the mass of the cliff. At one spot a rough stratification was observed, and it is
possible that stratified tuffs are there developed.
From a locality he termed the V- Cliffs Hogsback, Mr. Hodgson collected
specimens of coarse yellow tuff (783), red vesicular basalt (778) and a basaltic bomb
(77G). This locality is on the south-east side of Mount Terror, 20 miles south of Cape
Crozier and 30 from the ship. It is one of two exposures which have been found on
* Prior, Rep. ‘ Southern Cross’ Collections (British Museum), 1902, p. 322.
Fig. 5. — Cape Crozier and Mount Terror.
MOUNT TERROR. — WINTER QUARTERS.
11
the south side of the mountain. The other exposure has been termed the Sultans
Head, and Mr. Hodgson here obtained some bedded yellow tuffs. (785 to 793), also
some fragments of vesicular basalt (795).
At Cape Crozier (Fig. 5) itself, at sea-level, a stratified palagonite-tuff (228 to 230)
was seen bedded parallel to the present slope of Mount Terror and dipping to the west
beneath a basaltic lava-flow. This tuff is brown in colour and is very friable, crumbling
easily into rounded pellets about an eighth of an inch in diameter. The surface of the
rock was whitened by crystallisations of hydrated sodium sulphate (glauber salt). Bombs
and a great variety of volcanic rocks, also granites (197), dolerites (210) and sandstones
(214), were found lying about at this height, but these were usually ice-scratched (186).
A boss about 900 feet high appears to be a pipe or plug of some now defunct volcanic
vent; the rock (176) is a limburgite containing red and green olivine-augite nodules.
A rounded knoll of trachyte (224), half a mile east of this boss, attains a height of
1400 feet. Black pebbles of glassy basalt (217, 218, 219), quite similar to the
majority of the pebbles composing the terraces a short distance lower down the hill,
were scattered all over the surface of this trachyte-dome, and appeared to have
been included in it, but in the short time allowed on shore no trachyte actually
enclosing basaltic pebbles was met with. On the south side, the 900-foot boss of rock
mentioned above adjoins a mass of yellow tuff (231) through which a grey green
trachytic rock (188, see p. 114) seems to have been forced. From the ship many
other parasitic vents were seen on the slopes of Mount Terror, but were not visited.
The Winter Quarters.
Winter Quarters were taken up near the end of a long peninsula which juts out
southward from the base of Mount Erebus in latitude 77° 51' S., longitude 166° 45' E., and
in this district not much information could be obtained relative to the general geological
history of South Victoria Land (Plate II). The peninsula is 10 miles long by 2 miles
broad, and has an average height of 700 feet. It is entirely composed of recent volcanic
rocks, and only about four of its twenty square miles are free from snow.
At Hutton Cliffs, a stratified basalt-tuff occurs as a cliff 500 feet high. This tuff-
cliff is quite isolated, and is divided into two parts by the snow which falls over the
cliff as a small glacier. The northern part (452-457) is composed of rather coarse tuff
and is more definitely stratified than the southern (458-462) ; but for each mass the
dip is the same, and is about 60° to the north-north-west. The rock consists mainly
of fragments of vesicular basalt-glass and varies from yellowish-green to almost black
in colour, but some of the hand-specimens have reddish bands of palagonite, and others
have incrustations of calcium carbonate. These cliffs are about 5 miles distant from
Castle Rock and 10 miles from the Turk’s Head.
The Sulphur Cones lie on the north side of the peninsula at the foot of Castle
R,ock and at a distance of three miles from the ship, and are so called because native
c 2
12
II. T. FERRAR.
sulphur was found scattered over their surfaces. They rise 50 to 100 feet above the
ice, and consist of black hornblende-basalt and olivine-basalt (382-386, see p. 103)
which have apparently filled up volcanic necks now undergoing rapid denudation.
The sulphur (379) is found thickly distributed over the surface of the frost-riven rock
and is sometimes in quite perfect crystals.
Castle Rock (Fig. 4), 3 miles distant from Winter Quarters, rises to a height of
1400 feet as a bold crag. It is surrounded by vertical cliffs 400 feet high. On the
south the foot of the crag is snow-covered, but on the north the land falls sheer away
for 1000 feet. On the east and west sides black basalt can be seen forming the lower
part of the crag, which consists above entirely of palagonite-tuff (380). The tuff
varies much in texture ; sometimes it consists of yellow and black angular fragments
of olivine-basalt and basalt-glass half an inch across and very uniform in size ; in
other places the black masses attain as much as a foot in diameter ; sometimes they
are almost circular in section, and are often arranged in parallel rows. The summit
of the rock is flat and strewn with loose black fragments of olivine -basalt (319),
more than two inches in diameter and very uniform in shape and size. About a
mile to the northward of this rock occurs another crag consisting of tuff quite like
the former and possibly of similar age.
Crater Hill. Along the south-east side of the peninsula there are three
scoria- craters. Two of these are rather insignificant, but the third, Crater Hill,
rises to a height of over 1000 feet, and the crater-lip at its summit is almost perfect.
On the north side the lip rises about 200 feet above the bottom of the crater, but on
the south side it has been broken down. The rocks obtained from this hill include
black vesicular basalt (341) and red scoriaceous basalt-glass : the latter has obviously
flowed over the lip of the crater and now forms the highest point of the hill. Near
the south foot of Crater Hill, porphyritic olivine-basalts (656 and 659, see pp. 105-6)
rise sheer out of the sea and form a cliff. They appear to extend as horizontal sheets
right under both Crater and Observation Hills.
The Harbour Heights , or Ay-rival Day Heights, as they have been sometimes
called, include the three prominent eminences between Castle Rock and Hut Point
They rise over 100 feet above the general level of the snow-covered peninsula.
Numerous vesicular olivine-basalts (323) and basalt-bombs (367) have recently been
ejected from these vents, and an occasional flat space, bare of snow, exhibits massive
but vesicular lava-flows of olivine-basalt (366) (Plate 11). These volcanoes have a
general resemblance to the Pleistocene* volcanoes of Auckland in New Zealand.
Between the southernmost crater of the Harbour Heights and Crater Hill there is a
basin-shaped depression, the vent of another small volcano. Near this depression
occurs a large rock-mass measuring quite 15 feet across and 12 feet high. This rock
appears to be the remains of a dyke, and from it the specimens (369-378) were
taken. It is a limburgite with abundant foreign inclusions. Of these some are of
* Hutton, Trans. New Zealand Inst. (1899), 1900, vol. xxxii, p. 178.
OBSERVATION HILL.
13
pure transparent felspar, while others have the mineral composition of gabbro and
peridotite (see p. 107). Most of the coloured inclusions are quite angular, but a few
are rounded ; the largest of them are about three inches long.
On the south-east side of the crater a non-vesicular black olivine-basalt,
approaching limburgite in character (326, see p. 105), forms a rugged hillock about 100
feet high. The exposure of rock is quite 50 yards across, but the material is crumbling
rapidly away and fresh rock is only obtainable near the summit.
Observation Hill. At the extreme south-west point of Ross Island is Observation
Hill, which is separated from its neighbour, Crater Hill, by a narrow col called
The Gap. Observation Hill has very steep slopes which make an angle of 40° to the
horizontal, and, almost meeting in a point, produce a strikingly pyramidal hill (Plate II).
The south-west side slopes away more gradually and terminates in Cape Armitage.
This prolongation appears to be due to the presence of a sheet of rock which is bedded
horizontally. This rock (553) occupies but a small area, about 200 yards long and
50 yards broad. The rock is a porphyritic olivine-basalt, almost black in colour and
containing phenocrysts of green olivine up to one-eighth of an inch in diameter.
Observation Hill appears to have been built up of successive Hows of trachytic lava
which have welled up through one single outlet. These Hows are now to a great extent
denuded ; but on the south-east side the remains of a sheet occupy the greater part
of the hillside, and rest upon another similar sheet (412). The lower sheet spreads
out and forms the flatter south-east side of the hill.
On the north shoulder of the hill the trachytic lavas show a rather greater variety of
texture, especially near the 400-feet contour. A dark-grey hornblende- trachyte (273 and
281), with abundant lapilli-like inclusions (see p. 118), up to an inch in diameter, forms
the shoulder on which a perched block of black vesicular basalt is prominent. The
rock with these inclusions has a conspicuous platy structure, and the upturned edges of
the slabs into which it weathers may be traced across the Gap to the base of Crater
Hill. This platy rock is considerably contorted and its apparent strike is exceedingly
variable, sometimes turning through more than two right angles in 50 yards. It is
obviously older than, and unconformably overlain by, the yellow hornblende-trachytes
(278, 279, 280) of the higher part of the hill. Higher up the hill occurs a dyke of
grey hornblende-trachyte (277, see pp. 117 and 119). The dyke is not more than 10 feet
broad, but is traceable 100 feet vertically up the hill. The top of the hill consists of a
yellow trachyte ; locally it is streaked with grey ribbon-like bands (288) which follow the
How-structure. The weathered surface of the rock is honeycombed, but as here the wind
removes the snow immediately after its fall very little frost-action seems to take place.
On the southern side of the summit the darker rock begins to preponderate, and at a
point some 30 yards away from the top, and 50 feet below it, the yellow rock gives place
to a dark-grey hornblende-trachyte (290). Below this rock comes another dark-grey
hornblende- trachyte with spheroidal structure (655). The spheres which make up the
mass of this exposure sometimes attain a diameter of over 2 feet and are visible over
14
H. T. FERRAR.
an area of some 20G square feet. The spheres are all planed off to an even surface,
and there is no change in the slope of the hill to correspond with the junction of the
two rocks.
Turtle Back Island.
Turtle Back Island, low and insignificant in aspect, lies in the bay between the
Winter Quarters peninsula and the ice-tongue in Erebus Bay. It is less than a
quarter of a mile long and about 100 yards broad, is rectangular in plan, and rises to a
height of 50 feet. The loose rock-material on the surface of the island is bedded, and
the layers of dark rock form a small anticline, of which the axis is the longer or
N.E.-S.W. diameter of the island. Two boulders of kenyte (trachydolerite) with
lenticular crystals of felspar (447 and 484) were found on this island. They are similar
to the rock-specimens brought from the lower slopes of Mount Erebus, but are more
glassy and of a black colour. Black augite-olivine nodules (448 and 451) are
common, but the mass of the island consists of fine-grained fragments of olivine-
basalt (449).
Black Island.
In point of size Black Island comes next to Ross Island. It lies south of
latitude 78°, and is roughly triangular in plan, having a side about 15 miles long
(Fig. 32, p. 58). It shows two central peaks, each over 3000 feet high, and appears to
be composed entirely of volcanic rock. It is quite surrounded by glacier-ice, and is
therefore almost a nunatak. It is probably connected with White Island, situated to
the eastward, by an isthmus rising about 200 feet above sea-level. Specimens from
rock in situ were obtained from a hill, 900 feet high, near the north end of the island.
Compact and vesicular basalt-lavas (593, 594, 595) were obtained high up, but no
specimens of rock in situ could be obtained from the lower slopes, which were
completely covered with rock-debris. At the south-east end is a yellow trachytic
rock (609, 610); it appeared to be a dyke nearly a quarter of a mile wide breaking
through the black basaltic rock. The rock forms a bold headland nearly 400 feet
high. The major joints, which are vertical, strike north-wTest and south-east, and
notable variations in the appearance of the rock occur on either side of the joints.
There are two other apparently similar rock-exposures near this spot, but time did
not permit their examination.
White Island.
This island is 20 miles long, but is less than 5 miles broad. Its longer axis is
nearly north-and-south ; the island lies between the longitudes 167° and 168° E., and
is south of latitude 78° S. The land rises very suddenly out of the ice which surrounds
it, and attains a height of 2000 feet. The only rock in situ obtained from it is a
“BROWN ISLAND.”— THE DAILEY ISLANDS.
15
black hornblende-olivine-basalt (311), which occurs as a boss on the summit. One or
two crateriform depressions occur on the lower portions of the island, and there seems
little doubt that the whole island consists of volcanic rock.
“ Brown Island.”
This mass of land, about 15 miles long and 5 miles broad, is only apparently
an island, for it is connected by a narrow isthmus about 8 miles long to Mount
Discovery, a volcanic cone on the edge of the mainland. As the peninsula is so
nearly isolated, and bears so gi’eat a resemblance to the other islands, it is convenient
to include it in this chapter. When the ice was at a maximum “ Brown Island ” was
certainly cut off from the mainland, which lies to the west. If the moraines covering
the isthmus could be removed it is probable that even now an island would be
produced.
“ Brown Island” is 2812 feet high and entirely composed of volcanic rocks. The
northern end is comparatively low and flat. Since many patches of rock of a bright-
red colour occur scattered over it, we may presume that scoria-cones are present.
The southern and higher end consists of a single crateriform hill, and around the
crater are red vesicular basalt-lavas (605) which have flowed over the sides of the
rim. A hornblende-basalt (608, see p. 104) occurs on the east side at a height of about
2000 feet. This rock dips north at an angle of 63°, and from the fact that it ends
abruptly as a cliff there is little doubt that much of the original lava-stream has been
removed. On the north side of the crater a subsidiary peak of banded yellow rock
forms a massive hill 500 feet high. The crater is at least half a mile in diameter,
and a small shallow pool about 100 yards long occupies its centre. There is little
ice or snow in the crater, the lip of which is about 100 feet above the pond. On
the west side a white trachytic rock (606) has forced its way through the covering
of basalt-glass, and was found on the crater-lip. On the lower slopes no rock in situ
was observed, but the whole surface was covered with black smooth loose fragments
of basalt (603), like the black pebbles at (Jape Crozier described on p. 11.
The Dailey Islands.
A number of conical islands, the Dailey Islands, rise through the floating ice
at the head of McMurdo Sound. They are five in number, and all lie almost on the
same east-and-west line. Four of these are small and conical, and not more than a
quarter of a mile in diameter. The fifth is perhaps a mile long, half a mile wide, and
200 feet high ; it is the only one that is at all easily accessible. It is situated
on the western margin of the pinnacled ice* (Fig- 44, p. 79). The specimens
collected are all of basaltic rocks of limburgite type (510), but plutonic boulders
* Ferrar, Geog. Journ., April 1905, vol. xxv, plate, p. 374.
16
H. T. FERRAR.
lie on the surface and in the small crateriform hollow at the centre of the island.
There are one or two dykes which project slightly above the loose scoriaceous
matter of the general surface.
The Dellbridge Islands.
These four islands lie three or four miles south-west of the base of Mount Erebus
and twelve miles north of Winter Quarters ; although the nearest together are two
miles apart, they are probably all remnants of a once continuous land-mass.
Inaccessible Island, the most northerly of the group, is elongated in an east-and-
west direction, and has an almost sheer cliff facing the south. Its north side slopes at
about 40°, and is therefore too steep in places to hold the disintegration-products of
the rocks which form its higher peaks. The dimensions of the island are, roughly,
length two miles, breadth half a mile, and height 500 feet. Mr. Hodgson collected
specimens here, and tells me that the rocks are much confused, but generally dip to the
north. The specimens include a red vesicular trachytic lava (802), porphyritic basalts
(805), yellow trachytes (803), and trachydolerite of intermediate character (804) ; on
the south-east end there are many irregularly bedded chrome-yellow bands. Tent
Island is nearly rectangular in plan, and has sides about a mile long. Its greatest
height is about 400 feet, and the highest point is close above the steep north-west cliff.
The upper surface of the island slopes to the south-east and agrees with the dip of the
lava-beds. The lowest rock exposed is a basalt-agglomerate (817) which occupies the
lowest 100 feet on the north-west cliff. It is covered by sheets of a vesicular glassy
kenyte (463-466), with lenticular porphyritic crystals of felspar, like the rock of Cape
Royds. These sheets have a dip of about 20°, are parallel, and are each about 20
feet thick. Razor Back Island is merely a ridge of rock rising 100 feet above the
water. Its sides meet to produce a central ridge of which the angle is not much
greater than a right angle. The long axis is perhaps half a mile long, and along the
same straight line is the Little Razor Back Island. The specimens obtained are
vesicular lavas of olivine-basalt (470, 471) and of trachytic rocks (473, 476).
17
Chapter III.
THE MAINLAND OF SOUTH VICTORIA LAND.
Cape Adare.
It will not be out of place to supply here a few additional notes on the rocks of
Cape Adare, latitude 71° S. It may be pointed out that the Cape lies at the foot of
the gigantic Admiralty Range, and is formed of horizontal sheets of basalt and basalt-
Fig. 6. — Cape Adabe Peninsula, fbom Ross Sea.
agglomerate, similar to those which occur in Coulman Island and perhaps the other
islands oil’ the coast.
The peninsula of Cape Adare (Fig. 6) consists mainly of nearly horizontal sheets of
basaltic lava laid one above another to form a flat-topped promontory, which gradually
increases in height from north-west to south-east. Dykes occasionally cut across
these sheets.* The successive sheets are thinner and more numerous at the north-west
VOL. I.
Prior, Rep. 1 Southern Cross’ Collections (British Museum), 1902, p. 327.
D
18
H. T. FERRAR.
extremity, but they become thicker, and are slightly inclined upwards two miles or
so towards the south-east. The following sections may be of interest : —
(A) near the north end of the Cape —
Top. (6) 100 feet — red basalt-glass (8G5)
(5) 300 feet — black hornblende-basalt (859-861)
(4) 50 feet — tuff (857)
(3) 100 feet — basalt with vertical joints (856)
(2) 50 feet — vesicular basalt (854)
Bottom. (1) 200 feet — talus
(B) about two miles south of the end of the Cape —
Top. (6) 600 feet — (unexamined)
(5) 100 feet — tuff
(4) 50 feet — boulder-breccia (54)
(3) 60 feet — black olivine-basalt, weathering into vertical columns (49)
(2) 100 feet — red scoriaceous basalt (51)
Bottom. (1) 100 feet— scree-slope
This approximately horizontal structure appears to be characteristic of the steep
coast line between Cape Adare and Cape Jones, a distance of about 150 miles. This
part of the coast is a cliff varying between 1000 and 2000 feet in height. Sometimes
anticlinal and synclinal folds, whose axes appear to run east-and-west, are seen.
Occasional red bands, possibly like those on Coulman Island and Cape Adare, can be
distinguished.
The Volcanic Cones on the Mainland.
The number of volcanic cones on the mainland is less than has been hitherto
supposed,* but those seen are interesting from their occurrence on what is probably a
exeat line of fault. These cones all rise from the low foothills that form the coast, and
O
the latter is always parallel to the mountain-ranges. The volcanoes, being isolated
cones and having as a background the massive main range, stand out most
prominently.
The Summit of Cape Jones (Fig. 7).
The summit of Cape Jones may be taken as a type of these. The hill near
the end of the Cape rises to a height of over 3000 feet, and shows admirably the
even convex curves of mountains of accumulation. The whole is covered by a
deep pall of snow, which either breaks off at the edge of the high sea-cliff, or
blends gradually with the ice-sheet of Lady Newnes Bay. This volcano occurs in
latitude 73|° S., longitude 170° E. It lies to the west of Coulman Island, and on
the strip of foothills which is here nearly 20 miles broad. From the north these
* J. W. Gregory, ‘ Nature,’ 1901, vol. lxiii, p. 610.
VOLCANIC CONES ON THE MAINLAND.
19
foothills can be seen to be decreasing slightly in height westwards towards the
base of the Admiralty Range, and thus appear to mark off a great longitudinal
valley running parallel to the coast.
Cape McCormick.
On the end of Cape McCormick, latitude 72° S., there are two bare cones which
rise 1000 feet above the sea-cliff. They have crateriform summits which bear a
striking resemblance to Crater Hill near the Winter Quarters of the Expedition.
Fig. 7. — A Volcanic Cone on the Mainland; the Summit of Cape Jones. The ‘Discovery’ in a gulf in
the Lady Newnes “ Piedmont-afloat.”
Mount Brewster.
Mount Brewster, on the north-east side of Lady Newnes Bay, though it does not
attain a great altitude, is noteworthy. The mountain is about 3000 feet high, and
rises from the flat lowland at the base of the mountain range. The range with its
angular spurs towers above the foothills, and Mount Brewster with even outline rises
but little above the snow-covered land around it. The summit of this hill is slightly
flattened, and some part of the crater may still remain.
Mount Melbourne.
This mountain, 8000 feet in altitude, is situated in latitude 74|° S., longitude
165° E., and on three sides rises directly from the sea ; on the fourth side it is joined
to a range of higher peaks. The slopes of the mountain are not markedly convex, but
D 2
20
H. T. FERRAR.
appear symmetrical from all points of view. The base is less than 20 miles in
diameter, and the mountain, though 8000 feet high, is not nearly so voluminous as
Mount Erebus or Mount Terror.
Basalts* appear to be developed at its base, and but few parasitic vents or obvious
lava-flows are seen upon its sides. The only specimens brought back by the
‘Discovery’ Expedition are rounded pumice-fragments (899). These were obtained
from the floating ice of Wood Bay, and must have been transported by the wind
during the winter months.
Between Cape Washington and Cape Bernacchi, or in other words between
Mount Melbourne and Mount Evans, no volcanic cone has been noted. This is
important when we remember that here the foothills are absent, and that the coast is
straight and uniform for a distance of over 200 miles. South of Cape Bernacchi,
foothills composed of gneissic rocks are developed for some 50 miles. These are
separated from Mount Morning and Mount Discovery by the valley of the Koettlitz
Glacier, which' trends north-west and lies parallel to a line joining the two
volcanoes.
Mount Morning.
Mount Morning is a low dome 5779 feet high, and is almost circular. At
its base it is 10 miles in diameter. On the south side the mountain slopes down
to sea-level, and on the south-west it is separated from the main ranges by low
foothills. The Koettlitz Glacier, which opens out north-eastward, occupies part of the
above-mentioned depression between foothills and mountain -range. Radiating lava-
flows are a prominent feature of this mountain, but no specimens could be obtained
from them.
Mount Discovery.
Mount Discovery, the last volcanic cone which we shall note, is 9085 feet high,
and lies in latitude 78|° S., longitude 165° E. (Fig. 32, p. 58). It adjoins Mount
Morning on the west, but is cut oft' from the nearest mainland by Discovery Gulf,
the ice-filled gulf into which the Koettlitz Glacier flows. The mountain is
symmetrical in outline and has the form of a bell. The inflected curves of its sides
unite at the summit without indicating the presence of a crater, and they spread out
to give the mountain a circular base, some 15 miles in diameter at sea-level. This
mountain was not visited, but the moraines stranded in the land-locked bay on its
north-east side show that basaltic fragments are the commonest ejectamenta.
The Minna Bluff.
The Minna Bluff is a long and narrow promontory which projects south-eastward
from the foot of Mount Discovery. It seldom attains a height of more than 2000 feet.
It is 35 miles long, but its breadth is rarely greater than 5 miles. Its sides are
* Prior, Rep. ‘ Southern Cross ’ Collections (British Museum), 1902, p. 322.
THE CONTINENTAL RANGE.
21
very steep and almost parallel, and their wall-like appearance is unbroken by glacier
or ice-cascade. No structural features are very evident, but specimens of basaltic
and phonolitic rocks (619 and 622) obtained from two spots near its south-east
end prove its volcanic origin. On the north-eastern extremity there are lava-flows
quite like those of the east side of Black Island and of the Harbour Heights, Winter
Quarters. The outline is unbroken and therefore it is impossible to say whether
this peninsula is composed of lava-sheets, or is a series of scoria-cones like those
which make up the Harbour Heights.
The Continental Range.
South Victoria Land, as previously mentioned, consists of a great range of
mountains stretching in a nortli-and-south direction for 800 miles at least, and is appar-
ently the eastern edge of a
vast plateau, for between
latitudes 77° and 78° S.
Captain Scott travelled
200 miles westward over
a level region having a
uniform height of about
7600 feet above the sea.
The range maintains
a uniform high level. Any
peaks, such as Mount
Sabine, that rise to heights
of over 10,000 feet do so
from correspondingly high surroundings, so that there are practically no peaks rising
to great altitudes from low levels and towering above the surrounding land. In fact,
the land does not show great relief.
Surgeon McCormick, of H.M.S. ‘Erebus,’ considered the whole ran ac to be
volcanic ; but this is obviously not the case, for all the higher peaks are pyramidal in
outline, and exhibit a house-roof shape which could not have resulted from the
eruption of rocks from local centres. The Ross Expedition was less fortunate than
the ‘ Discovery,’ for the latter was able to steam in close to the land and see the peaks
from nearer points of view. Thus, just south of Cape Washington, a tabular mountain,
Mount Nansen (Fig. 8), was observed from the ‘ Discovery ’ to have apparently horizontal
bedding planes and almost perpendicular scarps showing plateau-structure. The earlier
explorers were too far from the land to perceive these characters.
The range, or chain of mountain-ranges, naturally divides itself into sections or
links, and these may be conveniently considered separately.
(1) The area between Cape Adare and Cape North, a distance of 100 miles, is
Fig. 8. — Mount Nansen, the tabular mountain south of
Cape Washington.
99
Li Li
IT. T. FERRAR.
more snow-covered than is the land further to the south, and the coast, which is
parallel to the mountains, faces north-east. The mountains here form the main part
of the Admiralty Range of Ross. They diminish in altitude as one passes westward
to Cape North, and with the decrease in height there is a corresponding increase in the
proportion of the snow-covered area. At Cape North itself the covering of snow is
almost uninterrupted. Here the peaks which form the horizon are all of the pyramidal
type, and they have their easterly shoulders truncated sharply at the shore. There
are no deep valleys, but the snow often exhibits prominent series of terraces, one above
another and parallel to the coast, and the whole is somewhat suggestive of the existence
of some horizontal structure in the rock beneath.
(2) The Range which occupies the 250 miles of coast between Cape Adare
and Cape Washington forms the highest, and perhaps the largest, land-mass.
This area lies to the south of area (1), the Cape North portion, and is continuous
with it. In the south its line of peaks recedes so far from the coast that the
connection between this area and the third, or Prince Albert, section is not yet known.
Here one sees the possibility of a division into two distinct geological areas, for low
foothills are almost continuous along the whole length of the coast from the Cape
Adare promontory to Cape Sibbal'd* or even Cape Washington itself. Behind these
foothills there appears to lie a depression, which takes the form of a series of valleys
running north-and-south, and behind the depression is a wall, or possibly a fault-face
or escarpment, which rises to heights of 10,000 feet, and has weathered into a series
of fine pyramidal peaks.
Many photographs illustrate the form of the range, and some show peaks,
such as Mount Minto and Mount Adam, which rise as enormous gables from a
plateau already high, and thus do not greatly overshadow the surrounding
mountains.
At the head of Robertson Bay the depression at the foot of the mountains
resolves itself into a valley, and even the bay itself may be considered a continuation
of this. On the south side of Mount Melbourne this depression is again prominent ;
here it resolves itself into a valley running out to the south-east, and having the
volcano (Mount Melbourne) as a part of its left bank, affording evidence that the
mountain is situated upon a line of fault.
(3) The Prince Albert Mountains, 200 miles in length * and trending due
north and south, is the lowest large area of land seen by the Expedition. This range
is important, not only because it is practically new, but because of its extreme uni-
formity of character. It is highest at the north end, where Mount Nansen, mentioned
above, rises to 8788 feet, is lowest about the centre, latitude 76° S., where it is only
about 3500 feet (Mount George Murray, 3591 feet), and rises again to 8000 feet (Horse-
shoe Mount, 8228 feet) on the latitude of Mount Erebus. It is remarkable that here
the eastern border is always steep and gives one the impression that it is only the
outlying edge of some great plateau from which streams of ice come down between
THE ROYAL SOCIETY RANGE.
‘23
the nunataks. Later, when the former extent of the Beacon Sandstone Formation is
considered, it will be seen that this uniformity of landscape is not surprising.
(4) The Royal Society Range has a length of some 50 miles ; this length is
almost bisected by the 78th parallel of latitude, and is the only part of South
Victoria Land which has been examined in detail. In the main all the structures
observed in the Admiralty Range are again seen, but are much more strikingly
developed (Fig. 9). There are foothills of insignificant height, a north-and-south
valley separating the foothills from the main mountain -mass, and a mountain-mass
Fig. 9. — Modnt Huggins and the Royal Society Range.
rising in a uniform cliff behind to a height of 10,000 feet and having occasional
peaks over 12,000 feet in altitude.
From our Winter Quarters this range could always be seen, though quite
50 miles away ; and so clear was the atmosphere that, even at this great distance,
the plateau-form was always evident and was rendered still more striking by the
broad band of lighter-coloured rock below. This band must be at least 2000 feet
thick ; it is apparently bedded horizontally and extends from end to end of the
range. The peaks rising above the plateau are of darker-coloured rock and in
strong contrast with it.
Thus, the form of the range appears to be determined by the horizontality
24
H. T. FERRAR.
of the rocks which compose it, a character abundantly proved by the sledge-parties
who passed along the deep glacier-tilled valleys that cross it eastward to the sea.
(5) The four separate ranges which determine the 300 miles of almost straight
coast to the south of latitude 79° S. appear to be exactly similar to those already
considered, and may be dismissed with the mention of the plateau-character which
is strikingly shown and beautifully illustrated by Dr. Wilson’s sketches, made
during the great journey to the south, when Captain Scott, Lieutenant Shackleton,
and Dr. Wilson reached latitude 82° 16' 33" S. These sketches are all the more
valuable in that they were made by an unprejudiced observer.
Each of the four ranges averages 8000 feet in height ; they are separated by
wide channels far below the level of the plateaux, and such channels, having straight
and exceedingly steep sides,
appear to be typical features
in the geography of all South
Victorian mountain ranges.
King Edward VII Land
(Fig. 10).
Fig. 10. — King Edward VII Land.
This land lies at the
eastern end of the Great
Ice Barrier of Ross, and is,
therefore, connected with the
mainland by ice at least.
It lies between latitudes 7G and 78° S., and longitudes 148° and 160 W. It
rises 2000 to 3000 feet high and is almost wholly snow-covered. The coast trends
N.E.-and-S.W., and appears to be banked with low foothills completely covered with
snow. It consists of two parts: (l) an isolated headland, some 1500 feet high and
10 miles long, standing well before (2) which, except for the low and isolated peak
at the north-east end, is smooth and tabular, but is completely snow-covered. The
two areas are separated by a comparatively low snow-covered depression.
The headland is unsymmetrical in shape and has a steep cliff on its north
side ; at three places, where too steep to hold the snow, there are good rock-
exposures. The land was not visited, but specimens obtained from a dredge-haul
(256-258) and from two icebergs consist of granites and gneisses and not of
volcanic rocks. As the current is south-west along the coast, the rocks dredged
up cannot have been carried far; it is also quite improbable that the icebergs can
have travelled any great distance.
25
Chapter IV.
THE GNEISSIC ROCKS AND CRYSTALLINE LIMESTONE.
As the gneissie rocks occur at sea-level at the foot of the highest part of the
Royal Society Range, and as they are found in the Range below a sequence of
rocks which is 12,000 feet thick, they may be safely regarded as forming the
ancient platform on which the central part of South Victoria Land is built.
Above the gneisses, come successively, over a very large area, granites, sandstones
and clolerites. Little is known of the field-relations of these except the order in
which the rocks occur. The important deposit of sandstone provides a convenient
stratigraphical datum-line, with reference to which the positions of the other rocks
will be considered. The above order seems, however, to be chronological, for,
where the junctions between any two rocks were examined, the lower rock usually
appeared to be the older. With regard to the gneissie series, the localities at which
they have been examined may be considered in three groups : —
(1) The Foothills of the Royal Society Range.
(2) The Kukri Hills.
(3) The Cathedral Rocks.
The Southern Foothills (Fig. 11).
South Side of the Blue Glacier.
On the hill J1;* 5400 feet high and 15 miles from the sea, occur masses of crystalline
limestone which rise at least 1000 feet above the snow. The limestone is almost pure
white, and the constituent crystals are often an eighth of an inch across. It becomes
so crumbly, on weathering, that it is difficult to get a hand-specimen from the
rounded surfaces that are exposed.
The rock (568) has important structural planes which dip at 70° to the east,
while the strike is north-and-south. The hill appears to be wholly composed of
limestone ; its western slope is very straight and steep, and is suggestive of a
fault. Parallel to the bedding-planes or, more probably, joint-planes are bands of a
dark fine-grained hornblende-biotite-granite (569) from 4 inches to 2 feet in thickness
and about 50 yards apart. At this spot there were no obvious metam orphic features
near the junction of the two rocks.
* For localities indicated by letters, see tlie map of the district near the 1 Discovery ’ Winter Quarters, and the
sections in Plate VII.
VOL. i.
E
20
II. T. FERRAR.
The south end of the Foothills.
Before leaving the discussion of this locality, mention must he made of the
specimens brought back by Dr. Wilson from 30 miles further south, close by the
Koettlitz Glacier. From the specimens themselves (834-853) and from what lie has
reported about their occurrence in the field, it would appear that the Foothills are
composed of the rocks of the metamorphic series.
Darne Inlet. Latitude 80° S. Longitude 1 G 1° E. (about).
At the entrance to this inlet, Lieutenant M. Barne collected fragments of
granite, gneiss and mica-schist (734) from a scattered moraine. These fragments must
Fig. 11. — This Crystalline Limestone on the hill J„ south side op the Blue Glacier.
have been derived from the mass of land lying south of the inlet and quite 170 miles
south of the Blue Glacier.
The Northern Foothills.
The Northern Foothills occupy an almost rectangular area about 12 miles long
and 10 miles broad. The Northern Foothills are separated from the Southern
Foothills by the Blue Glacier, which occupies a deep and rather steep-sided valley from
5 to G miles across. They appear to be mainly composed of crystalline limestone.
The west side of the foothills, on the north side of the Blue Glacier.
Two miles north of G4, the limestone is similar to that of J1( G miles distant
(Fig. 12). Here, however, the dominant structure-planes strike N.N.E.-S.S.W.,
DYKE-ROCKS OF THE NORTHERN FOOTHILLS.
27
and dark veins of a doleritic rock (56G) make an angle of about 30° with them.
From the floor of the Snow Valley, the snow sloped steeply up to the rock-
face at an inclination which was too great for us to get the sledges up.
Leaving these at the bottom of the slope for a couple of hours, a very risky
proceeding when no landmarks are available, we obtained specimens without mishap.
The fault-face on the Northern Foothills is more obvious than on the Southern ; it
is brought into prominence by the fine, bare and apparently glaciated peak (g), which
rises over 4000 feet high.
Half-way down the Blue Glacier.
Working down the Blue Glacier, Dr. Iyoettlitz collected somewhat similar
specimens from a hill 5 miles east of G4 ; here, as nearer G4, the bedding-places
dip to the E.S.E., and the dark
veins cut across the strike to
the E.N.E. On this hill one of
the structure-lines is remarkably
prominent and suggests a thrust-
plane, but no difference in the
characters of the rocks on the
opposite sides of the line was
obvious to the naked eye. The
specimens collected by Dr.
Iyoettlitz in this neighbourhood
include dioritic dyke-rocks (572,
The north-east end of the Blue Glacier.
At G4, the most south-eastern of the Northern Foothills, the arrangement of the
rocks is well seen from a distance. Here, however, the rock is so planed down by
ice-action that the rapid alternations of dark dyke-rocks and light-coloured limestone
are rendered evident. The dykes crossing the brow of the hill are plainly visible
from the glacier. A dyke of kersantite (579, see p. 130), 20 yards wide, cuts the
crystalline limestone (575, 570). The hill is 4000 feet high ; the snow wraps its base
and reaches quite up to the 1000-foot contour. Here again the sledges had to be
left more than a mile away from the exposure. On this occasion, owing to the snow-
storm that began during our absence, they were difficult to find when we returned.
The right hank of the Ferrar Glacier , between G2 and G;! (Plate I ).
The Northern Foothills, as stated above, form a rectangular mass with the north
side some 12 miles long, forming the terminal portion of the right bank of the Ferrar
E 2
574) and a schist (570)
Fig. 12. — The Crystalline Limestone on the north side of
the Bloe Glacier, at G,.
28
H. T. FERRAR.
Glacier. The hill G2 occurs at the north-west corner and sends out a shoulder four
miles to the west. This shoulder is cut off from the granite-hills, G3, by a glacier
which flows northward out of the Snow Valley. The shoulder runs out as a narrow
promontory along the same line as the north edge of the foothills, and rarely rises
much more than about 1000 feet. The tributary glacier flowing north causes an
inconvenient belt of hummocks two miles in width, and it is not till a height of 700 feet
has been ascended that rock is found in situ. The slope of the hill makes an angle
of between 30° and 40° to the horizontal, and is covered with loose morainic matter ;
but at a height of 700 feet a crag of gneiss (729) appears. The rock is dark,
Fig. 13. — The Gneiss at the east end op the Lower Kukri Hills, near the hill H.
fine-grained, and very streaky. The foliation dips to the south-west at an angle
of 60 , a fact of some importance, as we shall see when we consider the
Kukri Hills.
The Kukri Hills.
This name has been given to the hills lying immediately north of the Ferrar
Glacier, as in plan they have the outline of that implement. They separate the North
Fork from the East Fork, and are themselves divided, both topographically and
geologically, just at that point where the Ferrar Glacier floats off into its deep
fiord-like channel. The western and higher part includes all hills denoted by the
letter D on the map ; while the eastern and lower part is defined by the hills m and
II, at its western and eastern extremities, respectively.
GNEISS OF T1IE KUKRI HILLS.
29
The Eastern or Lower Kukri Hills hardly rise above 3000 feet, but maintain
this height most uniformly over the whole of their length between m and Id, a
distance of 15 miles. These hills form a narrow promontory about two miles in
breadth with steep, sometimes almost vertical, sides. They lie six miles or so
north of the Northern Foothills.
New Harbour Height (II) (Fig. 13).
At the extreme eastern foot of the hill H, or New Harbour Height, specimens
(730, 731) of hornblende-schist and gneiss were obtained. The gneiss belongs to the
dark variety, the structure-lines dipping at an angle of 30° to the north-east. About
Cathedral Bocks. B0 m
G G2 E D4
Fig. 14.— Looking dp the Ferrar Glacier, Northern Foothills on the left, Cathedral Rocks near the
CENTRE, AND THE KUKRI HlLLS ON THE RIGHT.
a mile west of this point, where a small hanging glacier on the south side of the hill
occurs, the dip suddenly changes to one of 20° to the west. The dip is emphasised by
the fact that the snow always lies only in sheltered hollows. The hanging glacier lies
on what appears to be a fault.
Below the hill D4.
Below m at the western end of the Lower Kukri Hills a white crystalline limestone
occurs, and the bedding-planes of this rock dip to the north-east at 70°. The apparent
thickness is about 1000 feet, and the strike is N.W. — S.F. Between this white
30
II. T. FERRAE.
limestone and the dark foliated gneiss at the east end, a long stretch of the valley-
side appears to be gneiss with foliation-planes dipping to the west, and therefore to
agree with that met with on the south side of the valley near G2 (Fig. 14). The white
limestone (728) abuts upon a grey augen-gneiss (727), but the actual junction could
not be examined as a hanging glacier lay upon it. The augen-gneiss appears to be
part of the great mass which forms the lower and greater part of the hill and rises
to quite 4000 feet.
Below Ho.
At Cape Bernacchi, 20 miles north of the Northern Foothills, the rock composing
the hill Ho has apparently the same structure as that of New Harbour Height, or
the hill H, while the elongate hill ht is a replica of the portion between II and in,
and has structural planes dipping to the west. The rocks were not examined at
this spot, but the serrated outline and the trend of the snow-water channels point
to the structural lines being exactly parallel to those on the Lower Kukri Hills.
The Cathedral Rocks.
These rocks lie 40 miles from the coast, and rise to heights of over G000 feet.
The glacier-surface at their base is about 1500 feet above sea-level, and the
summits of the Royal Society Range, of which the Cathedral Rocks are the northern
extremity, rise to altitudes of over 12,000 feet directly behind them. The Cathedral
Rocks slope steeply down to the Ferrar Glacier, and form its right bank for a
distance of 10 miles. They form three of the shoulders of the range just mentioned,
and are separated by tributary glaciers which run out northward along narrow and
steep-sided valleys. These shoulders project as aretes from the main plateau of the
Royal Society Range. They are composed of gneiss, granite and dolerite, and may
be topped by small outliers of sandstone (Plate III and Section II, Plate A 11).
There is an exposure of gneiss at the foot of the central shoulder, which is
designated E2 on the map. This exposure rises 500 and 600 feet above the ice
(Plate IV). The line dividing it from the granite is very sharp, and can be followed
for a distance of some 3 miles along the glacier. On the west it is hidden by
a sudden rise of the surface of the ice, and on the east is cut off by a boss of
diorite (715), which appears to have burst through from below.
The diorite forms the eastward half of this shoulder as well as the whole
lower portion of the eastern shoulder _Ej. The gneiss is overlain by a sheet of
pink granite, which, once known, is easily recognisable at a distance by the fact
that the latter forms screes, whereas the former produces a cliff. Idle upper
surface of the gneiss is a well-marked undulating line, cut oil short on the east
where it meets the diorite. In certain other places several much smaller dykes
transgressing the gneiss were observed. Of these dykes some are grey granite and
GNEISS OF THE CATHEDRAL ROCKS.
31
some are pink granite, and both kinds are more abundant towards the eastern
end of the exposure. Here the pink rock displays augen-structure (716), and
occasional isolated patches and wisps of the ordinary foliated gneiss were observed
in the middle of the masses of augen-rock. The dykes form a rough network
over the face of the gneiss, and their thicknesses vary from 6 inches to 12 feet.
The most prominent consists of a pink quartz-porphyry (709), cutting through the
gneiss perpendicularly to the foliation. The gneiss here is dark in colour, and its
alternating laminae are usually under a quarter of an inch in thickness. These
foliations are themselves folded into series of anticlinal or isoclinal folds, with
amplitudes of about 8 feet, the various bands remaining parallel.
Other specimens from later veins or dykes traversing the augen-rock may be
mentioned, namely : — •
(1) A green actinolite-rock (725).
(2) A white pegmatite-vein (723).
(3) A thin seam containing mica-plates up to half-an-inch across.
32
Chapter V.
THE GRANITES.
Though granitic rocks had not been found in situ in the Ross Sea area, the
frequency with which fragments had been dredged up by the various ships proved,
if not a wide distribution, at any rate a great local development of this kind of rock
in the area under consideration. If we neglect the occurrence in moraines such as
those found on the basaltic peninsula of Cape Adare or on the slopes of the volcano
Fig. 15. — Dolerite upon Granite on the north side of Granite Harbour.
Mount Terror, there are two localities where granite has actually been found in
place. At the first, which has been called Granite Harbour, in latitude 77° S., the
granite abounds ; though no other rock could be examined in the time available, a dark
rock was seen capping the granite (Fig. 15); that this is dolerite is almost certainly
proved by the plateau-like form of the hills on the side of the harbour
remote from our landing-place, and by the finding of dolerite-fragments on the
scree-slopes (see p. 54). The second locality is the Royal Society Range ; here the
granite has been examined at several spots over a distance of some 20 miles, and
appears to occupy an area of quite 200 square miles. This district may con-
veniently be subdivided into two areas — (l) the Snow Valley west of the Northern
Foothills, where the granite occurs in isolated hills, and no other kind of rock has
been seen ; (2) on the two sides of the Ferrar Glacier, where granite, and its relations
to the rocks both above and below, have been examined.
GRANITES OF GRANITE HARBOUR AND TIIE SNOW VALLEY.
33
Granite Harbour (Fig. 15).
In this harbour there is a prominent headland some 500 feet high and two miles
long ; in form it is distinctly like a bursting cabbage. Where a landing was made
the rock proved to be entirely granite. The rock-surface is absolutely bare of
snow, and is weathering under desert-conditions apparently analogous to those
described by Walther as obtaining in Sinai.* The joints tend to be platy and
parallel to the surface, but the edges of the joint-blocks are ragged, and the curve
is usually convex downwards. In other places the joints are vertical ; there the
rock breaks up more rapidly, and produces talus-slopes which extend almost the
whole height of the cliff-face. Dark circular patches, in rows which are often
parallel to the joint-planes, are seen on the surface, and it came as a surprise to
find that the rock is coarsely crystalline.
The greater portion of the boss consists of a grey biotite- granite (129), and the
larger talus-slopes always follow certain veins or dykes wrhich extend up the whole
face of the cliff. The centres of these veins consist of a coarse piuk granitic rock
(155) with idiomorphic crystals of red orthoclase up to half an inch in diameter,
but within a distance of some fifteen feet these phenocrysts become paler in colour,
the rock meanwhile becoming less porphyritic, and thirty feet from the centre it has
graded into the ordinary grey granite of the main mass of the boss. These pink
dykes are about one hundred yards apart ; it is noteworthy that in many cases
the change from grey to pink is not quite gradual, but takes place in stages at
the joint-planes, thus suggesting multiple dykes. These stages of the passage are
marked by bands, a foot or so across, which become successively coarser and pinker
as one passes from the sides towards the centre.
Thin seams of micaceous schist (96), narrow black basalt-dykes (113), and
numerous other varieties of rock, wrere met with, and specimens of these were
collected during our hasty scramble ashore (see p. 126).
The Snow Valley west of the Northern Foothills.
In the area between the Foothills and the Royal Society Range, a district
which I have called the Snow Valley, isolated hills just raise their heads above
the snow, and expose to view occasional masses of granite-blocks, which at first
sight would appear not to be in situ. There are five or six of such hillocks, with
summits about 3500 feet above sea-level, which form the watershed between the
Blue Glacier and the ice-cascade separating the hill G2 from the hill G3.
The points p1; 77.,, etc., on the map indicate the positions of these hillocks, but
of the four only ij2 was visited ; it proves to consist of grey hornblende-biotite-
granite (561, 562). The mass exposed is about 100 yards long, and rises 200 feet
VOL. I.
Walther, Abhand. math.-phys. Cl. d. k. sachs. Ges. Wiss., 1891, Bd. xvi, p. 364.
F
u
IT. T. FERRAR.
above the snow which surrounds it. When traced from west to cast the rock
becomes finer grained. In places it encloses many vertical quartz-veins (5G0).
Near E4, at a height of 5000 feet, Mr. Skelton obtained a specimen of grey
granite (626), and another from a boss of rock just peeping above the snow. In
this exposure the joint-planes dip to the south, and, in places, kersantite-veins (625)
cross the mass.
