Full text of "Proceedings of the Royal Society of Victoria. New series"

See other formats

L H Moon
& Son

Bookbinders

-Mil

PROCEEDINGS

OF THK

opl ^orietjj of §irtori<t.

VOL. XL. (New Series).
PARTS I AND II.

Edited under the Authority of the Council.

ISSUED 17th DECEMBER . tQ2 f /, and 14th JUNE, 1 928.

{Containing Papers read before the Society during the mouths of
March to December , 1027).

IHW AUTHORS OK I'll K SEVERAL PAPERS ARK INDIVIDUALLY RESPONSIBLE KOK iiik.
SOUNDNESS OF THK OPINIONS GIVEN AND FOR THK ACCURACY OF THK
STATEMENTS MADK THEREIN

ROYAL SOCIETY'S HALL,

VICTORIA STREET, MELBOURNE, C. 1.

FORD & SON PRESS PTY. LTD., DRUMMOND ST., CARLTON, MELBOURNE.

1 928

CONTENTS OF VOLUME XL.

Part I.

. _ . . „ _ . PAtf

Art. I. — Anatomy of Australian Coniferous Timbers. By

Reuben T. Patton, B.Sc., M.F. (Plates I.-V.) 1

II— Vitality of White Races in Low Latitudes. By

C. H. Wickens. .. ... ... ... 17

HI— The Stony Creek Basin and the Corinella Dyke.

By D. Orr, B.Sc. ... ... ... 25

IV. — An Olivine Anorthoclase Basalt from Daylesford.

By D. Orr, B.Sc, ... ... ... ... 34

V . — Notes on the Coastal Physiography of Port Campbell,
Victoria. By ,J. T. Jutson, B.Sc., LL.B. (Plates

VI., VII.) 45

Part II.

VI. — A Revision of the Genus Pultenaea. Part V. By

H. B. Wjt,liamson, F.L.S. ... ... ... 57

VII. — Some Tertiary Volcanic Minerals and their Parent

Magma. By D. J. Mahon y, M.Sc., F.G.S. ... 62

VIII- Experimental Error of Field Trials in Australia.

By H. C. Forster, B.Ag.Sc., and A. J. Vasey,
B.Ag.Se. ... ... ... ,.. ... 70

IX.— Contributions to the Flora of Australia, No. 34.
Additions to the Flora of the Northern Territory

and Locality Records. By Alfred J. Ewart,

D.Sc., Ph.D., P.L.S., F.R.S., and Phyllis H.
Jarrett, B.Sc, ... ... ... ... #i

X.— Fossil Plants of the Stony Creek Basin. By

Reuben T, Patton, B.Sc., M.F. (Plate VIII.) 88

XI. — The Staurog-raptus Bod of Victoria. By W. J.

Harris, M.A., and R. A. Keble. (Plate IX). 91
List of Members ... ... ... ... , |# 99

Index ... ... ... ... ... iQ7

[Pkoc. Roy. Soc. Victoria, 40 (N.S.), Pt. I., 1927.]

Art. I . — Anatomy of Australian Coniferous Timbers .

By R. T. PATTON, B.SC , M.F.

(With Plates I.-V.)

[Read 10th March, 1927.]

Introduction.

The conifers of Australia are unique in many respects, as has
often been remarked, and are on the whole widely different from
those of the Northern Hemisphere. The large and well-known
tribe, Abietineae, is entirely absent from Australia, and associated
with this is the entire absence of such characters as the fusiform
ray, the ray tracheide, and the resin canal. On the other hand,
the resin tracheide, which is entirely absent from the above men-
tioned tribe, appears in two of our genera, A gat his and Araucaria ,
belonging to the tribe Araucarineae, which is almost exclusively
southern.

Owing to the difficulty in obtaining some of our isolated species
a long time has elapsed since this work was begun. Specimens of
all the genera have been obtained, but the following species have
not been received: Callitris Rod Endl., C. M orris oni, R.T.B., C.
Driimmondii Benth. et Hook., Actinostrobus acuminatus Parlat.,
and Podocarpus Droiiyniana F.v.M.