From the “ 3500 feet Knoll,” e5, Mr. Skelton brought back a specimen of
a somewhat coarse-grained pink granite with phenocrysts of felspar up to a quarter
of an inch across (555, 556). The exposure is much weathered, and it is here that
the type (A) of hollowed rock* with
the white (calcium carbonate) incrusta-
tion (554) occurs (Fig. 16). This will
be referred to later (see p. 88).
At ec Dr. Koettlitz got a speci-
men of dark grey hornblende-granite
(563) with idiomorphic crystals of
pink felspar up to one inch in length.
The height at which this exposure
occurs is more than 3000 feet, and
the rock forms the eastern end of a
spur of the Royal Society Range.
The joint-surfaces are conspicuously
developed, and are arranged as a
syncline with east - and - west axis.
Other specimens from this locality
are a grey biotite-augen-gneiss
and a doleritic rock (565).
The Granite Hills between G2 and G3 (Plate IV).
The hill G3 rises to a height of 3500 feet above sea-level ; it is 1000 feet
above the level of the Snow Valley, and nearly 2000 feet above the ice in the
valley below. Eastwards the height decreases to 2000 feet, where this patch of
bare rock is separated from the gneiss of G2 by the ice-cascade previously
mentioned. As a whole this G3 block is a series of rounded hills ; as viewed from
the surface of the Ferrar Glacier (Fig. 43, p. 78), it has no very conspicuous valleys,
but presents an almost straight and even valley-wall.
The specimens (55 7, 558) from near the summit of G3 are all of hornblende-
granite with large pink porphyritic crystals of orthoclase. 1 1 is here that type 13 of
hollowed crystalline rock is found, and owing to the rapid weathering the ground
* Ferrar, Geol. Mag., Dec. V, 1905, vol. ii, p. 190.
Fig. 16. — Hollowed Granite-boclder in the Snow Valley
NEAR THE ROYAL SOCIETY RANGE.
GRANITES OF CATHEDRAL ROCKS.
35
around is covered with large loose fragments of felspar. At the foot of G3, 1500
feet above sea-level, the felspars in the rock are even larger than those on the
summit, and a dark dyke (714)
(see p. 131) with phenocrysts of
hornblende up to two inches long
is exposed a few feet above the
ice. In addition, below’ this “ dark
dyke,” thei’e is a green band of a
fine-grained rock (732) about five ^
feet thick. The dark dyke pro- Jj
duces a dark patch on the hillside, £
which, owing to the contrast, can §
be seen at least 10 miles away.
On the opposite side of the glacier,
above d3, there are three or four
similar patches which, though
larger, are probably due to a
similar occurrence of dykes.
The Cathedral Rocks.
These rocks already referred to
(see p. 30) form a very imposing m.
triple headland on the south side 8
of the Ferrar Glacier, and are as -s
important as they are picturesque,
for here, there seems no doubt, is °
contained the whole history of the
Royal Society Range. At the base,
as already stated, is banded gneiss.
Above it, and divided sharply from
it, is the granite, which must be
about 4000 feet thick. Above the
granite is a sheet of dolerite, which
o
is rendered conspicuous by its
weathering back faster than the
granite and leaving a prominent
ledge. Upon the. dark dolerite-
sheet is a yellow cap, presumably of sandstone, which forms the summits of all
three headlands (Fig. 17). (See Sections I and II, Plate VII.)
36
H. T. FERRAll.
At the foot of E2, dykes of fine-grained pink (711) and grey (710) granite force
their way into the gneiss and blend with the sheet of granite. East of the
shoulder E2, a tongue of granitic rock ends the gneissic exposure, as if here bursting
through from below. At the edge this tongue is mainly pink in colour, and
occasionally there are large patches of almost pure pink-felspar-rock in it. When
traced eastward it passes into a diorite, becoming gradually darker in colour (715)
and coarser in texture, while well-formed black crystals of hornblende appear and
increase in size up to a quarter of an inch in diameter.
The relation of this rock to the granite making up the hill G3 cannot be traced
owing to the great mass of snow in the Descent Pass, but the sharp dividing line
between the granite and the dark-coloured dolerite above it can be followed round all
three shoulders and across to the tabular hill E, and thence along the east side of
the South Arm for a distance of more than 10 miles, keeping almost exactly the
same level all the way. This sheet-like mode of occurrence of the granite appears
to be a constant feature in this area and is seen over the whole south side of the
Upper Kukri Hills (Fig. 18).
The Kukri Hills.
The Kukri Hills project as a wedge into the depression where the ice from the
South Arm meets that flowing east from the inland plateau. From them granite
has been actually obtained at three spots, namely, (l) the western extremity below D,
(2) near the middle of the south side below D2, and (3) the eastern end of the upper
portion below the peak D4. (See Section III, Plate VII.)
The hill D, as seen from the south, shows about 2000 feet of dark rock
(dolerite) occupying the whole of the cliffs, which fall sheer to the level of the ice,
here about 3000 feet above sea-level. In the middle of this great mass of dark rock
are three large triangular masses of a light-coloured rock which are plainly visible
eight miles away. At the western foot of D is a still larger mass of pale rock which
must be 1000 feet thick at least. Viewed from the north and west, this rock, which
proved to be a pink granite, could be seen sending tongues into the columnar dolerite,
and the junction of the two (699, 700, see p. 128) was seen to be quite irregular ;
it is quite clear that the granite is here the later intrusion. The joint-planes of the
granite dip to the north-east at an angle of nearly 30°; the granite (701 ), though
generally pink in colour, has occasional dark-grey masses (702-703) locally contained
in it.
The eastern end of Solitary Eock (D5a) shows four bands of rock with regular
horizontal junctions. Two of these bands are dark-brown and two are light-yellow
in colour, and the alternation of the colours suggests that the rocks are like those
seen on the north side of the North Fork, where yellow and black bands occur in
the same order and with similar thickness.
Granite Dolerite
GRANITES OF THE KUKRI HILLS.
37
When the Kukri Hills are observed from Knob Head Mountain, or from the
summit of Descent Pass, they present an almost sheer wall facing south. This wall is
broken at regular intervals by glaciers, which usually occupy hanging valleys. At the
mouth of each valley there is a well-marked junction of dark- colon red and light-
coloured rock, and in places the colours alternate regularly as before. The hill D
contains a straight yellow band near its summit which is formed by a dark rock. The
yellow band continues towards the east, and gives to the hill Dj a tabular outline.
On the hill D., it is only represented by a small outlier. Below this yellow band on
Fig. 18. — The Horizontal Upper Surface of the Granite on the south side of the Kukri Hills.
the hill Dj there is a horizontal black band about 1000 feet thick, wdrich appears to be
part of the dolerite of D, and this black band extends eastwards beneath the yellow
outlier of D2. Below the black band there is an attenuated wedge of yellow rock,
which begins about the middle of the cliff Dls and, rising slightly, reaches the top of
the cliff-face a little to the east of D». This yellow wedge shows prominent joint-planes
which dip to the east, and appears to weather in quite a different way to the outliers
on the summits of D, and D2. It is possible that this is part of the intrusive granite
of the promontory D. Below this again is a second dark band, which was subsequently
proved to consist of dolerite (704). This, too, is a part of the D mass, and maintains
38
II. T. FERRAR.
a uniform thickness of about 2000 feet. As it rises to the eastward it forms the
highest third of the hill D3 and caps the hill D4.
Below I)2 another light-coloured rock, a grey biotite-granite (708), protrudes
through the ice, and may be followed for a distance of over 10 miles along the side of
the valley. The upper surface of the granite is very well marked and forms an almost
horizontal straight line, but near the hill D3 it becomes somewhat undulating (Fig. 18).
This granite forms the greater part of D4, and finally forms the summit of the hill m.
It is probably at least 4000 feet thick. Below D.> the junction of the lower dolerite
with the grey granite is 500 feet above the ice, as measured with an aneroid barometer,
and about 3000 feet above sea-level, and it would seem that the surface of the granite
slopes west at an angle of about 2°. This spot is five miles east of the pink granite
at D, eight miles N.N.W. of the granite at Cathedral rocks, and ten miles N.W. of
the granite at G3.
Grey augen -gneiss forms the base of D4, and was again encountered at the foot
of m (727). At this last-mentioned spot, as stated in the foregoing chapter, the
augen-gneiss adjoins the metamorphic limestone, but a glacier completely covers their
junction. From a distance it was seen that the junction must occur just where the
higher and western part of the Kukri Hills ends and the lower and more uniform
eastern part begins. The augen-rock must be more than 3000 feet below the
dolerite-granite junction and at least eight miles east of the hill D2.
From a consideration of the above it would seem that the grey granite of these
hills is older than the dolerite which rests upon its even upper surface, but that the
pink granite of D is intrusive and later than the dolerite.
In Moraines.
Fifty miles inland, at a height of 4000 feet above sea-level, small and large
boulders of both grey and pink granite (G93) were found on the side of Beacon Height
West. They were resting upon a surface of the Beacon Sandstone. The spot where
these fragments occur is some distance up one of the Dry Valleys. As the land south
of the Dry Valleys rises to over 7000 feet in height, it is possible that among these
peaks granite occurs at a greater elevation than 4000 feet, and has been brought
down to its present place upon the sandstone by the ice which once occupied
the valleys.
On the slope of Knob Head Mountain (B,,) there were huge boulders of granite
at a height of 4000 feet above the sea, but no granite was found in the upper
part of the mountain itself. About one-tliird of the material of the moraines in
the South Arm consists of granite- blocks, and all varieties appear to be there
represented.
39
Chapter VI.
THE BEACON SANDSTONE FORMATION.
The existence of fossiliferous sedimentary rocks in South Victoria Land lias been
considered probable ever since H.M.S. ‘Challenger’ dredged up sandstones, limestones
and shales* in a high southern latitude, but as it was thought that the coastal belt
of the land was composed entirely f of volcanic rocks, there was little to encourage
the hope that fossiliferous strata would be met with in the course of the ‘ Discovery ’
Expedition.
In dredging off Coulman Island several small fragments of a white granular
quartz-grit were brought up, and when just south of the conical Mount Melbourne a
tabular mountain, Mount Nansen, was seen, our hopes of finding sandstone were raised
to a very high pitch. This mountain showed well-marked horizontal structure, and steep
scarp-slopes which vividly recalled Table Mountain at Cape Town in South Africa.
Further south, in about latitude 75° 57', many tabular hills with black caps could
be seen fronting the sea, and the possibility of such tabular mountains being composed
of plateau-basalt had to be considered. However, when the ‘ Discovery ’ anchored on
the south extremity of Ross Island, the Western Mountains (the Royal Society Range
of our present nomenclature) were seen to be made up of differently coloured horizontal
bands which run from end to end of the ran ire. These rock-belts are well brought
out in some of the photographs, and at a distance of 50 miles the contrasts of colour
were more obvious than in any of the photographs taken close at hand.
Lieut. A. B. Armitage’s pioneer-journey through these mountains proved that
horizontal structure and plateau-features are extremely constant. The specimens
(628, 630, 639-642) he brought back included a sandstone which is somewhat like
the Millstone Grit of the top of Ingleborough in Yorkshire, and suggested the
probability of the existence of fossiliferous sediments in the district.
Lieut. Armitage reported that the sandstones attained a height of nearly 8000 feet
and were accessible at a spot 60 miles inland on the very edge of the Inland-ice. The
photographs taken by Lieut. R. W. Skelton on this journey (Fig. 19) showed that the
sandstone has a marked effect on the scenery, and the name Beacon Sandstone Forma-
tion, which I propose to give to the deposit, is derived from the remarkable mountains
IL and B4 to which Lieut. Armitage has given the name Beacon Heights.
Accordingly Captain Scott arranged that I should go with him as far as the
edge of the Inland-ice and do as much geological work as was possible on the return
* Murray, Geol. Mag., Dec. IV, 1898, vol. v, p. 270 ; Prior, Mineralogieal Magazine, 1899, vol. xii,
p. 81, note.
t Gregory, ‘Nature,’ 1901, vol. lxiii, p. 609.
40
IT. T. FERRAR
journey. A second attempt had to lie made, owing to the sledges breaking down
on the first, and, even then, bad weather confined the parties to their tents for a
period of six and a half days ; when the weather had moderated I had but one month
in which to examine the 600 square miles of new country.
From what we had seen on the way out, plateau-dolerite would be found
overlying the Beacon Sandstone. The latter was not exposed at Depot Nunatak,
but in the moraine at the foot of the rock I found abundant sandstone blocks, and
the majority of these were locally blackened by carbonaceous matter (743, 744).
None of these blocks contained fossils, other than the small lenticles of carbonaceous
material which I thought suggestive of organic origin. These were our first
evidences of Antarctic life in the geological past, and as my companions, Kennar
(P. 0.) and Weller (A. B.) spread out our sodden gear in the sun under the lee of
the nunatak, hopes indeed ran
high, and all looked forward
O f
to the joy of further new dis-
coveries.
Next day therefore found
the camp near the foot of the
hill B1( where three hundred
feet or so of the sandstone
could be seen cropping out
below the overlying dolerite.
Imagine my delight when,
arriving with bag and hammer
at the rock-face, I found thin,
black, irregular bands in a
pure white sandstone. Though
the bands were two hundred
feet below the capping dolerite,
their carbonaceous material was much charred ; hence, after collecting a few specimens,
we left this promising locality, perhaps prematurely, and moved diagonally down
the valley to the vast exposures of the Inland Forts. II etc I had hoped to find
better specimens, but neither here nor elsewhere did we meet with anything
nearly so good as at our first locality near the dolerite-j unction. The sandstone
of these Inland Forts is quite 2000 feet thick, and, though we carefully sought
for its base, no indications of that base or of the relations to the underlying rocks
could be found.
The Beacon Sandstone is also present at the foot of Knob Head Mountain, which
is over 30 miles to the east of Depot Nunatak and about 3000 feet lower. The
localities at which the Beacon Sandstone was examined* will therefore be considered
in turn, in the order in which we came to them.
THE BEACON SANDSTONE AT Bt.
41
Below the Hill I>.
From a position on the side of the hill B1; we could see some miles away
striking alternations of dark and light bands just peeping up from below the brown
rock of a15. The dark bands are conspicuously paler than the overlying rock
(dolerite) and are presumably carbonaceous sandstone. If we take this into con-
sideration and the fact that carbonaceous sandstone is found in the Depot Nunatak
moraine, and bands of it occur below B1( it would seem that only the upper portions
of the Beacon Sandstone are fossiliferous.
The Beacon Sandstone at Bx is locally disrupted by the dolerite, but the
horizontal bedding is not materially disturbed, except at one place where huge
masses of the sandstone have been bodily upraised. One of these dislocated masses is
100 feet thick and a quarter of a mile long. It contains many small black iron-stained
nodules (663), which are set in a matrix of very coarse quartz-grains. Of the 300 feet
exposed near the camp, the major part is a pure, even- grained, coarse sandstone.
False bedding or current-bedding is displayed, and locally there are discontinuous
bands of quartz-pebbles (673, 674). The pebble-bands appear and disappear quite
suddenly in the ordinary sandstone, and they are never more than four inches
thick ; the pebbles themselves vary from the size of a sparrow’s egg to that of a
hen’s egg, and quite 99 per cent, consist of vein-quartz or quartzite (672). Some-
times the pebbles are very sparsely scattered, and a bed 12 feet thick may contain
only a single pebble. The quartzite (quartz-schist) pebble (675) was found under
such circumstances and measures 8x5x4 inches. The sandstone-blocks of Depot
Nunatak display abundant lenticular pieces of yellow mudstone up to two inches in
length, but these lenticles were not observed elsewhere.
The carbonaceous matter (743-762) only occurs in the lowest hundred of the 300
feet exposed ; the carbonaceous bands, like the pebble-bands, are there discontinuous,
and often follow the intricacies of the current-bedding. The black bands commonly range
from an eighth to a quarter of an inch in thickness ; some of them were found to extend
horizontally for quite 100 yards, others disappear completely within a very few feet.
Near this spot the sandstone is partially calcareous, and a blue limestone-band
formed a conspicuous shelf projecting from the cliff-face. Just below this was a
well-marked band of pebbles set in incoherent sand, or in sand only slightly
cemented by carbonaceous matter. The following sequence, in order of superposition,
will give some idea of the nature of the Beacon Sandstone at this spot.
Top. (7) 200 feet — almost pure sandstone with occasional pebbles (GG5)
(G) 2 feet — band containing carbonaceous substance (745-7G2)
(5) 12 feet — sandstone with brown bands
(4) 12 feet — hard white sandstone with a three-inch strip of fibrous
mineral (wollastonite) (G76)
(3) 12 feet — black shale and study sandstone (754)
(2) G inches — limestone-band (G71)
Bottom. (1) G feet — black shale (754)
G
VOL. I.
42
II. T. FERRAR.
The carbonaceous band (G) had been slickensided and baked to such an extent
that it has proved impossible to determine the fossils which it contains (see report by
Mr. E. A. Newell Arber, on p. 48).
The Inland Forts (Fig. 20).
Five miles west of the Inland Forts, at the spot marked y2 on the map, the
cliff forming the north side of the glacier is composed of two rocks, a yellow one
below and a dark one above. The junction as usual is regular and almost
horizontal. On examination the yellow rock proved to be sandstone. The
only accessible part was the base of the cliff where the rock is a sandstone barren
of fossils. It is subdivisible into a series of alternate yellow and white beds, and
a few pebble-patches were
noted. The following section
from the level of the ice up-
wards, shows the order of
succession : —
Top. (4) 100 feet — brown col-
li rnnar rock
(dolerite)
(3) 100 feet — yellow
sandstone
(2) 100 feet — uniformly
light - coloured
sandstone
Bottom. (1) 100 feet — a yellow
and much banded
sandstone.
At the Inland Forts, where the hills are at about the same level as those
below y2, 2000 feet of the Beacon Sandstone are exposed, and of this nearly
1500 feet have been examined. The Forts are four conspicuous hills mainly
composed of sandstone, but they are capped by dolerite (Fig. 20). They are separated
by well-marked cols through which the ice once forced its way northwards into the
adjoining drainage-system. The exposure is well illustrated by the photographs of
this side of the valley. The sandstone is part of a great deposit which is buried
westwards beneath the Inland-ice and determines the distinctive features of
the mountains on each side of the main valley of the Ferrar Glacier.
Extending southwards from C6 and C8 are two ridges of similar, sand-
stone. These ridges have rounded outlines and resemble groins built against a
sea-wall to break the force of the waves. Though nowr above the level of the
ice, they formerly acted like groins and thus collected rock-material ou their
westward side. These ridges, which I have termed the West and the East
East Groin Round Mount, C,
Fig. 20. — The Inland Fobts. Sandstone capped by Dolerite.
THE BEACON SANDSTONE AT INLAND FORTS.
43
Groin respectively, are low and attenuated spurs of the horizontally bedded
sandstone, which is here cut into cirques. The groins afford the most easily
accessible exposures of sandstone in the whole region. The slope was seldom
too steep to be climbed, and, as the horizontal structure is well etched out by
denudation, any particular bed may be traced along the whole length of the
ridge. Here, too, the rock was a white or yellowish sandstone with not so
much as a sign of a shale or a limestone-band. So far as could be ascertained,
only one of its horizons contained organic remains, and these of a most doubtful
nature. These were found on West Groin, where the surface of a sandstone-bed
was covered by what appeared to be cylindrical casts of
some organism (763-767). Possibly these cylinders may
be entirely a result of weathering, but, as they are all of
much the same diameter and cross and intercross in all
directions, I thought at the time that they are probably
more than this, and I still think that they may be of organic
origin. There is no sign of actual structure in the bound-
aries of the cylinders, but there is usually a slight depression
parallel to, and close along, their sides. The length of the
cylinders varies from six inches to three feet, and the
diameter is usually about half an inch ; they project nearly
half an inch above the smooth surface of the surrounding
sandstone.
At another spot on West Groin, 800 feet above the
level of the ice, there occurred an impression (763, 764) on
the surface of the sandstone. This appeared as a shallow
hollow, somewhat like the imprint of a flat crooked stick
with a blunt rounded end (Pig. 21). The impression was an
eighth of an inch deep, two inches wide, and one and a
half feet long.
O
Along the central line there are two markings „ _
o o of Sketch made in the Field.
parallel to the outer boundary and about a quarter of an
inch apart. These run nearly the whole length of the impression, and on either
side of them are rows of rather deeper pits about a quarter of an inch apart,
which alternate on the two sides of the central lines (see note by Mr. E. A. Newell
Arber, on p. 48).
Sundry other, but smaller, rod-like markings (765-769) occur on other specimens,
and with the same alternate pits, and I am inclined to think that these impressions
are, at least remotely, derived from bodies with organic structure. One of the smaller
impressions, which are 6 inches to 1 foot long and about half an inch across, still
retains fragmentary remains of dark carbonaceous matter. The following table of the
succession, from the bottom of West Groin to the top of its corresponding hill, shows
how uniform is the Beacon Sandstone Formation.
G
9
44
II. T. FERRAR.
Top. (14) 100 feet — dolerite, which caps the sandstone
(13) 200 feet — yellow sandstone
(12) 100 feet — sandstone with occasional yellow bands
(11) 100 feet — sandstone with ferruginous concretions (G77)
(10) 200 feet — yellow sandstone
( fl) 100 feet — sandstone with cylindrical casts (7 G 1—7 67)
( 8) 200 feet — yellow sandstone with ferruginous concretions
( 7) 50 feet — white sandstone
( G) 200 feet — yellowish sandstone
( 5) 100 feet — marble-like sandstone (679)
( 4) 50 feet — nearly white sandstone
(3) 10 feet — stalagmitic sandstone (678)
( 2) GO feet — almost white sandstone
Bottom. ( 1) 30 feet — variegated brown and yellow sandstone (hard).
The variegated brown sandstone (l) at the base appeared to be altered to a slight
extent ; it is harder than most of the higher beds, and the ferruginous concretions in
it are sometimes two feet across. They are flattened horizontally, and are sometimes
joined together.
The stalagmitic sandstone (3) is so called because stalagmites stand out between
successive beds on the rock-face, and it would appear that the rock had been locally
hardened by infiltration. It is made up of alternate hard and soft layers which are
each about a foot thick.
The marble-like sandstone (5) (G79) was harder than that above and below, and
locally its surface has a superficial glaze. The ferruginous concretions (11) (G77) in
the upper band often weather out as balls up to a foot in diameter ; sometimes,
however, the concretions have disintegrated faster than the rocks in which they were
imbedded and have left spherical hollows.
Finger Mountain (B) (Fig. 22).
Before we entered the district of the Dry Valleys, the Beacon Sandstone was
examined near the foot of Finger Mountain ; though 10 miles south of the preceding
area, it retains the same characters and appears to be barren of fossils throughout.
Near the contacts with the dolerite, variegated bands (635) have been produced. At
this spot the sandstone, like that at B1} has been dislocated ; but again its general
horizontally has not been disturbed, notwithstanding that intrusive sheets of dolerite,
up to 500 feet thick, have forced their way along joints and bedding-planes.
Finger Mountain (B) contains a wedge of sandstone which separates two sheets
of dolerite (Fig. 22). One of these sheets caps the hill ; the other separates the wedge
from the major portion of the sandstone which only just appears above the ice. The
whole sequence occupies a cliff of about 500 feet high ; the wedge of sandstone is about
100 feet thick at its eastward extremity, whence it thins westwards and disappears
in a distance of about two miles. One bed of sandstone after another is cut out by
the dolerite as it transgresses them upwards to join the mass which caps the hills to
45
the south of Finger Mountain. Immediately to the south of Finger Mountain the
wedge is considerably thinner. It is exposed in the valley, which, cutting back,
produces the sharp spur marked l, b2.
Again, on the north side of the glacier in Round Mountain (Cj) is a wedge
of yellow rock, which is probably a similar sandstone, and is also caught up by
the dolerite in exactly the same way. This mountain, however, differs from Finger
Mountain in having a small sandstone-outlier which caps and protects the dolerite
at the summit.
Along the right bank of the Ferrar Glacier from I!, to B the sandstone may
be seen above the level of the ice, but local disturbances prevent the upper surface
from appearing as a continuous line along this side.
Fig. 22. — Fingeb Mountain. Wedge op Sandstone in the Dolekite.
The Dry Valleys (Plate V).
These two valleys lie on the south side of the Ferrar Glacier and on the
west side of the Beacon Heights (B4, B3). Both have vertical sides 500 feet high,
which suddenly give place above to less steep slopes as the surrounding mountains
are approached. Both are tributaries of the main valley. The smaller lies
immediately south of Finger Mountain, and, narrowing the while, trends due west
for a distance of two miles ; at this point the valley suddenly turns southwards,
the ice which occupies it suddenly ends, and displays a bare, flat, stony bed. The
confining walls continually approach each other, and one mile above the ice-cliff
they suddenly come together in a veritable cul-de-sac.
The larger valley is about four miles long and also ends in a cliff. Its sides
are steep and parallel, and maintain the same height all the way round. The
46
H. T. FERRAR.
valley-bed here also is Hat and free from ice ; it is strewn with boulders of all sizes,
and is therefore exceedingly rough. The breadth is less than twQ miles, but near
the mouth, where it is joined by the smaller valley, it widens, and together they
open out into the main depression of the Ferrar Glacier. '
The Beacon Sandstone on the north side of the smaller valley is most
accessible at a spot west of b2, where the usual sandstone is capped by dolerite.
About 300 feet are exposed, and the beds can be traced horizontally all round the
left side until cut ofl' by the dolerite of the hill x. The main mass of x is dolerite,
but a small exposure of sandstone is visible at its western foot. In the middle
of the mass there are two other narrow strips of sandstone, each about 20 feet
thick and half a mile long, which seem to have been caught up by the intrusion.
At the west foot of this hill the Beacon Sandstone shows a new feature, for on the
under side of a large block there was a six-inch bed composed of angular quartz-
fragments (681). These pieces of quartz are fairly regular, almost cubic, and about
an inch long. They are set in a matrix of the usual sandstone, and it is worthy
of note that no rounded pebbles were here observed.
There are four very prominent buttresses south of x, which form the sides of
the larger valley, and in each of the buttresses two bands of yellow rock and two
of brown rock were seen alternating regularly. These alternations possibly represent
parts of once continuous intrusions of dolerite which follow the same bedding-planes
across the whole area. A similar arrangement also holds in the buttresses of the
eastern valley wall.
The Beacon Heights (B4, B3) (Plate V, and Section I, Plate VII).
On the western side of Beacon Height West (B4) there is a small outcrop of
the Beacon Sandstone. This, as before, has horizontal bedding-planes. The bulk
of the rock is coarse, even-grained in texture, and almost white in colour. The
greater part of the mountain appears to consist of sandstone, for the lower 2000 feet
shows a yellow rock, with horizontal joints or bedding-planes, where the even
covering of dark talus-products is wanting. The summit is a small cap of brown
rock, which is separated from a larger mass of the same brown rock by a band of
yellow about 500 feet thick, also bedded horizontally. The larger mass of brown
rock is continued in the summit of B3, and even extends to the summit of Knob
Head Mountain further to the east.
The sandstone crops out as a small cliff on the side of this mountain (B4) ; there
the cylindrical rugosities (see p. 43) are again developed and appeared to be quite
similar to those observed on West Groin. This outcrop was traced for a distance
of a quarter of a mile along the hillside, and the cliff is on an average 50 feet
THE BEACON SANDSTONE AT KNOB HEAD MOUNTAIN.
47
The Terra Cotta Mountains (Fig. 23).
The Terra Cotta Mountains (B6, B7, Bs) appear to be composed mainly of sand-
stone. They are abundantly riddled by dykes of dolerite which appear to have had
considerable effect on the sandstone. The sandstone has a pale-pink tinge, and in
the distance the hills have a clull-red colour, which contrasts strikingly with the
dazzling snow and the yellow sandstone elsewhere. Some of the specimens from
the moraine show that the sandstone has been altered to quartzite (697). The
dykes will be mentioned in the next chapter when the dolerite-rocks are considered.
Knob Head Mountain (B9) (Plate V).
The last spot where the Sandstone Formation was examined is on the east side
of Knob Head, 30 miles from the South-west Arm or the hill Bj and about
30 miles from the sea. Here
a small outcrop, similar to
that on the west of Beacon
Height, is found at an eleva-
tion of about 3500 feet. At
a distance the whole lower
portion of Knob Head Moun-
tain appears to consist of the
sandstone. The mountain,
like the Beacon Heights, has
a small cap of dark-coloured
rock (dolerite), which is
separated from a larger
sheet below by a narrow
and horizontal yellow band.
This similarity in the summits of the three mountains makes it probable that all
are parts of the same two sheets of dolerite.
The sandstone-outcrop on Knob Head is less than a quarter of a mile long
and about 100 feet high. Here again one bed was observed to have, all over its
surface, cylindrical prominences like those described on page 43. At this spot,
also, some beds contain alternate dark-coloured and light-coloured laminations, but
nowhere did the rock show bands at all like the black bands found near the foot
of Bj, nor was there found any structure in these dark laminae which would
suggest organic life.
48
APPENDIX TO CHAPTER VI.
REPORT ON THE PLANT-REMAINS FROM THE BEACON SANDSTONE.
By E. A. Newell Arber, M.A., F.L.S., F.G.S., University Demonstrator
in Palseobotany, Cambridge.
The remains collected by tlie ‘ Discovery ’ Antarctic Expedition, and regarded as
probably of the nature of fossil plants, are unfortunately of little value botanically.
The material was derived from two localities, viz., the hill Bx in the South-west
Arm of the Ferrar Glacier, and the Inland Forts. The specimens of Beacon Sandstone,
containing much carbonaceous material, from the bill Bj were collected by Mr. Ferrar
on the 12th and 13th of November, 1903, at a height of 50 feet above the level of
the ice (see p. 41). Several of these show fair-sized, carbonaceous impressions or
markings, which, in all probability, are of vegetable origin. One example somewdiat
resembles in appearance a piece of petrified wood, but a microscopic section made
from this material has failed to show any trace of organic structure.
The specimens from the Inland Forts are pieces of a pale yellow sandstone,
obtained by Mr. Ferrar on November 1G, 1903, at a spot some 800 feet up the West
Groin (see p. 43). Some of these show one or more series of irregular puckerings,
consisting of slight pits or depressions, sometimes lined by a small amount of
carbonaceous material. It appears, however, to lie impossible to form any opinion as
to whether these features are due to vegetable agency or otherwise.
The imperfect evidence presented by these specimens will neither permit of any
opinion as to the botanical nature or affinities of the fossils themselves, nor of the
geological age of the beds in which they occur. Their discovery may, however, lie
regarded as affording indications that, at some period or other in geological time,
vegetation flourished so far south as latitude 77k° ■ Such a conclusion i$ of great
geological interest, and is in harmony with the fact, now ascertained beyond doubt
by the discovery * of abundant evidence of varied vegetations belonging to several
different geological epochs, that the climate of the Antarctic, as of the Arctic regions,
has been much more genial at more than one period in the past than at the present
day.
Nathorst, A. G., Sur la flore fossile des regions antarctiques, Compt. Rend. Acad. Sci., 1904, cxxxviii, p. 1447.
49
Chapter VII.
THE DOLERITES.
The doleritic type of rock has been found in practically the same localities as the
Beacon Sandstone, and it will, therefore, be convenient to consider the localities in
the same order as before. The dolerite of Depot Nunatak is the highest point from
which rock of any kind has been collected in South Victoria Land. Dolerites occur here
at an elevation of 7000 feet, and, at the foot of Knob Head Mountain, about 30 miles
nearer the coast, they have also been seen only 3500 feet above sea-level. There is
no evidence of the presence
of surface-outpourings, and
as no vesicular or scoria-
ceous rocks were observed,
even in the moraines, it
would appear that these
rocks are wholly intrusive.
Depot Nunatak (A.)
(Figs. 24 & 25).
Lieutenant A. B.
Armitage on the first
journey through the Royal
Society Range obtained
weathered dolerite - frag-
O
ments (632, 633) at Depot
Nunatak, and at the same time Engineer-Lieutenant R. W. Skelton photographed
Fig. 24. — Depot Nunatak, from the East.
great
columns
through the snow.
The
the parent rock which rises as a mass of
rock (662)’ is an outlier, and protrudes through the snow at an elevation of about
6000 feet. The nunatak rises to a height of nearly 500 feet above the snow and
is exceedingly columnar throughout. Some of the columns are 12 feet in diameter
and, though broken, give the impression that they extend the whole height of the
cliff. Depot Nunatak is 60 miles from the coast and is entirely cut of}' from the
dolerite
capping
the sandstone eight miles to the east.
The Hill Bx.
Here the dolerite caps the sandstone and produces a cliff which rises vertically
for more than 500 feet. This cliff forms the east side of the South-west Arm for
a length of ten miles. As before, the columns which go to make up the sheet are
VOL. I.
H
50
H. T. FERRAR.
about 12 feet in diameter, and weather to a bright-chocolate colour. At its
junction (668) with the sandstone it becomes finer in texture, and it has obviously
altered the sandstone for a distance of two feet, at least, from the contact. At one
spot a mass of sandstone lias been caught up in the sheet, and, near by, a pipe
of dolerite 50 yards in diameter cuts vertically across the bedding-planes of the
sandstone. The sheet extends along the side of the valley towards Finger Moun-
tain and is interrupted by occasional small faults, the throw of which is always
less than 100 feet.
On the north side of
and are capped by a dark
or dyke of the dolerite
cuts through the sand-
stone and joins the mass
above, and at y:i there
are two sheets of dolerite
separated by sandstone.
Further east again, at
C7, the cap of dolerite
has been removed by de-
nudation, and the lower
sheet, which is reduced to
under 100 feet in thick-
ness, caps these isolated
hills. Near the Inland
Forts three dykes of the
dolerite were examined,
and a series of specimens
(682-688) of dolerite and sandstone was collected along a line transverse to
one of these dykes. These dykes are about 12 feet across and rise vertically
through the sandstone to join the dolerite-sheets above. Where the dolerite meets
the sandstone weathering has been accelerated, and the dykes lie in chimneys*
which are sometimes 20 feet deep. One of these pipes is on the south end of West
Groin, the other on the north end of East Groin and close to West Fort (C9).
Finger Mountain (B) (Fig. 22).
This mountain is 7084 feet high, and is composed of alternate layers
of sandstone and dolerite. The lowest rock visible at the base is Beacon
The Inland Forts (Fig. 20).
the Ferrar Glacier the hills are very uniform in height
rock for a distance of 20 miles. At the point y1 a pipe
Fig. 25.— Columnar Dolerite of Depot Nunatak.
* A. Geikie, ‘ Ancient Volcanoes of Great Britain,’ 1897, vol. ii. p. 120.
THE DOLEEITES OF FINGER MOUNTAIN.
51
Sandstone, of which not more than 100 feet appears. Above this there is a
sheet of columnar dolerite 200 feet thick, which on the west side of the hill
unites with another sheet of dolerite. These two sheets are separated by the
wedge of sandstone already referred to in Chapter VI. The bedding-planes of
this wedge are horizontal, and are made conspicuous by the intrusion of numerous
thin sills of dolerite along them. The wedge tapers to the west, and at this end
the columns of dolerite do not appear continuous throughout the cliff, but break
and bend over to the west at a line which follows the inclination of the upper
surface of the now absent wedge. Finger Mountain narrows eastward to a sharp
spur (bj-ba). The upper sheet of dolerite ends at a scarp at the summit of B,
while the lower continues some distance and is cut off by a structure-line, parallel
to the line of transgression followed by the sill above the sandstone-wedge. This
spur is capped by an outlier of the dolerite, which is separated by sandstone from
a lower sill of transgressive dolerite 200 feet thick.
Specimens were collected both from the last sheet (692) and from a dyke
(691) cutting across the bedding-planes between this and the one above. Dykes
and sills are numerous at this rather disturbed locality. A sill of dolerite 30 feet
thick extends for a hundred yards along a bedding-plane, then terminates suddenly
with a vertical end. Another sill 10 feet thick runs along a bedding-plane for
50 yards, breaks steeply downwards for 50 feet, and then, forcing its way along
a bedding-plane for 100 yards, finally thins out and disappears. A third sill,
2 feet thick, extends 50 yards along a horizontal bedding-plane, but gradually
decreases in thickness and ends as a wedge.
The specimens (695, 696 ; see p. 138) were collected from the base of x, from
a sheet of dolerite below 200 feet of sandstone. Here also the dolerite occurs
in sheets wdiich alternate with the layers of sandstone, and dykes and thin
sills are numerous.
Knob Head Mountain (Bg) (Fig. 26).
At a height of 3000 feet above the sea, and 30 miles inland, close under the
foot of Knob Head Mountain, which is over 8000 feet in altitude, there is a cliff,
100 yards in length and 200 feet high, composed of columnar dolerite. Above the cliff
the hillside slopes up more gently, and is covered with drift-blocks of granite and
dolerite ; the covering is broken only by this exposure of rock near its base. The outcrop
shows columns 12 feet in diameter, and from 20 to 200 feet in height. There are
occasional horizontal cross-joints, but cup-and-ball structure is not developed. The
Beacon Sandstone appeared to rest upon this dolerite-mass (661) ; it forms the
main mass of the mountain, but the junction of the two could not be found. The
hill B(;, three or four miles to the west, consists mainly of sandstone, but is
riddled by dykes which form a network on its surface. On its west side there is
h 2
52
H. T. FERRAR.
a pipe of dolerite about, 100 feet in diameter, which rises vertically through the
sandstone, but cannot be traced to a junction with any of the overlying sheets of
dolerite.
The Kukri Hills (Fig. 27).
The bluff D forms the western extremity of the Kukri Hills and, as already stated,
consists mainly of dolerite. If reference be made to the section along the Kukri Hills
(Section III, Plate VII) it will be seen that two parallel sheets of dolerite, each
Fig. 26. — Columnar! Dolerite at the foot of Knob Head. The large boulder on the sky-line is of Granite.
about 2000 feet thick, run together at D. These sheets dip to the westward, and
a specimen (704) obtained shows that the dolerite becomes finer in texture at its
junction with the granite. The specimen was got from just above the lower junction
and below D2. The junction here is most striking and extends in an absolutely
straight line for a distance of 10 miles along the side of the East Fork.
From the regular alternations of yellow and dark-coloured rock, I was at first
inclined to suppose that there are two sandstone-deposits, but further work proved
that the intrusive sheets cannot be continuous over the whole area. It would,
however, be interesting to know what structural weaknesses have induced the
EXTENT OF DOLERITE-FORMATION.
53
dolerites to maintain uniform horizons for so great a distance, and to remain always
separated by the same thickness of sandstone. The sandstone between the sills is
always about 500 feet thick.
The Former Extension.
On consideration of the facts stated above, it will be seen that wherever a
dark rock was encountered it proved to be dolerite ; further, all the abundant dark
fragments in the moraines belonged to that kind of rock. Dolerite has been shown to
cap the highest sandstone seen, and to be intrusive into it on each side of the upper
Ferrar Glacier. Some mountains are entirely composed of dolerite, and others, such as
the Beacon Heights or Knob Head Mountain, have mere caps of that rock, which may
be remains of a once con-
tinuous sheet. At the
Cathedral Bocks dolerite
must rest upon granite,
and apparently at one time
have been continuous with
the sheet which caps the
granite in the Kukri Hills.
Further, the Royal Society
Range, which is a faulted
crust-block * and is
higher than any of the
surrounding country, has
strongly developed
plateau -features (Fig. 9,
p. 23) ; the rock which
forms the highest peaks
is dark and therefore pro-
bably dolerite. If this be so, we may be sure that both dolerite and the Beacon
Sandstone Formation extend quite 50 miles in an east-and-west direction.
Next, dolerite caps the Beacon Sandstone at the Inland Forts, and the hills
for at least 10 miles north are capped by rock which cannot be other than dolerite.
The contrast in colour between cap and sandstone is always so strong that this
inference could be made even without regard to the evidence of abrupt changes
in the hill-outlines at the junctions. For similar reasons there can be no doubt
that the dolerite still caps the sandstone-hills, which extend 10 miles to the south
of the main Ferrar Glacier (Fig. 28). These facts render it extremely probable that the
Cathedral Rocks Kukri Hills
Fig. 27. — The dark band in the Kukki Hills on the bight shows
THE DOLERITE-SHEET BESTING UPON THE EVEN SUBFACE OF THE GbaNITE.
Gregory, 1 The Great Rift Valley,’ 1896, p. 220.
54
II. T. FERRAR.
sandstone-dolerite area is at least
50 miles long in a north-and-south
direction, and has an extent of 2500
square miles at least.
A few of the hills in this neigh-
bourhood, and especially the hill we
called Obelisk (C3), 10 miles east of In-
land Forts, are pointed (Fig. 39, p. 72)
and resemble the hills in the Torridon
Sandstone districts of north-west Scot-
land ; we may surmise that in these
cases the. cap of dolerite is lacking.
In Granite Harbour a dark rock
which weathers like columnar dolerite
may be seen above the granite (see
p. 32), and further up the coast to the
north the higher peaks of many of the
hills were formed of a dark rock and
stood out in striking relief over the
Inland-ice against the cloudless sky.
The black tabular formation is most
striking about latitude 77c 0' S., longi-
tude 164° 49' E., and again in latitude
75° 57' S., longitude 163° 5G' E.
The area occupied by the Sand-
stone Formation is a question still to
be solved. At the outskirts of the
area indicated above there seems
to be no doubt the sandstone has
an enormous thickness, roughly
2000 feet. Lieutenant M. Baene
records horizontal structures in lati-
tude 80° S., and Lieutenant E. II.
Shackleton has taken a photograph
still further south which shows the
plateau-features to be still prominent
there. Towards the north we need
only mention Mount Nansen in lati-
tude 75° 30' S., and the pyramidal
forms of the peaks of the Admiralty
Range discussed above.
55
Chapter VIII.
THE SEA-ICE AND THE SHORE-ICE.
The Sea-ice.
During tlie winter months the surface of the sea in high latitudes often freezes
in a uniform sheet which does not vary greatly in thickness. This covering has
had many names given to it, but on the whole Sea-ice is perhaps the most suitable,
as suggesting that the ice is derived directly from the sea.
Sea-ice requires to be distinguished from other floating ice (ice at sea) of
different origin, and this
can readily be done by the
close examination of even
a small fragment. The
Structure of sea-ice has
been dealt with by Dr.
Axel IIamberg,* Dr. von
Drygalski,! and others.
Dr. von Drygalski describes
sea-ice as being composed
of bundles of fibres packed
together perpendicularly to
the surface of cooling. A
point not mentioned by
him is, that, when the sea
first freezes, the upper two
inches consists of plates,
a quarter of an inch across
and a sixteenth of an inch thick, which lie horizontally, and only gradually do these
give place to the sheaves of vertical fibres which make up the greater mass of the
ice. Between the plates and the fibres is a layer of ice, about half an inch thick,
of which the structure is very confused. Fig. 29 shows a mass of ice-crystals which
have grown upon a fishing line. Mr. Hodgson records these crystals as occupying
a length of 17 fathoms of his line, and gradually diminishing in quantity from
the surface downwards.
The Salinity * f seems to depend more upon the rate of freezing than upon the
depth or distance from the surface. Both the authors quoted above have made
* Axel Hamberg, Bihang, K. Svenska. Vet.-Akad. Handl., 1895, Bd. xxi, Afd. 2, No. 2.
t Drygalski, 1 Gronland-Expedition,’ 1897, Bd. i, p. 424.
Fig. 29. — Crystals of Ice which have grown upon a Fishing Line
SEVERAL FATHOMS BELOW THE LOWER (OR FREEZING) SURFACE OF
THE SEA-ICE.
56
H. T. FERRAR.
observations in this connection, and have proved a great amount of variation. Our
observations show that the variation is even greater than they have recorded. We
may mention here that the average amount of salt in dry sea-ice is about 4 • 3 grams
a litre, whereas 32 *8 grams a litre is the average salinity of the sea.
Ice met with at sea is more variable, and the amount of salt contained
in it depends upon the previous history of the ice.
Near the shore, where floe-ice has buckled below the level of the sea-surface,
shallow ponds form, and a gradual concentration of the dissolved salts takes place.
Solutions containing as much as 266*6 grams of salt a litre have been found in
such pools.
When the open sea first freezes, part of the concentrated solution left yields
well-crystallized rosettes (ice-flowers) on the upper surface of the ice. The rosettes
are usually two to three
inches across and about an
inch high and are scattered
thickly over the surface ;
they are always extremely
saline. If the sea-ice be-
comes depressed by a great
accumulation of snow, the
original upper surface is
still always capable of
identification through this
local first salt-concentrate.
The Snow on the sea-
ice in McMurdo Sound does
not often accumulate to a
thickness greater than two
feet. Where any object
interrupts the uniform level of the ice, snow-drifts form. The amount of the drift
depends upon the magnitude of the object ; such an object as our vessel produced
a drift some 30 feet thick. The accumulation of local drifts has little ultimate effect
upon the depth of the lower surface of the ice, for the snow, as it accumulates, pushes
down the underlying layers into a region where the temperature of the water
approaches that corresponding to greatest density ; the lower ice then melts and is
transported by the currents of the water (Fig. 49, p. 85). The equivalent of more
than 18 feet of snow has been observed to be removed by this means during two
consecutive years. From this observation it would seem impossible that a very thick
sheet of ice could be entirely built up through continued deposition of snow on
sea-ice or oceanic ice.
At a hole near the ship the upper surface of the snow was three feet above
Fig. 30. — The Pack-ice, seen from the Crow’s Nest of the Ship.
PACK-ICE.
57
water-level, and the original freezing surface of the ice had been depressed to four
feet below water-level. The total thickness of snow accumulated at this spot during
the two years was more than 20 feet ; as the thickness here when the ice broke up
was only 15 feet, the whole of the original ice and the earliest deposits of snow
must have been entirely removed by the melting action of the sea-water previous
to the final break-up. By this means, water-vapour from low latitudes is condensed
in high latitudes and transported again to low latitudes, without taking part in the
glaciation of the polar land-masses.
Size of floes. Names * such as Pancake-ice, Bay-ice, Field-ice, Pack-ice,
Stream-ice, are given to sea-ice at certain periods of its short life. Thus Pancake-
ice is the first product of the frozen sea. It is an aggregation into roughly circular
masses of the ice-plates which first crystallize. The fiat cakes thus formed are
about an inch thick and one
to two feet across, and have
notably turned-up edges. In
a sheltered bay where these
cohere to form a thin sheet,
the result is called Bay-ice.