Although the work is not complete, it has been decided to pub-
lish what has been done owing to many requests for information
regarding the structure of our conifers. The structure of the
commercial coniferous timbers of x\ustralia has been studied by
Baker and Smith (1), but these authors did not investigate the
non-commercial species. These latter are of interest, however,
from a palaeontological point of view. Moreover, these authors
have used well recognised terms in somewhat unusual ways. The
present work, therefore, has covered all the species obtainable.

For the accompanying photomicrographs I am indebted to Mr.
H. Marriott, of the Anatomy Department, University of Mel-
bourne.

Pherosphaera.

This genus is endemic to Australia and Tasmania. One
species, P. Fitzgcraldi F.v.M., is found in New South Wales, and
the other, P. Hookeriana Arch., in Tasmania. Both are small
shrubs growing in restricted habitats.

Transverse. — Rings fairly well defined, walls thick, lumen of
cells of summer wood almost obliterated. Width of cells about
15 /x. Resin cells moderately abundant or absent.

Radial. — Bordered pits uniseriate, occasionally partly biseriate.
Pits on the lateral walls of the medullary rays one per tracheide,
large, elliptical and oblique.

R. T. Patton :

Tangential- — Medullary rays 1-5 cells high. Pits present on the
tangential walls of the tracheides.

Tracheides from 0-7 to 1-3 nun. long.

In P. Hookcriam the cells walls are not nearly so thickened as
in P. Fitzgcraldi. In the former the cells are squarish, with walls
3-5 ii in thickness, but in the' latter they are rounded and the
walls are 6-8 n thick. In both species the thickness of the wall
throughout the ring is fairly constant. In P. P itzgeraldi pits on
the tangential walls are very abundant, but infrequent in P. Hook-
eriana.

Material of both species authenticated by Professor Lawson;
material of P. Hookeriana from Government Botanist, Tasmania,
and from G. Weindorfer, Esq.

M icrocachrys.

A monotypic genus endemic to Tasmania, .17. tctragona Hook,
is a small shrub, rare.

Transverse. — Rings rather indefinite; tracheides not very regu-
larly arranged in radial rows; walls thin, squarish. Resin cells
absent.

Radial. — Bordered pits uniseriate, indistinct. Lateral walls of
medullary rays 1-2 pits per tracheide. These pits are verv irre-
gular as regards size and shape ; some are large and open without
any border, while some are narrow, elongated and bordered.
Medullary rays frequently irregular in outline. No parenchyma.

Tangential.— Medullary rays 1-4 cells high, maximum 7. Cells
elongated, thin walled.

Material authenticated by Government Botanist, Tasmania, and
by G. Weindorfer, Esq., Cradle Mt., Tasmania.

Dacrydium.

_ There is only one species, D. Franklin $ Hook. It is an average-
sized tree, and is endemic to Tasmania. The genus is also found in
the Malay area. New Zealand, and Chile.

Transverse, — Rings narrow, well defined. Summer wood 2-4
cells wide. Spring wood very open, cells usually somewhat hexa-
gonal or rounded. Thickness of walls of spring wood about 8 /*.
Tracheides not in regular radial rows. Resin cells absent.

Radial. — Bordered pits uniseriate, not crowded. Medullary
rays have the marginal walls somewhat sinuous. Lateral pits on
medullary rays one, rarely two, per tracheide, large, elliptical,
oblique, narrow or broad. Parenchyma absent.

Tangential.— Medullary rays 3-7 cells high, maximum 13. Cells
thick walled. Parenchyma absent.

Material from trade sources and from the Sydney Techno-
logical Museum.

In regard to this species Baker and Smith ( 1 ) state that one of
“ the most distinguishing characters of the wood is the fineness
of the wall structure.” This is by no means the case as their

Australian Coniferous Timbers .

photograph shows. This statement could well be used in regard
to Athrotaxis. As a matter of fact the walls are thick and there
is very little difference between the thickness of the spring and
summer cell walls.