Later, this thickening extends
to large areas and the ice is
then called Field- or Fast-ice.
During the summer Field- or
Floe-ice breaks up into floes,
which float out northwards
from McMurdo Sound and
join the belt of Pack - ice
(Drift-ice, Treib-eis ) encircling
the Antarctic regions (Figs.
30, 31). As the floes drift north they break up and dissolve away, and they are
met with at sea in elongate and scattered patches which are termed Stream-ice.
The size of the floes varies greatly; one may be 100 yards across, another may be
two miles or more, but the thickness in the Ross Sea is never more than six feet.
The floes are necessarily larger f on the south side of the belt of pack, as there
they are protected from the swell by the stream-ice. Outside, the swell is most
destructive and rapidly breaks up any large ice-field.
Hummocks are rather exceptional in the sea-ice of South Victoria Land,
and all that occurred in McMurdo Sound were less than three feet high. These
seemed to be caused by the ice in the outer part of the bay breaking for a time
* Markham, ‘The Antarctic Manual’ (Boy. Geogr. Soe.), 1901, p. xiv; H. Bink, ‘Danish Greenland,’
1877, p. 73.
t Colbeck, Geog. Journ., 1905, vol. xxv, p. 403.
Fig. 31. — The ‘Discovery’ brought to a standstill by Pack-ice.
VOL. I.
I
White Island Black Island Mount Discovery
58
H. T. FERRAR,
Cape Armitage Hut Point
Fig. 32.— Water-holes in Sea-ice at Cape Armitage and Hut Point in January, 1904.
THICKNESS OF SEA-ICE.
59
from the inner ice which is fixed, and then impinging on it again and again. In
the belt of pack-ice, hummocks are rare and usually less than 10 feet in height.
The Ross Ice-sheet presses against the sea-ice on the south-east side of the
Winter Quarters peninsula, and produces a series of hummocks some two miles long.
Some of the hummocks rise as buddings of the sea-ice, here 8 feet thick ; in others
the sea-ice breaks into pieces about 20 feet long and these are forced up on end. Four
parallel rows stretched out south-west from Pram Point and grew very gradually
before the movement of the ice-sheet, only becoming conspicuous at the end of the
first winter.
The Thickness. — The sea-surface freezes during the winter, but its ice rarely
exceeds a thickness of 8 feet. Our observations on the thickness of sea-ice are
rather exceptional for McMurdo Sound as a whole ; for in one case the ice-gauge
was placed where a strong
current was known to
exist, and in the other
case the ice - gauge was
placed in a sheltered bay,
in ice which was always
wind-swept and free from
snow. At the former spot
on March 1st, 1903, a
water-hole was open : on
the 24th of April the ice
in it was 3 feet thick ;
and by about mid-winter
(June 27th) it had grown
to 5 feet. On August
23rd the thickness was
6 feet 6 inches, and water
continued to freeze until December 5th, by which time it attained its maximum
(8 feet 5| inches). After this date the ice began to disappear from below, and by
January 28th was all gone. A water-hole off' Cape Armitage (Fig. 32) was observed
to open each year, which shows this melting action of the sea on sea-ice to be
important. In 1904 the ice which surrounded the ‘Discovery’ broke up naturally
and rapidly floated away, and the rate at which the break-up took place seemed to
be independent of the thickness (Figs. 33, 34).
Transpoi't. — During the winter-months cracks in sea-ice, radiating from both
Hut Point and from Cape Armitage, were formed. These cracks are produced by the
ice in the middle of the strait being pushed forward faster than the ice at the sides.
The pressure is always in one direction, and is caused partly by the movement of the
Ross Ice-sheet and partly by accumulations of snow along the shore. The cracks
Pig. 33. — Sea-ice breaking away from the Winter Quarters in 1902.
GO
II. T. FERRAR.
freeze up soon after they are formed, and no movement in the reverse direction can
therefore take place. Cracks made in this way seldom open much more than two
inches at any time, hut during the year have indicated a movement of more than
20 yards. Sea-ice therefore is removed in three ways: (l) through corrosion by
sea- water, (2) breaking up and floating away piecemeal, (3) creeping bodily away
from the land.
The “ creep ” of sea-ice is very small compared with the movement which takes
place when an ice-field breaks up at the end of the winter ; but it is important, as it
prepares the field-ice for the action of the ocean-swell which breaks it up during the
Pig. 34. — The Relief-ships forcing their way through the barrier of Floe-ice in 1904.
summer. As the field-ice breaks up, the floes that are formed drift northwards to
augment the pack-ice.
Thawing. — Virtually no surface-thawing of the sea-ice takes place, for the
air-temperature in the open is always far below the freezing point, and the snow
reflects most of the radiant heat. Locally dust and other extraneous particles sink
into the ice or snow, but the holes thus formed are filled by the silting action of
the snow. Mention has already been made of the action of the sea in reducing
the thickness of the floe-ice, and it is noteworthy that during sunny days the
sea-temperature itself is slightly raised and then the water melts more quickly the
ice at its seaward edge. This greater capacity for melting the ice has been observed
for at least two miles within the margin of the fast-ice of McMurdo Sound.
SHORE-ICE FRINGES.
61
The Shore-ice.
The shore of South Victoria Land is always fringed with ice, which ends sharply
seawards in a perpendicular wall (Plate II). The wall varies in height from
3 to 300 feet, and two
types of fringe may be
distinguished.
(l) Fringe due wholly
to the frozen spray. Such
fringing ice never attains
a height of more than
six feet. The fringe of
O
ice around the land in
Granite Harbour is perhaps
most characteristic, and
forms a typical ice-foot.*
It remained firmly frozen
to the land all the year
round. It is not materially added to by snow-precipitation, and dissolves rather
than increases during the few summer-days when the sea washes against it. As the
maximum rise and fall of the tide is less than three feet, a low ice-foot, as contrasted
with the high ice-foots of
Arctic regions, is to be ex-
pected. The breadth of the
fringe varies from 6 to per-
haps 60 feet, and the surface,
which is usually fairly flat,
often contains pools of water
during summer. The chief
action of the fringe is conserv-
ative.! It protects the land
from the action of eroding
breakers and floating ice, and
more especially protects the
rock-cliff by cementing to-
gether its talus.
(2) Fringe of glacier-ice
adherent to the land (Figs. 35, 36). The ice is free from salt, shows the well-known
blue bands, and has a well-marked granular structure. Such fringes vary much in
Fig. 36. — Shore-ice wrapping the Land near the foot of
Castle Rock.
Fig. 35. — The Ice-foot at Hut Point.
* Drygalski, ‘ Gronlaud-Expedition,’ 1897, Bd. i, p. 285.
f Bonney, Quart. Joum. Geol. Soc., 1902, vol. lviii, p. 699.
62
H. T. FERRAR.
their dimensions. The height varies from 6 feet to 300 feet, and the breadth from
10 yards to a mile. The surface usually dips at about 20° towards the ice-cliff.
As in the case of the first type, the action is wholly protective, but is even
more effective owing to the greater size of the fringe.
A fringe of glacier-ice wraps the west side of Winter Quarters peninsula from the
base of Mount Erebus to Crater Hill ; near Castle Rock, where the ice-cliff reaches
300 feet, the fringe is perhaps at its highest.
Around Winter Harbour the cliti' varies from 10 to 40 feet in height, and has
a breadth which is sometimes as much as 100 yards ; there, and in several other
places, this type of fringe forms a sort of ill-marked terrace about 100 feet above
sea-level, and slopes gently out towards its seaward face.
As in the case of glacier -terminations, it would seem that a lofty ice-cliff
has always correspondingly deep water immediately adjoining it.
63
Chapter IX.
THE LAND-ICE.
The following classification of the land-ice of South Victoria Land has been found
convenient. The headings are largely taken from Dr. E. von Drygalski’s ‘ Gronland
Expedition,’ and from Dr. Albert Heim’s ‘ Handbuch der Gletscherkunde,’ but have
been to some extent modified to meet local requirements.
(1) Ice-sheet * or Inland-ice* is the name applied to a mass of ice which
covers a continental area of land. In South Victoria Land the sheets
are of unknown extent, and enwrap and obliterate the inequalities of the
interior land-surface, leaving coastal land-fringes comparatively free from
ice.
(2) Local Ice-caps ,f Hochlandeis, f the ice covering partially or wholly a
limited land-mass. This ice may extend as an unbroken mass right down
to the sea, or may escape as ice-streams. Such a cap may be defined as
an ice-sheet on a 'Small scale. These terms are necessarily relative, for we
frequently speak of a polar ice-cap with reference to the earth as a whole,
or again of a “ local ice-cap such as that upon Hayes Peninsula, quite
a small area.
(3) Piedmont-glaciers J are formed by ice crowding on to a coastal plain at the
foot of a mountain-range. In South Victoria Land three types are
distinguished : (a) normal piedmonts-on-land ; (b) piedmonts-aground ;
(c) piedmonts-afloat.
(4) Glaciers of Greenland type , or Ice-streams, § drain an ice-sheet and end
in the sea.
(5) Glaciers of Norwegian type || consist of streams of ice flowing down well-
defined valleys (fiords) from a large firnfield.
(6) Glaciers of Alpine type* ] Valley-glaciers, drain small intermontane basins
(firnmulden) : seldom advancing far from their mountain-sources, they never
reach the sea.
* H. Rink, ‘ Danish Greenland ’ (1877), p. 39 ; Drygalski, ‘ Gronland-Expedition,’ 1897, Bd. i, Chapter IV ;
Heim, ‘ Handbuch der Gletscherkunde,’ 1885, p. 51 ; Garwood and Gregory, Quart. Journ. Geol. Soc., 1899, vol. lv,
p. 682 ; Nansen, ‘First Crossing of Greenland,’ Chapters XV to XVIII.
f H. Rink, ‘Danish Greenland’ (1877), p. 366; Drygalski, ‘Gronland-Expedition,’ 1897, Bd. i, p. 118;
Chamberlin, Journal of Geology, 1895, vol. iii, p. 66.
t I. C. Russell, ‘ Glaciers of North America,’ 1897, p. 2.
§ H. Rink, ‘Danish Greenland’ (1S77), p. 369; Heim, ‘Handbuch der Gletscherkunde,’ 1885, p. 51, ff. ;
Drygalski, ‘Gronland-Expedition,’ 1897, Bd. i, Chapter IV (Die Karajak-Eisstrome und ihr Nahrgebiet).
|| Heim, ‘ Handbuch der Gletscherkunde,’ 1885, p. 50 ; Drygalski, ‘ Gronland-Expedition,’ 1897, Bd. i, p. 298, ff.
If Heim, ‘ Handbuch der Gletscherkunde,’ 1885, p. 55 ; Drygalski, ‘ Gronland-Expedition,’ 1897, Bd. i, p. 298, ff. ;
A. Geikie, ‘ Textbook of Geology,’ 4th edit., 1903, vol. i, p. 535, ff.
64
H. T. FERRAR.
(7) Cliff-glaciers* Re-cernented glaciers, Glaciers remanies are broken glaciers of
the above types, which having a Thalweg, too steep in places to hold the
ice, have areas of bare land separating the firnfields and the final tongues
of ice which they produce.
(8) Hanging glaciers, f Corrie-glaciers, Hangegletscher, are masses of snow and ice
lying in cirques, corries or hanging valleys. The ice in these disappears by
melting, or by ablation, or by both processes, before the glacier can reach the
main ice-stream.
(9) Ice-slabs are apparently peculiar to South Victoria Land, and are glaciers
which, from the cessation of ice-supply, have slipped away from their
former firnfield.
(10) Icebergs J are common to both the polar regions, and may be appended to
this list as products of the breaking-up of all glaciers which reach the sea.
1. The Inland-ice.
We have seen that the mountain-belt of South Victoria Land, quite 60 miles
in breadth, buttresses a firnfield of vast but unknown extent. This ice-sheet has a
horizontal upper surface, which on the west of the Royal Society Range has an
elevation of 7650 feet. It Hows eastward through the range in a deep-sided valley
which bifurcates downwards in a most peculiar way. In the Prince Albert Range
the flow passes between nunataks, which are sometimes 20 miles long, in arms ten
miles across. The nunataks are usually broadside on to the present flow. The
passes are shorter than they are broad, and have a striking similarity to those
on the west coast of Greenland. §
In the ranges lying south of the Royal Society Range are deep and narrow
channels termed “ inlets ” by Captain Scott. These channels are ice-filled, but
lie much below the level of the adjacent mountain-ranges and have very steep
sides. A theodolite showed only a very slight rise in the horizon along these,
so that if the Hinterland rises at all high it must lie many miles west of the
coast. The Inland-ice flows through these channels to feed the Ross Piedmont
or Ice-sheet.
The surface of the Inland-ice, where observed, consists entirely of soft snow-
powder and shows no gradual passage through granular snow to compact ice.
The soft snow-powder, being readily taken up by wind and whirled about, is
removed in bulk and transported bodily into the Ross Sea.
* Heim, ‘ Handbuch der Gletscherkunde,’ 1885, p. 58.
f Heim, ‘ Handbueh der Gletscherkunde,’ 1885, p. 45.
t Heim, 1 Handbuch der Gletscherkunde,’ 1885, p. 273 ; Drygalski, ‘ Gronland-Expedition,’ 1897, Bd. i,
Chapter XIV.
§ Drygalski, ‘ Gronland-Expedition,’ 1897, Bd. i, Chap. IV. and maps at end of vol. i.
ICE-CAPS.
65
2. Local Ice-caps.
The icy covering of Mount Erebus provides, perhaps, the best example of an
Antarctic local ice-cap, and its features are exactly those of the Greenland ice-caps on
a small scale. Snow covers the greater part of the area, ice-streams flow down
between bare nunataks,* e.g. The Turk’s Head and The Skuary, and there is the bare
coastal fringe between Cape Royds and Gape Barne.
The streams of ice which radiate from the mountain are too ill-defined to be
called true glaciers. They have no snow-sheds, neither have they any well-defined
banks. Still, all the mountains of Ross Island are completely covered with snow, and
at definite points give off icebergs, which float away to the open ocean. The upper
parts of Mount Erebus are covered by snow so thin that the outlines of lava-streams
near the summit can be
recognised at a very great
distance. The middle
slopes have occasional
patches of bare rock pro-
truding through the snow,
and rising up or dipping so
steeply that snow cannot
long remain upon them.
The lower slopes are even
more broken by bare lava,
and the ice must always
average less than 700 feet
in thickness, for the ice
sea-cliff is never more than
100 feet high.
The surface of Mount
Erebus shows numerous
ice-falls in which the crevassed ice-surface stands out above the level of the more
even normal ice-covering. Mount Terror, Sturge Island of the Balleny Group,
Coulman Island and Mount Melbourne, all form centres of local ice-shedding (see
Plate I and Fig. 1 (p. 3), Fig. 3 (p. 5), Fig. 37).
3. Piedmonts.
Masses of ice, which have a breadth greater than the length measured along the
direction of flow, and lie at the foot of all large areas of high land, are conveniently
referred to as piedmonts. These masses in South Victoria Land differ from such a
* I. C. Russell, 1 Glaciers of North America,’ 1897, Tacoma, p. 62, Fig. B.
Fig. 37. — Ice-foot and Pack-ice in Wood Bay at foot of
Mount Melbourne.
vol. i.
K
66
H. T. FERRAR.
typical piedmont as the Malaspina Glacier * in that they ai-e supplied by driftings
and snow-cascades from the adjoining land-mass, whereas the Malaspina Glacier lies
below the snow-line and is only fed by distinct valley-glaciers.
Three types of piedmont are distinguishable.
(a.) Normal piedmonts, piedmonts-on-land.
On the west side of McMurdo Sound, between the moraines and the Northern
Foothills, there is a continuous ice-belt without apparent source of supply and lying
wholly upon low land. This ice-belt occupies an area 10 miles long and 5 miles
broad, and appears to be fed by the snow drifting over the lower passes of the
foothills. On the west it covers the hills to a height of quite 1000 feet, but on the
east it ends as an insignificant marginal sea-cliff less than 50 feet high. On the
north it slopes sharply down towards the depression of the Ferrar Glacier, but on the
south it merges into the Blue Glacier. On the whole the surface is convex, and
slopes more steeply around the outer edge. The whole of the ice rests on land, and
seems, by entirely burying the shore, to protect it from denudation.
Occasionally, along the base of the Prince Albert Range, the convex ice-slopes,
such as the one discussed, connect definite ice-streams, which, strangely enough, are
always at a lower level. From Cape Bernacchi to Granite Harbour, and from there
to Cape Gauss, there are two notable unbroken stretches of ice-covered land. These
areas may be regarded as series of land-piedmonts. Occasionally, conspicuous sea-
washed rock-cliffs protrude through them, and the ice-cliff facing the sea is obviously
the edge of a broken ice-lenticle. The length of the mass varies from 10 to 50 miles,
but the breadth cannot be more than about 10 miles. The evidence would seem to
suggest that piedmonts are rather relics of a former greater ice-supply than products
of the present conditions ; their action would appear to be now entirely protective.
(b.) Piedmonts-aground.
Piedmonts-aground are well represented along the sides of Coulman Island,
which has bare cliff-sides and a fiat snow-covered top (Fig. 3, p. 5). It is surrounded
by a comparatively low ice-wall produced by a talus of snow, which drifts off the
top of the cliff and accumulates along the cliff-sides to form a continuous belt.
Sometimes the talus is a mixture of rock and ice, but as a rule rock-matter was
conspicuously rare. Small cascades of ice fall over the rock-cliff along the dykes
and joint-cracks, which are seldom large enough to be called valleys.
This fringe has in section a convex upper surface. Near the rock-cliff it
becomes steeper. At the seaward edge the convexity increases, and the termina-
tion is a cliff 100 feet high. Sections parallel to the shore would show a series of
undulations, the crests being opposite to the main points of supply. Such fringes
* I. C. Russell, ‘ Glaciers of North America,’ 1897, p. 109.
PIEDMONTS- AFLOAT.
67
as that of Coulman Island are sometimes as much as 15 miles long, but are
rarely more than 2 miles broad. The snow encircles the rock-cliff up to heights
of 200 to 400 feet above sea-level, and the seaward edge is not often more than
70 feet above water. The distinction between “ piedmont-on -land,” and “piedmont-
aground” is to some extent hypothetical, for it is difficult to make sure that ice
at any particular point does not extend below sea-level (Fig. 56, p. 93).
On Sturge Island of the Balleny Group, a transition from “ piedmonts-
aground ” to “ piedmonts-afloat ” is also evident, for sometimes the undulating fringes
flatten out along definite lines parallel to the shore and extend at least 5 miles out
to sea, and so are probably partly afloat (Fig. 1, p. 3).
(c.) Pied monts-ajloat.
Piedmonts-afloat are represented by four important examples,' (l) the sheet of
ice which fills up Lady Newnes Bay, (2) the sheet at the foot of Mount Neumayer,
Drygalski Piedmont, (3) the sheet which extends from Cape Gauss eastwards for
20 miles at least, Nordenskiold Piedmont, and (4) the Ross Ice-sheet, or Great
Ice Bai’rier of Ross. All these are characterised by great extent with flat or slightly
undulating surface, and by a clitf-edge between 50 and 200 feet high, which has
enough water immediately in front of it to completely float the ice. The best
known of these floating piedmonts is the ice-mass which Ross in 1841 called the
Great Ice Barrier, but as this name entirely fails to convey the idea of vast extent,
we conclude not to adopt it as a general type, but prefer rather to class the Ice
Barrier of Ross and similar ice-masses as a subdivision of Russell’s term ‘ Piedmont.’ *
The Ross Piedmont has a seaward edge some 500 miles long, and its terminal edge
rises to an average height of 150 feet (Fig. 38). The depth of water close to this
ice-face is usually between 300 and 400 fathoms, and the sea-floor is covered with
fine rock-flour. If reference be made to the chart at the end of the volume,
details of heights and depths along it may readily be seen. If it be assumed
that aerated glacier-ice floats with at most six-sevenths of its volume immersed
in sea-water, f we may take it that the height in feet of the sea-cliff above is
equivalent to the depths in fathoms below, and hence that this ice-cliff must
be afloat. Further evidence that it floats is afforded by the tide-crack around
Mount Terror, White Island, and in many other places. The uniform horizontality
of the upper surface was proved by Captain Scott in his sledge- journey to
the south, and by Lieutenant Royds and Mr. Bernacchi in their trip for 155
miles to the south-east.
The intimate structure of Piedmont-ice shows that as far as water-level it consists
of normal glacier-ice. On the surface away from the land only fine snow was met
* I. C. Russell, ‘ Glaciers of North America,’ 1897, pp. 2 and 3.
t Heim, ‘ Handbuch der Gletscherkunde,’ 1885, p. 278, and H. Rink, 1 Danish Greenland,’ 1877, p. 358.
K 2
68
H. T. FERRAK.
with, but close to the shore crevasses and pressure-ridges show massive and vesicular
ice. The vesicular ice contains air * amounting on the average to about 8'5 per cent,
of its own volume, and the ice-grains are usually less than a quarter of an inch
across. Near White Island the grains are at least half an inch across and of a
very uniform size. This occurrence, however, is exceptional, for though no running
water was seen, the patches of bare ice are quite saturated with water, and it has
already been proved by Herr Em den f that growth of glacier-grains takes place
most rapidly at or near the freezing point. All observations made seem to show
Fig. 38. — The Ross Piedmont from the side of Mount Terror,
SHOWING THE CLIFF-EDGE AND FLAT Ul’I’ER SURFACE.
that the Ross piedmont is produced by ice-streams, not able to melt upon land, being
pushed out to unite in a shallow bay, after which the ice-mass floats off towards
the deep sea. It is remarkable that along the whole cliff-face from Cape Crozier
to Cape Colbeck no trace of foreign matter in the ice could be observed. At the
eastward end the land is completely buried in snow, but along the west side the
land is comparatively bare. Rock-rie/im was never met with more than a very few
miles from land. The chasm J that skirts the west side is in itself sufficient to
* Heim, ‘ Handbueh der Gletscherkunde,’ 1885, p. 113.
f Emden, Neue Denkschr. schweiz. nat. Ges. 1893, Bd. xxxiii, Abth. 1.
t Scott, Geog. Journ., April 1905, vol. xxv, p. 366, plate.
GLACIERS OF GREENLAND TYPE.
69
prevent boulders rolling on to the surface, but well-developed moraines on it are
seen where, on Minna Bluff and Black Island, the ice hugs the shore. These
moraines, however, will be discussed later ; they are mentioned here as evidence
that the ice-sheet does transport matter upon its surface. The movement of this
piedmont seems to be comparatively rapid. Where measured by Lieutenant Barne at
Minna Bluff, it was proved that a point moved through 608 yards * in thirteen and
a half months.
4. Glaciers of Greenland Type. (Plate III.)
Under this head will be included such ice-streams as flow from the Inland-ice.
In the Antarctic region this type of glacier is magnificently developed, and every
gradation, from streams 5 to 60 miles long, and 5 to 10 miles wide, is to
be seen.
The Prince Albert Mountains exhibit features so similar to those of the west
coast of Greenland that one description would suffice for both. Attention may,
however, be drawn to the fact that the Greenland ice-streams end in fiords and
come down between nunataks free from snow, whereas the ice-streams of the Prince
Albert Range, though they may, perhaps, lie in fiords, project as if towards the
edge of some coastal platform parallel to the mountains and well above the snow-
line. The nunataks are wholly encircled by snow-slopes which are rather higher
than the ice between them. The Ferrar Glacier is the only one that has been
entirely traversed. Situated as it is on one of the highest ranges of South Victoria
Land, it can hardly be considered quite typical, though steep inlets, which break
directly from the coast and back into the high mountains, are peculiar to the whole
resrion. The Ferrar Glacier has its source in the Inland- ice which lies at an
O
altitude of 7600 feet above sea-level to the west of the Royal Society Range. The
head of the glacier is an amphitheatre some 10 miles across, and is marked off by a
few small nunataks. The sudden rise to the Inland-ice is almost semicircular and
stretches round from Depot Nunatak to the North-west Nunataks, with a curve
concave to the east. This concave curve is marked by two parallel ice-falls, each
about 500 feet high. From the foot of this fall the ice moulds itself to the valley,
and between straight, parallel, and almost vertical, rock-walls flows off eastwards.
Near Finger Mountain the valley widens somewhat, the north wall continues its
straight course, but the south wall recedes irregularly to the base of Knob Head,
and in this way leaves the depression at the Solitary Rocks (D6) through which
the ice from Windy Gully and South Arm enters. The ice from South Arm splits
on a submerged water-shed and part flows first westwards and then northwards
into the North Fork, while the rest, considerably supplemented by the discharge
from the west of the Royal Society Range, fills up the East Fork, and eventually
floats in the narrow fiord between the Lower Kukri Hills and the Northern Foothills.
Scott, Geog. Journ., April 1905, vol. xxv, p. 363.
70
H. T. FERRAR.
At the same time, of the ice from the upper portion of the Ferrar Glacier, part
flows round the south side of Solitary Rocks and unites with that from Windy Gully
and South Arm ; part ends short of the col between the Solitary Rocks (D5 and DBa)
in a gradually attenuated tongue ; and the rest, which hugs the north side, probably
extends past D5a to be joined by that from the two tributaries from the south.
Captain Scott travelled down the North Fork, and tells me that the ice there ends in
an insignificant cliff some 12 feet high, leaving the lower portion of the valley strewn
with moraines and in part occupied by small frozen lakes.
At yY the height of the north wall is not more than 1000 feet above the ice,
but by about 10 miles farther eastward the valley-bed has fallen 1000 feet while
the adjoining mountains remain at about the same altitude. The valley deepens
continually ; near the Kukri Hills it has fallen nearly 5000 feet below the mountains
and is bare of ice. The south wall, west of Finger Mountain, averages 500 feet above
the ice, though the mountains are very much higher and culminate in the tabular
mass T.
Windy Gully and South Arm also probably come down from the Inland-ice,
which, about 20 miles further south, lies at an altitude of 7600 feet. The canon-like
form of the valley, therefore, is not so pronounced here as in the North Fork. Where
Cathedral Rocks face the Kukri Hill, the East Fork is 6 miles wide, and, having sides
4000 feet high, is remarkably canon-like. It is about 20 miles long ; further
eastwards it passes into the long fiord (nearly 15 miles long) which lies between
the Lower Kukri Hills and the Northern Foothills.
The surface of the ice is locally crevassed, and it is noticeable that the crevassed
areas, as on Mount Erebus, stand up above the general ice-surface. Crevassed areas
are found: (1) in the middle of the glacier, six miles north-east of Depot Nunatak ;
(2) close to the foot of Finger Mountain, where the valley-wall begins to retreat
southwards ; (3) close to b2, where the ice is forced sharply round into the Dry
Valleys ; (4) on the south side of Solitary Rocks, where the ice enters North Fork ;
and (5) in the middle of East Fork, opposite Cathedral Rocks and three miles from
them. Marginal ice-cliffs are a constant feature of the glacier, and moraines are rarely
entirely absent.
5. Glaciers of Norwegian Type.
The Norwegian type of glacier is well represented by the Blue Glacier referred
to in connection with the gneissic series of rocks. The Snow Valley, which lies
parallel to the base of Royal Society Range, has at one time fed five or six valley-
glaciers which flowed out eastward into Discovery Gulf. All except the Blue Glacier
have been broken across by diminution of ice-supply ; and the Blue Glacier, draining
a very extensive firnfield, is so nearly stagnant that, where measured, it had not
moved more than about three feet in the year.* Its length, measured from the
* Ferrar, Geog. Journ., April 1905, vol. xxv, p. 381.
GLACIERS OF NORWEGIAN AND ALPINE TYPE.
71
point wkei’e definite banks begin to be evident, is some twelve miles ; but, if measured
from the foot of the mountain-range, is about double that amount.
The surface of the glacier is snow-covered, and on the north side the snowdrifts
of the foothills quite blend with those of the glacier-ice. No definite Bergschrund
coukl be made out. For the last four or five miles the north side of this glacier adjoins
the land-piedmont described above ; but here again no evidence of movement was
seen. The south side of the glacier is bare ice, and below the hill Jt crevassed areas,
as in the ice of the Ferrar Glacier, stand in relief. Again, on the south side of the
glacier-snout, the ice stands off from the land and leaves the conspicuous channel
noted so frequently. The ice- wall adjoining this channel shows sections of enclosed
dirt-bands and moraine, and these are specially abundant in the lower part. As the
cliff forming the snout is very clean and free from rock -debris, such matter as is now
being carried must be close down upon the sole of the ice. The Koettlitz Glacier
may belong to this class, but little is known of its upper reaches, and it apparently
ends to the south of the Southern Foothills. At this point an ice- tongue from the
Koettlitz Glacier breaks into the lower part of one of the minor valleys of the
foothills, stagnates, and is no longer joined by the local glacier of the valley. The
latter still exists as an ice-slab higher up. The main glacier advances a little further
along its main valley and feeds the floating ice of Discovery Gulf (see Plate VI).
Two glaciers flow into Granite Harbour and join at the sea-edge, but as their
sources are unknown they also cannot be classified with certainty. They are peculiar
in that they too lie in narrow valleys which have very steep and straight sides, and
from a distance are very like the Ferrar Glacier. All these valleys lie approximately
at right angles to the main trend of coast and mountain-range, and seem to be a
characteristic structural feature of the region. Glaciers of similar type are quite
numerous among the snow-covered foothills of the Admiralty Range ; some of them
come down from valleys in the main range, while others arise in the snowfields of
the foothills themselves. Into Robertson Bay flow some ten or twelve great glaciers
which appear to drain the Inland-ice to the west of the mountains.* The Cape
North portion of the mountains is traversed by at least two of the trough-like
valleys carrying glaciers, but as the whole region is completely covered with snow
and ice it is not easy to distinguish the types of the glaciers.
6. Glaciers of Alpine Type : Valley-glaciers.
The most picturesque glacier of Alpine type is that of the deep and narrow
valley between the hills E2 and E3 of Cathedral Rocks. This glacier is partly
supplied with snow from the plateau south of Cathedral Rocks. It is about five
miles in length and one in breadth. It is crevassed from end to end. It joins the
Ferrar Glacier about 2500 feet above sea-level, and causes at least three transverse
buddings on the surface of the latter.
* Bernacchi, ‘The Antarctic Manual’ (Roy. Geogr. Soc.), 1901, p. 503.
A Dolerite Sheet.
72
H. T. FERRAll.
On the north wall of the North Fork, three glaciers drain out of one firnfield
and end about 1000 feet above the ice of the main valley (Fig. 39). Two of these are
cliff-glaciers, for the ice breaks off at the edge of a cliff and falls in avalanches
which are lost in the main glacier at the foot of the cliff. The third has lately
been a cliff-glacier, but its present loss by ablation exceeds the supply, and it now
ends some distance from the edge of the cliff and therefore is of Alpine type. The
south side of the Upper Kukri Hills has a numerous series of hanging valleys
distributed at regular intervals between D and D4. There are at least eight in a
The Obelisk C3
Fig. 39. — Three Ice-tongues falling into North Fork.
distance of 10 miles. Of the eight between D and D4, five have glaciers of Alpine
type which, keeping their continuity, fall as cascades into the main valley below, and
give rise to little or no disturbance in the latter.
7. Cliff-glaciers.
The three other glaciers between D and D4 are true Cliff-glaciers and, ending some
700 or 800 feet above the main glacier, discharge only as avalanches down the face of
the cliff. The width of these glaciers is usually less than a quarter of a mile. They
extend most of the way down the cliff, which is here about 4000 feet high. The
GLACIERS AND ICEBERGS.
73
hanging valleys all hang about 300 feet above the ice of the main valley, and
therefore at about 2000 feet above sea-level. In some of them the hanging lip is very
evident, while in others the Thalweg is very nearly uniform all the way.
8. Hanging Glaciers ; Corrie-glaciers (Fig. 20, p. 42).
Four Corrie-glaciers are worthy of mention. These lie on the south side of
the Inland Forts and occupy the cirques below the cols which link up the Forts.
Three of these glaciers are quite isolated, but the fourth is joined by a tributary
from the west side of Round Mountain (Cj). All flow southwards, but fail to reach
the ice of the main valley, and are now building up crescentic moraines at their
terminations. The interest of these glaciers lies in the fact that, though they now
flow southward, they were formerly forced northward by the Ferrar Glacier into
another drainage-system. Their supply is local from the Inland Forts, and the cols
at their heads are completely bare.
9. Ice-slabs (Plate VI).
Ice-slabs are found in all valleys on the east side of the Southern Foothills
of the Royal Society Range. These slabs are masses of ice, four to six square miles
in extent and about 50 feet thick. They are the relics of glaciers which once
drained the Snow Valley ; but, owing to diminution of ice-supply, this has now
become an inland basin, and its overflows have slipped away from it, leaving a
subsidiary watershed bare. The surface of the ice- slabs is quite clean, and free
from mud or stones. It is convex and slopes gently outwards from a centre. The
ice-cliffs which bound ice-slabs show abundant dirt-bands and scattered morainic
matter in their lower parts. In each valley in the Southern Foothills there is a
glacier of this type, and it would seem that their development is due to the peculiar
forms of the hills and their surroundings.
O
10. Icebergs (Fig. 40).
Icebergs have been defined by Heim as masses of glacier-ice floating in the
sea. They are common to both polar regions. The icebergs of South Victoria
Land are usually tabular in form, and the vast majority seem to be derived from
some common source. Shore-ice is not prolific in the formation of bergs, as such
ice remains firmly fixed to the land throughout the year. Blue Glacier was under
observation for over sixteen months, and during that time no berg broke away
from it, the snout remaining firmly frozen on to the sea-ice from the year 1902
to the year 1904. Within a mile of Blue Glacier, however, five bergs were aground,
and could only have been derived from it.
VOL. I.
L
74
II. T. FERRAR.
If no ice-stream moved faster than does the Ferrar Glacier the number of
bergs would be almost negligible. Doubtless the numerous ice-streams crossing
the Prince Albert Mountains must give rise to a certain number of bergs, but even
then the number is probably small. The piedmonts-on-land, lying on a flat shore
and receiving only a small supply of snow, can seldom provide enough ice to form
icebergs. Cliffs that encircle shore-ice hold snowdrifts which are sometimes as
much as GO feet thick. These always float out to sea during the summer, and
produce small bergs which may easily be mistaken for the broken-up masses of
Fig. 40. — An Iceberg, over 150 feet high, tilted through nearly 90°.
tabular bergs. Piedmonts-aground do appear to produce icebergs which have
irregular shapes and are produced in the manner of text-book icebergs.
On the Balleny Isles, -where the snowfall is much greater, the piedmonts creep
further out to sea and therefore supply larger bergs. It is, however, to floating
piedmonts that we must ascribe the great majority of Antarctic bergs, and when
we remember that the edge of the Ross Piedmont, whose ice is advancing north-
ward faster than any other ice in South Victoria Land, has lost a belt averao-inc
15 miles in width during the last sixty-five years, we see that it must have given
rise to innumerable bergs.
O
Piedmonts-afloat, from their configuration, can only supply tabular bergs, and
the sizes of these vary enormously. Although the Ross Piedmont rises over 200
ICEBERGS.
75
feet above sea-level, bergs over 150 feet high were seldom seen. Most bergs are
less than a quarter of a mile long and about 70 feet high. The largest bergs seen
were near King Edward VII Land, where there were many over 150 feet high ;
they had grounded in places where soundings showed 100 fathoms of watei Local
ice-caps supply few bergs. Mount Melbourne, for example, has its ice-cliffs
cavernous and overhanging, and we may conclude that years have elapsed since
bergs separated from them (Fig. 37, p. 65).
Distribution. — Icebergs float northward along the coast of South Victoria Land
bearing with them their burden of mud and stones, and soundings seem to show
that most of this is dropped within the Antarctic Circle. In latitude 67° S. the
sea-floor proved to consist of mud and ice-scratched stones, whereas only diatom-
ooze had been obtained in the deeper water of latitude 60° S. The distribution of
bergs in latitudes which are being constantly navigated is represented on the
Admiralty Ice-chart, No. 1241, and a short paper by Mr. H. C. Russell* gives
some measurements as to the sizes of the bergs. For our purpose, however, the
northerly migration along the coast is the point of interest ; the long string of
bergs grounded near Cape Adare seemed to have formed a banner-shoal ; melting
there, they must deposit the moraine brought by them from higher latitudes.
Icebergs are destroyed by two agents, the sea and the sun. Some bergs float
north into the warmer waters of more temperate latitudes and are there quickly
melted. As the berg is undermined by the warm sea-water it becomes top-heavy
and sometimes turns over, large pieces being broken oft’ in the process. Some
bergs ground in high latitudes and are only slowly dissolved. At certain stages
these may float off the shoal and go to swell the mass of drifting ice.
Bergs containing mud, sand or gravel absorb radiant heat, and some ice is
melted. The water produced distributes the mud over the surface and the rate of
destruction is increased; on December 7th, 1903, a hot day, a berg was seen to have
rivulets over all the sides turned towards the sun. A berg becoming inverted may
carry mud and stones from the sea-floor above water, and these the sun immediately
utilizes for the disintegration of the berg. In latitude 77° S., thawing of ice is of
little importance except in December and January. Melting begins in the middle
of November, becomes comparatively rapid by the middle of December, continues
through January, and virtually ceases about the middle of February.
L 2
H. C. Russell, Journ. Roy. Soc., New South Wales, 1898 (1897), vol. xxxi, p. 221.
7 G
Chapter X.
THE LAND-ICE — continued.
Englacial rock-debris may usually be seen near the terminal ends of glaciers. It
occurs in well-defined bands interlaminated with bands of almost pure ice. The
ice-walls which form the edge of a glacier also show rock-matter, and, as near the
Pig. 41. — Uplift of Morainic Material in the Ice at the foot of Knob Head.
snout, the rock-matter is more abundant in the lower layers. In the upper
reaches of the Ferrar Glacier, the ice-cliffs though 100 feet high show only
occasional small stones, but in the middle reaches, e.g., at the base of Knob Head
Mountain, boulders up to four feet across were observed low down in the ice-cliff.
These boulders are ice-scratched and sub-angular : they are mixed with numerous
small stones and some sand. At this spot also two streams of ice meet, and at
their junction the englacial matter is forced up 70 feet and appears as a normal
medial moraine (Figs. 41, 42).
ENGLACIAL ROCK -DEBRIS.
77
Below the hill D4 the chasm between ice and rock is only about 50 feet deep
and some 30 feet wide. The lower 20 feet of the ice-cliff is heavily charged with
rock -debris of all shapes and sizes. Bounded and apparently water-worn boulders
up to 12 feet across are exuded from the ice, together with fine sand and mud.
Fig. 42. — The Dark Band of Ice-without-grain, below Normal Glacier-ice,
AT THE FOOT OF KNOB Head.
The sun is wasting the cliff away so rapidly that a lateral moraine is appearin
below the general level of the glacier, and a small stream flowing along it is
O O' O O
carrying away the finer material which is being dropped into it by the rapidly
thawing ice.
O
‘ OQ
78
II. T. FERRAR.
On the south side of the snout of the Blue Glacier and 50 feet below the
upper surface, there occur a channel and ice-cliff, like those of the Ivarajak
ice-stream figured by Dr. Drygalski ; * along the channel, bands of mud and
stones are visible in the ice-cliff. In one of the ice-slabs of the Southern Foothills
sandy cnglacial matter occurs in much the same way as in the Ivory Glacier of
Spitzbergen.f
Suprciglacial matter is remarkably scarce. In this respect the glaciers of the
Antarctic region stand in contrast with those of Switzerland or New Zealand. The
latter are so much covered with angular rock-debris that no ice can be seen within
three miles of the actual snout. The lateral and medial moraines on the Ferrar
Glacier are not often as regular in distribution as moraines of other regions.
Sometimes they begin suddenly about five miles from the nearest rock-exposure,
and, after extending some
way down the valley, end
as suddenly as they began.
Again, though bare rocks,
such as Depot Nunatak,
shower talus upon the ice, the
moraine produced can only
be followed some two or
three miles down the valley.
In the Dry Valleys, the
moraines appear to be melt-
ing off and to be falling
back into the channels be-
tween ice and rock. A few
isolated moraines are also
scattered at random over
the surface of the invading
Ferrar Glacier.
The moraines which are brought down by South Arm are perhaps the most
striking in the region. Five of them are very prominent. One in the middle of this
tributary extends north past the base of Knob Head, then turns for a time north-
west, and finally curves round to the north-east, and on entering the North Fork
is lost. Two pairs of moraines, rather nearer the east side of the glacier, bend
round into East Fork and following the ins and outs of the valley, eventually find
their way on to the floating ice at its mouth (Plate III). This double pair, as seen
from Descent Pass, looks like the single pair of wheel-tracks of a waggon, but, as a matter
of experience, at least half a mile of bare ice separates one pair from the other.
* Drygalski, ‘ Griinland -Expedition,’ 1897, Bd. i, plate 14, p. 64.
t Garwood, Quart. Journ. Geol. Soc., 1898, vol. liv, plate 16.
Pig. 43. — Moraine on the Feerab Glacier.
Kukri Hills on the left, Granite-hills (Gs, G3) on the right.
MORAINES.
79
The moraines consist of three-foot boulders of dolerite, granite and sandstone,
which are accompanied by very little fine material (Fig. 43). The moraines are
30 feet wide, and the boulders are scattered thickly over the whole width. At the
base of Cathedral Rocks another moraine commences and accompanies the other
four. They produce parallel furrows iu the surface of the ice which increase in
depth as the boulders disappear. About 10 miles east of Cathedral Rocks these
moraines are still represented by the parallel depressions, but the boulders are few
and isolated, and numerous patches of coarse-grained ice mark the positions of those
which have sunk through. Out on the floating portion of the Ferrar Glacier the
boulders are even more scattered, and solitary survivors, about 100 yards apart,
are all that represent the moraines of the upper reaches.
It was among these moraines, at heights between 3000 and 4000 feet, that we
found numerous mummi-
fied carcases of the crab-
eating seal ( Lobodon car-
cinophagus). These are
interesting in that the
movements of seals ashore
are always slow and la-
boured ; how they could
have travelled 40 miles
uphill over rough ice and
soft snow is an unsolved
mystery. A pecten shell
was also met with in
gravel 10 miles up the
Ferrar Glacier and 20 feet
above the sea. The gravel
had formed a glacier-table
(Fig. 44), and the ice around was all glacier-ice, but is not above the reach of some
exceptional tidal wave.
On floating ice at the head of McMurdo Sound there are great quantities of
moraine (Figs. 45, 47) ; the latter completely covers the .ice and makes it difficult
to make sure that this great mass is really afloat. There is, however, the tide-
crack, which, following the land-boundary, marks off this debris- strewn area as a
stagnant but floating mass. Our observations seem to show that the ice is really
an overflow from the Ross Piedmont. In Discovery Gulf, also, the surface of the
floating ice carries much rock and some organic (512, 513) debris, and extends for
a distance of more than 20 miles from land. Between Black Island and Brown
Island the morainic matter is unworn, its stones being usually angular. The
moraines occur in long trains of cones which often rise 50 feet above the general
Fig. 44. — Glacier-table formed by a Layer of Gravel.
80
II. T. FERRAR.
level of the ice. Sometimes the cones blend with one another, and produce a series
of ridges whose direction follows the former direction of movement.
On rounding the north
end of Black Island, the
lines of cones curve west-
ward, and are further con-
tinued northward to the
“ pinnacled ice ” or old ice-
edge. Occasionally large
boulders up to four feet in
diameter are found, but these
disappear and are replaced
by great quantities of coarse
sand (261), which is often
blown about by wind. It
is this sand which, by
inducing melting, produces
the rivulets. These give
rise to the fantastic
“pinnacled ice” which presents so insuperable a difficulty to the sledge-traveller.
Re-sorted moraines were observed at Cape Adare ; the “ beach,’ from which so
varied an assortment of pebbles has been taken, is one mass of such moraines. The
average height of the beach is about 20 feet above the sea, but only 30 yards of the
northern fringe has recently
undergone modification by water.
In detail the beach consists of
parallel series of ridge-and-furrow
with amplitude of about four feet.
The ridges curve with the rock-
wall. Sometimes the fine material
appears stratified, but the cover-
ing of pebbles usually hides all
evidence of structure. The.ridges,
which are occupied by penguins,
flatten northward ; and the depres-
sions which contain stagnant water
sometimes join up and form
large digitating ponds. At an
elevation of more than 800 feet
on Cape Adare are other moraines. Fl°- 46,-Moraine-cone op Ice-scratched Stones, on which the
r _ Balanus Shells were found, on the floating Glacier-ice
These cross the peninsula to the IN THE BAY BETWEEN WHITE ISLAND AND BLACK ISLAND.
Fig. 45.— Moraines on Floating Ice at the Head of McMurdo Sound.
MORAINES.
81
north-east ; they consist of small stones, some ice-scratchecl boulders, and a few
blocks of rock up to 12 feet across. The beaches of Possession Island, Wood Bay
and Franklin Island, appear to be similar to the Cape Adare beach.
Stranded moraines also occur above Cape Crozier on the slopes of Mount Terror
at heights of 300 to 500 feet. Others on the south-east side are very striking.
They lie 800 feet above sea-level and are 200 or 300 feet above the level of the
Ross Piedmont. Other moraines occur on that shoulder of Mount Erebus which
terminates in Cape Royds. They occur up to a height of 1000 feet, and are very
well seen near the 800-foot contour. They cover some three square miles of
area. Granites, and rocks like the dolerites of the Royal Society Range, are
the most conspicuous components.
The moraines on the west side of
McMurdo Sound are developed on a
larger scale than any other moraines in
South Victoria Land. An area there, 5
miles by 3, is one mass of debris- cones,
some of them being as much as
100 feet high (Fig. 47). These cones
rest in some cases upon land, in other
cases upon fixed ice, and occur on a
flat which is not more than 4 feet
above the sea. The cones are more
or less in lines, and the lines appear
to radiate irregularly from two points,
one set of them converging near the
snout of Blue Glacier. Though the
cones rise considerably above the edge
of the land-piedmont described on
p. 66, they follow its eastward border.
On the south side of Blue Glacier these cones are replaced by a continuous line
of moraine, and this hugs the edge of the Southern Foothills for a distance of
some 30 miles.
Before leaving the subject we must briefly mention the ice which supports the
cones and occasionally protrudes through the covering of debris. In particular cases
it is often impossible to say what part of the material is ice and what part is
rock-debris, and hence no attempt has here been made to distinguish between
moraines still being carried by ice and moraines now resting upon the land.
Even on Cape Royds water oozing from some of the ridges showed that the
latter contained ice. During summer the fine material is continually being
separated from the coarse by the water from melted ice (Fig. 52, p. 89). In
some cases, however, the cloak of debris is too thick to allow the heat of the
Fig. 47. — Moraines supported by Ice, on the west side
of McMurdo Sound.
82
H. T. FERRAR.
summer suu to get through, and the ice beneath it may then be preserved
almost indefinitely.
The characteristic sheer ice-walls bounding the glaciers of South Victoria Land
show that under present conditions the sides are receding from the land. The interven-
ing channels often contain frozen ponds, which in some cases, though only 50 yards
broad, are more than a mile in length. The large pond at the base of Knob
Head may be contrasted with the Marjelen See, in that it follows the straight
side of the main valley instead of merely occupying the dammed-up end of a
tributary valley (Fig. 41, p. 76).
The structure of the ice between the bands of intra-glacial material at the
base of Knob Head shows remarkable variations. The uppermost 40 feet appears to
be quite normal vesicular glacier-ice and is free from rock -debris (Fig. 42, p. 77).
Below this are several notable dirt-bands, and among them other bands from
2 to 10 feet thick, perfectly clean, and clear as rock-crystal. On melting small
fragments from these bands no granular structure could be seen, and it is
suggested that they are, in part at least, due to intrusive thaw-water. Other
bands showed air-vesicles elongated at right angles to the banding. The ice
o o o o
which contains the majority of boulders has a structure comparable to that of
ordinary rock-fault breccias, and it would appear that the ice here glides forward
as a series of rigid sheets along parallel thrust-planes.
Up-thrust of morainic material similar to the up-thrust in Spitzbergen described by
Professor Garwood,* was also observed at Depot Nunatak, where Beacon Sandstone
boulders are brought up to the surface. Up-thrust was again in evidence behind the
Solitary Rocks on the Ferrar Glacier. Up-thrust produced by impact of two streams of
ice is further seen at the foot of Knob Head, where the dirt-bands with large boulders
bend up and appear on the surface 70 feet above their usual position (Fig. 41, p. 76).
Tee-movement. — Owing to the great distance which separated Winter Quarters
from any glacier, our observations on the rate of ice-movement have been few
in number. The rate at which the ice from South Arm forces its way into
East Fork of the Ferrar Glacier is probably less than six feet per month. Other
observations made at its snout indicate that the rate is extremely small. The Blue
Glacier moves less than four feet a year, while the Ross Piedmont, as measured by
Lieutenant Barne from the depot off Minna Bluff, moved no less than 608 yards
in 13 J months. The movement of Ferrar Glacier or Blue Glacier causes little
disturbance of the sea-ice ; slight movements are transmitted to the latter and
become lost in its more ordinary movement. Where the Ross Piedmont abuts
against Mount Terror, three parallel and well-defined ridges appear. These are at
least 50 miles long and usually some 50 feet high. They have been traced by
Lieutenant Royds towards the north end of White Island, but gradually flatten out
and fan. At Pram Point four lines of parallel hummocks, each about 15 feet high,
* Garwood and Gregory, Quart. Journ. Geol, Soc., 1898, vol. liv, p. 219.
MOVEMENT OF GLACIERS.
83
are caused by an overflow of Eoss Piedmont. Captain Scott found that the ice of
such channels as Mulock Inlet pushes the piedmont-ice away from the land and leaves
a chasm,* some 100 feet deep, in the intervals between them. At such outstanding
points as Minna Bluff, cracks and crevasses radiate outwards, particularly towards
the east and north-east ; but a sledge-party, by giving the land a wide berth,
was able to avoid most of these. Near the north end of White Island also, series
of radiating cracks are found. It would therefore seem probable that the Eoss
Piedmont is moving northwards bodily.
The ice-falls of Ferrar Glacier indicate movement, but, as the crevasses always
remain drifted up with snow, the rate must be exceedingly slow. In the channel at
the foot of Knob Head, where the evidence of up-thrust is recorded above, the banks
of the frozen ponds have several small ridges alongside and parallel to the glacier-
side. These ridges, which are occasionally broken along their length, indicate a
certain amount of movement ; as this is the only spot where rupture caused by
shearing movement is obvious, the fact is noteworthy.
Near the sea, where the ice-tongue floats in its valley, the tide-crack follows the
side of the glacier for a distance of at least 10 miles. Near the foot of the hill
G2 the crack trends towards the centre, and, gradually disappearing, is replaced by
other cracks which trend inwards up the valley. It would seem that the point of
replacement indicates the floating of the ice, and that the oblique cracks show a
slightly more rapid forward movement of the mass of ice behind.
In the amphitheatre or depression of the Ferrar Glacier, two miles from the foot
of Knob Head, the ice shows a network of ribbon-like cracks or fracture-lines ( liisse j).
These are often less than two inches apart, and, without opening more than a hair’s
breadth, extend for great distances. Parties camped on this ice have observed on
several occasions that very rapid splitting or bursting asunder takes place with
loud report, as soon as the hills cast their shadows on the ice. The reports
which accompany the splitting are loud and frequent, and often resemble the noise
of independent rifle-firing. The noises frequently continue for an hour and a half at
a time. Kupturing has also been observed at several other spots, and seems to be
caused by strain set up by changes of temperature in the ice. That the ice is in a
state of strain is proved by the fact that a blow from an iron-shod ski-stick has
produced cracks which have extended 50 yards across the surface of a mass of ice
not less than 100 feet thick.
Snow. — The usual accumulation of snow took place during violent blizzards
when the air became thick with fine snow-dust (Fig. 48). On a few occasions in the
summer, however, large flakes fell gently from a cloudy sky. Sometimes soft hail in
rounded pellets and soft woolly hexagonal snow-crystals descended from an overcast sky.
Occasionally also, during summer, hexagonal ice-crystals up to half an inch across fell
* Scott, Geog. Joum. April 1905, vol. xxv, p. 366, plate.
f Drygalski, ‘ Gronland-Expedition,’ 1897, Bd. i, p. 80 ; Heim, ‘ Handbucli der Gletscherkunde,’ 1885, p. 202.
M 2
84
11. T. FERRAR.
from a dear sky ; on the surface of the Ross Piedmont, they supplied much of
the superficial ice of the area. In the course of a day or two the crystal-plates
break up into grains, which drift hither and thither with the wind. After a
blizzard, soft snow usually becomes tightly packed, and the snow-dunes which have
been formed show a smooth and hard surface. Later this dune-snow granulates,
notwithstanding that the temperature remains constantly below 0° F. If a wind
springs up, the grains are carried away and the dunes disappear. Graduated pegs
were set up in the snow to determine the changes which take place in its sur-
face, and the observations show that during two years much snow drifted past
them. Wind carries the snow bodily away ; the importance of this factor in
reducing the height of the inland-ice of South Victoria Land will be appreciated if
we recall the six and a half days during which the sledge-parties were weather-
bound on the edge of the
inland - ice. During that
week, the air, which passed
at an average rate of 50
miles an hour, was so
charged with fine snow that
objects 10 yards away were
indistinguishable. That this
is not unusual may be in-
ferred from the fact that at
Winter Quarters the days
on which no silting, or
surface-drift, of the snow
took place were few.
The winds by carrying
snow on to the surface of
sea-ice help to drain the
land ; and the sea-ice, as it breaks up and floats north, takes away much superfluous
water-substance which has had no opportunity of glaciating the land (Fig. 49).
The snow-dunes usually took the form of crescents and symmetrical elongated
domes, never more than three feet high. The longer axes of the domes were parallel
to the direction of the prevailing winds, those of the crescents transverse. As soon
as their substance becomes granular, the winds remove and obliterate all trace of them.
During the process of destruction, snow-surfaces resemble a wind-worn surface
of false-bedded and slightly indurated sand. The less granulated layers are the
more indurated, and stand out beyond the coarser and less resisting bands, thus
giving the appearance of stratification. The forms assumed by the disintegrating
dunes are very variable,* and some become very fantastic. The silting snow helps
* Vaughan Cornish, Geog. Journ., August 1902, vol. xx, p. 137.
Fig. 48. — Undulating scepace of hard “ Marbled ” Snow.
SNOW.
85
the wind and behaves like a sand-blast, cutting away both the soft and the
hard layers.
No transformation from snow to glacier-ice could be observed. Present climatic
conditions are such that thawing, even partial thawing, only takes place very locally,
and all the surfaces encountered were either granular white snow or compact ice. Even
at the head of the Ferrar Glacier the change from snow to ice is absolutely sudden,
and along the base of the great cascades the ice presents its characteristic rippled
surface. Local accumulations of snow do occur in the larger depressions, but the line
separating granular snow from glacier-ice is always sharp. A few snow-dunes were also
seen, consisting of opaque white snow, too hard to be cut even with an iron spade.
In 1902 Lieut. Arm it AGE travelled up Ferrar Glacier over soft snow, and at one
of his camp-sites left pieces of spun
yarn, a tin and a piece of wood ;
they were found by our party
a year later and lay loose upon
hard transparent ice ; the tracks of
his men and sledges could have
been followed all the way up the
glacier. The sledge-tracks appear
as two parallel ridges, standing in
relief nearly an inch above the
general ice-level. The footprints
of the men also stood in relief,
but the dark objects left lying
about were not so raised. From
these facts it would seem that in
this locality loss by ablation exceeds
gain by precipitation.
A surface of white snow absorbs
little incident radiant heat. Owing;
to the low temperature of the air, the growth of the grain can therefore only take
place slowly. Iu sheltered spots or near bare rock, snow and ice melt rapidly during
summer, and even in the open long furrows filled with water * appear. The best
example of this was seen among the hummocks near Black Island, where long furrows
filled with fresh water separate rows of hummocks ( Hug el f) from one another.
Temperatures at fixed depths in the ice were determined during 1903, and the
observations show that the variations from day to day are surprisingly small. It
will suffice here to note that the highest temperature recorded at a depth of six feet
was — 9 C. and the lowest —24 '4° C. The change was gradual throughout the year.
* Drygalski, ‘ Gronland-Expedition,’ 1897, Bd. i, p. 78, plate.
t Drygalski, 1 Gronland-Expedition,’ 1897, Bd. i, p. 86, plate.
Fig. 49. — The two lower men are standing upon the
UPPER SURFACE OF SEA-ICE DEPRESSED BY SNOW BELOW
WATER-LEYEL.
86
n. T. FERRAR.
The minimum reading; was taken after mid-winter and the maximum occurred in
January, and hence a considerable lag in temperature is produced by the six feet of
ice. Temperatures at greater depths in the crevasses * show that there the lag in
temperature is even greater, and also that the maximum temperature reached by the
ice is far below the melting point. The following observation from a crevasse near
the junction of the ice of South-west Arm with that from inland is of interest : —
November 3rd, 1903, 7 p.m. Depth of crevasse 30 feet.
Temperature of the air +20° F. ( — 6 *7° CL).
Temperature of the ice —21° F. ( — 29 ‘4° G).
* Drygalski, 1 Gronland-Expedition,’ 1897, Bd. i, p. 450 ; Heim, 1 Handbuch der Gletsckerkunde,’ 1885,
p. 288 ; Nansen, Meteorological Report, 1894, 1895, 1896.
87
Chapter XI.
DENUDATION.
Wind-action. — The winds in South Victoria Land prove to be as strong and as
constant as any oceanic trade-wind. Around Winter Quarters the bare land-surfaces
are usually covered to a depth of six inches by a loose cloak of rook-debris. Below
this the earth is permanently frozen throughout the year, and here rock-surfaces due to
fracture often seem to remain quite unweathered. The layer of rock -debris consists of
a mixture of occasional boulders,
abundant small stones and rock-
chips, embedded in a matrix of
impalpable flour, and all would
seem to be rapidly disintegrating.
Many of the boulders seem to have
no very definite outer boundary,
and the protected surface may be
seen to pass gradually through a
state of crumbling (547) into im-
palpable powder (446). This cloak
is usually damp for a week or two
in summer, but: becomes dry and
loose when frozen during winter.
Here decomposition and disintegra-
tion proceed simultaneously, and
any particles loosened by frost from
the upper surfaces are at once blown
away by the wind ; the fine material
which remains is always an inch or so below the loose layer of stones at the top
of the deposit.
The loose stones are often smoothed and pitted (325) in the manner peculiar to
the wind-worn stones of desert regions, * and some of the harder ones have a superficial
glaze. Some of the boulders are too granular to receive polish ; gradually crumbling
away, they for a time leave patches of small angular fragments, still too large to be
transported by the wind, to mark the spots they once occupied.
The wind has carried away the smaller rock-fragments from the summits of
Observation Hill and Castle Bock. Those which remain are upwards of two inches
in diameter. The summit of the former, which is composed of trachytic lava, is
* Walther, Abhand. math.-phys. Cl. d. k. sachs. Ges. Wiss., 1891, Bd. xvi, p. 447. (Dreikaoter.)
Fig. 50. — Hollowed Granite-boulder with Incrustation of
Calcium Carbonate, near Descent Pass.
88
II. T. FERRAR.
honeycombed to a remarkable extent. The boss of trachyte above Cape Crozier, the
kenyte of Cape Royds and basalts of other areas, show similar wind-effects.
The Beacon Sandstone of the Royal Society Range, likewise, was almost free
from the fine disintegration -products, particular beds being often bare for lengths
of a mile or more. The loose quartz-grains derived from these seem only to
remain in crevices or below projecting rock-shelves. Dolerite-columus, too, are quite
smooth, and are coated with a bright chocolate-coloured crust (670) rarely more
than one-eighth of an inch thick.
Hollowed granite-boulders (Fig. 16, p. 34 ; Fig. 50) were observed at the foot
of Royal Society Range near Descent Pass, and two types may be distinguished.
(a) In fairly normal granite. The rock (555, 556) is a grey to pink granite
with felspars usually about a quarter
of an inch long ; it appears to be
quite fresh even on the surface, and
has a marked superficial glaze on
both convex and concave surfaces.
The most striking cavity is on the
south and weather-side of a large
block and therefore faces away from
the sun ; it is about eighteen inches
across at the opening, and the
diameter increases inwards to at
least two feet. The depth of the
cavity is a little more than a foot,
and the back wall is partially covered
with a hard mamillated or botryoidal
crust (554), consisting mainly of
calcium carbonate, the surface being
white, and harsh to the touch. The
crust was lamellar, scarcely more
than one-eighth of an inch thick, but the projecting botryoids, which are sometimes
partially hollow, may be more ; it was firmly fixed to the granite-face, so that it
was impossible to decide whether the surface beneath was or was not glazed.
(b) In a very coarse granite containing abundant large crystals of orthoclase.
The hollowed blocks (557, 558) are rounded, but the surface, owing to the rapid
disintegration, is roughened rather than glazed. The largest cavity is in a block
6 feet by 4 feet by 4 feet, which is hollowed almost to a shell. The cavity is
four feet long, three feet deep, and two feet high, and has four apertures varying
from a foot to eighteen inches in diameter, one on each side of the block. The lip
of the apertures is exceedingly sharp, the angle being certainly not greater than
30°. No incrustation was seen on the walls of this cavity, but on the floor is a
Fig. 51. — Saline Pond in Moraines on west side op
McMordo Sound.
DENUDATION BY WATER,
89
sprinkling of the finer disintegration-products of the granite which abundantly litter
the surrounding area.
These cavities in crystalline rocks apparently resemble the cavities in granite
observed by Mr. F. F. Tuckett * and Professor T. G. Bonney f in Corsica and by the
Rev. R. Baron J in Madagascar, but the incrustation of calcium carbonate shows that v
wind is not the only factor involved in then- formation. As in Corsica, many
saucer-like depressions and a few potholes were observed, and seem to mark stages
in the development of the completed cavities. Internal incrustations do not seem
to be recorded, but Mr. Baron mentions a “ white powder alkaline to the taste ”
as occurring in the hollowed blocks of Madagascar.
Water-action. — Water, as an agent of denudation in South Victoria Land, is
at present a factor of limited import-
ance (Figs. 51, 52, 53). On glaciers
it merely washes away the finer
material already thawed out of the
ice. On bare rock it seldom appears,
but along the south side of the Kukri
Hills and in other places a marked
water-channel occupies the marginal
ice-area, and in summer water flows
along the junction of ice with rock.
Water, therefore, may undercut rock-
cliffs and tend to widen and terrace
the sides of valleys. Any ice thawed
away by water is at once replaced by
the advance of fresh ice, a process
which tends to render permanent the
course of the water-channel. Actual
undercutting of rock-cliff was only
seen occasionally, but at the foot
of the hill D, along the Cathedral Rocks and along the foot of the granite hills
G3, was most evident.
During the summer, water everywhere distributes mud and sand over
the ice. Much of this mud is derived from the moraines which protect
ridges of ice, and the running thaw-water sorts sand from gravel and fine mud
from sand. In this way stratified and false-bedded sands and gravels may be
derived and appear among morainic accumulations (Fig. 52). Channels cut in the
floating glacier-ice are common at the head of McMurdo Sound, and during summer
Fig. 52. — Water separating Mud prom Gravel in the
Moraines on Minna Bluff.
vol. i.
* Tuckett, Geol. Mag., Dec. Y, 1904, vol. i, p. 12.
f Bonney, Geol. Mag., Dec. V, 1904, vol. i, p. 389.
t Baron, Geol. Mag., Dec. V, 1905, vol. ii, p. 17.
N
90
II. T. FEERAR.
the water flowing through them often spreads sand and mud over the surface of
the sea-ice.
The most notable effects of water-action in the area were seen on the north-east
side of Brown Island, an
island which retains practi-
cally no snow on its surface.
In January, 1902, a very
warm clear day followed a
summer snow - storm, and
caused a rapid melting of
the snow just deposited.
Rivulets becoming confluent
produced comparatively
large streams, which coursed
in straight and narrow
trenches down the hillside.
The slope here is very
steep, and trenches some-
times 20 feet deep had
been eroded. The coarser
material washed off the hillside spreads out as a delta on land, but much of
the finer material is carried further by the muddy stream, in and out among the
lines of moraines, and distributed over the surface of the floating glacier-ice. We
observed a stream increase
in depth from one foot to
three feet in the space of
a few hours. The swollen
stream cuts into the pro-
tected moraines and, over-
flowing the more level
areas, there deposits its
silt. Finally, the stream
coursing northwards into
McMurdo Sound passes off
the pinnacled or floating
glacier-ice* on to the sea-
ice which its finest sediment
then sullies (Figs. 53, 54).
A similar flood must
have occurred early in
* Ferrar, Geog. Joum., April 1905, vol. xxv, plate, p. 374.
J, Blue Glacier G
Fig. 54. — Seaward edge op the Glacier-ice floating in
McMurdo Sound.
Fig. 53. — Water-channel on Floating Ice in McMurdo Sound.
DENUDATION BY CHEMICAL ACTION.
91
December, 1903, for an area of sea-ice, six square miles in extent, was found with
an average of eighteen inches of muddy water upon it. Some of this water may
possibly have been produced in place, for this inundated area lay along the north edge
of the floating glacier-ice, and during the winter-gales must receive foreign matter.
Chemical action. — Chemical decomposition of rocks is more obvious in the dry
climate of South Victoria Land than in other areas, for rain can usually remove
soluble salts as fast as they form. On Hut Point all rock-fragments have thin
incrustations of sodium sulphate (398). The incrustation is sometimes so abundant
that the rocks look as if they have been dusted over with lime or flour ; if the loose
surface-matter be scraped away, thin discontinuous beds of the pure salt may be seen
dipping gently into the hill. The surfaces of many boulders in The Gap are covered
by a lace-like network of white lines (262, 263) consisting of calcium carbonate.
Near the north end of
White Island a great
quantity of perfect crystals
of sodium sulphate (298)
was obtained on a mound
of the floating glacier-ice
(Fig. 55). The percentage
of water in this salt, as
determined by Dr. Prior,
was 55 '86, which is virtu-
ally identical with that
characteristic of pure Mira-
bilite or Glauber Salt.
Near the isolated
moraines in the bay be-
tween White Island and
Black Island, on floating glacier-ice, there are five, or six mounds, two feet high
and up to five feet across, of the same white salt (623). The mounds are entirely
composed of the salt, which is in well-formed crystals, though the outer ones have
effloresced to some small extent. The moraines near these mounds contain halanus
shells (612) together with ice-scratched granite and other boulders (Fig. 46, p. 80).
In one of the moraine-cones on the west side of McMurdo Sound, a bed of
this salt (741), about eighteen inches thick, is traceable horizontally for about
ten yards. This bed is at least 50 feet above a pond of brackish wTater which
occurs at the foot of the moraine. Dr. E. A. Wilson also found this salt (740)
near the head of Discovery Gulf, and Mr. T. V. Hodgson (742) on Inaccessible
Island. As many of the ponds among the moraines are much too saline for
drinking, it is possible that this peculiar and abundant concentration of soluble
salt may be due to a former crystallization from similar ponds.
Fig. 55. — Fractured Dome in the Floating Glacier-ice, near the
spot where Sodium Sulphate Crystals were found, two miles
FROM THE NORTH END OF WHITE ISLAND.
92
II. T. FERRAR.
Frost-action. — Owing to the very low temperatures prevalent in high southern
latitudes, the denuding action of frost is not strikingly conspicuous. As a rule, the
wind removes all snow from bare rocks, and a marked line always divides the local
snow-fields from the areas free from snow. Thawing and freezing only occur
near the edge of snow-fields, and, therefore, owing to the general absence of water,
frost-action is once more rendered impotent. Castle Rock was perhaps the best
example of a frost-riven mass near Winter Quarters. It rises sheer above the
snow-covered peninsula ; but the side facing the sun (the north side) slopes steeply
down to shore-ice nearly 1000 feet below. Snow is drifted by the prevailing easterly
wind on to the north side, and in summer large riven blocks fall down and litter the
area below. The north side of Cathedral Rocks is similarly shattered ; from its
pinnacled outline it would appear to be more subject to frost-action than the isolated
peaks further to the west.
The dolerite, though forming no prominent talus-slopes, appears more prone to
split than other less-jointed rocks and a conspicuous ledge is always left at its
contact-junctions. Where dolerite occurs above sandstone, a terrace of the sandstone
stands out in front of the steep dolerite-cliff. Fans of dolerite- talus are very
conspicuous in the smaller of the Dry Valleys, also below the hill D, and along the
south side of the Kukri Hills.
No screes encroach upon the upper parts of the Ferrar Glacier, and the dolerite
usually rises perpendicularly from the ice. Along the ice-streams talus-fans are
rarely abundant enough to become continuous even at their base. The sandstone
undergoes little frost-action. Often its surfaces still retain the rounded outlines
which have been produced by ice-action. All transport of rock-material is now
accomplished by wind, which carries off the sand-grains as fast as they are loosened.
An important agent in wearing down the sandstone, and one which can hardly
be classed with any of the ordinary agents of denudation, may be included here.
The columns of dolerite, in falling down the cliffs, break away the softer sandstone-
beds and produce a sort of “ chimney talus-shoot,” which conveys the debris to the
fan at the bottom. Other fragments follow this line of descent, and thus the
deepening of the gully is accelerated. On the hill x several such gullies may be
seen. At a height of about 500 feet above the ice, the edges of the sandstone-beds
which have been caught up by the dolerite (see p. 46) are serrate, and at one
spot a groove or gully, 20 feet wide, of U-shaped section, has been produced. This
groove has perpendicular sides, and the uppermost bed of sandstone has been cut
back 20 feet from the edge of the cliff.
The granites of Antarctica, as of other regions, seem prone to form screes.
In the metamorphic limestone area, the hills are usually so rounded that there is
seldom an opportunity for a loosened block to change its position, and no transport
takes place until the rock is so finely disintegrated that it can be carried oft’ by
the wind.
DENUDATION BY ICE-ACTION.
93
Ice-action. — Adopting the same plan as before, we shall briefly review the general
action, as a geological agent, of each specified form of ice. The sea-ice, as already
pointed out, seldom runs aground, and is therefore negligible as an agent for striating
or abrading rocks or for contorting beach-deposits. Sea-ice forced up on to the land
has been observed at only one spot. This was the very exposed north-east corner
of the stranded moraines on the west side of McMurdo Sound, where sorting' and
rearrangement of the moraines is so constantly happening that no permanent effects
of sea-ice could be traced. As a transporting agent, sea-ice is not very effective.
Sometimes a boulder may roll across the fringe of shore-ice on to it and be taken
out to sea. The original boulder may be angular or it may be ice-scratched ; its
condition when on the ocean-floor can hardly be said to indicate its method of
transport. Dust and fine sand are often blown on to sea-ice and may then be
further transported. In
Wood Bay great numbers
of pumice-pebbles (899) had
been blown on to sea-ice,
which would be drifted far
to the North by the preva-
lent ocean-currents before
it melted.
The shore-ice has a
conservative * effect upon
the land. It binds together
the talus of the hills, and
so protects talus and rock
from the eroding action of
drifting ice-floes or waves.
When a piece of an ice-foot
floats out to sea it usually
carries a great load of debris. Stones roll on to the surface, and, through the
melting of the ice around, work downwards. Pockets of dust ( Kryokonit )f are
exceedingly numerous ; probably also much rock-material is held within the sole.
All these must be transported. As ice which has left the shores of South
Victoria Land seldom grounds, the abrading or striating action of shore-ice there
must be small.
The glaciers taken as a whole are not now modifying the form of the land to
any great extent. The corrie-glaciers and ice-slabs appear to be aggrading rather
than excavating the valleys in which they lie (Fig. 56). The corrie-glaciers at the
Inland Forts have a Beryschrund, but, judging from the small amount of terminal
* Bonney, Quart. Journ. Geol. Soc., 1902, vol. Iviii, p. 699 ; Bonney, Geog. Journ., 1893, vol. i, pp. 481-499.
t Drygalski, ‘ Gronland-Expedition,’ 1897, Bd. i, p. 94, ft.
Fig. 56. — A Glacier descending from the top of Coclman Island
into the Sea.
94
H. T. FERRAR.
moraine encircling them, no serious “ plucking ” of large rock-masses * can now
be taking place. This plucking does seem to be illustrated in the ice-free cirque
to the north of the Forts. There several blocks of Beacon Sandstone, as much
as 20 feet in diameter, have been extracted and transported about 50 yards ; the
sockets from which the boulders are derived are still very evident, and contain frozen
water-ponds.
The ice is everywhere retreating ; some few valleys are now quite bare and are
moraine-covered. No obvious ploughing effects of ice were seen, and roches moutonnees
are not by any means conspicuous. In Granite Harbour a few perched blocks and
a few ice-planed rock-surfaces were observed. This harbour is fiord-like, and has
depths of over 100 fathoms within a quarter of a mile of the shore. At Gj of the
Northern Foothills, close to the Blue Glacier, the metamorphic limestones are
beautifully rounded more than 1000 feet above the present ice-level and perched
blocks are everywhere abundant. Observations at the snouts of several glaciers seem
to show that no regular shedding of bergs is going on. Bergs from the Blue Glacier
would contain a great quantity of rock-matter, and in former times must have
transported an enormous quantity of debris.
Of the many icebergs met with, few showed rock-debris on their surfaces,
and, as piedmonts supply the vast majority of bergs, this freedom from rock-
material is not surprising. A few bergs showed angular rock-fragments on their
upper surfaces, and one or two had coloured dirt-bands interstratified with the
snow-layers. Icebergs aground often capsize and bring up material from the local
sea-floor ; this may be further transported, for through the melting of the
berg its draught diminishes. At the same time the rock-dour, which is a very
wide-spread deposit in the Ross Sea, is likely to be contorted by the moving
berg. On the whole, then, we may conclude that owing to the form of the coast
of South Victoria Land, rock-surfaces abraded or scratched by doating ice must
there be exceptional.
It would also seem probable that as the sea-ice diminishes during the summer,
so are the doating piedmonts now diminishing. The numerous soundings taken by the
‘ Discovery ’ along the edge of the Ross Piedmont, at places which at the time of the
voyage of Ross (1841) were beneath the ice, show that the sea-door is covered with a
stiff yellow clay (soundings 10-41), which contain tests of foraminifera, many diatom-
frustules and a few sponge-spicules. A somewhat similar clay (soundings 176, 177,
178) was found 10° further north near Balleny Island, also from 368 fathoms,
(sounding 13) oft' the Nordenskiold Piedmont. In water shallower than 100 fathoms
oceanic currents apparently remove the dne material and, as in other regions,
deposit it beyond the littoral zone. The whole of the door of the Ross Sea
seems to consist of rock-dour milled by the great glaciers of South Victoria
Land.
* Willard D. Johnson, Journal of Geology, 1904, vol. xii, p. 573.
95
APPENDIX TO THE REPORT ON THE FIELD-GEOLOGY.
NOTES RELATIVE TO MACQUARIE AND AUCKLAND ISLANDS,
OUTSIDE THE ANTARCTIC CIRCLE.
Macquarie Island (Fig. 57).
Macquarie Island is situated in the South Indian Oceau in latitude 54° 30' S.,
longitude 158° 30' E., and it has been suggested that it is part of the “zone of
Fig. 57. — The Strand and the Steep Coast-line op the east side op Macquarie Island.
fire ” * which connects New Zealand with Mount Erebus and Mount Terror. It is
about 23 miles long and five miles broad. It has an average height of 500 feet,
but one of its peaks rises to quite 2000 feet. The longer axis is in a north-east
and south-west direction and the south-east side is a precipitous cliff almost 200
feet high. The foreshore is narrow and the cliffs are remarkably straight ; they
* Judd, ‘ Volcanoes,’ 3rd edit., 1885, p. 230; Scrope, ‘ Volcanos,’ 2nd edit., 1862, p. 471 ; Bonney, ‘ Volcanoes,’
1899; map of distribution of Volcanoes at end of the volume.
96
IT. T. FERRAR.
extend for about 10 miles on either side of Lusitania Anchorage, where the
‘ Discovery ’ remained for about four hours.
The cliffs do not rise directly from the foreshore but from a slightly raised
platform or strand about 100 yards wide, and, in many places, more or less
covered with talus. In places the cliff is broken by steep gully-like water-courses
coming down from the plateau behind. The exposures in these gullies show that
the rocks all dip at about 10° to the north-west.
Small peat-bogs occur behind the harder rock-outcrops which hold back the
streams ; in the bogs round pebbles and sandy gravel occur. On the seaward side
of the hard outcrops a terrace at the level of the peat-bogs behind, at least 20 feet
above the level of the stream, extends some way towards the sea. The terrace,
sometimes 100 yards or more in breadth, consists of stratified clays, sands and
gravels. In plan it has the form of a delta which has been cut into by the
present stream and it may possibly be a raised beach. The rock-specimens obtained
are dolerites and basalts (see p. 109) which show little relationship to those from
South Victoria Land.
All the specimens are volcanic, many of them are slickensided, and others,
such as (9), have obviously been considerably crushed. The specimens (1) to (6)
inclusive were collected successively on our way up the gully. The specimen (11)
came from a height of about 1000 feet, from a bold crag overlooking the peat-
bogs.
A few dykes are seen crossing the strand. One of these (13) is 20 feet across,
and, with another (12), runs out to sea as a dangerous reef.
Auckland Islands (Fig. 58).
This group of islands was visited by Sir James Clarke Ross, and the specimens
collected have been described by Dr. Prior. *
The hills surrounding Ross Harbour rise over 1200 feet; they appear to
be built up of series of basaltic sheets, f but owing to the extreme density of
the vegetation few exposures could be found. These occur as “scars” or small
cliffs, over which streams sometimes plunge as waterfalls, but the scars are seldom
high enough to rise above the brushwood. The lowest rocks exposed along the
shore are all basalts, and are much more porphyritic than those (879 and 880) from
the summit of Mount Eden. All the basalts at sea-level are prominently columnar ;
the columns are about two feet in diameter and are sometimes, e.g. Deas Head,
300 feet high.
The eastern coast of the main island is a maze of fiords into which flow
the streams coming down from the western peaks. All the higher peaks lie
* Prior, Mineralogical Magazine, 1899, vol. xii, p. 71.
t Hector, Trans. New Zealand Inst., 1870 (1869), vol. ii, pp. 179-183.
AUCKLAND ISLANDS.
97
close to the western shore, where the land rises sheer from the sea as an enormous
cliff’. This cliff’ would seem to be undergoing rapid denudation by reason of the
prevailing westerly winds. The basalt-sheets dip slightly to the eastward, and at
sea-level occasional dykes (8 77, 892, 893) may be seen.
Another point of interest is the delta at the head of Laurie Harbour, the inner
land-locked part of Boss Harbour. This consists mainly of sand and shingle, but
has occasional shell-layers which in some cases are several feet above the high-tide
mark. The main stream of the Laurie Harbour valley now cuts into this and it
would therefore appear, that here, as in Macquarie Island, we have some evidence
of very recent elevation. Again, near Erebus Cove, a low spit of rock is covered
by about 3 feet of clay (874) and boulders. All the boulders are rounded and
Fig. 58. — The south side of Ross Harbour, Auckland Islands, showing Submerged Valleys.
water-worn they vary from two inches to a foot in diameter. The smaller boulders
occur in layers as if stratified, and the whole is overlain by a bed of peat.
Basalt-outpourings were found in Enderby Island, and in addition curious deposits
of clay and sand also occur. The clay seems to cap the basalts of the interior
and is easily distinguished by its covering of tussock-grass. The sand appears in
Sandy Bay as a bare hill edged with trees.
VOL. I.
98
SUMMARY.
Although the geological work of the ‘ Discovery ’ Expedition was confined to a
limited area, the collections of rocks and photographs which have been obtained
provide materials for forming some definite conclusions as to the geological history
of the region. The other expeditions, which lately entered the South Polar Regions,
worked in localities much more than 1000 miles distant from the ‘ Discovery ’ area
and from each other, and information obtained in one area may not hold for all.
Chapter 1 deals with most of the islands which occur at intervals along the
straight north -and-south coast of South Victoria Land, and also with various islands
lying between New Zealand and Cape Adare and within the Antarctic Circle. They •
arc bounded by inaccessible cliffs, and the surrounding sea is comparatively shallow.
Apparently they all consist of recent volcanic rocks.
Chapter II deals with the islands in the vicinity of Mount Erebus and our
Winter Quarters. This group I have spoken of as the Ross Archipelago, and of
the greatest of the group as Ross Island. This island has been built up by the
volcanoes Erebus and Terror, of which the former is still active ; only steam, never
any lava or solid matter, was seen to be emitted from the vent at its summit,
12,000 feet above the sea.
In Chapter III the relations of the conical volcanoes on the mainland are
considered. The conical volcanoes lie at the foot of a great wall-like range of
mountains, which in latitude 78° S. (the Royal Society Range) has a simple tabular
structure. This range is at least 800 miles long, trends due north -and-south, and
occasionally rises to peaks 13,000 feet high. On the east it ends abruptly in the
open Ross Sea, and on the west, for a distance of 200 miles at least, it forms a
great plateau some 7000 feet above sea-level.
In Chapters IV, V, VI and VII, the rocks which build up this great range
are considered in the order in which they occur in the field.
The gneisses have been found at sea-level and at the base of a series of rocks
quite f 2,000 feet thick, and may safely be regarded as forming the ancient platform
on which the central part of South Victoria Land is built.
The granites belong to two periods, one older and one younger than a certain
sheet of dolerite. The older granite lies upon the gneissic rocks at the foot of the
Cathedral Rocks, and dykes from the former ramify into the latter. A peculiarity
of this mass of granite is that it has a nearly horizontal upper surface which can
be traced for many miles along the sides of the Ferrar Glacier.
The Beacon Sandstone Formation is a deposit about 2000 feet thick and
remarkably uniform in texture. It proved to be quite barren of organic remains
SUMMARY.
99
except near the top, where, at a height of nearly 7000 feet above sea-level, fossil
plant-remains were found. Unfortunately, owing to decay of the plants and to
changes produced by a neighbouring sheet of dolerite, their characters are almost
indeterminate. This intrusive dolerite, though it gives no evidence of surface-flows,
forms the highest peaks of the Royal Society Range. The plateau- features are still
obvious, but the original plateaux seem to have been dissected prior to the earth-
movements which dislocated the sandstone.
Chapters VIII, IX and X, describe the ice as met with in the Ross Sea area.
The thickness, salinity and behaviour of the sea-ice, the shore-ice (ice-foot), and the
glaciers are described. The inland-ice, local ice-caps and piedmont-glaciers are con-
trasted with those that have been observed in the Arctic regions. Temperatures in the
ice at various fixed depths were determined ; they show that at these depths the
ice remains permanently some degrees below its melting point. Ice-slabs, or glaciers
which have slipped away at their heads owing to decrease in the supply of
water-substance, occur among the foothills of the Royal Society Range, and appear
to be of a type not yet observed elsewhere.
There is one fact on which most of the observers, in both the Arctic and Antarctic
regions, seem to agree, viz., the recession of the ice. Tyndall* foretold that the
ice would be tending towards a minimum when the condensation on both poles
was about equal ; whereas J. D. Whitney f maintains that only general glaciation
can occur when there are bi-polar ice-caps, and that we are now entering upon
a glacial epoch. The ice in the ‘ Discovery ’ area was found to be developed on
a comparatively small scale, the steep-sided valleys providing almost ideal rock-
exposures.
In Chapter XI the agents of denudation are discussed. The wind plays a
comparatively important part in this dry area, while the effects of water-action are
conspicuous by their absence. Chemical action is very pronounced in some localities,
while frost-action, owing to the small amount of precipitation, is almost quite absent.
The geological action of the ice is only briefly touched on.
In an Appendix are given some brief notes relative to Macquarie and Auckland
Islands, at which brief stays were made during the voyage.
It is my pleasant duty to express my thanks to the many kind friends who
have assisted me in my work. To Captain R. F. Scott, R.N., C.Y.O., D.Sc., and
the officers of the ‘ Discovery,’ my thanks are due for the interest taken in my work
* Tyndall, ‘Heat: a Mode of Motion,’ 1898, 11th edit., p. 231, and ‘The Forms of Water,’ 1892,
11th edit., p. 154.
f Whitney, ‘ The Climatic Changes of later Geological Times,’ Mem. Mus. Harvard Coll., 1882, vol. vii, p. 321.
100
SUMMARY.
and the ever ready help they accorded me ; all assisted me in collecting, and the
photographs taken by Engineer-Lieutenant R. W. Skelton, R.N., are invaluable; the
arrangements made for me by Captain Scott were all that I could have wished.
I am indebted to Dr. G. T. Prior for the names of the rocks and minerals
mentioned in the text.
Lastly, 1 am very grateful to Mr. W. G. Fearnsides, M.A., F.G.S., Fellow of
Sidney Sussex College, for kindly reading the whole of the manuscript and suggesting
many improvements in the text.
REPORT ON THE ROCK-SPECIMENS
COLLECTED DURING THE
‘ DISCOVERY ’ ANTARCTIC EXPEDITION, 1901-4.
By G. T. Prior, M.A., D.Sc., F.G.S.,
Assistant in the Mineral Department , British Museum.
Introductory.
The rock-specimens brought back from the Antarctic regions by the ‘ Discovery ’
Expedition are listed under about 1,000 numbers. Amongst them is much transported
material obtained from moraines, scree-slopes and soundings, but most of the specimens
were collected in situ by Mr. Ferrar, the geologist to the Expedition.
The specimens obtained from Cape Adare, Coulman Island, and Franklin Island
consist of hornblende-basalts and limburgites, similar to those which were brought
back by the ‘ Southern Cross ’ Expedition and were described in the Report relating
thereto.*
The following notes, therefore, refer mainly to the specimens from the Ross
Archipelago, and from the opposite mainland which was the scene of most of Mr.
Ferrar’s work in the field.
For purposes of description they will be taken in the following order, which
corresponds fairly closely with that adopted by Mr. Ferrar for the Report on the Field-
geology :
Chapter I. — Volcanic rocks.
These include the eruptive rocks of the Ross Archipelago, Scott Islands,
Auckland Islands, and Macquarie Island.
Chapter II. — The crystalline-limestones, gneisses and granites, which form the
basement-rocks of South Victoria Land.
Chapter III. — The lamprophyric and other dyke-rocks, intrusive in the basement-
rocks.
Chapter IV. — The Beacon sandstone and other sedimentary rocks.
Chapter V. — The dolerites intrusive in the Beacon sandstone.
Prior, Rep. ‘ Southern Cross ’ Collections (British Museum), 1902, pp. 321-332.
102
Chapter I.
VOLCANIC ROCKS.
Basalt appears to be the prevailing lava which has been erupted by the volcanoes
of this Antarctic region. The ‘ Southern Cross ’ collection showed that Cape Adare
and, in all probability, also the islands off the north coast of South Victoria Land,
are mainly composed of this rock ; but the presence, in that collection, of one or
two specimens of a phonolitic character was sufficient to suggest that in this Antarctic
region there is a similar association of basalts with more-acid rocks rich in alkalies to
that which prevails in East Africa and along the Atlantic volcanic chain generally.*
This suggestion is amply confirmed by the specimens brought back by the
‘ Discovery ’ from the Ross Archipelago. Basalts of a basic type are the prevailing
lavas, but accompanying them are phonolitic trachytes and kenytes (trachydolerites of
Rosenbusch) very rich in alkalies. To this latter type belong perhaps the most
remarkable specimens in the collection, viz., the rocks from the slopes of Mount Erebus,
showing conspicuous lenticular crystals of anorthoclase and exhibiting characters almost
precisely identical with those of the rhomb-porphyries of Norway and the more recent
kenytes of Mounts Kenya and Kilimandjaro in East Africa.
Basalts of the Ross Archipelago.
The basalts of the Ross Archipelago are very similar to those of Cape Adare. They
present the same two types, viz., hornblende-basalts with few and small phenocrysts,
and olivine-basalts with plentiful and fairly large porphyritic crystals of olivine and
augite.y
Hornblende-basalts.
The hornblende-basalts are dark gray, slightly vesicular rocks, showing only
sparingly small black porphyritic crystals of hornblende and augite. They are rather
more coarse-grained than the very compact rocks of Cape Adare.
Under the microscope small phenocrysts of pale purplish-brown augite and deep
reddish-brown basaltic hornblende (or more often magnetite-pseudomorphs after horn-
blende) are seen in a ground-mass of felspar-laths, magnetite-grains, purplish augite in
grains and needles, and small olivines.
The felspar-laths are mostly of labradorite, giving symmetrical extinctions in twin
lamellae of about 22°. Apatite is plentiful in the ground-mass and as inclusions in the
hornblende-pseudomorphs (see Fig. 59) ; it has often a pinkish tinge and is cloudy
with black inclusions arranged in lines parallel to the sides of hexagonal sections.
In most of these basalts the hornblende is only represented by pseudomorphs, but
* Prior, Rep. ‘ Southern Cross ’ Collections (British Museum), 1902, p. 328, and Mineralogieal Magazine,
1903, vol. xiii, p. 261.
t Prior, Rep. ‘ Southern Cross ’ Collections (British Museum), 1902, p. 326.
HORNBLENDE-BASALTS.
103
in a specimen (218)* from the top of the 1 300-fb. knoll of Mount Terror it is for the
most part unaltered. Most of the phenocrysts in this rock, however, show two stages
of growth, in which variations of chemical composition are indicated by differences of
optical characters ; a rounded and corroded nucleus of cossyrite-like hornblende, having
pleochroism from brown (for vibrations across the length) to black (for vibrations
along the length) is surrounded by the more usual barkevikite-like hornblende with less
absorption, from bright yellow to deep reddish-brown. The augites in this rock also
show similar variations in composition, and have generally a pale-green nucleus
surrounded by a pale-purple zone ; but in this case the change has been continuous, and
there is no evidence of corrosion of the first-formed green nucleus.
The hornblende-pseudomorphs in these rocks are similar to those which have
been often described.f They consist mainly of grains
and rods of magnetite, sometimes in radiating groups,
but generally arranged in lines roughly parallel to the
length of the crystal, with purplish augite and a little
felspar crystallised about them. With the magnetite is
usually associated a cossyrite-like hornblende, showing
pleochroism from deep brownish-red for vibrations
across the length of fibres to nearly opaque for §5
vibrations along the length. The pseudomorphs are
generally surrounded by a narrow, sharply defined
border of purplish augite like that of the ground-mass.
According to Becke’s theory f the alteration and
partial absorption of the hornblende probably took place during phases in the eruption
in which the pressure diminished much more rapidly than the temperature.
A quantitative chemical analysis of a hornblende-basalt from the Sulphur Cones
(385) gave the following result §
Fig. 59. — Pseudomorph after Horn-
blende in Basalt (691) from
Castle Rock, showing inclusions
of Apatite. (Magnification, 10 diam.)
SiO, =43-92
mol. ratios.
•727
TiO, = 4-19
•052
AL03 = 17-42
•170
Fe203 = 4-09
•02G
FeO = 8-83
•123
MuO = 0-09
CaO = 9-53
•170
MgO = 4-89
■121
Na20 = 4 -GO
•074
K20 = 2-17
•023
P,05 = 0-07
•005
H20 at 110° = 0-0G
H20 above 110° = O'll
100-57
* The numbers refer to Mr. Ferrar’s List of Specimens.
t For bibliography see Hyland, Tsehermak’s Min. Petr. Mitth., 1889, Bd. x, p. 238; also Washington, The
Volcanoes of the Ivula Basin in Lydia, New York, 1894.
t Beeke, Tschermak’s Min. Petr. Mitth., 1896-7, Bd. xvi, p. 335.
§ For discussion of the results of analyses see p. 119.
104
G. T. PRIOR
Hornblende-basalts were found on the 1300-ft. knoll on Mount Terror (218),
at the Sulphur Cones near Winter Quarters (382, 385, 391), and also on Brown Island
(G08). The latter specimen is almost precisely identical in microscopic characters with
the rocks from the Sulphur Cones.
Olivine-basalts.
To the basalts with porphyritic crystals of augite and olivine belong, apparently,
most of the more basic lavas of the Ross Archipelago. They are dark-gray to black
rocks, showing to the naked eye numerous crystals of these minerals. By increase of
glass and decrease of felspar in the base they pass into limburgite-like rocks, in some
of which the clear yellowish-green olivine-phcnocrysts reach a considerable size (up to
1 in. in length).