In microscopic section Dacrydium Franklinii cannot be satisfac-
torily distinguished from Phyllocladus rhomboidalis. In the latter
the pits on the medullary rays are usually larger than in the
former, and in the latter also two pits per tracheide are common.

In Phyllocladus rhomboidalis the tracheides are more rectangu-
lar, seen in cross section, than in Dacrydium Franklinii .

Podocarpus.

This genus is represented in Australia by six species, five of
which have been studied. The genus is very widespread in the
Pacific regions, extending from New Zealand to Japan. In habit
they range from dwarf shrubs, as P. alpina, to large trees, as P.
•data.

Transverse. — Growth rings more or less indefinite, the spring
wood gradually passing into summer wood, the latter consisting
of one to few rows of very narrow cells. Walls thick, cells sub-
rectangular to round, arranged in fairly definite radial rows.
Medullary rays resinous or non-resinous. Resin cells present or
absent ; when present, widely scattering or zoned. This character
is widely variable both in Australian and extra-Australian species.
Kanehira (4), speaking of P. philip pincnsis Foxworthy says,
44 Resin cells numerous, often connected tangentially ” ; while of
P. macro phyllus Don (5) he says, “Resin cells present rather
scarce.” Again, speaking of P. Nagi Zoll et Moritz, he says,

Resin cells evenly distributed in both early and late wood.” The
character is not generic.

Radial. — Rays resinous, uniform in character. The outer upper
and lower walls somewhat irregular in outline and thicker walled.
Inner horizontal walls not strictly parallel. End walls more or
less vertical, unpitted, frequently curved. Lateral walls bearing
1-3 bordered pits per tracheide; pit aperture narrow or broad,
oblique. Bordered pits um seriate or incompletely biseriate, and
then opposed, not alternate. Pits scattered, generally, but when
crowded, indistinct Bars of Sanio are visible. Resin cells con-
spicuous when present.

Tangential. — Rays uniseriate, moderate in height ; cells ellipti-
cal to round, thin to thick-walled, sometimes resinous.

This genus with its somewhat rounded cells, scattering resin
cells, and single rows of bordered pits, is very similar to Callitris.
No species of Podocarpus has, however, the peculiar plate of
thickening across the bordered pits as seen in some species of
Callitris. The cells of Callitris , as seen in transverse section, are
always more rounded. In Podocarpus the medullary rays never
have more than three pits per tracheide.

R. T. Patton :

PODOCARPUS ELATA R.Br.

Transverse. — Rings of growth broad, summer wood few cells
thick, not very distinct ; the cells being about half the width of the
spring wood cells. Resin cells abundant, widely scattered.

Radial. — Resin cells conspicuous; rays resinous.

Tangential.- — Rav cells resinous, elliptical.

Specimen authenticated by the Sydney Technological Museum.

P. PEDUNCULATA Bail.

Transverse. — Rings very indefinite, thin walled. Resin cells
absent.

Radial. — Ray cells devoid of resin; resin cells absent. Pits on
medullary rays moderately large, widely open.

Tangential. — Medullary rays devoid of resin; cells varying
greatly in size ; thin walled.

Specimen authenticated by the Technological Museum, Sydney.

P. spinulosa R.Br.

Cells rounded, thick walled ; resin cells absent. Rings of
growth not determinable. Rays devoid of resin.

Specimen authenticated by the Technological Museum, Sydney.

P. ladei Bail.

Similar to P. spinulosa.

Specimen authenticated by Government Botanist, Queensland.

P. alpina R.Br.

Transverse. — Rings very narrow, only a few cells wide, very
irregular in width, fairly distinct. Resin cells abundant.

Radial. — Resin parenchyma present. Pits on lateral walls of
medullary rays usually two, large, elliptical.

Tangential. — Rays very short, one cell high common, maximum
four.

Specimens authenticated by Government Botanist, Tasmania;
G. Weindorfer, Esq. ; and by the author.