Under the microscope (Plate VIII. Fig. 1) the olivines are seen to be mostly of
very irregular outline and to be often deeply indented and corroded : they are clear,
colourless and fresh, showing generally only a few inclusions of magnetite. A marked
exception to this rule, however, is presented by the phenocrysts of a basalt from
Harbour Heights (366), which resemble closely the pseudomorphs after hornblende
described above (Fig. 59), since they are rendered almost opaque with magnetite
arranged in parallel lines. They may possibly mark a stage in the conversion of
hornblende into olivine.* In some of the more limburgite-like rocks ( e.g ., 176 from
Cape Crozier) the olivine shows small brown octahedral inclusions of picotite, and
also numerous vermiform inclusions similar to those in the augite and olivine of the
nodules described in the following section (see Fig. 60, p. 108).
The augite-phenocrysts are of the pale purplish-brown titaniferous variety common
to basaltic rocks. As in the hornblende-basalts, variations of composition are indicated
by changes in tint and by zonal structure exhibited between crossed nicols.
In the ground-mass occur olivines and augites of the same character as the
phenocrysts, together with magnetite and ilmenite in grains or rod-like skeletal
crystals, sharply defined felspar-laths, and, in many cases, brown glass.
The structure varies from pilotaxitic to hyalopilitic ; in some of the rocks fluidal
structure round the larger phenocrysts of olivine is well marked.
The felspar-laths are of labraclorite, giving symmetrical extinctions of about 25°.
Only rarely is a much-corroded small phenocryst of felspar seen in these rocks. In
some fine-grained basalts {e.g., 180, from the top of the 900-foot knoll at Cape Crozier,
and 311, from White Island) the phenocrysts are very sparingly distributed. In these
rocks olivine occurs very plentifully in the ground-mass, either as small rhombic sections
or (in specimen 180) as long prismatic crystals, not easily to be distinguished in ordinary
light from the lath-shaped felspars. In some specimens (222) augite-phenocrysts are
seen to have been formed round magnetite-pseudomorphs after hornblende. In others
(335) unaltered hornblende occurs as phenocrysts in addition to the augite and olivine.
* See Mtigge, Petrog. Untersuch. an Gest. v. tl. Azoren. Neues Jalirb. 1883 (ii), p. 224.
OLIVINE-BASALTS.
105
A basalt (431) from tlie foot of Castle Rock contains curious opaque white
inclusions up to 2 inches in diameter. They appear to be fragments of sandstone
caught up in the lava ; but if so they have suffered extreme metamorphism. The
basalt has permeated them in thin black vesicular glassy veins. The inclusions consist
of a hard, minutely vesicular, glass which is colourless except at the actual contact with
the basalt, where it becomes brown and nearly opaque. The glass incloses a few small
scattered irregular grains of quartz and felspar, and is crowded with minute needle-like
microlites, and, nearer to the basalt, with larger needles of colourless augite, such as
have been described in the case of sandstone metamorphosed by basalt.*
The results of chemical analyses of the olivine-basalt (656) from near the Gap, and
of a limburgite-like rock (326) from Ridge Road, near Winter Quarters, are as follows
under I and II respectively f
I.
IA.
II.
III.
IV.
(656)
mol. ratios.
(326)
(Graveneck)
(Hiirtlingen.)
Si02
—
42-14
•098
42*10
41-17
44-14
TiO,
=
4-90
•001
4-93
3-08
1-34
alo3
-
14-95
•140
14-87
13-24
13-87
Fe203
r
2-90
•018
3-20
3-50
11-73
FeO
=
9-71
•135
9-70
12-50
H-
CO
MuO
=
0-12
0 • 07
CaO
=
10-32
•184
10-03
10-24
10-80
MgO
=
9-47
•235
8-88
8-21
7-23
Na20
-
3-27
•053
3-20
2-57
3-25
k2o
=
1-80
•019
1-80
1-G0
1-54
PA
=
0-40
•003
0-58
0-53
0-80
H.,0 at 110°
—
0-12
0-11
H20 above 100°
=
0 ■ 10
0-12
3-21
1-87
S
=
0-09
C02
=
0-04
Ol
o
o
r-H
100-31
100-04
101-41
The two results are almost identical, showing that probably most of the limburgite-
like rocks only differ from the other olivine-basalts by their much more glassy base,
in which little felspar has been developed.
For comparison with these results are given under III an analysis by Senfter of
a “ hornblende-diabase ” of Devonian age from Graveneck, Nassau, which was described
by Streng ; J and under IV an analysis by Sommerlad of a Tertiary hornblende-
basalt from Hartlingen, Westerwald.§ The close similarity between I and III extends
even to the very high percentage of titanic acid.
Olivine-basalts were found on Mount Terror, at Cape Crozier (830, 222, 218) ;
and near Winter Quarters, at the Sulphur Cones (383) accompanying the hornblcnde-
* Dannenberg, Tschermak’s Min. Petr. Mitth., 1895, Bd. xiv, p. 53.
t For discussion of the analyses see p. 119.
t Streng, Ber. oberhess. Ges. f. Nat. u. Heilkunde, 22, 1883, p. 248.
§ Sommerlad Neues Jahrb., 1883, Beil.-bd. ii, p. 165.
von. i.
P
G. T. PRIOR.
106
basalts, Castle Rock (431, 319), Crater Hill (341), Harbour Heights (323, 325, 3GG),
aucl between the Gap and Horseshoe Bay (G5G).
Specimens of the more glassy limburgite-type were obtained from the same
localities, and also from Turtle Back Island (449) and Dailey Islands (510).
The basalt (553) from Cape Armitage has microscopic characters precisely similar
to those of the basalt (65G) from between the Gap and Horseshoe Bay, and is probably
the prolongation of that rock mentioned on p. 12 of the Report on the Field-geology.
The bombs found on Harbour Heights (3G7, etc.) consist of olivine-basalt like that
of the lavas.
The results of the analyses both of the hornblende-basalt and of the olivine-basalts
indicate that nepheline is present in these rocks, either in the interstices of the felspar-
laths or potentially in the glassy base. Generally it could not be recognised with
certainty under the microscope, but in a basalt (718) from the scree-slope below
Cathedral Rocks could be distinguished in the ground-mass small colourless patches
of isotropic analcite (?), and of doubly -refractive nepheline of which the refraction
was about the same as that of Canada balsam. This rock, which showed under the
microscope small phenoci’ysts of pale-purple augite and rounded olivines in a ground-
mass of felspar-laths, augite and magnetite, has a much closer relationship to the
recent basalts of the Ross Archipelago than to the dolerites of the Ferrar Glacier.
Analcite is probably present in some of the other basalts which show small amounts
of colourless isotropic material in the base, e.g., specimens from Cape Adar-e (49), from
Little Razor Back Island (471), and from the top of Castle Rock (319).
As a connecting link between the olivine-basalts with no porphyritic felspars and
the alkaline rocks (kenytes) with porphyritic anorthoclases, to be described in a later
Section (p. 110), are a few specimens of basalt showing conspicuous phenocrysts of felspar.
These come from Black Island (593), the debris-heap off Minna Bluff (619),
Inaccessible Island (805), Turtle Back Island (484, a boulder), and Cape Barne (814).
They show numerous phenocrysts of clear glassy felspar in a ground-mass of felspar-
laths, magnetite in grains and rods, pale-purplish augite and a little olivine. The
rocks resemble the kenytes, but the felspar-phenocrysts have not the characteristic
shape of the anorthoclases ; instead, they consist of an acid labradorite or andesine,
having a specific gravity of 2 ‘68, and showing an extinction on & (010) of about
16°. In the case of the boulder (484) from Turtle Back Island, these phenocrysts are
so flattened that the rock appears to be composed of alternating white and black
layers.
Coarse-grained Felspathic and other Nodules in the Basalts.
A striking feature in the basalts of the Ross Archipelago, especially in those of
limburgite-type, is the number of included coarse-grained nodules. These were found
plentifully, not only enclosed in basalts but also in loose lumps, in the neighbourhood
of Winter Quarters.
NODULES IN TIIE BASALTS.
107
Some of them are of the usual type common to many basalts, and consist mainly
of olivine and enstatite, with brilliant green chrome-diopside, like the nodules in the
limburgite of Franklin Island.*
Others, however, differ from most of those previously described in consisting to a
large extent of plagioclastic felspar, and thus have the character of gabbros rather than
of peridotites.f
Such gabbro-like nodules were found in the hornblende-basalt (385) from the
Sulphur Cones, and in the limburgites from the neighbourhood of Winter Quarters
(408, 316, 335, 415), as well as in the basalt (375) from Harbour Heights to which
reference has been made in the Report on the Field-geology (p. 13).
In the latter rock many of the inclusions, as stated by Mr. Ferrar, are quite
angular. In fact, some of them appear to be only loosely held in the basalt, like
fragments caught up in the molten mass. In most of the other specimens, however,
the nodules present the more usual rounded appearance, and are completely
enclosed by the basalt. In some cases (385) the junction is perfectly sharp, and the
rock shows under the microscope no variations in structure or composition in the
neighbourhood of the nodule. In other cases (335) the basalt has obviously permeated
the nodule in thin veins. This nodule (335) is composed mainly of pale-green augite,
enstatite and olivine, but contains also a little basaltic hornblende and biotite, and
some plagioclastic felspar in broad plates. Where the basalt has permeated the nodule,
the augite-plates have been converted into aggregates of small grains which bear some
resemblance to the bronzite-chondrules in meteorites. The hornblende in the nodule
may possibly have been derived from the basalt, which contains phenocrysts of similar
hornblende ; the biotite also may have resulted from the interaction of basalt and
nodule ; but the felspar can hardly be conceived to owe its origin to the basalt, since
the latter is of distinctly limburgite-type, showing under the microscope little or no
felspar. Moreover, the most felspathic nodules of those examined, viz., 385 and 316,
show no signs of having been attacked by the enclosing basalt.
These two felspathic nodules consist of granular aggregates of plates of labradorite,
giving symmetrical extinctions of 20°-30°, and irregular crystals and corroded patches
of nearly colourless augite having a roughly parallel arrangement ; a little olivine is
alsQ present in specimen 385. Under the microscope the pale yellowish-brown augite
has a rough, dirty appearance due to numerous inclusions. These consist for the most
part of gas-bubbles, glass and opacite, but some fantastically-shaped ones are probably
liquid inclusions. One crystal showed long black needles arranged in parallel lines,
like those observed by Dannenberg in augite-nodules in the basalts of the Rhine. |
The felspars in these nodules also enclose numerous gas-bubbles.
* Prior, Kep. ‘ Southern Cross ’ Collections (British Museum), 1902, p. 328.
t Felspathic nodules in the basalts of the Siebengebirge have been described, see Dannenberg, Tschermak’s
Min. Petr. Mitth., 1895, Bd. xiv, p. 35; Laspeyres. Verh. naturh. Ver. pr. Bheinl., 1900, Jahrg. lvii, p. 194; and
Zirkel, Abhand. math.-phys. Cl. d. konigl. sachs, Ges. d. Wiss,, 1903, Bd. xxviii, p. 165.
% Dannenberg, l.c. p. 40,
p 2
108
G. T. PRIOR.
A nodule (342) found near the ship, and consisting of pale-green diopside, grains
of clear olivine and plates of labradorite, showed a few dull-green isotropic grains of
picotite. In another somewhat similar nodule (478), from below Castle Rock, occurs
a little hypersthene, showing straight extinction and pleochroism from pale-pink to
pale-green.
The nodule (415), which is represented in section in Plate VIII, Fig. 2, was enclosed
in a hornblende-limburgite from between the Gap and the ship. It consists of a granular
aggregate of labradorite and pale-green diopside, with smaller grains of altered olivine
and a little basaltic hornblende. The labradorite shows symmetrical extinctions of
23°-26°, and encloses numerous gas-bubbles. The olivine grains are red with oxide of
iron, which is often arranged in bundles of wavy threads crossing each other at right
angles ; probably these threads were originally of magnetite, like the inclusions in the
olivine-nodules described below.
Fig. GO. — Inclusions in Augite op Nodule Fig. 61. — Olivine with Magnetite-
from Turtle Back Island. inclusions, prom Castle Rock.
(Magnification, 100 diam.) (Magnification, GO diam.)
Of the non-felspathic nodules, one (180, from the 900-ft. knoll on Mount Terror)
consists of olivine, enstatite, and purple titaniferous augite ; another (438, from Harbour
Heights) almost entirely of clear, colourless olivine and enstatite ; while others are
composed wholly of olivine impregnated with magnetite. A nodule (or possibly a
bomb) from the south slope of Turtle Back Island consists mainly of smoky-black
augite, and shows only a little yellow olivine in bands. Under the microscope the
augite shows some pleochroism from pale smoky-purple to brownish yellow, and has
numerous gas and vermiform liquid (?) inclusions arranged in lines ; in one crystal a
halo of fantastically-shaped inclusions surrounds an included dark rounded grain
(see Fig. GO).
Some of the nodules in the basalt near Winter Quarters (375 and 437) are of
dark ironstone-like appearance. They were found to consist wholly of olivine-grains
densely impregnated with magnetite, and thus resemble the olivine-phenocrysts in the
basalt (3GG) from Harbour Heights (see p. 104). Under the microscope one of these
BASALT-TUFFS.
109
grains showed an obtuse negative bisectrix, being cut parallel to 010. In this case the
magnetite is seen to be arranged in lines parallel to the (100) cleavage, and also in
more wavy lines approximately at right angles to the first ; along the latter lines the
magnetite occurs in blobs, which are flattened out at right angles along the more
definite cleavage (see Fig. 61). This olivine with included magnetite is very similar to
that figured by Tschermak in the Sierra de Chaco meteorite.*
The origin of the so-called “ olivine-nodules ” in basalts has been the subject of
much discussion.f The presence in these Antarctic rocks of felspathic gabbro-like
nodules as well as those having the composition of peridotites is of interest, since the
general absence of felspar in “ olivine-nodules ” has been used as an argument against
the theory that they are intratelluric separations from the magma.
Basalts and Dolerites from Auckland Islands and Macquarie Island.
The specimens brought back from Auckland Islands are olivine-basalts similar
to those collected by the Ross Expedition and described by the present writer in
Mineralogical Mag. 1899, vol. xii., pp. 70-73. Those with conspicuous phenocrysts of
olivine and augite resemble the olivine-basalts of Cape Adare and the Ross Archipelago.
The dyke-rock (893) from Williamson Point shows under the microscope large and
much-corroded crystals of colourless olivine and pleochroic (purple to yellow) titani-
ferous augite in a base of felspar-laths, purple augite, and magnetite in grains and
skeleton-crystals.
The basalts from Macquarie Island are of a somewhat different type. They are much
more altered than the rocks of the Auckland Islands, and appear to be of greater age.
The more coarse-grained rocks (l, 5, 6) are diabasic in character, and consist of a
medium-grained aggregate of felspar-laths, colourless augite (sub-ophitic), large
magnetite-grains, sparingly distributed, and interstitial green chloritic and hornblendic
alteration -products. The rock (4) from “ 100 yards up the stream” (see p. 96) shows
large phenocrysts of labradorite and a few chloritic pseudomorphs after olivine. The
crushed rock (9) is a much-altered andesitic basalt showing large phenocrysts of labra-
dorite in a very fine-grained altered base.
Basalt-tuffs.
Fragmental basaltic rocks were found in most of the localities on Ross Island
where bare rocks are exposed, viz., Cape Crozier, V-Cliffs Hogsback, Sultan’s Head,
Castle Rock, The Turk’s Head and Hutton Clifts.
The pale-yellow tuff from Sultan’s Head contains numerous black fragments of
vesicular basalt which show under the microscope a few sharply defined felspar-laths
and grains of olivine in a glassy base dense with magnetite. The tuff, however,
* Tschermak, Mikr. Beschaff. Meteoriten, 1885, Lief. Ill, FI. XXIII, Fig. 4.
t See Zirkel, Lehrbuch d. Petrographie, 1894, Bd. ii, p. 931 ; and Abhand. math.-phys. Cl. d. kgl. sacks.
Ges. d. Wiss. 1903, Bd. xxviii, p. 103.
110
G. T. PRIOR.
consists mainly of fragments of vesicular basalt-glass imbedded in a base of angular
grains of orange-yellow palagonite with a little magnetite and small felspars ; in parts
this framnental material is cemented with calcite. In the fraernents of basalt-e;lass,
all of which have been converted into palagonite, magnetite has separated in fairly
distinct crystals, and the felspar- laths and olivine-grains which the glass contains are
fresh and unaltered.
The tuff from Castle Rock is of similar character ; in the fragments of basalt-
glass, however, much purplish augite, besides olivine and felspar, has been developed ;
the felspar also occurs in thin, broad plates as well as in laths. In the basalt-glass of
the tuff from Cape Crozier, on the other hand, no felspar has been developed, but it
shows sharply defined crystals of olivine and purplish augite.
The tuffs from the “ Bare Rocks ” at Hutton Cliffs in Erebus Bay differ remarkably
from the others in appearance and general character. They are of a dull-green colour,
Fig. 62. — Kenyte with Poephyeitic Anoethoclase, booldeb from Tuetle Back Island.
(Natural Size.)
and are much more compact, so much so that it is not easy with the unaided eye to
recognise their fragmental nature. They have the appearance of greater age than the
other tuffs, and show under the microscope that they have been much altered. They
consist mainly, like the other tuffs, of a very vesicular basalt-glass, but in this case the
glass has been converted into a dull-green chloritic or serpentinous alteration -product
instead of into yellow palagonite, and most of the crystals and mierolites of augite and
olivine have been obliterated ; only rarely is seen a fragment of vesicular basalt-glass
in which small prismatic purplish augites can be recognised. The tuff contains larger
fragments of a less glassy non-vesicular basalt, showing under the microscope lath-
shaped felspars and magnetite in skeleton-crystals, with green altered ferromagnesian
minerals.
Kenytes (Trachydolerites, Alkaline-basalts, Rhomb-porphyries).
From the slopes of Mount Erebus and the islands in Erebus Bay come the
remarkable rocks mentioned on p. 102. These are distinguished from the basaltic
KENYTES OF MOUNT EREBUS.
Ill
rocks of Mount Terror and Winter Quarters by their numerous and large (2-3 cm. in
length) porphyritic lozenge-shaped crystals of anorthoclase, resembling those of the
well-known rhomb-porphyries of Norway (see Fig. 62). They are almost precisely
identical in characters and chemical composition with the kenytes of Mount Kenya,*
and the description recently published by Dr. Finckh of the rhomb-porphyries
(kenytes) of Kilimandjaro f could be followed almost word for word in an account of
these Antarctic rocks. They are alkaline basalts or trachydolerites (Rosenbusch),
intermediate in type between ordinary basalts and phonolites.
In colour they vary from dark-gray, to nearly black in the more glassy varieties.
Under the microscope they show, besides the large phenocrysts of anorthoclase, small
rounded olivines and pale-gray or brown to pale-purple augites sparingly distributed.
In different specimens the ground-mass varies in texture, and to some extent
also in mineral composition. In some (812
from the Skuary) it is quite holocrystalline, and
consists of a trachytic mesh of interlacing fel-
spar-laths (mainly anorthoclase) with interstitial
shreds and small prismatic crystals of a pale-
green augite and grains of magnetite ; in
others (464 from Tent Island) it is a brown
glass, dense with magnetite in grains or rod-like
skeleton-crystals, but showing in clearer pale-
brown streaks minute microlites of felspar and
augite. In the boulder 447 from Turtle Back
Island, the glassy base is confusedly spherulitic
with minute magnetite - grains arranged in
radiating wavy lines. In Fig. 63 is represented
a spherule with the black cross as seen between
crossed nicols. The glass in the base of specimen 541 from the slope of Mount
Erebus is almost colourless, but in parts is rendered nearly opaque with fine
dusty magnetite ; in the clear brown glass, however, which is included in the
anorthoclase-phenocrysts of the same rock there has been no separation of magnetite
except round the edges, from which project long, colourless, needle-like microlites
of augite.
O
Fig. 63. — Spherule with Magnetite, in
Glassy Kenyte, prom Turtle Back Island.
(Magnification, 200 diam.)
The rocks from Cape Royds (818, 820) present a distinct variety characterised by
the presence of leucite. They show large phenocrysts of anorthoclase and, very
sparingly, small rounded olivines, like those in the other kenytes, but contain in the
ground-mass fairly numerous, small, rounded, isotropic crystals of leucite having-
characteristic central and marginal inclusions of the magnetite and augite of the base (see
Plate VIII, Fig. 3). These rocks are therefore similar to the leucite-rhomb-porphyries
* Gregory, Quart. Journ. Geol. Soc., 1900, vol. Ivi, p. 205.
f Finckh, Festschr. z. siebzigsten Geburtstage von Harry Rosenbusch, 1906, pp. 373-397.
G. T. TRIOR.
112
of Kilimandjaro described by Finckli. * The leucites show a refraction markedly
less than that of Canada balsam. In most of the small crystals the central inclusion
of base occupies the greater part ; in many also the outer edge is invaded by the
minute felspar-laths which generally form a fringe round the crystal. That the leucitic
material in these rocks is only in small amount is indicated by the result of the bulk-
analysis (see p. 113), and more especially by the partial analysis of the part soluble
in nitric acid, the result of which is as follows, under I : —
(Nephelino.)
Si02
= 43 • 12
43-74
Al,Oa, Fe.,0,
=- 33-32
34-48
CaO
= 2-G3
—
Na20
= 14-96
1G-G2
K.,0
= 3-73
4-55
Cl
-- 0-70
—
H20 etc. (diff.)
= (1-54)
0-8G
100-00
100-25
The rock-powder gelatinised readily with acids, and as much as 25% was decom- .
posed by half an hour’s digestion with dilute (1:5) nitric acid. Under II is given, for
comparison, the result of an analysis by Clarke of a nepkeliue from Litchfield, Maine.
The result of the above analysis suggests, therefore, that the soluble portion of the
rock consists mainly of nepheline.
An attempt was made to separate the leucites in the rock-powder by means of
heavy liquids, but the grains which floated in a liquid in which moonstone sank and
leucite floated were in very small amount, and of these only a few remained dark
between crossed nicols.
The anortkoclase-phenocrysts in all these rocks have the characteristic lozenge-
shape determined by the development of the faces M (llO), m(110), and y (201).
They almost invariably show minute twin-striations, occasionally in two rectangular
directions according to the albite- and pericline-laws. Like the similar phenocrysts in
the kenytes of Mount Kenya, they contain numerous inclusions of glassy base, and also
occasionally inclusions of olivine, augite, apatite, and magnetite or ilmenite. On this
account no chemical analysis was attempted. In physical characters they agree
perfectly with the anorthoclases of Pantelleria and Kilimandjaro. The specific gravity
is 2-G2. Cleavage-flakes parallel to b (010) show a positive bisectrix, slightly inclined,
and give an extinction-angle with the trace of c(00l) of about G°, while cleavage-flakes
parallel to c (001) have extinctions of about 1°. The refraction was about the same as
that of clove-oil (1-538). Wolff (quoted by Finckh) found the value of yna for the
anorthoclase of Kilimandjaro to be 1 • 537G.
Nepheline was not found as phenocrysts in these rocks, neither could it be with
certainty detected in the base, but that it is present is indicated by the results of the
bulk-analysis.
* Finckli, l.c., p. 382.
KENYTES. — PHONO L1TIC TRACHYTES.
113
The olivine-phenocrysts occur very sparingly, and are generally less than 1 mm.
in diameter ; they are clear and colourless, and often contain inclusions of apatite and
magnetite. They are generally of irregular and sometimes of perfectly round outline :
oidy occasionally is one seen with distinct crystalline shape. Intergrowths of augite
and olivine occur, and in one case augite was seen to be enclosed in olivine.
Augite-phenocrysts occur even more sparingly than the olivines. They are
generally without distinct crystalline outline, and consist of a pale-gray or brown
augite having angles of extinction of over 40°. In a specimen (837) of glassy kenyte
from a moraine at the “mouth of 2nd Alpine Valley,” in Discovery Gulf, the augite-
phenocrysts are in larger amount, and attain a length of 2 to 3 mm. : usually they are
less than 1 mm. in length. The augite of the ground-mass occurs generally in small
prisms of a pale-gray to green colour, but in some specimens this is replaced to some
extent by grass-green segirine-augite.
Apatite is fairly plentiful in these rocks ; often it is of a pink colour, and dense
with dark inclusions arranged in lines parallel to the length of the needles.
A chemical analysis of the leucite-kenyte (818) from Cape Royds gave the
following result under I*, as compared with that of a kenyte from Mount Kenya
under Ilf, a leucite-kenyte (“ leucite-rhomb-porphyry ”) from Kibo (Kilimandjaro)
under II If, and a nepheline rhomb-porphyry from Vasvik, Norway, under IV:—
I.
Ia.
II.
III.
IV.
(Cape Royds.)
Mol. ratios.
(Mt. Kenya
•)
(Kibo.)
(Vasvik.)
SiO.,
56 • 09
•929
53-98
53-44
56-04
TiO.,
=
1-23
•015
0-57
0 1 69
0-65
A1203
=
20'7i)
•203
19-43
20-39
21-50
Fe203
=
1-54
*010
4-39
4-22
1 -06
FeO
=
3-84
•054
2-05
1-76
3-28
MnO
=
0-05
0-26
trace
CaO
=
3-18
•057
2-04
2-13
2*42
MgO
=
1-26
•031
1-07
1-12
1-12
Na.,0
=
7-33
•118
8-81
8-70
8-39
K.,0
=
3-91
•041
5-27
5-75
5 " 03
PA
=
0-38
•003
0-30
0-49
01
=
0-17
H.,0 at 110°
=
0-19
0*13)
0-97
0-67
11. ,0 above 110°
=
0-39
1-663
SO,
0-22
Zr02
0-27
100-35
99-96
100-21
100-16
Phonolitic
Trachytes
(Trachydolerites)
AND
PlIONOLITES.
Most of the rocks which are here included under the name of trachyte are of a
somewhat basic type, and so closely related in chemical composition to the preceding
alkaline basalts or kenytes that some hesitation is felt in separating them ; with the
kenytes they might all be classed under the trachydolerite group of Rosenbusch.
These trachytes are pale ash-gray compact rocks ; but weathered specimens (in
* For discussion of analyses, see p. 119.
t Prior, Mineralogical Magazine, 1903, vol. xiii, p. 247.
| Finclsh, l.c., p. 392. q
vol. i.
G. T. PRIOR.
114
which the augite of the ground-mass, together, apparently, with some of the magnetite,
has been altered, with development of orange-yellow epidote) are of a pale-yellow or
pink colour. They differ from the kenytes in showing either no phcnocrysts of
anorthoclase or only few and small ones.
Those which approximate most closely to the kenytes show under the microscope
an occasional small rounded phenocryst of olivine and augite in a ground-mass consisting
of a mesh of felspar-laths, with interstitial irregular grains and shreds of dull-green
augite, needles of pleochroic (yellow to dull-green) gegirine-augite, and only a few
magnetite grains. The felspar-laths are not sharply defined, and are often associated
with more platy felspars; they have a refraction not greater than clove oil ( 1 • 538),
and consist mainly of anorthoclase. That nepheline is present in these rocks is
probable, although it cannot with certainty be recognised under the microscope.
To this type belong specimens 804, from Inaccessible Island, and 473, from Little
Razor Back Island.
The specimens collected by the * Morning ’ from Scott Island and a trachyte (188)
from the top of the 900-ft. knoll at Cape Crozier are of similar character, but show
no porphyritic olivines and augites in the thin slices examined. The rock from Cape
Crozier is of a lighter-gray colour and of more salic character than the specimens from
Scott Island ; it shows the peculiar surface-shimmer, due to the parallel arrangement of
the platy felspars, which is so characteristic a feature of many of the alkaline rocks
of Norway (solvsbergites) described by Brdgger.*
A coarse-grained inclusion in this trachyte (188) consists of an aggregate of stout
felspar-prisms, with some pleochroic (yellow to grass-green) gegirine-augite and a little
cossyrite-like hornblende showing pleocliroism from reddish-brown for vibrations across
the length to black for those along the length ; the felspars are partly oligoclase with
symmetrical extinctions of about 8°, and partly anorthoclase.
The results of chemical analyses of the rock from Scott Island, and of the trachyte
from -Cape Crozier (188) are given under I and II respectively.! They show the close
chemical relationship between these rocks and the leucite-kenyte, the analysis of which
is given on p. 113.
I.
Ia.
II.
IlA.
(Scott Island.)
Mol. ratios.
(Cape Crozier.)
Mol. ratios.
SiO„
=
55-93
•926
57-95
•959
Tio;
=
0-64
•008
0-40
•005
Ab03
=
19-61
•192
20-43
■200
Fe.A
=
1-75
•on
3-43-
•022
FeO
=
G-32
•090
1-35
•020
MnO
=
0-13
0-07
CaO
=
3 ■ 53
•063
1-90
•034
MgO
=
0-50
•013
0-26
•006
Na.,0
=
7"75
•125
8-32
• 134
K,0
=
3-67
■039
5-96
•063
P.,0,
=
0-12
•001
0-07
•001
ILO at 110°
=
o-io
0-23
TLO above 110°
=
0-19
0 • 39
100-24
100-76
* Broggcr, Eruptivgesteine dcs Kristianiagebietes, 1894, I, p. 76.
f For discussion of the analyses see p. 119.
PHONOLITIC TRACHYTES.
115
Closely related to the preceding, but of a more definite trachytic type, are other
ash-gray rocks, which show under the microscope small, sharply defined rectangular
phenocrysts of anorthoclase in a ground-mass consisting of a trachytic mesh of felspar-
laths, with interstitial dull-green segirine-augite and a little magnetite ( see Plate VIII,
Fig. 4). The felspar-phenocrysts show the minute twin-striations characteristic of
anorthoclase, and similar striations can also be detected in some of the larger felspar-
laths of the base. The segirine-augite is in small, prismatic crystals, and shows
pleochroism from brownish yellow to dull-green.
To this type belong most of the ash-gray or yellow (when weathered) trachytic
rocks from Cape Crozier (248, 224, 251, 243, etc.), as -well as specimens from
Inaccessible Island (802, 803, 807), Brown Island (598, 600), and Black Island (610).
Of almost precisely similar character to the rock (248) from Mount Terror
(Plate VIII, Fig. 4) are trachytic rocks (55) obtained from the dredge off Cape
Wadworth, Coulman Island.
Of extremely salic character is the phonolitic trachyte (607) from the middle of the
crater of Brown Island. It is a white, friable rock, resembling a domite. It consists
of a fine-grained aggregate of small felspars (mostly short rectangular, but some lath-
shaped) without flow-structure, in the interstices of which are distributed a few
magnetite- grains and, very sparingly, needles of dull-green segirine-augite ; a few grains
of segirine and small crystals of sphene are also present. The felspars have a refraction
near that of Canada balsam, and are doubtless mainly anorthoclase. With a high
power, under the microscope, are seen numerous minute and very thin hexagonal and
square sections of an undetermined mineral, which is isotropic or only very feebly
doubly-refracting, and has a refraction markedly less than that of Canada balsam.
That nepheline is present in the base, although it cannot be definitely distinguished,
is indicated by the chemical analysis of the rock, the result of which is as follows
under I* : —
I.
Ia.
II.
(Brown Island.)
Mol. ratio.
(Mont Miaune.
Si02 = 58-G4
•977
CO
—
TiO, = 0-28
•003
trace
Al.A = 22-55
•221
19-G6
Fe203 = 0-97
•00G
3-43
FeC) = 0-99
•014
—
MnO = trace
—
trace
CaO = 1-43
•023
1-53
MgO = 0-10
•004
0-31
Na,0 = 9-87
•159
10-04
ICO = 4-98
■053
4-71
P205 = trace
H20 at 110° = 0-081
—
—
H20 above 110° = 0- 35)
1-00
S03 = 0-27
100-30
99-4G
Q 2
For discussion of the analyses, see p. 119.
116
G. T. PRIOR.
Under II., for comparison, is appended the analysis of a phonolite from Mont
Miaune, Velay, France.*
More closely approaching phonolites are rocks from the debris-heaps at Minna
Bluff (613) and from Black Island (530).
Specimen 613 from the Minna Bluff is a dark greenish-brown compact rock
showing no phenocrysts. Under the microscope are seen feathery flakes of pale-
green gegirine-augite, and ragged tufts of cossyrite and of a brown altered mineral
(probably another soda-hornblende or pyroxene). These are thickly and uniformly
distributed in a rather unindividualised base, showing indistinct felspar-laths and
a few magnetite-grains ; nepheline is probably present in the base, but could
not be definitely identified. The segirine-augite shows extinctions up to 37°, and
pleochroism from pale grass-green for vibrations along the length to pale pinkish-
yellow for those across the length. The cossyrite has pleochroism from nearly
colourless to deep purplish-brown or opaque ; the shreds were too irregular to admit
of a more precise determination of the optical characters.
The rock (530) from Black Island is nearly colourless. Under the microscope it
shows ( see Plate VIII, Fig. 5) a fine-grained trachytic felt of felspar-laths, with much
deep-blue riebeckite-like hornblende scattered in minute shreds through the section,
and accumulated in patches round small altered crystals of what was once probably
nepheline. The riebeckite shows pleochroism from pale-brown to colourless across
the length of fibres to indigo-blue along the length. These two rocks are similar
to some of the phonolitic rocks of the Rift Valley, East Africa. I
Of phonolitic rocks closely related to the kenytes two specimens deserve mention,
viz., a rolled pebble (25) from the beach at Cape Adare, and a dark-green compact
rock (622) from Minna Bluff (debris- heap).
The pebble from Cape Adare is a dark-green compact phonolite showing numerous
porphyritic felspars having the characteristic lozenge-shape of anorthoclase. Under the
microscope, besides the large (up to 1 cm. in length) phenocrysts of anorthoclase
showing minute twin-striations, are seen a few small phenocrysts of deep reddish-brown
hornblende, and more numerous pseudomorphs, after hornblende, consisting of grains of
magnetite and deejngreen aegirine-augite. The ground-mass consists of numerous
shreds and small prismatic crystals of segirine-augite, like that in the hornblende-
pscudomorphs, evenly distributed in a base of felspar-laths (not sharply defined) and
interstitial nepheline. Magnetite-grains occur very sparingly in the base. A few
small crystals of sphene are also present.
Of somewhat similar composition is the dark -green phonolitic rock (622) from the
Minna Bluff. It shows under the microscope numerous long prismatic phenocrysts of
orthoclase and anorthoclase in a very fine-grained base consisting of a mesh of minute
felspar-needles with globulites and microlites of green segirine-augite. Small crystals
* Emmons, On some Phonolites from Velay and the Westerwald. Dissertation, Leipzig, 1874.
t Prior, Mineralogical Magazine, 1903, vol. xiii, p. 237.
THE TRACHYTIC ROCKS OF OBSERVATION HILL.
117
of this mineral are also included in the porphyritic felspars, and occur sparingly as
phenocrysts. The anorthoclase-plienocrysts show no twin-striations, but have mottled
extinctions ; the refraction is not appreciably higher than that of Canada balsam.
The trachytic rocks of Observation Hill present a special type which approaches
very closely to the “ tephritic trachyte ” of Forodada, Columbretes, described by
Becke * and placed by Rosenbusck in the group of trachydolerites.
The mineral which especially characterizes them is a reddish-brown basaltic horn-
blende similar to that which is present in so many of the basalts. This mineral occurs
in small prismatic crystals, only occasionally large enough to be detected by the
naked eye.
These trachytic rocks practically compose the whole of Observation Hill, near
Winter Quarters. All the rocks described on p. 13 of the Report on the Field-geology,
both the slabby rocks (281, 291), the spheroidal rock (655), the dyke-rock (277), and
the yellow rocks (279, etc.), are very similar in microscopic characters, and only differ
in the more or less glassy nature of the base or in the size and number of the horn-
blende-needles. The yellow rocks from the top of the hill appear to be only weathered
gray rocks. In specimens showing alternating bands of gray and yellow (see p. 13)
these bands show under the microscope a precisely similar and continuous mesh of
felspar-laths, and only differ in the yellow bands containing plentiful yellow grains of
epidote in place of the minute magnetite-grains and small dull-green augites of the
gray bands. These trachytes show no phenocrysts of felspar, augite or olivine. Under
the microscope are seen small prismatic crystals of basaltic hornblende (or of black
magnetite-pseudomorphs after it) in a trachytic mesh of minute felspar-laths, showing
generally well-marked flow-structure, with thickly disseminated grains of magnetite
and small needles of pale dull-green augite.
In the dyke-rock (277) and in the dark rock (290) on the S. side of the hill the
hornblende is less altered and in larger amount than in most of the other specimens,
and the flow-structure of the lath-shaped felspars is less pronounced (see Plate VI II,
Fig. 6).
A black streaky rock with greasy lustre (264), half-way up the hill, is a glassy
variety presenting some interesting features. Under the microscope it shows a few
small prismatic hornblendes in a base consisting of a nearly colourless glass enclosing
felspars, partly in sharply defined laths, but mainly in very thin hexagonal or nearly
rectangular plates : small prismatic crystals, both of pale dull-green augite and of
basaltic hornblende, and minute needle-like microlites of augite with a very little
magnetite, are also present. In parts of the slide occur groups of minute colourless
circular or roughly six-sided and eight-sided isotropic crystals (Fig. 64). Though they
appear to have a high relief, the Becke-eflect shows that their refraction is less than
that of Canada balsam, and they are here referred to leucite. Minute icositetrahedra
* Becke, Tschermak’s Min. Petr. Mitth., 1896-7, Bd. xvi, p. 174.
118
G. T. PRIOR.
of leucite were found by Becke in the glassy base of the trachyte from Forodada.* The
platy felspars in this rock have a refraction a little lower, and the lath-shaped felspars
a refraction a little higher, than that of Canada balsam. Most of the felspar is probably
anorthoclase.
Fig. 64. — Leucite-crystals in Base op Glassy
Hornblende - trachyte (264), prom Obser-
vation Hill. Tlie prismatic crystals are
mostly augite : the long one at the top is
hornblende. (Magnification, 150 diam.)
Fig. 65. — Dendritic Magnetite in Glassy Base
op Trachyte from Observation Hill. (Magni-
fication, 200 diam.)
Another very similar glassy variety (442) from between the ship and Observation
Hill shows under the microscope, throughout the slide, small dendritic patches
(see Fig. G5) of magnetite (possibly altered riebeckite or other soda-hornblende).
A noteworthy feature of these trachytes of Observation
Hill is the presence in many of them of dark “ lapilli-like ”
enclosures. In the yellow rocks on the top of the hill these
dark rounded enclosures, about the size of a hazel-nut, stand
out on the weathered surface like fragments of black basalt
caught up in the trachyte. Under the microscope, however,
they are seen not to have the character of basalt, but to
consist of a dense mesh of interlacing prisms of basaltic horn-
blende, similar to that in the trachyte, with magnetite-grains
and only a little interstitial felspar (see Fig. GG). These
enclosures are therefore somewhat similar to those described
by Becke f in the case of the similar tephritic trachyte of
Ferrera, Columbretes. The hornblende in these enclosures
and in the trachytes has the characters of barkevikite and shows pleochroism, a —
yellow. /3 and y — deep reddish-brown ; the long prisms have straight extinction and
compensate aci'oss their length with a quartz-wedge cut parallel to the optic axis.
* Becke, 1. c. p. 176.
t Becke, 1. c., p. 16, 8 and Taf. Ill, fig. 4.
Fig. 66. — Hornblende-inclu-
sions in the Trachyte op
Observation Hill. (Magni-
fication, 150 diam.)
CHEMICAL RELATIONS OF THE VOLCANIC ROCKS.
119
A chemical analysis of the hornblende-trachyte (277) from Observation Hill gave
the result under I.* With this is compared the analysis of the tephritic trachyte from
Forodada under II, f and that of a trachydolerite (haiiyne-phono.lite) from Campanario,
IH-t
I.
IA.
II.
III.
(Observation Hill.)
Mol. ratios.
(Forodada.)
(Campanario.)
SiO„
=
55-47
•918
56-19
55-40
TiCi,
=
1-32
•017
0-57
0-43
ALA
=
20-67
•202
20-25
21-03
FeA
=
2-83
•018
2-76
1-64
FeO
1-86
•026
2-32
3-04
MnO
=
0-02
trace
CaO
=
3-43
•061
4-30
3-57
SrO
=
0-01
MgO
=
1-43
•035
1-12
0-91
Na.,0
=
8-33
•134
6-33
7-64
Iv.,0
=
4-8G
•051
4-19
4-42
PA
=
0-03
0-54
0-23
H.,0 at 110°
=
0-08
0-65
0-95
FLO above 1 lo°
=
0-12
(
Cl,S03etc. O' 25 *
0-57
100-46
99-47
99-83
As in the case of the kenytes and the other trachytes, the analysis indicates the
presence of nepheline which, however, could not be detected with certainty in the base
of the rock.
A determination was made of the silica-percentage of the altered yellow trachyte
(278) from the top of Observation Hill and gave the result 56 "96.
Chemical Relations of the Preceding Volcanic Rocks.
For convenience of comparison the results of the analyses of seven of the rocks
described in the preceding pages are brought together in the following table.§
Si02
a.)
(2-)
(3.)
HI
(5.)
(6.)
HI
=
42-14
43 ■ 92
55-47
55-93
56-09
57 ’ 95
58-64
TiO.,
=
4-90
4-19
1-32
0-64
1-23
0-40
0-28
A1A
14-95
17-42
20-67
19-61
20-79
20-43
22-55
FeA
=
2-90
4-09
2-83
1-75.
1-54
3-43
0-97
FeO
=
9-71
8-83
1-86
6-32
3-84
1-35
0-99
MnO
=
0-12
0-09
0-02
0-13
0-05
0-07
trace.
CaO
=
10-32
9-53
3-43
3-53
3-18
1-90
1-43
SrO
= 1
o-oi
MgO
=
9-47
4-89
1-43
0-50
1-26
0-26
0-16
Nn„0
=
3-27
4-60
8-33
7-75
7 * 33
8-32
9 "87
K20
=
1-80
2-17
4-86
3 " 67
3-91
5-96
4-98
PA
=
0-40
0-67
0-03
0-12
0-38
0-07
trace.
Cl
=
0-17
H.,0 at 110°
=
0-12
0-C6
0-08
o-io
019
0-23
0-08
11,0 above 110°
=
0-16
0-11
0-12
0-19
0-39
0-39
0-35
100-26
100-57
100-46
100-24
100-45
100-76
100-30
* For discussion of analyses see below,
f Becke, l.c., p. 177.
t Sauer, Untersuch. ii. pkonolit. Gesteine der Canarischen Inseln, Inaug. Diss., Halle, 1876, p. 60.
§ The analysis of the glassy limburgite 326 is here omitted, since the result is almost identical with that of
the holoerystalline olivine-hasalt 656.
<;. T. PRIOlt.
120
(1) . Olivine-basalt, dill between the Gap and Horseshoe Bay (656).
(2) . Hornblende-basalt, cone below Castle Hill (385).
(3) . Hornblende-trachyte (trachydolerite), 500ft. up Observation Hill (277).
(I). Phouolitie trachyte (trachydolerite), Scott Island.
(5) . Leucite-konyte, “ Keep” of Cape Royds, slope of Mount Erebus (818).
(6) . Phonolitic trachyte, top of 900-ft. Knoll, Mount Terror (188).
(7) . Phonolitic trachyte, crater of Brown Island (607).
As an experiment in the use of the American quantitative system of classification*
the “ norms ” (percentage mineral composition by weight) of the seven rocks were
calculated, and are given in the following table : —
G)
(2)
(3)
(4)
(5)
(G)
(7)
KAlSi308
=
10-55
12-79
28-95
21-76
22 - 92
35-60
29 ' 47
NaAlSiA
=
4-19
12-05
29-30
35-24
41-47
34-82
39-30
CiALSiA
=
20-57
20-29
4-69
7-81
12-23
0-77
2 • 50
NaAlSiO.,
=
12-78
14-48
21-56
16-53
11-07
18-72
23-66
NaFcSiA
=
1-88
1-58
CaSiOs
=
11-71
9-51
5-23
3-71
0-46
3-64
1-62
FeSi03
=
2-51
2-90
9-37
0-39
0-79
MgSiOj
=
8-20
6-00
3-60
1-30
0-30
0-66
0-40
FcaSi04
=
3-67
2-34
3" 67
Mg.,Si04
=
10-78
4-27
1-96
FeO ■ Fe A
=
4-17
6-03
2-25
2-55
2-32
3-41
1-39
FeO • TiOa
=
9-27
7-90
2-54
1-22
2-28
0-77
O' 16
Ca32P04
=
0-93
1-55
0-20
0-31
0-93
The names which the rocks would receive in this new classification arc as follows
(1) Limburgose.
(2) Limburgose.
(3) Laurdalose.
(1) Esscxose.
(5) Laurvikosc.
(6) Miaskose.
(7) Miaskose.
The result shows that the classification supplies a variety of names to rocks not
differing very widely in chemical composition.
Another system for the chemical classification of igneous rocks, which has been
recently brought forward, is that of Osann.f
This classification is based on the molecular percentages as calculated from the
results of analysis. For flic seven analyses the molecular percentages are as follows:—
(i)
(2)
(3)
0)
(5)
(6)
(7)
SiO.,
=
45-06
48-92
62 • 78
63-17
63 " 66
66-49
66 • 92
TiOj
=
3-96
3-52
1-13
0-54
1-05
0-35
0-20
A1A
9-44
11-46
13-81
13-11
13-96
13-85
15-14
Fc.A
—
117
1-72
1-21
0-75
0-66
1 • 49
0-41
FeO
=
8 * 73
8-27
1-79
6-12
8-71
1 *37
0-96
CaO
=
11-87
11-42
4-18
4-29
3-89
2-34
1-58
MgO
—
15-15
8-15
2 "42
0-84
2-14
0-45
0-27
Na,0
=
3 • 40
4-99
9-17
8-52
8-09
9-28
10-89
K..< )
=
1-22
1 • 55
3-51
2 • 66
2-81
4-38
3-63
* Quantitative Classification of Igneous Rocks, Cross, hidings, Pirsson and Washington, Chicago, 1903.
t Osaun, Tsohonnak's Min. Petr. Mitth., 1900, Bd. xix, p. 351, and 1901, 13d, xx, p. 899,
CHEMICAL RELATIONS OE THE VOLCANIC ROCKS.
I ‘21
In the Osann formulae, which represent the chemical composition of the rocks, the
number of molecules of 8i02 (including Ti02) are denoted by s; the alkalies arc united
to ALO:! (as in the felspars and felspathoids) in a group (NaK)2Al204, which is denoted
by A ; the remainder of the Al2Oa is united with CaO (as in anorthite) in a group
CaAl204, which is denoted by C ; and the rest of the CaO with the other metallic
oxides (FeO, MgO) are united in a group RO represented by F. The number of
molecules of Na20 is calculated on the assumption that Na2O + K2O = 10 and is
denoted by n. Moreover, as the absolute magnitudes A, C, F arc not necessary, since
2A + 2C + F =100 — .'.', their ratios a, c, /, where a + c + / = 20, are used. The ratios
a, c, /give an approximate idea of the relative amounts in which the alkali-felspars,
anorthite and dark constituents, enter into the composition of the rocks.
For the seven rocks the Osann formulae are as follows under ( I ) (7)
(1)
^4804
Ju ^7 35
I.
S00'77 ^2 ^2/10 ^7H
(2)
5B244
^3J ^/l4 ^703
II.
'^U0'78 ^3 CvJ IB ^7'2
(3)
,<f <13-1)1
^11 ^1 fs ^7 23
nr.
,<f03'34 ^10 ^1 ft fhl
(4)
•Vi* 72
«10 ('Vifn J ^7*02
IV.
,<?03'78 fh ^4 ./«J
(6)
S(H7l
^J0 ^2i./74 ^7 40
V.
503'7O ^104 ^7 2
(6)
^13j ('o /ilj ^0 70
VI.
^08'20 ^14 ^0 ./ IS ^7
(7)
,S‘(I7T2
^10 ^1 fs ^'7 '80
VII.
^0714 ^1(1 roJl ^7 6
For comparison, opposite the formula of each Antarctic rock, is
formula (I-Vll) of the following : — -
T. Nepheline-basanifce from llundskopf, Salzungcn.
II. Lirnburgite from Heldburg, near Coburg.
III. Pkonolite from Miidstcin, Bohemia.
IV. “ Rhomb-porphyry” (kenyte) from Kibo, Kilimandjaro.
V. “ Haiiyne-tephrite ” from Campanario, Pahna.
VI. Pbonolite from Bull Cliff, Colorado.
VII. Pbonolite from Mont Miaune, Velay.
appended the
The analytical results show that these Antarctic volcanic rocks do not form
anything in the nature of a rock-series, but that they may be divided fairly sharply
into two groups, a basic one consisting of hornblende- and olivine- basalts of limburgite-
type (1 and 2), and one of medium basicity consisting of kenytes and phonolitic
trachytes (3-7) very rich in alkalies.
These two groups appear to represent the two main products into which the
magma has been differentiated in this Antarctic region.
A graphical representation (according to Brogger’s method)* of these two dif-
ferentiation-products, as illustrated by rocks (l) and (5), is shown in Bigs. < >7 and 08
respectively.
In an Osann triangle, f the rocks of the first group fall into Division III, and
those of the second group into the upper part of II, in a similar position to that
assigned by Becke J to the tephritic and phonolitic rocks of the Bohemian Mittelgcbirge.
* Briiggor, Eruptivgestcine des Kristianiagebietes, ISOS, J 1 1, p. 2 AO.
t Osann, Tschermak’s Min. Petr. Mitth. 1900, Bd. xix, p. 868, and PI. IV.
% Becke, Tsehermak’s Min. Petr. Mitth. 1908, Bd. xxii, p. 214, and PI. II.
von. i.
It
122
G. T. PRIOR.
In both the basaltic and alkaline groups, the ratio of soda to potash is nearly the
same, with Na>K. A similar basaltic hornblende occurs in certain rocks of each
group, and not in others in which its place is taken by olivine. The presence or
absence of hornblende in both groups probably depended to a large extent upon the
conditions of eruption ; in the dyke-rocks, which cooled while still under a high pressure,
the mineral was more likely to survive than in the lavas.
Altogether, this Antarctic region, from Scott Island in the north to the Minna
Bluff in the south, has good claim to be regarded as a definite petrographical province
CaO
Fig. 67. — Graphical Representation op the Chemical Composition op the
Olivine-basalt (656) from near the Gap.
CaO
Fig. 68. — Graphical Representation of the Chemical Composition op the
Leccite-kentte (818) prom Cape Royds.
characterised by the association of basalts of limburgite-type with alkali-rich rocks of
medium basicity in which anorthoclase is the prevailing felspar.
In a wider sense, these Antarctic rocks belong undoubtedly to the great Atlantic
as opposed to the Pacific type of eruption. The marked contrast between the younger
volcanic rocks of the Atlantic volcanic chain and its lateral branches, and those of the
volcanoes encircling the Pacific Ocean, was pointed out by the author amongst the
conclusions drawn from a study of the igneous rocks of the Great Rift Valley of East
RELATIVE AGES OF THE VOLCANIC ROCKS.
123
Africa.* In the same year similar ideas were brought forward with greater elaboration
by Becke in a striking contrast which he drew between the volcanic rocks of the
Bohemian Mittelgebirge and those of the Andes, j He suggested that these two Pacific
and Atlantic types were intimately connected with the two tectonic processes of Suess
which have mainly affected the earth’s crust, viz., faulting by tangential pressure, and
fracture by radial contraction ; where young volcanic rocks occur along a faulted
mountain-chain like the Andes, they belong to the Pacific group ; where, on the other
hand, they occur on block-fractures (“ Schollenbriiche”), they belong to the Atlantic type.
In the case of these Antarctic eruptions, we have volcanic rocks of undoubtedly
Atlantic type developed along a coast which has been described as typically Pacific. J
They form, therefore, apparently an exception to Becke’s rule ; but it may be pointed
out that the volcanoes of South Victoria Land, from which specimens of lava have been
obtained, are not generally ranged parallel to the coast, but appear to occur along lines
of weakness directed nearly at right angles to it. §
As regards the relative ages of the basalts and alkali-rich rocks of the Boss
Archipelago, the observations in the field lead to no very conclusive result. From the
intrusive appearance of the trachyte on the south-east end of Black Island ( see p. 14),
Mr. Ferrar was inclined to regard the trachytes as younger than the basalts, but
he was unable to examine closely the actual junction of the two rocks at this
spot. In the case of the kenytes of Mount Erebus and the islands in Erebus Bay,
Mr. Ferrar is of opinion that they are older than the basalt-flows of Winter Quarters,
since on Kazor Back Island they are bent into a sharp anticline which has not affected
the other rocks. It seems probable, therefore, that the trachytic rocks having a
chemical composition so similar to that of the kenytes are also older than the basalts.
This idea is also supported by the fact that the bombs scattered over trachytic rocks
are of the same character as the basalts. Mr. Ferrar also describes the trachyte-bosses
(e.g., that exposed in the Gap) as occurring like “islands,” without off-shoots into the
basalt. Thus Observation Hill is probably not intrusive in the basalts, but stands out
from them as an older rock, just as the phonolite-peak of Fernando Noronha projects
above the younger basalts.
On the whole, therefore, it seems probable that in these Antarctic volcanic rocks
the sequence of eruption has been from rocks of medium basicity to basic, and that the
trachytes and kenytes preceded the basalts.
* Prior, Contributions to the Petrolog. of British East Africa, etc., Mineralogical Magazine, 1903, vol. xiii,
p. 260.
f Becke, Tschermak’s Min. Petr. Mitth. 1903, Bd. xxii, p. 248.
| Gregory, Nature, 1906, vol. lxxiii, p. 300.
§ See, however, footnote on p. 140.
R 2
124
Chapter II.
THE BASEMENT-ROCKS OF SOUTH VICTORIA LAND.
Crystalline Limestone.
The crystalline limestones, which appear to constitute the prevailing rock of the
Northern and Southern Foothills, are coarse-grained aggregates of calcite-crystals
loosely held together. For the most part they are very pure, and show, in addition
to the calcite, only a few small flakes of graphite and of a nearly-colourless phlogopite.
Amongst the specimens, however, which were found by Dr. Wilson along Discovery
Gulf, are fragments of crystalline limestone containing numerous rounded yellow crystals,
which analysis has shown to be of chondrodite. The rock (see Fig. 69) is very
similar in appearance to the chondrodite-limestones of Burma and Finland.
Fig. 69. — Crystalline Limestone with Chondbodite,
ebom Southern Foothills. (Natural Size.)
Amongst the boulders found on the East of Mount Terror, just above the Barrier,
is one of a schistose crystalline limestone, or calc-schist, containing quartz in fine grains
in alternating bands with the calcite.
Gneiss.
Of gneisses there are not many specimens in the collection. They are gray and
red foliated medium-grained rocks, some of which show conspicuous augen-structure.
Like the Arctic gneisses of Greenland* these Antarctic rocks have the characters of
metamorphosed granites and diorites (orthogneiss). They consist of orthoclase,
oligoclase, quartz, biotite, and almost invariably hornblende. Microcline was not
detected in any of the gneisses examined, but it occurs in large plates in a felspathic
grit on Finger Mountain, which had obviously been derived from the decomposition
of gneissic and granitic rocks.
In the augen-gneiss, orthoclase generally forms the “ eyes ” which enclose biotite,
idiomorphic oligoclase, and occasionally rounded blebs of quartz. f The “ eyes ” are
mostly altered and red with oxide of iron. They are enclosed in a mosaic consisting of
* Belowsky, Beit. z. Petrog. west. Nord-Gronlands, Zeit. d. deut. geol. Ges., 1905, Bd. lvii, pp. 15-90.
f Evans, Quart. Journ. Geol. Soc., 1906, vol. lxii, p. 96.
GNEISS AND GRANITE.
125
interlocking grains of clear quartz and slightly altered oligoclase and orthoclase, with
biotite and hornblende in strings marking the foliation. Both quartz and orthoclase in
the mosaic occur in very irregular interpenetrating patches, which in ordinary light
appear to belong to individual crystals, but break up between crossed nicols into a
number of differently orientated grains. Cataclastic structure of this kind is well marked
in the augen-gneiss 727 (see Plate IX, Fig. 1) from D4 in the Kukri Hills (pp. 30 and 38).
In this rock the interspaces between the large “ eyes ” of orthoclase are occupied partly
by a quartz-mosaic showing marked undulose extinctions, and partly by shattered
felspars crowded with small blebs of quartz (“ quartz de corrosion,” as described by
Lacroix in the case of the charnockites of India*), or forming with quartz a micro-
pegmatitic intergrowth (see Fig. 70).
The red augen-gneiss (716) from Cathedral Rocks has somewhat similar characters,
but the quartz-felspar-mosaic between the large “eyes” of pink orthoclase shows less
pronounced cataclastic structure, and the quartz and
felspar are in more distinct grains ; effects of pressure,
however, are evident in the bent twin-lamellse of the
oligoclase, and quartz of corrosion is also present.
This red granite-gneiss appears to be intrusive in
the gray, just as in Greenland the fine-grained red
gneisses are intrusive in the gray mica- and horn-
blende-gneisses.
O
From Cathedral Rocks comes a hornblende- or
diorite-gneiss (724) consisting of a granular aggregate
of quartz and plagioclastic felspar with much pleo-
chroic (green to brownish-yellow) hornblende. The
quartz is not in large amount, and the felspar is more basic than the oligoclase of
the other gneisses, since it shows symmetrical extinctions of 18°-21°.
Associated with the gneisses are hornblende-schists (730, 704) which help to give
the dark streaky appearance to the rocks referred to on p. 28.
Granites and Diorites.
The granites of South Victoria Land are for the most part typical hornblende-
biotite-granites similar to those from Cape Adare brought back by the ‘ Southern Cross ’
Expedition.
Of these rocks the more noteworthy will be considered under the particular
localities from which they come, and as far as possible in the order in which they are
mentioned in the Report on the Field-geology.
Granite Harbour. — The gray biotite-granite (129) which forms the greater part of
the boss at Granite Harbour (see p. 33) is somewhat gneissic in character, and shows
signs of parallel structure in the arrangement of the shreds of biotite. It consists of
* Lacroix, Rcc. Geol. Surv. Ind., 1891, vol. xxiv, p. 168.
Eig. 70. — Micropegmatite surrounding
Felspar in Augen-gneiss (727) from
the Kukri Hills. (Magnification, 25
diam.)
126
G. T. PRIOR.
large plates of orthoclase and oligoclase, with interstitial quartz-mosaic (cataclastic
structure) and flakes of biotite. The oligoclase is mostly idiomorphic, and shows both
albite- and pericline-twinning. A little sphene and small rounded zircons are present
as accessory minerals. The rock shows evidence of pressure in the strain-shadows in
the quartz, as well as in slight bending of the twin lamelhc and fracture of the crystals
of oligoclase. The dyke-rock (155), except for the large porphyritic red felspars, is
similar to 129, and also shows parallel structure in the arrangement of the biotite-flakes,
but the quartz-mosaic is of coarser grain. The large porphyritic orthoclases enclose
rounded crystals of oligoclase.
Other veins in the gray granite are of quartz-porphyry, and are doubtless
apophyses of the granite (155), since they show similar large red porphyritic
crystals of orthoclase and oligoclase. In specimen 168 these phenocrysts occur with
large rounded quartz-crystals and small pleochroic (grass-green to yellow) hornblendes
in a cryptocrystalline felsitic base.
The rock-specimens collected from the scree-slopes in Granite Harbour include : —
augen-gneiss ; epidosite ; granites with large red porphyritic felspars ; pegmatite ;
very beautiful quartz-porphyries, showing large pink porphyritic felspars and rounded
quartz in a fine-grained felsitic base ; diorites with large porphyritic hornblendes, some-
what similar to the dyke-rock (715) from Cathedral Rocks described below; sandstones
and dolerites (seep. 138), precisely similar to those of the Ferrar Glacier; and also
gabbros, showing under the microscope large ophitic plates of colourless augite and
green uralitic hornblende in a coarse-grained aggregate of plagioclastic felspars with
much pleochroic (colourless to rose-red) sphene.
The Southern Foothills.— The gray rock (569) from the Southern Foothills, which
occurs in bands parallel to the joint-planes of the crystalline limestone (see p. 25), consists
of a medium-grained allotriomorphic aggregate of oligoclase, orthoclase and quartz,
with shreds of green to brown hornblende and biotite showing well-marked parallel
structure. Grains of honey-yellow sphene are very abundant.
The Snow Valley. — Of the granites from the Snow Valley between Cathedral
Rocks and the Northern Foothills (p. 33), the porphyritic rock (563) from c6 is the
most noteworthy. It shows large porphyritic pink crystals of orthoclase, around which
lines of small hornblende-crystals appear to flow. Under the microscope the ground-
mass is seen to consist of allotriomorphic oligoclase and orthoclase and pleochroic
(yellowish-brown to black) hornblende, with quartz in quite subordinate amount.
The coarse-grained porphyritic granite (555) from the knoll eb shows fairly idiomorphic
oligoclase and less sharply defined orthoclase embayed by quartz, which has been
obviously the last mineral to consolidate ; hornblende and biotite are not in large
amount : some sphene is present. A dark patch in this granite has the composition of
a basic diorite or essexite somewhat like the rock of the “tongue” (715) described
on the next page. It consists of a coarse-grained aggregate of plates of altered
plagioclase, large ophitic reddish-brown hornblende and colourless diopside.
DIORITE FROM CATHEDRAL ROCKS.
127
Cathedral Rocks. — Of the specimens from Cathedral Rocks, the rock 715 of the
‘ tongue ’ or ‘ pipe ’ east of the shoulder E2 presents noteworthy features and has the
characters of a basic diorite or hornblende-gabbro. It consists (Plate IX, Fig. 2) of fairly
idiomorphic crystals of labradorite, showing symmetrical extinctions of about 22°, and
large ophitic plates of hornblende, with interstitial quartz in quite subordinate amount.
The hornblende has pleochroism from brownish-yellow to deep reddish-brown like that
of barkevikite, but the extinction in some crystals is as high as 15° ; it encloses small
altered felspars and much apatite. The section shows also two plates of hypersthene,
one enclosed in hornblende and the other altering to fibrous actinolite ; it gives straight
extinction and is faintly pleochroic, from pale-green for vibrations along the length to
pale-pink for those across. Other specimens (712 and 72 L) from the same ‘ tongue ’
show similar ophitic plates of hornblende, but contain also some biotite and orthoclase
and much more quartz. They approach, therefore, more closely to hornblende-granites.
The hornblende, too, in these specimens is of somewhat different composition, showing
pleochroism from dax-k-green to brown. Prisms of allanite occur in these rocks and
apatite is very plentiful.
A chemical analysis of specimen 715 gave the following result under I : — -
I. Ia. II. III.
(Cathedral Rocks.) Mol. ratios. (Hurricane Ridge.) (Sweet Grass Creek.)
Si02
=
53 -0G
•884
53-71
53-48
Ti02
=
1-60
•020
0'74
1-07
AIA
18-65
•183
18-00
19-35
FeA
=
1-44
■009
3-99
2-37
FeO
=
7*58
•105
4-05
4-90
MnO
=
0-05
•001
0-24
0-06
CaO
=
OO
•147
6-88
7-55
MgO
=
3-78
•094
5-19
3-67
Na.,0
=
3-20
•051
3-50
4-07
K.,0
=
1-55
•016
3-10
1-41
PA
=
0'38
•003
0-38
0-62
H..0 at 110°
—
0-16
jO-16
ILO above 110°
=
0-94
0-55
(0'80
SrO etc 0-38
100 -61
100-33
99-89
The Osann formula of this rock is : —
S59J5 a3 Gj fl\ W7'61
which is between that of Osann’s Hurricane Ridge type : —
S58'91 *4 CA fit n0-3
and his Sweet Grass type : — ■
S60 05 (,'i CS,fm W8 1-
The analysis of the mica-gabbro from Hurricane Ridge, Yellowstone National
Park, is given under II,* and that of the quartz-diorite from Sweet Grass Creek,
Crazy Mountains, under IILf
* Iddings, Monogr. U.S. Geol. Surv., 1899, No. 32, Part II, p. 340.
t Clarke, Bull. U.S. Geol. Surv., 1897, No. 148, p. 143.
128
G. T. PRIOR.
Calculation of the “ norm ” gave the following result : —
KAlSi308
= .
8-90
NaAlSijOg
=
2G-72
CaALSiA
=
32-25
SiO.,
=
3-48
CaSi03
=
2-55
FeSiOs
=
10-03
MgSi03
=
9-40
FeO. FeA
=
2-09
FeO. TiO„
=
3-04
Ca32P04
=
0-93
In the American quantitative system the rock would be classed as Ilessose.
Blue Glacier. — Of somewhat similar character to the rock of the “ tongue ” are
the dyke-rocks (572-574) collected by Dr. Koettlitz half-way down' the Blue Glacier.
One of these, in which the hornblende occurs in large ophitic plates, is almost
identical with the rock of the “ tongue,” but the others are less coarse-grained, and
show more numerous and better crystallised phenocrysts of hornblende. In this
respect they resemble camptonitic rocks from Montreal, but are of coarser grain and
approach to essexites. Under the microscope (Plate IX, Fig. 3), they show long,
prismatic crystals of hornblende in large amount, with interstitial plates of altered
felspar ; hexagonal sections of apatite are very plentiful, while grains of magnetite and
ilmenite occur very sparingly. The hornblende is similar to that in the “ tongue ”
with pleochroism : a = brownish-yellow, /3 and y = deep reddish-brown, and extinction
as high as 15° ; most of the sections show a dark-green margin.
Kukri Hills. — The specimens (G99, 700) from the Kukri Hills, illustrating the
intrusion of granite into dolerite (see p. 36), show pink granite in contact with and
almost surrounding fragments of a dark-gray rock. Under the microscope the latter
is seen to consist of felspar, in interlocking grains and indistinct prisms, with shreds
and irregular patches of green hornblende and a little quartz. Accessory constituents
are one or two small crystals of sphene and a few shreds of chlorite enclosing magnetite.
The hornblende shows pleochroism : a = pale yellowish-brown, (3 = dark greenish-
brown, y = dull green. If this rock, therefore, represents a dolex-ite like those
described in a later section (p. 136), it has suffered as extreme a metamorphism as
some of the old dolerites of Cornwall.
Amongst the pebbles from the dredge off King Edward VII Land are horn-
blende-biotite -granites and gneisses, and diorites with large ophitic plates of hornblende.
One pebble of coarse-grained biotite-granite differs from the others in containing
large crystals of microcline.
A boulder of hornblende-biotite-granite from the 500-ft. slope on Mount Terror
deserves mention owing to the peculiar character of the biotite, which occurs in small
thick crystals standing out conspicuously from the white ground-mass.
129
Chapter III.
DYKE-ROCKS (LAMPROPHYRES, etc.).
The dyke-rocks in the crystalline limestones and granites form an interesting series
ranging from kersantites to rocks allied to the banakites of Wyoming. Some of the
dykes show numerous phenocrysts of basaltic hornblende and approach to camptonites
such as occur generally in association with nepheline-syenites. No specimens, however,
of the latter rocks appear to have been met with in this Antarctic region.
Camptonites.
A typical camptonite is the dark-gray rock (839) found in situ by Dr. Wilson at
the south end of the Southern Foothills, near the Koettlitz Glacier. To the naked eye
this rock shows only a few small phenocrysts of hornblende. Under the microscope
(see Plate IX, Fig. 5) the rock is seen to consist of small prismatic crystals of reddish-
brown hornblende thickly distributed in a base of lath-shaped felspars with a little
quartz. The hornblende has the same optical characters as that in the diorite described
above (p. 127). The felspar-laths are of labradorite with refraction markedly greater
than that of Canada balsam, and symmetrical extinctions as high as 19°. Magnetite
and ilmenite are virtually absent. The section shows one or two foreign enclosures of
quartz containing liquid-inclusions with bubbles.
To the camptonites must also be referred a specimen (G29) brought by Lieut.
Armitage from the foot of Cathedral Rocks. It is a dark greenish-gray rock showing
to the naked eye no porphyritic crystals. Under the microscope small green uralitic
hornblendes are seen in a base of felspar-laths and thickly-distributed ragged prisms
and minute needle-like microlites of basaltic hornblende with a little biotite. The
felspar-laths appear to be of oligoclase : one rhombic section cut nearly parallel to
b (010) gave a positive extinction of about 10°. The hornblende-prisms show well-
marked flow-structure.
A specimen (498) from a moraine at New Harbour Height (S. foot) shows the
junction of a gray hornblende-biotite-granite and a black fine-grained camptonitic rock.
The latter consists of numerous small sharply-defined prismatic crystals of basaltic
hornblende, in a base showing a few felspar-laths and a little quartz but rendered
dense with long hornblende-microlites. Besides the hornblendes a few prismatic
crystals of colourless augite are also present. None of the small phenocrysts exceed
0 1 3 mm. in length. Near the junction the base becomes more dense and glassy and
shows no well-defined hornblende-microlites.
VOL. I.
S
130
G. T. PRIOR.
Kersantites.
The twenty-yards-wide dyke (579) cutting the crystalline limestone at Gj in the
Northern Foothills [see p. 27) is an augite-biotite-kersantite. It is a speckled gray
rock showing to the naked eye numerous flakes of biotite. Under the microscope
irregular pale-green to pale-purple augites and much biotite in shreds and straggling
ophitic patches are seen in a medium-grained mesh of plagioclase-laths with a little
interstitial quartz (see Plate IX, Fig. 4). A little reddish-brown hornblende of the
same character as in the preceding rocks is also present. The felspars are altered and
kaolinised ; some with refraction less than that of Canada balsam are doubtless
orthoclase, but most have higher refraction and are probably oligoclase.
A chemical analysis of this rock gave the following result under I. For com-
parison is appended, under II, an analysis of a kersantite from Stengerts, Spessart,*
and under III that of a shoshonite from Beaverdam Creek, Yellowstone Park.f
I. IA. II. III.
(Northern Foothills.) Mol. ratios. (Stengerts.) (Beaverdam Creek.)
SiO,
50-71
•840
51-80
53-49
TiOa
2-71
•034
0-71
A1,0.,
17-08
•107
10-65
17-19
Pe.jOj
1-38
•009
4-93
4-73
FeO
8-71
•122
2-14
3-25
MnO
o-oo
0-2!)
0-14
CaO
5’75
• 102
7-35
0-34
MgO
3 • 03
•090
0 ' 90
4-42
Na,0
3-82
•001
3-08
3-23
K..0
3-03
•037
4-05
3-80
PA
0-57
•004
0-43
H.,0 at 110° =
o-io
11,0 above 110° =
1-75
S 1,32
2-17
CO,
trace
0-50
BaO = O'OO
99-99
99-61
100-02
The Osann formula of this rock is : —
®B9'66 G / [2
which is near that of the shoshonite from Beaverdam Creek : —
soo a Cs| fia ^s a
Calculation of the “norm” gave the following result:—
KAlSisOa
=
20-57
NaAlSiA
31-90
CaALSiA
=
18-03
CaSi03
=
2-67
FcSiOj
=
1-45
MgSiO,
=
1-20
* Goller, Die Lamprophyrgiinge des siidlichen Vorspcssart. Neues. Jalirb., 1889, Beil.-bd. vi, p. 566.
f Iddings, Geol. of the Yellowstone National Park. Monogr. U. S. Geol. Surv., 1899, No. 32, Part ii,
p. 340. Analysis by Dakins.
DYKE-ROCKS RELATED TO BANAK1TE.
131
Fe2Si04
=
G • 94
Mg2Si04
=
5'32
FeO. TiO.,
=
5-17
FoO. Fe203
=
2-09
Ca32P04
=
1-24
In the case of this micaceous rock the “ norm ” diverges from the “ mode,” since
much of the potash occurs in biotite and not in orthoclase.
In the American quantitative system the rock would be classed as Shoskonose.
Thus both chemical systems refer the rock to the same type, from which, however, it
differs somewhat in mineral composition. As a shoshonitc it is chemically closely
related to the banakite-like rocks described in the next section.
Another kersantite appears to be the dark-gray dyke (625) intrusive in the gray
granite at E4 at a height of about 5000 feet. Under the microscope it shows long
shreds of biotite and much-altered green hornblende (mostly uralitic after augite, of
which a little still remains) in a ground-mass of altered prismatic plagioclastic felspars.
Dykes Chemically Related to Banakite.
Under this heading are here included certain dyke-rocks which bear some relation
to the camptonites in containing basaltic hornblende as phenocrysts and also occasion-
ally in the ground-mass, while they differ from them by the presence of numerous
phenocrysts of felspar, some of which are of orthoclase.
To this group belongs the brown dyke (714) in the granite at G3 in the Northern
Foothills. Besides the large porphyritic crystals of basaltic hornblende referred to on
p. 35 it shows also numerous phenocrysts of felspar and a few of augite. The felspar-
phenocrysts are mainly of labradorite, with symmetrical extinctions in albite-lamellse
of 26°. Others, however, with low refraction and showing no twin-striations are of
orthoclase, while some with mottled extinctions and refraction about the same as that
of Canada balsam are probably of anorthoclase. A few rounded phenocrysts of analcite
are also present (see Plate IX, Fig. G).
The ground-mass consists of a holocrystalline medium-grained aggregate of
plagioclastic felspars (mainly in rectangular sections, but also in laths), prismatic
crystals of pale-purple augite, and magnetite. Borne of the rectangular felspars in the
ground-mass are of labradorite giving symmetrical extinctions as high as 25°.
Apatite-needles are plentiful, and shreds of brown altered hornblendie material are
scattered through the slide.
A coarse-grained inclusion consists of an aggregate of felspar with refraction near
that of Canada balsam, pale-green to purple augite, basaltic hornblende, and analcite,
with large needles of apatite and much magnetite.
A somewhat similar dyke-rock (720) comes from Cathedral Rocks at E2. It shows
porphyritic crystals of labradorite (symmetrical extinctions of about 28°) and green
uralitic pseudomorphs, in a ground-mass of rectangular felspars and long needles of
s 2
132
G. T. PRIOR.
brown hornblende. Some of the felspar-plienocrysts are much decomposed, but show a
narrow margin of quite unaltered material.
A chemical analysis of specimen 714 from G3 gave the following result under I
I.
Ia
11.
III.
(714.)
mol. ratios.
(Banakite.)
mean of— (1) and (5), (p. 119)
Si02
:
48-22
•804
51-46
49-11
TiOo
-
2-09
•026
0-83
3-06
A1.A
=
18-47
•181
18-32
19-10
Fe203
=
5-28
•033
4-61
2-22
FeO
= •
3-90
•051
2-71
6-78
MnO
=
o-io
•001
0-17
0-08
CaO
6-02
•108
6-03
6‘75
MgO
::
2-07
•052
2-91
5-36
Na20
=
4-94
•080
4-11
5-30
K.,0
=
3-47
•037
4.48
2-85
PA
=
0-88
•006
0-86
0-39
H.,0 at 110°
—
0-44
ILO above 110°
2-89
}
3-89
C02 and loss
=
(1-23)
100-00 100-38
The Osann formula of this rock is : —
S60'44 an Gi/ll
The formula for a banakite-dyke from Ishawooa Canyon, Wyoming,* the analysis
of which is given under II, is : — -
S60 01 Chi ??6'8'
Calculation of the “ norm ” gave the following result : —
KAlSi.,0,
= 20-57
NaAlSi/R
= 25-15
NaAlSiO,
= 8-90
CaALSi.A
= 17-79
CaSi03
= 3-12
FeSi03
= 3-30
MgSiOs
II
to
o
Fe203
II
Cl
rc
GC
FeO.TiO.,
= 3-95
Ca32P04
= 1-86
In the American quantitative system, the rock would be classed as Akerose, but
a very slight difference in the ratio
K20 -f- Na20
~ CaO
would refer it to Shoshonose, like the
kersantite described above (p. 130).
Under III is appended the result of taking the mean of analyses (l) and (5) (see
p. 119) of the olivine-basalt (656) and the leucite-kenyte (818) of the Eoss Archipelago.
Although it may have no real significance, it seems worthy of note that the chemical
composition of this ancient dyke-rock (714) should be so near the mean of the two
Idlings, Monogr. U.S. Gcol. Surv., 1899, No. 32, Tart II, p. 347. Analysis by Eakins.
DYKE-ROCKS RELATED TO B ANA KITE.
133
principal products into which the magma appears to have been differentiated in the
recent lava-hows. If the FeO and Fe203 are taken together, then no item in the two
results, except that of the MgO, differs by more than 1 per cent. In these banakite-
like dykes, too, occurs the basaltic hornblende which is so ubiquitous a mineral in the
recent lavas. The ratio of Na20 : K20 is also nearly the same as in the younger rocks.
Doubtfully to be referred to under the heading of banakite is the dyke-rock (566)
intrusive in the crystalline limestone of the Northern Foothills at G4. It is a dark-
greenish-gray speckled rock, showing to the naked eye no phenocrysts. Under the micro-
scope it is seen to be microporphyritic, with numerous small, sharply-defined phenocrysts
of purplish augite, green chloritic and serpentinous pseudomorphs (possibly after
hornblende) and a few altered felspars. The ground-mass consists of altered felspar
prisms with shreds of biotite and hornblende and sparingly scattered magnetite-grains.
Of very similar character is the dyke (565) in the granite at e6 above the head of
Blue Glacier.
The specimen (646) found by Lieut. Skelton in a moraine, ten miles beyond
Cathedral Bocks, deserves mention here. It is a brownish-grey compact rock,
conspicuously porphyritic, and showing to the naked eye small prismatic felspars and a
few crystals of augite and hornblende. Under the microscope are seen phenocrysts of
oligoclase (symmetrical extinctions of 10°) and sharply -defined, nearly-colourless,
augites in a base of felspar-laths ; phenocrysts of basaltic hornblende are also sparingly
distributed. The rock is altered, with development of much epidote in the phenocrysts
of felspar and augite.
134
Chapter IV.
the beacon sandstone and other sedimentary rocks.
The Beacon Sandstones are medium-grained rocks, for the most part remarkably
free from ferruginous or other coloured impurities. The quartz-grains which compose
them have doubtless been derived from the granitic and gneissic rocks, as they show
liquid inclusions with bubbles, and occasionally numerous hair-like inclusions (rutile?)
such as are often found in the quartz of granites. In some of the grains these
needles occur in lines perfectly straight and parallel to the directions of extinction.
Accessory constituents are rare in these sandstones, but in some specimens opaque and
kaolinised white and pink felspars are plentiful. In
a fine-grained quartz-grit (641) at the base of Finger
Mountain are small irregular grains of pink garnet
and rounded crystals of zircon and rutile, and in
coarse-grained felspathic grits (640, 642) from the
same locality occur large angular fragments of micro-
cline. In these grits the quartz-grains are quite
angular, but most of those in the ordinary sandstones
are fairly well rounded.
The cementing material is siliceous and usually
is not in large amount, so that the grains are loosely
cohering, but in the “ stalagmitic ” and “ marbled ”
sandstones, mentioned on p. 44, narrow seams of the
rock have been converted into compact quartzite, owing
probably to local infiltrations of siliceous material.
Under the microscope such parts of the sandstone
(679) show rounded grains of quartz cemented by
accretions of quartz in crystalline continuity with the grains, as in the case of the
stiperstones of Shropshire. Between crossed nicols the slide has the appearance of
interlocking irregular quartz-grains, but in ordinary light the perfectly rounded oval
outlines of the original grains are clearly seen (see Fig. 71, in which the dotted lines
show the outlines of the original rounded grains).
Of the larger pebbles in the Beacon Sandstones, most consist of granitic quartz
with liquid inclusions and moving bubbles, but one specimen (638) from below Finger
Mountain appears to have been derived from an earlier sandstone, as it consists of
rounded and sub-angular quartz-grains cemented by quartz. Another pebble (675)
from the sandstone under B, consists of a quartz-schist showing, under the microscope,
irregular interlocking grains of quartz, with strings of carbonaceous material marking
Fig. 71. — Quartz-grains in Beacon
Sandstone (679) from Inland Forts.
The dotted lines show the original
rounded outlines of the grains. (Mag-
nification, 20 diam.)
MICACEOUS SCHISTS.
135
the foliation. Pebbles of fine-grained quartz-grits, obviously derived from granitic
material, were obtained in the dredge off King Edward VII Land.
From various localities come specimens of dark micaceous schists, which under the
microscope have much the appearance of contact-metamorphosed sediments.
One of these specimens (96) comes from Granite Harbour. It consists of sub-
angular grains of quartz and some plagioclastic felspar, with strings of strongly
pleochroic (from nearly colourless to deep reddish-brown) mica ; grains of sphene are
also plentiful. A very similar rock (578), but of finer grain, was found near the
contact of the kersantite (579) with the crystalline limestone (576) at the north-east
end of the Blue Glacier. Somewhat similar rocks, but of still finer grain and more
slaty in appearance, were found amongst the rock-fragments from the dredge, both oft'
Balleny Islands (866) and off King Edward VII Land.
Black shaly to slaty rocks (836, 496) were obtained from moraines on the Blue
Glacier and at the south-east end of Black Island (525), and also from the dredge off
Balleny Islands (866).
136
Chapter V.
THE DOLERITES.
The dolerites, intrusive in the Beacon sandstone and the granite, are remarkably
uniform in appearance and in microscopic characters, from Depot Nunatak at the head
of the Ferrar Glacier to the Kukri Hills near its mouth.
The rock (662) from Depot Nunatak is fairly typical of all the specimens (see
Plate X, Fig. 1). It is a mottled gray-brown medium-grained dolerite, showing no
porphyritic crystals. Under the microscope it is seen to be made up mainly of
colourless augite, partly in long prismatic crystals, and partly in irregular sub-ophitic
plates, and plagioclastic felspar (labradorite chiefly) in stout prisms and lath-shaped
crystals. Grains of magnetite and ilmenite are very sparingly distributed.
A characteristic feature of most of these dolerites is the
presence, in patches and in the interstices of the augite and
felspar, of more acid material, showing quartz in radiating
(spherulitic) and micropegmatitic intergrowth with felspar.
In the section of specimen (662) quartz is seen to have
crystallised round prisms of felspar, from the end of which
springs a micropegmatitic intergrowth (see Fig. 72).
In this respect, as well as in their general characters,
these dolerites bear a striking resemblance to the so-called
“ augite-diorites with micropegmatite,” which are intrusive
in the gneisses and pyroxene-granulites of Southern India
(Madras Presidency), and have been described by Dr. T. H.
Holland.* In both cases the coarseness of grain of the
micropegmatite varies directly with the texture of the rock, and thus one of the
principal arguments used by Dr. Holland in favour of the primary origin of the micro-
pegmatite is applicable also to these Antarctic rocks. In some specimens, however,
( e.g ., 692 from Dry Valleys) this more acid and (in the case of this specimen) finer-
grained felsitic material occurs in such distinct patches (up to 2-3 mm. in diameter),
with interspaces nearly free from it and consisting simply of the felspar-augite aggregate,
as to suggest an intermingling of two rocks (granophyre and gabbro), such as occurs so
commonly in Skye.f It is remarkable that in this rock the magnetite (in rod-like
skeleton-crystals) is mainly confined to the more acid patches. Some of the specimens
of dolerite found (but not in situ ) at Depot Nunatak are gabbro-like in coarseness of
grain, and in these the micropegmatite is also coarse-grained, and constitutes practically
the ground-mass of the rock. One of these gabbro-like rocks (632) contains olivine
intergrown with the augite, which in this case is of the purplish titaniferous variety.
* Holland, Quart. Journ. Geol. Soc., 1897, vol. liii, p. 405.
t Harker, The Tertiary Igneous Hocks of Skye. Mem. Geol. Surv. of the United Kingdom, 1904, p. 169.
Fig. 72. — Micbopegmatite in
Dolebite (662) fbom Depot
Nunatak. (Magnification, 100
diam.)
DOLEPJTES.
137
In certain specimens (698 from Dry Valleys and 661 from Knob Head) the
micropegmatitic patches are not so prominent, but quartz is still present in the
ground-mass. These rocks are of somewhat finer grain, and consist of a base, of small
stout felspar-prisms with a little quartz, from which the colourless augites stand out
conspicuously in large irregular ophitic plates.
The augite in these dolerites gives extinctions of over 40° ; the sections are for the
most part without definite crystalline outline ; most of them show the usual twinning
on 100, and many of them also exhibit the “ herring-bone ” structure due to fine
striations parallel to 001 in the two halves of a twin.*
The felspar-prisms in these rocks generally show Carlsbad-twinning. Albite-
twinning is not so common, but in some sections showing twin lamellae symmetrical
extinctions of 26° were observed.
The result of a chemical analysis of specimen 661 from Knob Head is as follows
under I. Under II is given, for comparison, the analysis of the “ augite-diorite ” from
Madras referred to above.
I.
Ia.
II.
(Knob Head).
Mol. ratios.
(Madras).
SiO.,
= 53-26
•882
51-15
Ti02
= 0-70
•009
0-44
alo3
= 15-64
•153
15-92
FeA
= 0-24
•001
9-34
FeO
II
•105
2-87
MnO
= 0-11
—
0-09
CaO
= 12-08
•215
10-40
MgO
= 8-64
•214
6-48
Na./>
= 1-25
•020
1-19
K„0
= 0-58
•006
1-61
PA
= 0-04
0 • 06
H.,0 at 110°
= 0-35
- |
O'll
H20 above 110°
= 0-41
- I
100-74
99-66
The Osann formula for this rock is : —
s55'46 C'iifui 'h e;!
which is near the formula for a basalt from Etna belonging to Osann’s “ Royat ” type,
viz. : —
S54'21 ^sfu ^8
Calculation of the “ norm ” gave the following result : —
KAlSiA
==
3-34
NaAlSiA
=
10-48
SA
=
3-36
CaALSiA
=
38-09
CaSi03
=
9-05
FeSi03
=
13-72
MgSA
=
21-40
FeO. TA
=
1-37
FeO. FeA
=
0-35
VOL. I.
Schaub, Neues Jahrb., 1905 (i), p. 100, and Plate VI, Fig. 5.
T
1.38
G. T. PRIOR.
In the American quantitative system the rock would be classed as Auvergnose.
The contact-effects produced in the dolerites intrusive in the sandstone and granite
respectively are of interest. In each case the effects are similar, and due mainly to the
sudden cooling. In place of a coarse-grained dolerite, the rock a few feet from the
contact has the characters of a basalt, which becomes finer-grained and more glassy the
closer it approaches the sandstone or granite ; at the same time the felspar and augite
become long prismatic, with the former often enclosed in the latter (see Plate X, Fig. 3).
At one or two inches from the contact the dolerite changes from black to a pale-green
colour. That the production of this compact green aphanitic material is not due to any
absorption of silica from the sandstone is shown by a determination of the silica in it,
which gave a result 51 '96, actually lower than that obtained in the above analysis of
the dolerite from Knob Head.
In Plate X, Figs. 2-5 represent, as seen under the microscope, thin slices of
specimens of dolerite collected at different distances from the contact with sandstone.
Specimen 696 (Plate X, Fig. 2), from Dry Valleys, at 2ft. from the sandstone, still
shows some augite in ophitic patches enclosing felspar- laths, but most of the pyroxene
is in long prisms scattered through a base of felspar-laths, with interstitial felsitic
material crowded with magnetite in rod-like skeleton-crystals.
Specimen 687 (Plate X, Fig. 3), from Inland Forts, at 6in. from the sandstone,
shows long interlacing felspar-laths and colourless augite dispersed about them and
sometimes enclosing them, with interstitial patches of brown glass, dense with magnetite
in rods and grains.
Specimen 695 (Plate X, Fig. 4), from Dry Valleys, at 2in. from the sandstone,
shows a further stage in the passage to a glass. The rock is variolitic, and much finer-
grained than the preceding ; it shows a few porphyritic felspars, but consists mainly
of radiating sheaves of felspar-needles, with interstitial glass dense with magnetite.
Finally, at the actual junction with the sandstone, the rock (specimen 669, from B,)
is a pale-brown glass with dark-brown clouded patches arranged in wavy lines roughly
parallel to the line of contact (Plate X, Fig. 5). Close to the junction the magnetite is
in feathery tuffs surrounded by clearer halos, but at a few millimetres distance it occurs
in more distinct grains ; here also are seen one or two small porphyritic felspars and
altered augites in a base which is confusedly crystalline, with radiating sheaves of
minute felspar-needles, as in the preceding specimen but on a finer scale.
Specimens from the contact of dolerite and granite at D2 in the Kukri Hills show
similar characters, except that in this case the chilling process has not proceeded so far.
Thus specimen 705 shows, at the actual junction, no glass, but exhibits characters
intermediate between those of specimens 687 (Plate X, Fig. 3) and 695 (Plate X,
Fig. 4).
The dolerite (119) from Granite Harbour has the same characters as those of the
dolerites of the Ferrar Glacier, and shows similar patches of acid material. Specimen 154
from this locality is a peculiar hybrid rock. It is a dark dolerite with numerous small
SUMMARY.
139
(from 2 to 3 mm. in diameter) rounded red patches of granitic material fairly uniformly
distributed through it. Under the microscope (Plate X, Fig. 6) the main mass of the
rock is seen to consist of small purple sub-ophitic augites, felspar-laths, magnetite-rods,
and green pseudomorphous hornblende. The red patches are composed of stout red
and kaolinised felspar-prisms, with a little interstitial quartz and a few shreds of biotite.
Of interest in connection with the wide extension of the dolerite-sandstone
formation (see p. 53) is the fact that amongst the pebbles dredged up off Cape
Wadworth, and also to the south of the Balleny Islands, are some of dolerite like that
of the Ferrar Glacier, and showing similar spherulitic and micropegmatitic patches.
Further, a pebble (866) from the dredge south of the Balleny Islands shows characters
almost precisely similar to the “ chilled ” dolerite (687) from Inland Forts at the
junction of dolerite and sandstone.
The dolerite fragments found in moraines on Mount Terror are similar to the
dolerites of the mainland ; some are very coarse-grained and gabbro-like, with coarse-
grained micropegmatite, as in the specimens from Depot Nunatak described above.
In a moraine off White Island was found a fragment (307) of typical olivine-
gabbro. Under the microscope it shows broad plates of labradorite with glomero-
porphyritic groups of pale yellowish-brown diallage in large irregular plates and small
olivines in ter grown with the diallage.
In connection with the dolerites, a peculiar conglomerate found on the west
promontory of Black Island deserves mention. It consists mainly of rounded grains of
quartz and a little felspar (including microcline), cemented by calcite ; but it also
contains numerous fragments of dolerite, similar to that of the mainland.
SUMMARY.
The basement-rocks of South Victoria Land consist of crystalline limestones,
gneisses and granites.
The crystalline limestones are remarkably free from accessory constituents, but
some contain chondrodite.
The gneisses are metamorphosed granites (orthogneiss) and are often characterised
by prominent augen-structure and by cataclastic effects.
Above the gneisses occur masses of red and gray granites which pass occasionally
into more basic diorites or hornblende-gabbros.
Intrusive in these basement-rocks are dykes, chiefly of lamprophyric rocks
including camptonites and kersantites, but comprising also quartz-porphyries and
rocks chemically allied to banakite.
Upon the granite has been deposited an extensive sandstone-formation in which
some obscure charred plant-remains have been found.
Through the sandstones have been intruded very widespread sills and dykes of
dolerite. This dolerite is very uniform in mineral-composition throughout wide areas,
T 2
140
G. T. PRIOR.
and is characterised generally by the presence in the ground-mass of patches of a
micropegmatitic or spherulitie intergrowth of quartz and felspar. In this and other
characters it strikingly resembles the so-called “ augite-diorite,” dykes of which break
through the pyroxene-granulites and gneisses of Southern India.
The sandstone-dolerite formation of South Victoria Land appears to be very
extensive, for fragments of the characteristic dolerite were met with in most of the
localities within the Antarctic Circle visited by the ! Discovery,’ and were found in the
dredgings even as far north as near the Balleny Islands.
Finally, the islands off South Victoria Land, from the Scott Islands in the north to
the Ross Archipelago in the south, as well as Cape Adare on the mainland, and in all
probability many of the mountains, such as Mount Melbourne, which fringe the coast
below the main plateau, consist of volcanic rocks of comparatively recent date.
These volcanic rocks may be regarded as belonging to one petrographical province
characterised by the association of hornblende- and olivine-basalts approaching the
limburgite-type, with very alkali-rich rocks of medium basicity comprising phonolitic
trachytes and phonolites, and also alkaline-basalts or kenytes almost identical in mineral
and chemical composition with the ancient rhomb-porphyries of Norway and the very
similar recent lavas of Mount Kenya and Mount Kilimandjaro in East Africa.
A striking feature in the basalts of limburgite-type is the high percentage of
titanic acid and the number of included coarse-grained nodules, many of which are
felspathic and gabbro-like in character.
These volcanic rocks of South Victoria Land belong undoubtedly to the Atlantic
group, although they occur along a coast which has been described as distinctly of
the Pacific type.*
* While these pages were in the press, Band II, Theil I of the Deutsche Siidpolar Expedition (1901-1903)
was published: in No. 2, ‘Geologische Beschreibung des Gaussberges,’ by E. Philippi, the author brings forward
arguments in favour of the view that the coast of South Victoria Land is really of Atlantic type.
INDEX.
A, see Map and Section I (Plate VII) ; and also
Depot Nunatak.
a15, see Map.
„ sandstone and dolerite, 41.
Ablation, of glaciers, G4.
Actinolite, in diorite, 127.
Actinolite-rock, Cathedral Eocks, 31.
Adam (Mount), 22.
Adare (Cape), Fig. 6 (p. 17).
„ 7,21,22, 98,140.
„ analcite in basalt, 106.
., basalts, 2, 17, 18, 101, 102, 106,
109.
„ described, 17, 18.
,, granites, 2, 32, 125.
„ icebergs aground, 75.
„ pkonolite, 116.
„ re-sorted moraines, 80.
„ geological sections, 18.
Adelie Land, granites, 1, 2.
Admiralty Eange, 17, 19, 23.
„ described, 22.
„ glaciers, 71.
„ pyramidal peaks, 54.
Aegirine, in trachyte, 115.
Aegirine-augite, in kenyte, 113.
„ in phonolite, 116.
„ in phonolitic trachytes, 114-116.
Ages, relative, of basalts and trachytes, 123.
Akerose, 132.
Alkalies, ratio of, in rocks, 122, 133.
Alkaline basalts ( see Kenytes).
Alkali-rich rocks, association with basalts, 102, 121,
122, 140.
Allanite, in granite, 127.
Analcite, in banakite, 131, and Plate IX, Fig. 6.
„ in basalts, 106.
Analyses, “ augite-diorite,” Madras, 137.
„ banakite (714) from G3, 132.
„ banakite, Ishawooa Canyon, 132.
„ basalts and trachytes, 119, 120.
„ diorite (715), Cathedral Rocks, 127.
„ dolerite (661), Knob Head, 137.
„ hornblende-basalt, Hiirtlingen, 105.
Analyses, hornblende-basalt (385), Sulphur Cones,
103.
„ “hornblende-diabase,” Grilveneck, 105.
„ hornblende-trachytes (277, 278), Obser-
vation Hill. 119.
„ leucite-keny te (81 8), Cape Royds, 1 12, 1 1 3.
„ kenyte, Mt. Kenya, 113.
„ kersantite (579) at G1; 130.
„ kersantite, Stengerts, 130.
„ limburgite (326), Ridge Road, 105.
„ mica-gabbro, Hurricane Ridge, 127.
„ nepheline, 112.
„ olivine basalt, near the Gap (656), 105.
„ phonolite, Mont Miaune, 115.
„ quartz-diorite, Sweet Grass Creek, 127.
„ rhomb-porphyry (kenyte), Kibo, 113.
„ rhomb-porphyry, Vasvik, 113.
„ shoshonite, Beaverdam Creek, 130.
., trachyte (607), Brown Id., 115.
„ „ (188), Cape Crozier, 114.
„ „ Forodada, 119.
„ „ Scott Id., 114.
„ trachydolerite, Campanario, 119.
Andes, volcanic rocks of, 122.
Anne (Cape), Coulman Id., 5.
Anortboclase, in banakite, 131.
„ kenytes and trachytes, 102, 111,
112, 122.
„ phonolites, 116, 117.
„ phonolitic trachytes, 114, 115, 118.
Antarctic Continent, evidence for, 2.
„ Expedition (German), 140.
„ Manual, refs, to, 1, 2, 57, 71.
Anticline, Razor Back Id., 123.
Apatite, in banakite, 131.
„ in basalts, 102, and Fig. 59 (p. 103).
„ in diorite, 127, 128.
„ in kenyte, 113.
Arber, E. A. Newell, Report on plant-remains, 42,
43, 48.
Arctic Regions, ice-foot, 61.
„ ice-caps, etc., 99.
Armitage, Lieut. A. B., dolerite from Depot
Nunatak, 49.
142
INDEX.
Armitage, Lieut. A. B., camptonite from Cathedral
Rocks, 129.
„ „ pioneer-journey up Ferrar
Glacier, 39, 49, 85.
Armitage (Cape), Fig. 32 (p. 58).
„ „ basalt, 13, 106.
„ „ cracks in sea-ice, 59, GO.
„ „ water-lioles, 58, 59.
Arrival Bay, Plate II.
Arrival Bay Heights, see Harbour Heights.
Atlantic type, Antarctic rocks of, 102, 122, 140.
Auckland, volcanoes, 12.
Auckland Islands, Fig. 58 (p. 97).
„ „ described, 96, 97.
„ „ basalts, 96, 97, 101, 109.
Augen-gneiss, Cathedral Rocks, 30, 31, 125.
„ described, 124, 125.
,, Granite Harbour, 126.
„ Kukri Hills, 30, 38, 125, and
Plate IX, Fig. 1.
Augen-structure, in gneiss, 124, 125, 139.
Augite, in banakites, 131, 133.
„ in basalts, 102, 103, 104, 106, 109.
„ in camptonites, 129.
„ in contact-dolerites, 138.
„ in dolerites, 136, 137.
„ in kenytes, 111, 113.
„ in kersantites, 130. 131.
„ in nodules in basalts, 107, 108.
„ in phonolitic trachytes, 114-117.
“ Augite-diorite,” Madras, 136, 140.
„ „ anal., 137.
Auvergnose, 138.
B, see Finger Mountain.
B, (hill), see Map and Sections (Plate VII).
47
„ dolerite and sandstone, 40, 41, 45, 49, 50,
138, and Plate X, Fig. 5.
„ fossil plant-remains, 40, 41, 42, 48.
„ pebbles in sandstone, 41, 134.
Bs, B„ see Map and Sections (Plate VII), and also
Beacon Heights.
Bs, New Mount, see Map and Plates V, VII.
Bg, B7, Bg, see Map and Sections (Plate VII), and
also Terra Cotta Mountains.
B0, sandstone and dolerite, 51.
B9, see Map and Sections (Plate VII), and also
Knob Head Mountain.
B9a (mountain), see Map and Sections (Plate VII),
and Fig. 28 (p. 51).
bj, see Map and Plate V.
bl5 dolerite, 51.
„ sandstone, 45.
b2, see Map and Plate V.
„ crevasses, 70.
„ dolerite, 46, 51.
„ sandstone, 45, 46.
Balanus Shells, in moraines, 80, 91.
Balleny, Capt. John, discovery of Balleny Islands,
1, 2.
Balleny Islands, described, 1, 2.
„ dolerite, etc., from dredge, 3, 135,
139, 140.
„ icebergs, 74.
„ soundings, 3, 94.
Banakite, 129, 139.
„ described, 131-133.
„ Cathedral Rocks, 131, 133.
,, Ishawooa Canyon, anal., 132.
„ Northern Foothills, 27, 35, 131-133, and
Plate IX. Fig. 6.
„ Northern Foothills, anal., 132.
Banner-shoal, Cape Adare, 75.
“ Bare Rocks,” basalt-tuffs, 110.
Barkevikite, in basalt, 103.
„ in trachyte, 118.
Barne, Lieut. M., records horizontal structure, 54.
„ movement of barrier, 69, 82.
„ rocks from Barne Inlet, 26.
Barne (Cape), coastal ice-fringe, 65.
„ described, 9.
„ basalt, 9, 106.
„ basalt-agglomerate, 9.
Barne Inlet, gneiss and granite, 26.
Baron, R., cavities in granites, ref., 89.
Barrier (Great Ice), Fig. 3.8 (p. 68).
„ 10, 24, 64, 79, 81, 124.
„ described, 67.
„ hummocks produced by, 59.
„ icebergs from, 74.
„ moraines, 69.
„ movement of, 69, 82, 83.
„ pressure-ridges, 82, 83.
„ snow on, 84.
„ soundings along, 94.
Basalt, association with alkali-rich rocks, 102, 121,
122, 140.
,, Auckland Islands, 96, 97, 109.
., Black Id., 14, 106.
„ Brown Id., 15, 104.
„ Cape Adare, 2, 17, 18, 101, 102, 106, 109.
„ Cape Armitage, 13, 106.
„ Cape Barne, 9, 106.
INDEX.
143
Basalt, Cape Crozier, 10, 11, 104, 105.
,, Castle Eock, 12, 105, 106.
„ Chemical relations, 119, 120.
„ Coulman Island, 5, 6, 101.
„ Crater Hill, 12, 106.
„ Crozier Cliffs, 10.
„ described, 102-106.
,, fragments in tuffs, 109, 110.
„ Franklin Island, 1, 6.
„ Granite Harbour, 33.
„ Harbour Heights, 12, 13, 104, 10C, 108.
„ Inaccessible Island, 16, 106.
„ Macquarie Island, 96, 109.
„ Minna Bluff, 21, 106.
„ Mount Melbourne, 20.
,, near the Gap, 104-106, 121, 122 (Fig. 67),
and Plate VI II, Fig. 1.
,, ,, anal., 105, 119, 120.
„ Horseshoe Bay, 106.
,, Mount Terror, 10, 11, 103, 105.
„ nodules in, 12, 13, 106-109, 110.
„ Observation Hill, 13.
„ Possession Islands, 1, 4, 5.
„ Eoss Archipelago, 102-109.
„ Eazor Back Island, 16.
„ Sulphur Cones, 11, 12, 104, 105.
„ „ anal., 103, 119, 120.
„ Sultan’s Head, 11, 109, 110.
„ Turtle Back Island, 14, 106.
„ V-Cliffs Hogsback, 10.
„ White Island, 15, 104.
„ wind-effect on, 88.
„ Winter Quarters, 11-13, 104, 105, 123.
„ Young Island, 1, 2.
Basalt-agglomerate, Cape Barne, 9.
„ Coulman Island, 5, 6.
„ Mount Terror, 10.
„ Tent Island, 16.
Basalt-glass, Crater Hill, 12.
„ in tuffs, 110.
Basalt-tuffs, 11, 109, 110, see also Tuffs.
Basement-rocks, South Victoria Land, 25-38, 101,
124-128, 139.
Bay-ice, 57.
Beacon Heights (B3, B4), see Plate V, and Sections
(Plate VII).
„ „ 39, 45, 46.
„ „ caps of dolerite, 53.
„ „ granite in moraines, 38.
,, „ sandstone and dolerite,
46.
Beacon Sandstone, 23, 82, 88, 94, 98, 99, 101.
Beacon Sandstone, Beacon Heights, 38.
„ described, 39-54, 134.
,, extent of formation, 53.
„ fossil plant-remains in, 40-43,
46-48, 99, 139.
„ ( see also Sandstone.)
Beaufort Island, described, 7.
Beaverdam Creek, shoshonite, anal., 130.
Becke, F., Atlantic and Pacific type of volcanic
rocks, ref., 123.
„ hornblende-pseudomorphs, ref., 103.
„ rocks of Mittlegebirge, ref., 121, 122.
„ trachytes of Columbretes, ref., 117, 118.
Bergschrund, Inland Forts, 93.
Bernacchi, L.C., glaciers of Eobertson Bay, ref., 71.
„ south-east journey, 67.
Bernacchi (.Cape), 20, 30.
„ piedmont, 66.
Biotite, in banakite, 133.
„ gneiss, 124, 125.
„ granites, 125, 126, 128.
„ kersantites, 130, 131.
„ nodule in basalt, 107.
Bird (Cape), 8.
Bird (Mount), 8.
„ described, 10.
Black Island, Fig. 32, (p. 58).
„ 21.
„ basalts, 14, 106.
„ boulders, 80.
„ conglomerate, 139.
„ described, 14.
„ meltins of snow, 85.
„ Glauber salt, 91.
„ moraines, 69, 79, 80, 91.
„ phonolitic trachytes, 14, 115, 123.
„ plionolite, 116, and Plate VIII,
Fig. 5.
„ slaty rocks in moraines, 135.
Block-fracture, 123.
Blue glacier, Fig. 54 (p. 90).
„ 25, 26, 27, 33, 66, 133, 135.
,, debris- cones, 81.
„ described, 70, 71.
„ diorites, 128.
„ icebergs, 73, 94.
„ rate of movement, 82.
„ rock -debris in, 77, 78.
„ slaty rocks in moraines, 135.
‘ Bluff,’ The ( see Minna Bluff).
Bombs, basaltic, 123.
„ Cape Crozier, 11.
144
INDEX.
Bombs, Harbour Heights, 12, 106.
„ V-Cliffs Hogsback, 10.
Bonney, T. G., cavities in granites, ref., 89.
„ conservative action of ice-fringe,
ref., 61, 93.
„ “Volcanoes,” ref., 95.
Borchgrcvink, C. E., rocks from Cape Adare, 2.
„ „ „ Possession Ids., 4.
Borrodaile Island, described, 2,
Boulders, Auckland Islands, 97.
„ Ferrar Glacier, 79.
„ Knob Head, 38, 76, 77.
„ near Black Island, 80.
„ Winter Quarters, 87.
Brewster (Mount), volcano, 19.
Brogger, W. C., graphical representation of rocks,
ref., 121.
„ solvsbergites, ref., 114.
Brown Island, basalt, 15, 104.
„ described, 15.
„ moraines, 15.
„ phonolitic trachyte, 15, 115.
„ water-action, 90.
Buckle Island, 7.
„ active volcano, 3.
Bull Cliff, plionolite, anal., 121.
Burma, chondrodite in limestone, 124.
C, North-West Mountain, see Map.
Cd see Map and Fig. 20 (p. 42), and also Round Mt.
C3, see Obelisk.
C6, see Map and p. 42.
C7, see Map and p. 50.
C8, see Map and p. 42.
C9, see Map and West Fort.
Calc-schist, Mount Terror, 124.
Calcite, in basalt-tuffs, 110.
„ crystalline limestones, 124.
Calcium carbonate, crust on hollowed granite,
88, 89.
., „ deposit on boulders, 91.
Camel’s Hump, see Plate III, and Fig. 17 (p. 35).
Campanario, trachydolerite, anal., 119, 121.
Camptonite, 129, 139.
„ Cathedral Rocks, 129.
„ Montreal, 128.
„ New Harbour Height, 129.
„ near Koettlitz Glacier, 129, and Plato
IX, Fig. 5.
Cape Adare, etc., see Adare, etc.
Carbonaceous matter in sandstone, 40-42, 48, see
also Fossil.
Castle Rock, see Fig. 4 (p. 9) ; Fig. 36 (p. 61).
>, 11.
„ basalt, 12, 105,106.
„ basalt-tuffs, 109, 110.
„ described, 12.
„ frost-riven, 92.
„ hornblende - pseudomorphs, Fig. 59
(p. 103).
„ ice-fringe, 61, 62.
olivine-nodule, 108.
„ wind-action, 87.
Cataclastic structure, in gneiss, 125, 139, and Plate
IX, Fig. 1.
„ „ in granite, 126.
Cathedral Rocks (E!-E3), see Plates III, IV, V, VII
(Sections), and Fig. 14 (p. 29),
Fig. 17 (p. 35), Fig. 27 (p. 53).
„ 25, 38, 70, 98.
„ described, 30, 31, 35, 36.
„ banakite, 131, 133.
„ basalt from scree-slope, 106.
„ camptonite, 129.
„ diorite, 30, 36, 127, and Plate
IX, Fig. 2.
„ dolerite and granite, 30, 35 (Fig.
17), 36, 53.
„ frost-action, 92.
„ glacier, 71.
„ gneiss, 30, 31, 35, 36, 125.
„ granites and diorites, 30, 35, 36,
53, 127.
„ moraine, 79.
„ pegmatite, 31.
„ sandstone, 30, 35.
„ undercutting by water, 89.
Cavities, in granite boulders, 34, 87-89, and Figs.
16 (p. 34) and 50 (p. 87).
Cirques, 64.
„ Inland Forts, 73.
‘ Challenger ’ Expedition, dredgings, 2, 39.
Chamberlin, R. J., ice-caps, ref., 63.
Charnockites, India, 125.
Chemical action, in denudation, 91, 99.
Chemical relations of volcanic rocks, 119-123.
“ Chimneys ” of dolerite at Inland Forts, 50.
“ Chimney talus-shoot,” 92.
Chlorite in altered dolerites, 109, 128.
Chondrodite in crystalline limestone, 124, 139.
Chrome-diopside in nodules in basalts, 107.
Clarke, F. W., analysis of nepheline, 112.
„ analysis of diorite, 127.
Classification of rocks, 120, 121.
INDEX.
145
Clay, Auckland Islands, 97.
„ from sea-floor, 94.
„ Macquarie Island, 95.
Cliff-glaciers, G4, 72, 73.
Climate, Antarctic regions, 48.
Colbeck, Capt. W., discovery of Scott Islands, 3.
„ floes, ref., 57.
Colbeck (Cape), 68.
Cols at Inland Ports, 42, 73.
Columnar structure, in basalts of Auckland Islands,
96.
„ in dolerites of Ferrar Glacier,
49, 51, 52 (Pig. 26).
Concretions, ferruginous, in sandstone, 44.
Conglomerate, Black Island, 139.
Contact-effects, dolerite and granite, 52, 138.
„ dolerite and sandstone, 50, 138, and
Plate X, Figs. 2-5.
Continental Range, described, 21.
Cornish, Vaughan, forms of snow-dunes, ref., 84.
Cornwall, altered dolerites of, 128.
Corrie-glaciers, aggrading, not excavating, valleys,
93 (Fig. 56).
„ described, 64, 73.
„ Inland Forts, 73, 93, 94.
Corsica, cavities in granites, 89.
Cossyrite in basalt, 103.
„ in phonolitic trachytes, 114, 116.
Coalman Island, Fig. 3 (p. 5), Fig. 56 (p. 93).
„ 7, 17, 18, 101.
„ basalts, 5, 6, 101.
„ corrie-glacier, 93.
„ described, 5.
„ horizontal structure, 5.
„ piedmont-glacier, 66, 67.
„ quartz-grit from dredge, 39.
„ trachyte from dredge, 115.
Cracks in sea-ice, 59, 60.
Crater Hill, see Plate II.
„ 13, 19.
„ basalts, 12, 106.
„ described, 12.
„ glacier-ice fringe, 62.
“ Creep,” of sea-ice, 60.
Crevasses, in Ferrar Glacier, 70.
„ Mount Erebus, 70.
„ temperature of ice in, 86.
Cross, W., classification of rocks, ref., 120.
Crozier (Cape), see Fig. 5 (p. 10).
„ 68.
„ basalts, 10, 11, 104, 105.
„ basalt-tuffs, 109, 110.
Crozier (Cape), bombs, 1 1 .
„ boulders of sandstone, etc., 11.
„ limburgite, 11, 104.
„ phonolitic trachytes, 11, 114, 115.
„ stranded moraines, 81.
„ wind-effect on trachyte, 88.
Crozier Cliffs, basalts and trachytes, 10.
Crust-block, faulted, Royal Society Range, 53.
Crystalline Limestone, see Limestone.
Crystals of ice on fishing-line, 55 (Fig. 29).
„ of Glauber Salt, see Glauber Salt.
Current-bedding, in sandstone at B„ 41.
D, Dj-D*, see Map and Sections (Plate VII), and
also Kukri Hills.
D (hill), dolerite-talus, 92.
„ granite and dolerite, 36, 37, 38, 52.
„ hanging valley, 72.
„ undercutting by water, 89.
D, (hill), granite and dolerite, 37.
D„ (hill), see Plate III.
„ granite and dolerite, 36, 37, 38, 52, 138.
D3 (hill), see Plates III, IV.
„ granite and dolerite, 38.
D4 (hill), see Plates III, IV, and Fig. 14 (p..29).
„ augen-gneiss, 30, 38, 125.
„ crystalline limestone, 29.
„ granite and dolerite, 36, 38.
„ hauging valley, 72.
„ ice-cliff with rock-rein's, 76, 77.
„ moraines, 77.
I)-, D6a, see Map and Plate VII, and also Solitary
Rocks.
D5, 69, 70.
D6u, 70.
d3, see Map.
„ dykes, 35.
Dailey Islands, described, 15.
„ limburgite, 15, 106.
Dannenberg, A., felspathic nodules in basalts, ref.,
107.
„ sandstone altered by basalt, ref.,
105.
David, T. W. E., rocks from Possession Islands,
ref., 4.
Deas Head, columnar basalt, 96.
Debris-cones, McMurdo Sound, 81.
Dellbridge Islands, described, 16.
„ rocks of, 10.
Delta at Laurie Harbour, 97.
Denudation, South Victoria Land, 87-94, 99.
Dep6t Nunatak (A), Fig. 24 (p. 49), Fig. 25 (p. 50).
VOL. I.
U
L 4 (j
INDEX.
Depot Nunatak, 40, 69.
„ crevasses, 70.
„ dolerite, 49, 50, 136, 139, and
Plate X, Fig. 1.
„ mudstone-lent icles, 41.
„ sandstone in moraines, 40, 41.
„ talus from, 78.
„ up-thrust of morainic material, 82.
Descent Pass, 36.
„ hollowed granite-boulders, 87, 88.
„ moraines seen from, 78.
Diallage, in gabbro, 139.
Diatoms, sea-floor along Barrier, 94.
Differentiation, 121.
Diorite, 139.
„ Blue Glacier, 27, 128, and Plate IX, Fig. 3.
„ Cathedral Rocks, 30, 36, 127, and Plate
IX, Fig. 2.
„ „ anal., 127.
„ described, 125-128.
„ King Edward VII Land, 128.
„ Sweet Grass Creek, anal., 127.
Diorite-gneiss, Cathedral Rocks, 125.
‘ Discovery,’ at Cape Jones, 19 (Fig. 7).
„ at Macquarie Island, 95.
„ break-up of ice, 59 (Fig. 33).
„ Soundings along Barrier, 94.
1 Discovery ’ Expedition, geological work of, 98.
Discovery Gulf, 70, 71.
„ Glauber Salt, 91.
„ kenyte in moraine, 113.
,, rock -debris on ice, 79.
Discovery (Mount), 7, 15, and Fig. 32 (p. 58).
„ described, 20.
Distances, Table of, 7.
Distribution, of icebergs, 75.
Dolerite, at B„ 40, 41, 45, 49, 50, 138, and Plate
X, Fig. 5.
„ at b2, 46, 51.
„ at x, 46.
„ Beacon Heights, 46.
„ Cathedral Rocks, 30, 35 (Fig. 17),
36, 53.
„ contact with granite, 52, 138.
„ contact with sandstone, 49-54, 138, and
Plate X, Figs. 2-5.
„ Cornwall, 128.
„ Depot Nunatak, 49 (Fig. 24), 50 (Fig.
25), 136, 139, and Plate X, Fig. 1.
„ described, 49-54, 136—139.
„ dredge off Balleny Islands, 3, 135, 139,
140.
Dolerite, Dry Valleys, 45, 46, 136-138, and Plate
X, Figs. 2 and 4.
„ Finger Mountain, 40 (Fig. 19), 41, 45
(Fig. 22), 50, 51.
,, fragments in conglomerate, 139.
„ frost-action on, 92.
„ Granite Harbour, 32 (Fig. 15), 33, 54,
126, 138.
„ gullies produced by, 92.
„ in moraines, 11, 53, 81.
„ Inland Forts, 42, 43, 44, 50, 53, 138,
139, and Plate X, Fig. 3.
„ Knob Head Mountain, 46, 47, 49, 51,
52 (Fig. 26), 53, 137.
„ Kukri Hills, 36, 37 (Fig. 18), 38, 52, 53
(Fig. 27), 128, 136.
„ Macquarie Island, 96, 109.
„ Royal Society Range, 53, 99.
„ Solitary Rocks, 36.
„ South Victoria Land, 25, 49-54, 98, 99,
101, 136-140.
„ Terra Cotta Mountains, 47 (Fig. 23).
„ weathered surface of, 88.
„ with granitic patches, 138, 139, and
Plate X, Fig. 6.
„ formation, extent of, 53, 54, 140.
Drift-ice, 57.
Dry Valleys, see Plates V and VII.
„ 70.
„ dolerite and sandstone, 45, 46, 136,
137, 138, and Plate X, Figs. 2
and 4.
„ dolerite-talus, 92.
„ granite in moraines, 38.
„ moraines, 78.
Drygalski, E. von, ico-cracks, ref., 83.
„ „ ice-fringes, ref., 61.
„ „ Karajak ice-stream, ref., 78.
„ „ Kryolconit, ref., 93.
„ „ land-ice, ref., 63, 64.
„ „ salinity of sea-ice, ref., 55.
,, „ temperature of ice, ref., 86.
„ „ thawing of ice, ref., 85.
Drygalski Piedmont, 67.
Dune-snow, 84.
Dyke-rocks described, 27, 129-133.
Dykes, basalt at Auckland Islands, 97.
„ „ at Cape Adare, 17.
„ „ at Coalman Island, 6.
„ „ at Macquarie Island, 96.
„ „ at Dailey Islauds, 16.
„ dolerite at Bu, 51.
INDEX.
147
Dykes, dolerite at Finger Mountain, 51.
„ „ at Inland Forts, 50.
„ „ at Terra Cotta Mbs., 47 (Fig 23), 51 .
„ Granite Harbour, 33.
„ granite at Cathedral Rocks, 81.
„ trachyte at Observation Hill, 13.
E (hill), see Map and Sections (Plate VII), Fig. 14
(p. 29), and also Table Mt.
„ granite and dolerite, 36.
Ex — E3, see Map and Sections (Plate VII), and also
Cathedral Rocks.
E! (hill), see Plates III, IV, and Fig. 17 (p. 35).
„ diorite, 30.
E2 (lull), see Plate III and Fig. 17 (p. 35).
„ banakite, 131.
„ diorite, 127.
„ granite and gneiss, 30, 35, 36.
„ glacier, 71.
E4 (hill), see Map and Sections (Plate VII), and
also Snow Valley.
„ granite, 34.
„ kersantite, 34, 131.
e5 and e6, see Map and Sections (Plate VII), and
also Snow Valley.
e5, granite, 34, 126.
e6, dyke-rock, 133.
,, granite, 34, 126.
rju rj2, see Map, and also Snow Valley.
„ granite, 33.
Earth-movements, affecting sandstone, 99.
East Africa, rocks of Atlantic type, 102, 122, 140.
East Fork, see Plates IV, VII.
„ 28, 52, 69.
„ crevasses, 70.
„ ice-movement, 82.
„ moraines, 78.
East Groin, see Fig. 20 (p. 42).
„ dolerite, 50.
„ sandstone, 42, 43.
Eden (Mount), basalt, 96.
Emden, R., growth of ice-grains, ref., 68.
Emmons, A. B., analysis of phonolite, 115, 116.
Enderby Island, basalt, 97.
Englacial matter, 76-78.
Enstatite in olivine-nodules, 107.
Epidosite, Granite Harbour, 126.
Epidote, in banakite, 133.
„ in weathered trachyte, 114, 117.
Erebus Bay, 14, 110.
„ kenytesfrom islands, 14, 16, 110-113.
Erebus Cove, clay and boulders, 97.
Erebus (Mount), see Fig. 4 (p. 9).
» 7,10,16,20,22,95.
„ active volcano, 98.
„ crevasses, 70.
„ described, 8-10.
„ glacier-ice fringe, 62.
„ ice-cap, 65.
„ kenytes, 9, 14,102, 110-113, 123.
„ moraines, 81.
‘ Erebus ’ and ‘ Terror ’ Expedition, 1.
Essexite, Blue Glazier, 128.
„ Snow Valley, 126.
Essexose, 120.
Etna (Mount), von Waltershausen, ref., 9.
Evans, J. W., rounded quartz in gneiss, ref., 124.
False-bedding, in sandstone at B4, 41.
Fast-ice, 57, 60.
Faulting, in dolerite at Bx, 50.
„ (tangential), 123.
Fearnsides, W. G., 100.
Felspar, in altered dolerite, 128.
„ basalt-tuff, 109, 110.
„ contact-dolerite, 138.
„ nodules in basalts, 107-109.
„ sandstones, 134.
„ schists, 135.
Fernando Noronha, 123.
Ferrar Glacier, see Plates III, IV, V, and Fig. 14
(p. 29), Fig. 43 (p. 78).
„ 27, 28, 30, 32, 34, 35, 42, 45, 46,
50, 53, 54, 66, 71, 73, 74, 76,
92, 99, 126, 138.
„ boulders, 79.
„ change from snow to ice, 85.
„ cracks, 83.
„ described, 69, 70.
„ ice-falls, 69, 83.
„ Lieut. Armitage’s journey, 39, 49,
85.
„ moraines, 70, 78 (Fig. 43).
„ rate of movement, 82.
Ferrar, H. T., movement of Blue Glacier, ref., 70.
„ pinnacled ice, ref., 15, 90.
„ relative ages of basalt and trachyte,
123.
„ Report on Field-geology, 1-99.
„ rock-specimens collected by, 101.
Ferrera, “tephritic trachyte,” 118.
Field-geology, Report on, 1-99.
Field-ice, 57, 60.
Finckh, L., analysis of kenyte, 113.
U 2
148
INDEX.
Finckh, L„ rhomb-porphyries from Kilimandjaro,
ref., 111-113.
Finger Mountain (B), see Plates Y, VII, and Fig.
19 (p. 40), Fig. 22 (p. 45).
„ 50, G9.
„ crevasses, 70.
„ . dolerite and sandstone, 44, 45,
50, 51.
„ grit, 124, 134.
Finland, chondrodite in limestone, 124.
Fiords, Auckland Islands, 96.
Firnfield, 63, 64.
Firnmulden, 63.
Floating piedmonts, see Piedmonts-afloat.
Floe-ice, 56, 57, 60 (Fig. 34).
Flowers (ice-), 56.
Fluidal structure, in basalts, 104.
Folding of rocks, 6, 18, 31.
Foothills, between Cape Adare and Cape Washing-
ton, 22.
„ Cape Jones, 18.
„ Mount Morning, 20.
„ Royal Society Range, 23.
„ S. of Cape Bernacchi, 20.
See also Southern Foothills, Northern
Foothills.
Foraminifera, sea-floor along Barrier, 94.
Formulae (Osanu), of rocks, 121, 127, 130, 132,
137.
Forodada, “ tephritic trachyte,” 117, 118.
,, ,, ,, anal., 1 1 9.
Fossil plant-remains, in sandstone, 40-42, 43, 46-
48, 99, 139. •
Franklin Island, 7, 101.
„ basalt, 1, 6.
„ described, 6.
„ limburgite, 6, 101, 107.
„ olivine-nodules, 107.
„ stranded moraines, 81.
Fringe, of glacier-icc, 61, 62.
Frost-action, 92, 99.
G, G„ G2, G.„ see Map, and also Northern Foothills.
G (hill), see Figs. 14 (p. 29) and 54 (p. 90).
G] (hill), kersantite-dyke, 27, 130, 135.
„ limestone, 94.
„ micaceous schist, 135.
„ perched blocks, 94.
Go (hill), see Plate IV, and Fig. 14 (p. 29).
’ „ 33.
G2 (hill), gneiss, 27, 28, 30, 34.
„ ice-cracks, 83.
G3 (hill), see Map and Plate IV.
,, banakite-dyke, 35, 131.
„ gneiss, 27, 28.
„ granite, 33, 34, 35, 36, 38.
„ undercutting by water, 89.
G4 (hill), crystalline limestone, 26, 27 (Fig. 12).
g (hill), see Map and Plate IV.
„ 27.
■ylt see Map and Plate V.
„ 50, 70.
y2, see Map.
„ dolerite and sandstone, 42.
y3, see Map.
„ dolerite and sandstone, 50.
Gabbro, Granite Harbour, 126.
„ Hurricane Ridge, anal., 127.
„ moraine off White Island, 139.
Gabbro-like nodules, in basalts, 107-109.
Gap (The), see Plate II.
„ 13.
„ basalt, 104-106, 119-122 (Fig. 67), 132, and
Plate VIII, Fig. 1.
„ „ anal., 105, 119, 120.
„ calcium carbonate on boulders, 91.
„ nodule in limburgite, 108.
Garnet, in sandstones, 134.
Garwood, E. J., ice-sheet, ref., 63.
„ Ivory Glacier of Spitsbergen,
ref., 78.
„ up-thrust of morainic material,
ref., 82.
Gas-bubbles, in augites and felspars, 107, 108.
Gauss (Cape), piedmont, 66, 67.
Gauss (Mount), 7, 140.
Geikie, A., land-ice, ref., 63.
George Murray (Mount), 22.
German Antarctic Expedition, 140.
Glacier-ice, fringes, 61, 62. *'
„ McMurdo Sound, 90 (Fig. 54).
Glacier-table, Fig. 44 (p. 79).
Glaciers, 93.
„ Admiralty Range, 71.
„ Cape North, 71.
„ Coulman Island, 93 (Fig. 56).
„ Granite Harbour, 71.
„ Kukri Hills, 37.
„ New Zealand, 78.
„ nomenclature, 63, 64.
„ of Alpine type, 63, 71, 72.
„ of Greenland type, 63, 69, 70.
„ of Norwegian type, 63, 70, 71.
„ piedmonts, 63, 65-69, 74, 99.
INDEX.
149
Glaciers, Prince Albert Mts., 69.
„ Robertson Bay, 71.
„ Round Mountain, 73.
„ rock -debris in, 76.
„ Switzerland, 78.
Glaciers remanies, 64.
Glass, in basalt, 104, 105.
„ in contact-dolerite, 138.
„ in kenytes, 111.
Glauber Salt, deposits of, 11, 91.
„ water-determination, 91.
Glaze, on weathered stones, 87.
Gneiss, 98, 101, 139.
„ Barne Inlet, 26.
,, Blue Glacier, 70.
„ Cathedral Rocks, 30, 31, 35, 36, 125.
„ ‘ Challenger ’ dredgings, 2.
„ Greenland, 124, 125.
„ King Edward VII Land, 24, 128.
„ Kukri Hills, 30, 38, 125.
„ Madras, 136, 140.
„ New Harbour Height, 28 (Fig. 13), 29.
„ Northern Foothills, 27, 28, 30.
„ Royal Society Range, 25.
Gneisses, described, 25-31, 124-125.
Goller, E., analysis of kersantite, 130.
Granite, 25, 98,99, 101, 131, 139.
„ Adelie Land, 1, 2.
„ Barne Inlet, 26.
„ boulders at Knob Head, 51.
,, boulders in moraines, 78, 81, 91.
„ Cape Adare, 2, 32, 125.
„ Cape Crozier, 1 1.
„ Cathedral Rocks, 30, 35 (Fig. 17), 36,
53, 127.
‘ Challenger’ dredging, 2.
contact with dolerite, 52, 138.
described, 32-38, 125-128.
dyke at J), 25.
Granite Harbour, 32 (Fig. 1 5), 33, 125, 126.
hollowed, 34 (Fig. 16), 87 (Fig. 50), 88.
King Edward VII Land, 24, 128.
Kukri Hills, 36, 37 (Fig. 18), 38, 53
(Fig. 27), 128.
Mount Terror, 32, 128.
New Harbour Height, 129.
Possession Islands, 1, 4.
Royal Society Range, 32, 81.
screes from, 92.
Snow Valley, 33, 34, 126.
Solitary Rocks, 36.
South Arm, 36.
Granite, Southern Foothills, 25, 126.
Granite Harbour, see Fig. 15 (p. 32).
„ 54.
„ basalt-dykes, 33.
„ described, 32, 33.
„ dolerite, 32, 33, 54, 126, 138.
„ gabbro, 126.
„ glaciers, 71.
,, gneiss, 126.
„ granites, 32, 33, 125, 126.
„ ice-fringe, 61.
„ micaceous schist, 135.
„ pegmatite, 126.
„ perched blocks, 94.
„ piedmont, 66.
„ quartz-porphyries, 126.
„ sandstone, 126.
Graphite, in crystalline limestones, 124.
Gravels, Macquarie Island, 96.
„ 89 (Fig. 52).
Graveneck, hornblende-diabase, anal., 105.
Great Ice Barrier, see Barrier.
Great Rift Valley, Atlantic type of volcanic rocks,
122.
Greenland, glaciers, 69.
„ gneiss, 124, 125.
„ ice-cap, 65.
„ inland-ice, 64.
Gregory, J. W., ice-sheet, ref., 63.
„ kenytes from Mt. Kenya, ref., 111.
„ Pacific Coast of South Victoria
Land, 122.
„ up-thrust of morainic material,
ref., 82.
,, volcanoes of South Victoria Land,
ref., 18, 39.
Grit, Finger Mountain, 124, 134.
Gullies, in sandstones at x, 92.
II (hill) see Map and Sections (Plate VII), and
also New Harbour Height.
„ gneiss, 29, 30.
Ha (hill), see Map, 30.
b4 (hill), see Map, 30.
Haddington (Mount), 7.
Haggitt’s Pillar, 3.
Hail, 83.
Hamberg, A., structure of sea-ice, ref., 55.
Hangegletscher, 64.
Hanging glaciers, 64, 73.
Hanging valleys, 37, 72, 73.
Harbour Heights, see Plate II.
21
150
INDEX.
Harbour Heights, basalt, 12, 13, 104, 106, 108.
„ bombs, 12, 106.
„ described, 12.
„ limburgite, 12, 13, 107.
„ nodules in basalt, 107, 108.
Harker, A., igneous rocks of Skye, ref., 136.
Hiirtlingen, hornblende-basalt, anal., 105.
Hayes Peninsula, local ice-cap, 63.
Hector, basalts of Auckland Islands, ref., 96.
Heim, A., air in glacier-ice, ref., 68.
„ icebergs, ref., 73.
„ ice-cracks, ref., 83.
„ nomenclature of land-ice, ref., 63, 64, 67.
„ temperature of ice, ref., 86.
Heldburg, limburgite, 121.
Hessose, 128.
Hochlcmdeis, 63.
Hodgson, T. V., Glauber Salt, 91.
„ ice-crystals on fishing line, 55
(Fig. 29).
„ rocks of Inaccessible Island, 16.
„ rocks from Sultan’s Head, 11.
„ rocks from Y-Cliffs Hogsback, 10.
„ tuffs of Turk’s Head, 9.
Holland, T. H., “ augite-diorite ” of Southern
India, ref., 136.
Hollowed granite boulders, see Granite (hollowed).
Horizontal structure, at Cape Adare, 18.
„ Inland Forts, 43.
„ Mt. Nansen, 21 (Fig. 8), 39»
54.
„ Royal Society Range, 39, 98.
Horn (Cape), 7.
Hornblende, in banakites, 35, 131-133.
„ in altered dolerite, 128.
., in basalts, 102, 104, 122.
„ in camptonites, 129.
„ in diorite, 127, 128.
„ in gneiss, 124, 125.
„ in granite, 126.
„ in kersantites, 130, 131.
„ in nodules in basalts, 107, 108.
„ in phonolites, 116.
„ in trachytes, 117, 118, 122.
Hornblende-basalt, analyses, 103.
„ described, 102-104. See also
Basalts.
“ Hornblende-diabase,” Graveneck, anal., 105.
Hornblende-pseudomorphs, see Pseudomorphs.
Hornblende-schist, 125.
Hornblende-trachytes, Observation Hill, 117-119.
See also Trachytes (phonolitic).
Horseshoe Bay, basalts, 106, 120.
Horseshoe Mount, 22.
Huggins (Mount), Fig. 9 (p. 23).
Hummocks, 57, 59, 82, 85.
Hundskopf, nepheline-basanite, 121.
Hurricane Ridge, gabbro, anal., 127.
Hut Point, see Plate II and Figs. 32 (p. 58), 35
(p. 61).
„ 12.
„ cracks in sea-ice, 59.
„ ice-foot, 61.
„ incrustation of Glauber Salt, 91.
„ water-holes at, 58.
Hutton Cliffs, basalt-tuff, 11, 109, 110.
Hutton, F. V., volcanoes of Auckland, ref., 12.
Hyalopilitic structure, in basalts, 104.
Hybrid rock, Granite Harbour, 138, 139, and
Plate X, Fig. 6.
Hyland, J. S., bibliography of hornblende-pseudo-
morphs, ref., 103.
Hypersthene, in diorite, 127.
„ in olivine-nodule, 108.
Ice, depressed by snow, 85 (Fig. 49).
„ retreating, 94, 99.
„ of glacier-ice fringes, 62.
„ in Ross Sea area, 99.
See also Sea-ice, Pinnacled ice, Land-ice, etc.
Ice-action, in denudation, 93, 94, 99.
Ice-barrier, see Barrier.
Ice-belt, McMurdo Sound, 66.
Iceberg, tilted, Fig. 40 (p. 74).
Icebergs, 64, 73-75, 94.
„ Balleny Islands, 74.
„ Blue Glacier, 73, 94.
„ Cape Adare, 75.
„ distribution, 75.
„ King Edward YII Land, 24, 75.
„ melting of, 75.
„ Ross Island, 65.
„ Ross Piedmont, 74.
„ size of, 75.
„ South Victoria Land, 73, 75.
Ice-caps, 63, 65, 99.
Ice-crystals, 83.
„ on fishing fine, 55 (Fig. 29).
Ice-falls, Ferrar Glacier, 69, 83.
„ Mount Erebus, 65.
Ice-field, 60.
Ice-flowers, 56.
Ice-foot, Arctic Regions, 61.
„ Granite Harbour, 61.
INDEX.
151
Ice-foot, Hut Point, Fig. 35 (p. 61).
„ Wood Bay, Fig. 37 (p. 05).
Ice-fringe, Castle Rock, Fig. 36 (p. 61).
„ conservative action of, 61, 62.
Ice- gauge, 5!).
Ice-sheet, 63, 64.
Ice-slabs, 64, 99.
„ aggrading, not excavating valleys, 93.
„ englacial matter in, 78.
„ Southern Foothills, 71, 73, Plate VI,
and Map.
Ice-streams, 63, 65, 74.
Ice-tongues, in Erebus Bay, 14.
„ North Fork, 72 (Fig. 39).
Ice-without-grain, at Knob Head, 82, and Fig.
42 (p. 77).
Iddings, J. P., shoshonite, ref., 130.
,, classification of rocks, ref., 120.
„ gabbroof Hurricane Ridge, ref., 127.
Ilmenite, in basalts, 104.
., in dolerites, 136.
Inaccessible Island, basalt, 16, 106.
„ described, 16.
„ Glauber Salt, 91.
„ phonolitic trachytes, 16, 114,
115.
„ trachydolerite, 16.
Inclusions, coarse-grained, in basalts, 12, 13, 106-
109.
„ „ in banakite, 131.
„ „ in trachyte, 114.
„ hornblendic, in trachytes, 13, 118
(Fig. 66).
„ in anorthoclase, 112.
„ in minerals of nodules in basalts, 107-
109.
„ in quartz of sandstone, 134.
India, charnockites, 125.
Inland Forts, see Fig. 20 (p. 42).
„ corrie-glaciers, 73, 93, 94.
„ dolerite and sandstone, 40, 42-44, 50,
53, 138, 139, and Plate X, Fig. 3.
„ fossils (?) in sandstone, 43, 48.
„ moraines, 73.
Inland-ice, 39, 42, 54, 63, 64, 69, 70, 84, 99.
Inlets, 64.
Ishawooa Canyon, banakite, anal., 132.
Ivory Glacier, Spitzbergen, 78.
Jj (hill), see Map and Sections (Plate VII) and
Figs. 1 1 (p. 26), 54 (p. 90).
„ crevasses in Blue Glacier, 71.
J! (hill), crystalline limestone, 25, 26.
„ granitic dyke, 25.
Jenssen, Captain, rocks from Possession Island, 4.
Johnson, W. D., “plucking” of rock-masses, ref.,
94.
Joints, in granite, 33.
Jones (Cape), Fig. 7 (p. 19).
Judd, J. W., Volcanoes, ref., 95.
Karajak ice-stream, 78.
Kennar (P. O.), 40.
Kenya (Mount), kenytes, 102, 111-113, 140.
Kenyte, Cape Rovds, 9, 111-113, 132, and
Plate VIII, Fig. 3.
„ „ anal., 113, 119-122.
„ islands in Erebus Bay, 14, 16, 110-113.
„ moraine in Discovery Gulf, 113.
„ Mount Erebus, 102, 123, 140.
„ Mount Kenya, 102, 111, 112, 140.
„ „ anal., 113.
„ Mount Kilimandjaro, 102, 111, 112, 140.
„ „ anal., 113.
„ Skuary, 9, 10, 111.
„ Tent Island, 16, 111.
,, Turtle Back Island, 14, 110 (Fig. 60), 111
(Fig. 63).
„ wind-effect on, 88.
Kenytes, described, 110-113.
Kersantite, dyke at E4, 34, 131.
„ Northern Foothills, 27, 130, and Plate
IX, Fig. 4.
„ „ anal., 130.
,, Stengerts, anal., 130.
Kersantites, described, 130, 131.
Kibo, analysis of kenyte, 113, 121.
Kilimandjaro (Mount), kenytes, 102, 111-113, 140.
King Edward VII Land, Fig. 10 (p. 24).
>> * ‘
„ „ described, 24.
„ „ icebergs, 24, 75.
„ „ pebbles of granite,
etc., 24, 128, 135.
Knob Head Mountain (B9), see Plates V, VII, and
Figs. 14 (p. 29), 26
(p. 52), 28 (p. 54), 41
(p. 76), 42 (p. 77).
,, ,, 37, 6,h
„ „ boulders of granite, etc., 38,
51, 7 6, 77.
„ „ dolerite and sandstone, 40,
46, 47, 49, 51—53, 137.
„ „ frozen pond, 82.
IN DUX.
Knob Head Mountain, ice-ridges, 83.
„ „ icc-without-grain, 77 (Fig.
•12), 82,
„ „ moraines, 76 (Fig. 1 1), 78.
„ „ uplift of morainic maturial,
7(i, 82.
Koettlitz, l)r. It., rocks from llluo Glacier, 27, 128.
Koettlitz Qlonior, see. I’lato VI.
„ 20, 71.
,, camptonites, 1 2D.
KryokonU, 1)8.
Kukri 1 1 i I Ih ( I >, Dl-D4), see Platon III, IV, VII,
and Figs. I I (p. 21)), 18 (p. 37),
27 (p. 58).
„ 25, 70.
„ crystalline limestone, 28-80, 88.
„ doBoribod, 28-80, 86-88.
„ dolorito and sandstone, 86-88, 52, 58,
128, 186.
,, dolorito-talus, 1)2.
,, glaciorB, .87.
„ gneiss, 80, 88, 125.
„ grauito and dolorito, 86-88, 68, 128.
„ banging valleys, 72.
„ water-channels, 89.
„ (Lower), 29, 80, 69, 70.
Lubriulorito, in banakito, 181.
„ in basal tH, 102, In I, 106, 109.
„ in camptonites, 129.
„ in diorito, 127.
„ in dolorito, 186, 1.87.
„ in gabbro, 189.
Lacroix, A. F. A., “ quartz tie common," ref., 125.
Lady Newues Hay, see Fig. 7 ( p. 19).
„ piodniont-alioat, 18, 19, 67.
Lamprophyres, in I, 129- 181, 189.
I sind-ieo described, 68 86.
Lnspoyres, K. A. II., folspathic nodules in basalts,
ref., 107.
Laurdalose, 120.
Laurie Harbour, delta, 97.
Laurvikose, 120.
Leiieile, in glassy trachytes, 117, 118 (Fig. 61).
„ in kenyto, 111 — 1 1.9.
Lcueite-konyte, Gape Royds, see Kenyte.
Liniburgito, 101, l to.
„ Cape Orozior, II, 104.
„ Dailey Islands, 15, 106.
„ Franklin Island, 6, 101, 107.
„ Harbour Heights, 12, 18, 107.
„ Ridge Road, anal., 105.
Limburgite, Turtlo Rack Island, 106,
Limbiirgoso, 120.
Limestone, band in sandstone, 1 1 .
,, dredged by ‘Challenger’ 2, 89.
„ (crystalline), 101, 189.
„ „ denudation, 92.
„ „ described, 25-81, 124.
„ „ Kukri Hills, 29, 80, 88.
,, „ Northern Foothills, 26,
27 (Fig. 12), 91, 121,
180, 188.
., „ Southern Foothills, 25,
26 (Fig. II), 124
(Fig. 69).
Liquid inclusions in nodules in basalts, 107, 108
(Fig. 60).
„ in quartz of sandstone, 134.
Lister (Mount), see Plate V.
Little Razor Back Island, basalt, 16, 106.
„ „ described, 16.
„ „ phonolithio trachyte, 16,
Lobotlon rarcino/>haf/us, in moraines, 79.
Local ice-caps, see Ice-caps.
LongslalV (Mount), 7.
Lower Kukri Hills, see Kukri Hills (Lower).
Lusitania Anchorage, Macquarie Island, 95.
m (hill), see Map and Sections (Plate VII) and
Fig. 1 I (p. 29).
„ crystalline limestone, 29.
„ granite and gneiss, 88.
Macquarie Island, see Fig. 57 (p. 95).
„ 97, 101.
„ basalts and dolcrites, 96, 109.
„ described, 95, 96.
Madagascar, cavities in granites, ref., 89.
Madras, “ augite-dioritos" 186, I 10.
Miidstein, phonolito, 121.
Magnetite, dendritic, 1 18 (Fig. 65).
„ in banakito, 181.
,, in basalts, 102—104, 106, 109.
„ in basalt-tuff, 109, 110.
,, inclusions in olivine, 108 (Fig. 61).
„ in contact-dolcrites, 138.
„ in dolerites, 136.
„ in phonolito, 116.
„ in phonolitio trachytes, 11 1, 116, 117.
,, in spherules of glassy kenyto, 1 1 1 (Fig.
63).
Malaspina Glacier, piedmont, 66.
Marble-like sandstone, West Groin, 1 1, 131.
INDEX.
I f).'!
“ Marbled " snow, Fig. IS (j>. 81).
Marjolou Soo, 82.
Markham, Sir A., on soa-ice, vof., .r»T.
McCormick, Ur. R., 21.
McCormick (Cape), volcanic cones, I’.*.
McMurdo Sound, 15, 57.
,, Climber Salt in moraine-cone, 91.
„ hummocks, 67.
„ melting of ice, 60.
„ moraines, 66, 7!', 80 (Fig. 45),
81 (Fig. 17), 88, 81), 1)1, l)!!.
,, piodmont-on-land, fill.
„ saline ponds in moraines, 88 (Fig.
51), 89.
„ sea-ice forced up on land, 98.
,, snow on sea-iee, 66.
„ thickness of ice, 59.
„ water-channels, 89, 90 (Figs. 58
and 54).
Melbourne (Mount), see Fig. 87 (p. 66).
„ 7,39,140.
„ basalt, 20.
,, centre of ice-shedding, 66.
,, described, 19, 20.
., fault, 22.
„ iee-eap, 75.
„ ice-foot, 65.
„ pumice, 20, 93.
Meteorite, Sierra do Chaco, 109.
Meteorites, similarities with nodules in basalts,
107, 109.
Miaskose, 120.
Miauno (Mont), phonolitc, anal., 115, 1 16, 121.
Mica, Cathednd Rocks, 31.
Micaceous schists, 185.
Mica-schist, from ‘ Challenger’ dredging, 2.
Mieroeline, in granite, 128.
„ in grit, 124, 184.
Micropcgmatito, in augen-gneiss, 125 (Fig. 70)
„ in doloritos, 186 (Fig. 72), 140.
Millstone-grit, like Beacon Sandstone, 39.
Minna Blulf, see Fig. 52 (page 89), 122.
„ basalt, 21, 106.
„ described, 20, 21.
„ moraines, 69, 89.
„ movement of barrier, 69, 82.
„ phonolitc and trachyte, 21, 116.
„ radiating cracks, 83.
Minto (Mount), 22.
Mirabilito, see Glauber Salt.
Mittelgebirge, volcanic rocks of, 121, 122.
“ Mode,” 131.
Molecular percentages of basal I sand trachytes, 120.
Montreal, camptonites, 128.
Moraine-cones, near White Island, 80 (Fig. 16).
., McMurdo Sound, 8.1 .
Moraines, Barrier, 69.
„ Black Island, 69, 79, 80, 91.
„ Brown Island, 16.
„ Cape Adare, 80.
„ Capo Crozier, 81.
„ Cape Boyds, 81.
„ Cathedral Rocks, 79.
„ dolerites in, 53, 81.
„ Dry Valleys, 38, 78.
East Fork, 78.
„ Ferrar Glacier, 7 0, 78 (Fig. 13).
,, granites in, 38, 78, 81, 91.
Hill Ih, 77.
„ Inland Forts, 73.
,, Knob Head, 76 (Fig. II), 78.
,, McMurdo Sound, 66, 79, 80 (Fig. 15),
81 (Fig. 17), 88 (Fig. 51), 89, 91, 93.
,, Minna BlulT, 69, 89 (Fig, 62).
„ Mount Erebus, 81.
,, Mount Terror, 81.
,. North Fork, 78.
„ re-sorted, 80.
„ Skuury, 9.
„ South Arm, 78.
„ Terra Cotta Mts., 17.
,, White Island, 79, 80, 91.
‘ Morning,’ at Franklin Island, 6.
„ at Possession Islands, I.
„ specimens from Scott Islands, III.
Morning (Mount), see Plate VI.
„ described, 20.
Morrison, .1, I)., basalts from Franklin Island, 6.
„ rocks from Possession Islands, I.
„ rocks from Scott Islands, 3.
Mount Adam, etc., see Adam, etc.
Movement, of ice, 69, 82, 83.
Mudstone, lonticlcs in sandstone, 1 1.
Miigge, .1. 0. O., alteration of hornblende info
olivine, ref., lol.
Muloek Inlet, ice -pressure, 83.
Murray (Mount George), 22.
,, Sir .1,, dredgings by ‘Challenger,’ ref., 39
Nansen, F., ice-sheet, ref., 68.
„ temperature of ice, ref., 86.
Nansen (Mount), sec Fig. 8 (p. 21), 22.
„ horizontal structure, 21, 89, 51.
Natlmrst, A. 0., Antarctic fossil Horn, ref., 48.
VOI
154
INDEX.
Nepheline, in basalts, 106.
„ in kenytes, 112.
„ in phonolites, 116.
,, in phonolitic trachytes, 1 14-116, 119.
„ Litchfield, anal., 112.
Neumayer (Mount), piedmont-afloat, 67.
New Harbour Height (H), 28 (Fig. 13), 29, 30.
„ camptonite and granite, 129.
New Mount (B5), see Map and Plates Y, VII.
New Zealand, 95, 98.
„ glaciers, 78.
Nodules, coarse-grained, in basalts, 12, 13, 106-
109, 140, and Plate VIII, Fig. 2.
Nodules, ferruginous, in sandstone, 41.
Nordenskiold Piedmont, 67.
„ „ soundings, 94.
‘ Norm ’ of banakite, 132.
„ of basalts and trachytes, 120.
„ of diorite, 128.
„ of dolerite, 137.
„ of kersantite, 130.
North (Cape), 21, 22.
„ glaciers, 71.
„ pyramidal peaks, 22.
North Fork, see Sections (Plate VII), and Fig. 39
(p. 72).
„ 28, 69, 70.
„ ice-tongues, 72.
„ moraine, 78.
Northern Foothills (G, G1; G2, G4), see Plate IY
and Fig. 14 (p. 29).
„ 29, 30, 33, 66, 69, 70, 94.
„ banakites, 35, 131, 133, and
„ Plate IX, Fig. 6.
„ banakites, anal., 132.
„ crystalline limestones, 26,
27 (Fig. 12), 94, 124, 130,
133.
„ described, 26-28.
„ kersantites, 27, 130, and
Plate IX, Fig. 4.
„ perched blocks, 94.
North-weA Mountain (C), see Map.
North-west Nunataks, 69.
Norway, rhomb-porphyries, 111, 140.
Nunataks, 49, 64, 69.
Obelisk (C3), see Sections (Plate VII) and Fig. 39
(p. 72).
„ 53.
Observation Hill, see Plate II.
1 9 1 23
55 ^ J
Observation Hill, basalt, 13.
,, described, 13, 14.
,. trachytes, 13, 117-120, and
Plate VIII, Fig. 6
„ wind-action, 87, 88.
Oligoclase in banakite, 133.
„ in camptonite, 129.
„ in gneiss, 124, 125.
„ in granite, 126.
„ in kersantite, 130.
Olivine in basalt-tuff, 109, 110.
„ in basalts, 102, 104, 106, 109, 122.
„ in gabbro, 136, 139.
„ in kenytes, 111, 113.
„ in nodules in basalts, 107-109.
„ in phonolitic trachytes, 114, 122.
„ with magnetite-inclusions, 108 (Fig. 01),
109.
Olivine-basalts, described, 104-106 ; see also Basalts.
Olivine-nodules, in basalt of Franklin Island, 6.
„ in limburgite of Cape Crozier, 11.
„ origin of, ref., 109.
„ Turtle Back Island, 14, 108
(Fig. 60).
Orthoclase in banakite, 131.
„ in gneiss, 124, 125.
„ in granite, 33, 34, 126.
„ in kersantite, 130.
„ in phonolite, 116.
Orthogneiss, 124, 139.
Osann, A., classification of rocks, 120, 121.
„ formulas of rocks, see Formulas.
Pacific Ocean, volcanoes encircling, 122.
Pacific type of volcanic rocks, 122, 140.
Pack-ice, 56 (Fig. 30), 57 (Fig. 31), 59, GO.
„ Wood Bay, 65 (Fig. 37).
Paligonite-tuffs, 1, 4, 5, 11, 12, 109, 110.
Pancalce-ice, 57.
Pantelleria, anorthoclase, 112.
Peat, Auckland Islands, 97.
Peat-bogs, Macquarie Island, 96.
Pebbles in Beacon Sandstone, 41, 134.
Pecten, shell in gravel on Ferrar Glacier, 79.
Pegmatite, Cathedral Rocks, 31.
„ Granite Harbour, 126.
Perched blocks, 94.
Petrograpkical province, 122, 140.
Philippi, E., description of Gaussberg, ref., 140.
Phlogopite in crystalline limestone, 124.
Phonolite, Black Island, 116, and Plate VIII,
Fig. 5.
INDEX.
155
Phonolite, Bull Cliff, anal., 121.
„ Cape Adare, 116.
„ Minna Bluff, 21, 116.
„ Mont Miaune, anal., 115, 116, 121.
„ Possession Islands, 1.
Phonolites, described, 116.
., in 1 Southern Cross’ collections, 102.
Phonolitic trachytes, see Trachytes.
Picotite, in nodule in basalt, 108.
„ in olivine of basalt, 101.
Piedmont-glaciers, 63, 65-69, 71, 99.
Piedmont-ice, structure of, 67, 68.
Piedmonts-afloat, 63, 67-69, 91.
„ icebergs from, 7 1.
„ Lady Newnes Bay, 18, 19, 67.
„ Sturge Island, 3.
Piedmonts-nground, 63, 66, 67.
„ Coulman Island, 5.
„ icebergs from, 71.
„ Sturge Island, 3.
Pilotaxitic structure, in basalts, 101.
Pinnacled ice, 15, 80, 90.
“ Pipe ” of diorite, 127.
Pirsson, L. Y., classification of rocks, ref., 120.
Plant-remains, see Fossil.
Plateau of South Victoria Land, 21, 98.
Plateau-dolerite, Ferrar Glacier, 10.
Plateau-features, 21, 51, 99.
,, Royal Society Range, 23 (Fig. 9),
39, 53.
„ Mount Nansen, 21 (Fig. 8).
See also Horizontal Structure.
“ Plucking ” of rock-masses, 91.
Polar ice-cap, 63.
Possession Islands, see Fig. 2 (p. 1).
„ L7. "
,, described, 1, 5.
„ stranded moraines, 81.
Potholes in granite, 89.
Pram Point, hummocks, 59, 82.
Prince Albert Mts., described, 22.
„ glaciers, 69.
„ ice-slopeSj 66.
„ ice-streams, 71.
„ inland-ice, 61.
Prior, G. T., 100.
„ analysis of kenyte from Mt. Kenya,
ref., 113.
„ Atlantic and Pacific types of volcanic
rocks, ref., 102, 122.
„ basalt of Franklin Island, ref., 1, 6,
107.
Prior, G. T., basalt of Mouut Melbourne, ref., 20.
,, basalt of Mount Terror, ref., 10.
„ basaltic dykes at Cape Adare, ref., 17.
„ basalts from Auckland Islands, ref.,
96, 109.
„ dredgings of ‘ Challenger,’ ref., 39.
„ Glauber Salt, anal., 91.'
„ rocks of Coulman Island, ref., 5.
„ rocks of Great Rift Valley, ref., 116,
122.
„ rocks from Possession Islands, ref., 1.
„ Report on the Rock-specimens, 101—
142.
„ ‘ Southern Cross ’rock-collection, ref.,
2, 101, 102.
Pseudomorphs, after hornblende, 102, 103 (Fig.
59), 104.
Pumice, Mount Melbourne, 20, 93.
Pyramidal peaks, Admiralty Range, 54.
Pyroxene-granulite, Madras, 136, 140.
Quartz, angular fragments in sandstone, 46.
,, in diorite, 127.
„ in dolerite, 136.
„ in gneiss, 124, 125.
,, foreign inclusions in camptonite, 129.
“ Quartz de corrosion ,” in gneiss, 125.
Quartz-grains, in Beacon Sandstone, 134 (Fig. 71).
Quartz-grit, Finger Mountain, 124, 134.
„ in dredge off Coulman Island, 39.
„ King Edward VII Land, 135.
Quartz -porphyries, 139.
„ Cathedral Rocks, 31.
,, Granite Harbour, 126.
Quartz-schist, pebble in sandstone, 41, 134.
Quartz-veins, in granite of Snow Valley, 34.
Quartzite, pebbles in sandstone, 41.
,, seams in Beacon Sandstone, 134.
„ Terra Cotta Mts., 47.
Raised beach ?, Macquarie Island, 95.
Razor Back Island, anticline, 123.
„ basalt, 16.
„ described, 16.
Re-cemented glaciers, 64.
Relief ships, 60 (Fig. 34).
Report on Field-geology, 1-99.
,, Plant-remains, 48.
„ Rock-specimens, 101-142.
Re-sorted moraines, 80.
Rhomb-porphyry, 102, 110-113, 140 ( see also
Kenyte).
Ridge Road, limburgite, anal., 105.
15G
INDEX.
Riebeckite, in phonolitic trachyte, 11G.
Rift Yalley, phonolitic rocks, 11G.
Rink, H., land-ice, ref., 63, 67.
„ sea-ice, ref., 57.
Robertson Bay, glaciers, 71.
„ slates, 2.
„ valley, 22.
Roches moutonnees, 94.
Rock -debris, at Winter Quarters, 87.
„ englacial, 76, 77.
Rock-flour, deposit in Ross Sea, 94.
Rosenbusch, H., trachydolerite group, 102, 113, 117.
Ross, Sir James Clarke, 1, 94.
„ „ description of Beaufort
Island, ref., 7.
„ „ description of Mt. Erebus,
ref., 8.
„ „ discovery of Coulman
Island, 5.
„ „ discovery of Franklin
Island, 6.
„ „ discovery of Possession
Islands, 4.
„ „ rock-speoimens from Auck-
land Islands, 96.
Ross Archipelago, 98, 101, 140.
„ basalts, 102-109.
„ described, 8-16.
„ relative ages of basalt and
trachyte, 123.
„ rocks of Atlantic type, 102.
Ross Expedition, at Auckland Islands, 109.
Ross Great Ice Barrier, see Barrier.
Ross Harbour, 96, 97 (Fig. 58).
Ross Island, 10, 13, 14, 39, 98.
„ basalt-tuffs, 109.
,, described, 8.
„ icebergs, 65.
„ snow-covered mountains, 65.
„ soundings, 8.
Ross Piedmont, see Barrier.
Ross Quadrant, 1.
Ross Sea, 1, 17 (Fig. 6), 64, 98.
„ granites dredged up, 32.
„ rock-flour deposit, 94.
„ thickness of floes, 57.
Round Mt. (Ci), see Fig. 20 (p. 42).
„ dolerite and sandstone, 45.
„ glacier, 73.
Royal Society Range, see Plates III and VII, and
Fig. 9 (p. 23).
„ 30, 33-35, 69, 70, 73, 98.
Royal Society Range, a faulted crust-block, 53.
„ described, 23.
„ dolerite, 53, 81, 99.
„ gneiss, 25.
„ granite, 32, 81.
„ hollowed granite-boulders, 88.
„ horizontal structure, 39.
„ ice-sheet, 64.
„ ice-slabs, 99.
„ sandstone, 88.
Rowe Island, 2.
Royds (Cape), coastal ice-fringe, 65.
„ described, 9.
„ keuyte, 9, 10, 111-113, 119-121,
132, and Plate VIII, Fig. 3.
„ „ anal., 113.
„ moraines, 81.
Royds, Lieut. C. W. R., south-east journey, 67.
,, pressure-ridges, 82.
Russell, II. C., size of icebergs, ref., 75.
Russell, I. C., ice-streams, ref., 65.
„ piedmont-glaciers, ref., 63, 66, 67.
Rutile, in sandstone, 134.
Sabine (Mount), 21.
Saline ponds, in moraines, 88 (Fig. 51).
Salinity, of sea-ice, 55, 56.
Salt, concentration of, in pools, 56.
„ in sea-ice, 56.
Sands, 89, 90.
„ Laurie Harbour, 97.
„ Macquarie Island, 95.
Sandstone, 40, 53, 92, 139.
„ area and thickness, 53, 54.
„ at Bj, 40-42, 45, 48-50.
„ Beacon Heights, 46.
„ boulders in moraines, 78.
„ Cape Crozier, 11.
„ Cathedral Rocks, 30, 35.
„ contact with dolerite, 49-54, 138.
„ Depot Nunatak, 40, 41.
„ described, 39-54, 134.
„ dolerite-sill in, 40 (Fig. 19).
., dredged by ‘ Challenger,’ 2, 39.
„ Dry Valleys, 45, 46.
„ earth-movements in, 99.
„ Finger Mountain, 44, 45 (Fig. 22), 50,
51.
„ fossil plant-remains in, 40-43, 46-48,
99, 139.
„ frost-action, 92.
„ Granite Harbour, 126.
Sandstone, inclusions in basalt, 105.
„ Inland Forts, 10, 42-11, 50, 53.
„ Knob Head Mt., 17, 51, 52.
„ Kukri Hills, 52.
„ stalagmitic, 41, 131.
„ Terra Cotta Mts., 17.
See also Beacon Sandstone.
Sandy Bay, 97.
Sauer, G. A., trackydolerite from Campanario, ref.,
119.
‘ Scars,’ Auckland Islands, 96.
Schaub, L., twinning of augite, ref., 137.
Schist, Barne Inlet, 26.
„ Blue Glacier, 27.
„ Granite Harbour, 33.
„ (hornblende), 29, 125.
„ (micaceous), described, 135.
Schofield, J. A., rocks from Possession Islands, 4.
Scott, Capt. R. F., 67, 83.
„ Great Ice Barrier, 67-69.
„ “ inlets,” 61.
„ journey south, 24.
„ journey west, 21, 39.
„ on North Fork, 70.
Scofft Islands, 101, 122, 110.
„ phonolitic trachyte, 3, 111
„ described, 3.
Scrope, G. P., Volcanoes, ref., 9, 95.
Sea-floor, composition of, 75.
„ along Barrier, 91.
Sea-ice, described, 55-60.
„ as transporting agent, 93.
„ forced up oil land, 93.
„ removal of, 60.
„ thickness, 59.
Seal, remains of, in moraines, 79.
Sections, at Cape Adare, 18.
„ in sandstone at B„ 11.
„ „ at Inland Forts, 42.
,, „ at West Groin, 41.
Sedimentary rocks, 101, 131, 135.
Senfter, R., analysis of “ diabase,” 105.
Sequence of eruption, volcanic rocks, 123.
Shackleton, Lieut. E. II., journey south, 21, 51.
Shale, 11, 135.
,, from ‘ Challenger ’ dredging, 2, 39.
Shingle, Laurie Harbour, 97.
Shore-ice, 61, 62.
„ conservative effect of, 93.
See also Ice-fringe.
Shoshonite, Beaverdam Creek, anal., 130.
Shoshonose, 131, 132.
INDEX. 157
Shropshire, stiperstones, 131.
Sierra de Chaco, meteorite, 109.
Silica, in contact-dolerite, 138.
Sills of dolerite at Finger Mountain, 51.
Silting of snow, 81.
Sinai, weathering of granite, ref., 33.
Skelton, Lieut. R. W., 100.
„ „ granites from Snow Valley,
34.
„ „ photographs, 39, 19.
„ „ rock-specimens found by,
133.
Skuary described, 9.
„ ice-streams, 65.
„ kenyte, 9, 10, 111.
„ moraines, 9.
Skye, granophyre and gabbro, ref., 136.
Slate, Robertson Bay, 2.
Slaty rocks, described, 135.
Smeeth, W. F., rocks from Possession Islands,
ref., 1.
Snow, accumulation of, 83-85.
„ on sea-ice, 56.
„ thickness of, 57.
Snow-crystals, 83.
Snow-drifts, 56, 81.
Snow-dunes, 84.
„ converted into ice, 85.
Snow Valley, 73.
„ granites, 33, 31, 126.
„ glaciers, 28, 70.
Sodium Sulphate, see Glauber Salt.
Solitary Rocks (D5, D5a), see Plates V, VII.
„ 69, 70.
„ crevasses, 70.
,, granite and dolerite, 36.
„ up-thrust of morainic material, 82.
Solvsbergites of Norway, ref., 111.
Sommerlad, H., analysis of basalt, 105.
Soundings, 3, 91.
South Arm, see Plates V and VII, and Fig. 28
(p. 51).
„ 69, 70.
„ granite and dolerite, 36.
„ granites in moraines, 38.
„ ice-movement, 82.
„ moraines, 78.
South Indian Ocean, 95.
South Victoria Land, 2, 11, 39, 57, 69, 71, 82, 87,
89, 91, 91, 95, 98, 123.
„ „ basement-rocks, 25-38, 101,
121-128, 139.
158
INDEX.
South Victoria Land, coast of Pacific type, 123, 140.
„ „ crab-eating seal, 79.
„ „ glaciers, 63, 64, 94.
„ „ icebergs, 73, 75.
„ „ ice-slabs, 64.
„ „ islands off coast of, 1-7, 98,
140.
„ „ land-ice, 63.
„ „ mainland, 17-24.
„ „ mountain-belt, 64.
„ „ shore-ice, 61, 93.
South-west Arm, see Plate VII.
„ 47, 49, 86.
‘Southern Cross’ Expedition, 101, 102, 125.
„ ,, at Franklin Island, 6.
„ ,, basalts from Mount
Terror, 10. '
„ „ rock-collection, 2.
„ „ rocks from Coulman
Island, 5.
Southern Foothills, see Plate VII.
„ 71,81.
„ camptonite, 129.
„ crystalline limestone, 25, 26,
124.
„ granite, 25, 126.
„ ice-slabs, 73, 78.
Sphene, in altered dolerite, 128.
„ in gabbro, 126
„ in granite, 126.
„ in phonolite, 116.
„ in schist, 135.
„ in trachyte, 115.
Spheroidal structure, in trachyte, 13.
Spherulitic structure, in dolerite, 136, 140.
„ in glassy kenyte, 111 (Fig. 63).
Spitzbergen, Ivory Glacier, 78.
„ • up-thrust of morainic material, 82.
Sponge-spicules, sea-floor along Barrier, 94.
“ Stalagmitic” sandstone, 44, 134.
Stengerts, kersantite, anal., 130.
Stiperstones, Shropshire, 134.
Strain in ice of Ferrar Glacier, 83.
Strain-shadows in quartz of granite, 126.
Stream-ice, 57.
Strong, A., “hornblende-diabase,” ref., 105.
Structure, of ice at Knob Head, 82.
„ of piedmont-ice, 67, 68.
„ of sea-ice, 55.
Sturge Island, see Fig. 1 (p. 3).
„ centre of ice-shedding, 65.
„ described, 3.
Sturge Island, piedmonts, 67.
Suess, F. E., tectonic processes, 122.
Sulphur, Sulphur Cones, 12.
Sulphur Cones, basalts, 11, 12, 103-105, 119, 120.
„ felspathic nodules in basalt, 107.
Sultan’s Head, tuffs and basalts, 11, 109, 110.
Summary of Report on Field-geology, 98, 99.
„ of Report on Rock-specimens, 139, 140.
Supraglacial matter, 78.
Sweet Grass Creek, diorite, anal., 127.
Switzerland, glaciers, 78.
T, see Map and Tabular Mountain.
Table Mountain (E.), see Plate VII and Figs. 14
(p. 29), 28 (p. 54).
Table of Distances, 7.
Tabular Mountain (T.), see Plate V, 70.
Tabular Structure, Royal Society Range, 98.
„ South Arm, 54 (Fig. 28).
See also Horizontal Structure.
Talus-fans, from frost-action, 92.
Talus-slopes, Granite Harbour, 33.
Temperature of ice, 85, 86, 99.
Tent Island, described, 16.
„ kenyte, 16, 111.
Terra Cotta Mountains (B0, B7, B„), see Plate VI 1
and Fig. 23 (p. 47).
„ „ described, 47.
Terra Nova (Mount), 8, 10.
Terror (Mount), see Plate I and Fig. 5 (p. 10).
„ 7, 8, 20, 68, 95, 98.
„ basalts, 10, 11, 103-105, 111.
„ boulder of calc-schist, 124.
„ centre of ice-shedding, 65.
„ described, 10, 11.
„ dolerite-boulders, 11, 139.
„ granite-boulders, 11, 32, 128.
„ moraines, 81.
„ olivine-nodule, 108.
„ parasitic vents, 11.
„ pressure-ridges, 82.
„ sandstone-boulders, 11.
„ tide-cracks in barrier, 67.
„ trachytes, 10, 114, 115, 120,
Plate VIII, Fig. 4.
See also Crozier (Cape).
Thalweg of glaciers, 64, 73.
Thawing of sea-ice, 60.
Thickness of sea-ice, 59.
Tide-crack, 67, 79, 83.
Titanic acid, high percentage in basalts, 103, 105,
140.
INDEX.
159
Tongariro, 7.
‘ Tongue ’ of diorite, 127.
Torridon Sandstone, 53.
Trachydolerite, Campanario, anal., 110, 121.
n Inaccessible Island, 10.
of Rosenbuscb, 113.
See also Kenyte and Trachyte (pkonolitic).
Trachyte (phonolitic), 102, 110.
Black Island, 11, 115, HO,
123.
Brown Island, 15, 115.
„ anal., 115.
n Cape Crozier, 11, H4, H5.
anal., 111.
Crozier Cliffs, 10.
Ferrera, 118.
Forodada, 117-119.
„ anal., 119.
Inaccessible Island, 10, 111,
115.
Mount Terror, 11, 111,115,
120, and Plate YIH, Fig. 4.
Observation Hill, 13, 117-
119, and Plate YIII, Fig- 0.
Razor Back Island, 10, 111.
Scott Islands, 3, 114.
.. anal., 111.
wiud-effect on, 87, 88.
Trachytes (phonolitic), described, 113-119.
Transport of sea-ice, 59.
Treib-eis, 57.
Triangle (Osann), 121.
Tschermak, G., Sierra de Chaco Meteorite, ref., 109.
Tuckett, F. F., cavities in granites, ref., 89.
Tuffs, Cape Adare, 18.
„ Cape Crozier, 10, 11, 109, 110.
„ Castle Rock, 12, 109, 110.
„ Hutton Cliffs, 11, 109, 110.
„ Skuary, 9.
„ Sultan’s Head, 11, 109, HO.
„ Turk’s Head, 9, 109.
Y-Cliffs Hogsback, 10, 109, 110.
Turk’s Head, 11.
„ basalt-tuffs, 9, 109.
ice-streams, 65.
TurtleBack Island, augite-olivine nodule, 11, 108
(Fig. 60).
v basalt, 11, 106.
described, 14.
kenyte, 11, 110 (Fig. 60), Hi
(Fig. 63).
Tyndall, J., on glaciation, ref., 99.
Upper Kukri Hills, see Kukri Hills.
Up-thrust of morainic material, 76 (Pig. 11), 82.
V-Cliffs Hogsback, basalt-tuffs, 10, 109, 110.
„ , bombs, 10.
Valley-glaciers, 63, 71, 72.
Variolitic structure in contact-dolerite, 138.
Vasvik, rhomb-porphyry, anal., 113.
Vegetation in Antarctic regions, 18.
Vein-quartz, pebbles in sandstone, 41.
Vesicular ice, 68.
Victoria Land, sec South Victoria Land.
Volcanic cones, Cape McCormick, 19.
,, mainland of South V ictoria Land,
18, 98.
„ Mount Brewster, 19.
„ Mount Discovery, 20
„ Mount Melbourne, 19
„ Mount Morning, 20.
Volcanic rocks, chemical relations of, 119-123.
„ described, 102-123.
„ two groups of, 121, 122.
Volcano, active, Mount Erebus, 8.
„ on Buckle Island, 1,3.
Volcanoes, Auckland, 12.
Wadworth (Cape), 5, 6.
n trachyte from dredge, 115.
Walterskausen, S. IV. von, Mount Etna, ref., 9.
Waltlicr, J., desert-weathering, ref., 87.
„ weathering of granite, ref., 33.
Washington (Cape), 20-22.
Washington, H. S., bibliography of hornblende
pseudomorphs, ref., 103.
sj classification of rocks, ref.,
120.
Water-action, 89-91, 99.
„ in moraines, 89 (Fig. 52).
Water-courses, Macquarie Island, 96.
Water-holes, 58 (Fig. 32), 59.
Weathering, of granite, 33.
n of trachytes, 113, 117.
See also Denudation.
Weller (A. B.), 40.
West Fort (C9), 50.
West Groin, doubtful fossils, 43 (Fig. 21), 48.
)i pipe of dolerite, 50.
sandstone and dolerite, 42-44.
Western Mountains, see Royal Society Range.
White Island, see Fig. 32 (p. 58).
, basalt, 15, 104.
160
INDEX.
White Island, described, 14.
„ gabbro in moraines, 139.
„ Glauber Salt, 91.
„ moraines, 79, 80, 91.
„ pressure-ridges, 82.
„ tide-cracks in barrier,' 67, 83.
„ vesicular ice, 68.
Whitney, J. D., on glaciation, ref., 99.
Williamson Point, basalt, 109.
Wilson, Dr. E. A., basalts from Mount Terror, 10.
„ Glauber Salt, 91.
., observations on Mount Erebus, 9 .
„ rocks from near Koettlitz
Glacier, 26, 129.
„ rock-specimens from Discovery
Gulf, 124.
„ sketches on journey south, 24.
Wind-action, 87, 88, 99.
„ on sandstone, 92.
„ removing snow, 84.
Wind-worn stones, AVinter Quarters, 87.
Windy Gully, see Plate VII, 69, 70.
AVinter Harbour, glacier-ice fringe, 62.
AVinter Quarters, see Plate II. and Figs. 32 (p. 58),
and 33 (p. 59).
„ 8, 14, 16, 19, 23, 82, 92, 98.
„ basalts, 11-13, 104, 105, 111, 123.
AVinter Quarters, boulders, 87.
„ denudation, 87.
„ described, 11-14.
„ nodules in basalts, 106-109.
„ snow-drifts, 84.
Winter Quarters Peninsula, glacier-ice fringe, 62.
„ „ hummocks in sea-ice,
59.
AYolff, F. von, refraction of anorthoclase, ref., 211.
AVollastonite in sandstone, 41.
AVood Bay, see Fig. 37 (p. 65).
„ ice-foot, 65.
„ pumice, 20, 93.
„ stranded moraines, 81.
Wyoming, banakite, 129.
x (hill), see Map and Plate V.
„ dolerite and sandstone, 46, 51.
„ gullies, 92.
Young Island, described, 1, 2.
Zircon, in granite, 126.
„ in sandstone, 134.
Zirkel, F., felspathic nodules in basalts, ref., 107.
„ origin of olivine-nodules, 109.
‘ Zone of fire,’ connecting New Zealand and Mount
Erebus, 95.
LONDON : PRINTED BY WILLIAM CLOWES AND SONS, LIMITED,
DUKE STREET, STAMFORD STREET, S.E., AND GREAT WINDMILL STREET, W.
National Antarctic Expedition, 1901-1904. British Museum
Report.
Arrival Bay.
National Antarctic Expedition, 1901-1904. British Museum Report. Fiej
The Gap.
Arrival Bay.
National Antarctic Expedition, 1901-1904. British Museum Report. Eie]j .d-geology .
Plate II.
Observation Bill.
Crater Hill.
Hut Point.
A PANORAMA OP WINTER QUARTERS, SHOWING HARBOUR HEIGHTS, CRATER HILL AND OBSERVATION
Harbour Heights.
d-geology.
FHE ROYAL S'
Looking up Fen-
Royal Society Range.
Camel’s Hump.
Ej E2
Cathedral Rocks.
d2
Perrar Glacier.
„/ ■ # V
’ aSk-Sill
:
Kukri Hills.
'
.
Rational Antarctic Expedition. 1901-1904. British Museum Report. Field-geology.
THE NORTH END OF THE ROYAL SOCIETY RANGE AND THE SOUTH SIDE OP THE KUKRI HILLS.
Looking up Ferrar Glacier from Descent Pass. See pp. 30, 69, 78.
Plate IV
27, 30, 34.
Plate IY,
d3
d4
g
G>
B,
Kukri Hills.
Northern Foothills.
Cathedral Rocks.
VIEW DOWN THE EAST PORK OF THE FERRAR GLACIER.
Showing the low granite-hills between G2 and G3 and the gneiss-exposure at the foot of Cathedral Rooks. See pp. 27, 30, 31.
National Antarctic Expedition, 1901-1904. British Museum Report. Field-geology.
Plate Y.
45-41
7i
Upper Beach of the Glacier,
Plate V.
T.
Cathedral Books, Art-. T.icf.or Tabular lilt.
National Antarctic Expedition, 1901-1904, British Idnseuin Report. Field-geology.
PANORAMA OF THE SOUTH SIDE OF THE FERRAR GLACIER AS SEEN FROM A POINT ABOVE THE SOLITARY ROCKS.
See pp. 45-47, 69-70.
Plate YI.
The Ends of Ice-slabs.
the background).
TO A TRIBUTARY VALLEY COI>
i
Mount Morning (in the background). The Ends of Ice-slabs.
THE OVERFLOW OF THE KOETTLITZ GLACIER INTO A TRIBUTARY VALLEY CONTAINING AN ICE-SLAB. See pp. 71, 73.
National Antarctic Expedition , 1901-1904. British Museum Report, Field-geology.
/
Plate VII.
sty
SOUTH-WEST ARM
jj (fossil plant-remains.)
* t
DEPOT NUNATAK
A (7,650ft)
jfliiiiimmiiiHiningfa-
INLAND-ICE
2 (7,650 ff)
EAST
Drawn by H. T. Fekrab.
Plats VII.
INLAND-ICE
EXPLANATION OF PLATE VIII.
Fig. 1. — Olivine-basalt (65(5) from cliff between Gap and Horseshoe Bay (p. 104).
All the clear olivines, except the three crystals at the top, belong to oue individual : the more
shaded crystals are augite.
Magnification, 30 diarn., 1 inch objective.
Fig. 2. — Gabbro-like nodule (415) in liuiburgite, from Winter Quarters (p. 108).
The dark grains are altered olivine, the shaded grains are pale-green diopside, a rather more deeply-
shaded grain on the left below is hornblende, the clear crystals showing twin-striations are
felspar.
Magnification, 30 diarn., 1 inch objective.
Fig. 3. — Leucite-kenyte (818) from Cape Royds (p. 111).
The large phenocryst of anorthoclase shows crossed twin-striations as seen between crossed nicols ;
numerous small leucites with central inclusions are seen in the base ; on the left towards the
bottom is a small phenocryst of olivine.
Magnification, 20 diarn., 1^ inch objective.
Fig. 4. — Phonolitic trachyte (248) from Mount Terror (p. 115).
A rectangular phenocryst of anorthoclase is seen in a trachytic mesh of felspar-laths with dark
grains of mgirine-augite.
Magnification, 30 diarn., 1 inch objective.
Fig. 5. — Phonolite (530) from Black Island (p. 11G).
Hornblende near to riebeckite, in moss-like patches, in fine-grained trachytic felt of felspar-laths.
Magnification, 30 diarn., 1 inch objective.
Fig. 6. — Phonolitic hornblende-trachyte (277) from Observation Hill (p. 117).
Prismatic crystals of basaltic hornblende in trachytic mesh of felspar-laths.
Magnification, 30 diarn., 1 inch objective.
Plate VIII.
v*»-
x- NSW- w^n. ,f*-
,j^k. w./ - . *■« &B*?L
National Antarctic Expedition, 1901-1904.
British Museum Beport.
ROCK-SPECIMENS.
E. Drake del.et lith.
EXPLANATION OF PLATE IX.
Fig. 1. — Augen-gneiss below D4, Kukri Hills (p. 125).
Cataclastic structure as seen between crossed nicols.
Magnification, 20 diam., 11 inch objective.
Fig. 2. — Iliorite (715) from Cathedral Rocks (p. 127).
Large ophitic plates of brown hornblende, and labradorite.
Magnification, 20 diam., li inch objective.
Fig. 3. — Diorite to essexite (572) from the Blue Glacier (p. 12.s).
Large sharply-defined crystals of brown hornblende with altered felspars, and much apatite in
small hexagonal sections.
Magnification, 20 diam., 11, inch objective.
Fig. 4. — Kersantite (579) from Northern Foothills (p. 130).
Biotice (dark, straggling crystals) and nearly colourless diopside (irregular, shaded plates) in a
base of kaolinised felspar-laths.
Magnification, 20 diam., 1 1-inch objective.
Fig. 5. — Camptonite (839) from Southern Foothills (p. 129).
Small, sharply-defined, crystals of reddish-brown hornblende in a base of felspar-laths.
Magnification, 20 diam., 1 1-inch objective.
Fig. 6.— Dyke-rock (714) related to banakite, from the Northern Foothills (p. 131).
Phenocrysts of labradorite (in centre), orthoclase (below, to the left), brown hornblende (below, to
right), analcite (round section, below the labradorite to the right), and purplish augite
(above the labradorite to the left), in a base of rectangular and lath-shaped felspars.
Magnification, 20 diam., 1 1-inch objective.
Plate IX.
National Antarctic Expedition, 1901-1904.
British Museum Report.
ROCK-SPECIMENS,
E. Drake del. et lith.
EXPLANATION OF PLATE X.
Fig. 1. — Dolerite ((702) from Depot Nunatak (p. 13G).
The shaded crystals are of augite ; some of the darker ones are altered and coloured brown with
oxide of iron. The clear sections are of labradorite. On the left and in the interstices of
the felspars and augites are seen patches of spherulitic material.
Magnification, 20 diam., l^-inch objective.
FiG.2. — Dolerite (696), 2 ft. from junction with sandstone, Dry Valleys (p. 138).
The augite is mainly in long prismatic crystals, the felspars in small laths : interstitial felsitic
material is crowded with magnetite.
Magnification, 30 diam.. 1-inch objective.
Fig. 3. — Dolerite (687), G in. from junction with sandstone, Inland Forts (p. 138).
Long prismatic felspars and augites, and interstitial glass black with magnetite.
Magnification, 30 diam., 1-inch objective.
Fig. 4. — Dolerite (695), 2 in. from junction with sandstone, Dry Valleys (p. 138).
Radiating sheaves of felspar-laths, and interstitial glass dense with magnetite ; a few porphyritic
felspars.
Magnification, 30 diam., 1-inch objective.
Fig. 5. — Junction of dolerite and sandstone (669) at B, (p. 138).
Brown glass, with dark, cloudy patches ; a few porphyritic felspars.
Magnification, 20 diam., 1 (-inch objective.
Fig. 6. — Dolerite (154) with granitic patches, Granite Harbour (p. 139).
Dolerite of purplish augites, felspar-laths and magnetite, with coarse-grained patch of altered
felspars and quartz.
Magnification, 30 diam., 1-inch objective.
Plate X.
National Antarctic Expedition, 1901-1904.
British Museum Report.
ROCK-SPECIMENS.
E. Brake del. et lith.
Oarded
LONGITUDE WEST OF GRE E NWICrl
LONGITUDE EAST OF GREENWICH
G.6o'.di<
"solurn
27 QI02
°5'O202~
* VICTORIA
>NOUt
OF THE
franV
30" Longitude
10* East, of 20* Greenwich 30*
BETWEEN - LATITUDES 66°S. AN D 83° S.
AND -LONGITUDES I50°E.AND I50°W.
SHOWING THE. LAND TO THE SOUTH OF 74“S.
SURVEYED UNDER THE DIRECTION OF THE R.G.S.
Captain R. F. Scott. R.N..C.V.O.
ENDERBY
J TERRA de
J0,CMorn
y FU EGO
LAN O'
COMMANDING THE
LAND
Shetland
Dl SCOVERY
Islands
Ht*"1
GRAHAM
NATIONAL ANTARCTIC EXPEDITION 1901-1902-03 04
Track of the Discovery y ^
Soundings in Fathoms -those noton ship's track were -taken by Ross in 1841
ALEXANDER
Nares (Chi
Feet above Sea.Level, ofthe Barrier*. marLecl rn blue
Inlet.
12-01 02i
CaptaJn Scott's Southern Sledge Journey 021102 to 03 02 03.
Belgica 1898
Western
2610 03 to 2412-03
Gauss, Winter Qr
Peter I .
Bellingshausen 1821
Lieut. Armitage
291102 to 19 01 03
061003 to 1312-03.
C.Dowt
Lieut. Royds.R.N. S.E.
10-11-03 to 12-12 03.
sebTS0H'
bay
KNOX LAND
blk. blcLch
br. brown. f. fit ne. g. grave 1/
S. soft sh. sh-ell «s sm. small/
lories y. yel Lorw
C Ada?*
oa-oi-o:
JTH •
ScotC'SO.Nov. 1903
Scale of Miles
'la n d.
l fo ovary "
V ICTORIA
31 Jan
Natural Scale. 1‘- 5,500,000 or 1 ItlcTl = 86’3 Mile
27-02-04
LAND
CLARIE LAND
AOELIE LA'fJD
Wi I kes 1840
05-01 02
Campbell Iv
Truck oi' the
oulAveu'd' )
retia-ri/ )
TASMANIA
03-01-02
Antipodes |-
Scale of Miles
N EW
ZEALAND
Nat. Scale, 1 40,000,000 or 1 Ln.cli=63l Mj
A U S T-R"A L I A
port Lyttelton',
Chatham
WEST LONGITUDE 180° EAST LONGITUDE
ConstrucGKL by Lxeutcncoxo &FA lifiJoch, R AT
Stanfords Gtcg }Escab^ London.
1 7 IB
1
1 1
/ 1
250 1
1
1
nr
i l
r | _
75"
i I
I
l
i
1 I
i
~i r
NATIONAL ANTARCTIC EXPEDITION (i9O/~/904), BRITISH MUSEUM REPORT.
W« CLOWES & SONS, L7? LITH. LCNOC
This text was extracted using OCR (Optical Character Recognition) and may contain errors. For the most accurate text, please refer to the original document.
Download Full Text File
Download the complete OCR text for this item as a plain text file.
Related Items by Subject
A voyage of discovery and research in the southern and Antarctic regions, during the years 1839-43
Additions to the floras of Southampton and Mansel islands, Hudson Bay
Alberta Geological Survey
Alpine and arctic plants [microform] : a lecture delivered before the Young Men’s Christian Association of Montreal, February, 1862
![Alpine and arctic plants [microform] : a lecture delivered before the Young Men’s Christian Association of Montreal, February, 1862](https://archive.org/services/img/cihm_53274)
Animal book