Phyllocladus.

This genus is not found on the mainland of Australia, and our
only species, P. rhomboidalis Rich., is endemic to Tasmania. The
genus is also found in New Zealand and the East Indies. The
timber of P. rhomboidalis is known commercially as Celery Top
Pine.

Transverse. — Rings comparatively narrow, well defined. Sum-
mer wood 2-4 cells broad, lumen almost obliterated. Spring wood
open, thick walled, rectangular ; cells up to 40 n in length, radially
arranged. Resin cells absent.

Australian Coniferous Timbers .

Radial. — Bordered pits nniseriate, not crowded, often scattered;
no Bars of Sanio present. Resin cells absent. Upper and lower
walls of the medullary rays straight or slightly sinuous, outer
walls thinner than the inner. Pits on the lateral walls 1-2 per
tracheide, large, oblique, elliptical, faintly bordered. Occasion-
ally the aperture is narrow and then it is distinct.

Tangential. — Medullary rays generally 4-12 cells high, maxi-
mum 20. Cells elliptical, thin walled. Bordered pits on the tan-
gential walls.

Authenticated material from the Sydney Technological
Museum, and from trade sources.

In regard to this species Baker and Smith (1) say, “Similar
anatomical characters ” to those of Dacrydium Frank linii. This
is so. These authors would separate the two species by the pre-
sence of bordered pits on the tangential walls of Phyllocladus
rhomb oidalis , and their absence from Dacrydium Franklinii. This
does not hold absolutely.

Agathis and Araucaria.

Agathis is represented in Australia by three species, only one of
which, A. robusta C. Moore, Queensland Kauri, is well known.
I he other two, A. Palmcrstoni F.v.M., and A. microstachys Bail,
et White, are also restricted to Queensland, but are confined to
the far north of that State.

Araucaria is represented by two species. One, A . Cnnninghamii
Sweet, which extends down from Queensland into the north-east-
ern part of New South Wales, gives the well known Hoop Pine
of commerce. The other, A. Bidimlli Hook., is confined to
Queensland. It may be noted that all these five species are re-
stricted to the coastal regions of North-Eastern Australia.

Botanically these two genera are very definite, but microscopi-
cally the timbers are very similar Penhallow (2) sought to dis-
tinguish the two genera by the distinctness of the annual rings
and for Araucaria he states, “ Growth rings not determinable,”
and for Agathis he says, “ Growth rings obvious but poorly de-
fined. ’ 1 he definiteness of the growth rings is not a generic charac-
ter, and Penhallow himself, under Araucaria Bidwilli , states,

“ Growth rings broad, poorly defined,' 1 almost the same words as
for the genus Agathis . Baker and Smith would separate the two
genera by means of the presence or absence of resin in the medul-
lary rays. They say ( 1, p. 329 ) that this “ is an important generic,
specific and phylogenetic character.” An examination of Baker
and Smith s own material, however, shows that this substance is
present in all species. Baker and Smith themselves, under A.
Bidwillh, say, loc. cit “ The medullary rays have their cells filled
with the brown or dark substance.” It is true that resin is far
more plentiful in Agathis than in Araucaria as a general rule, but
Araucaria BidwilUi contains more resin than Agathis Palmcrstoni.
Baker and Smith also attempt to separate these two genera by the

R. T. Patton :

absence of bordered pits from the tangential walls of species of
Agathis. This latter genus, however, has pits on the tangential
walls just as has Araucaria. Kanehira (4) notes pitting on the
tangential walls of A. alba. Pitting on the tangential walls is a
character easily overlooked.

It has been found impossible by the author to find a single
character that will separate the two genera. In the hand speci-
men. kauri timbers are somewhat denser, and usually darker, due
to a greater amount of resin present, than Araucarian timbers.
The darkness of the timbers is shown by A. australis Steud (New
Zealand), A. robusta (C. Moore ) F.v.M. (Queensland) , A. mic-
rostachys Bailey et White (Queensland), and A. lanceolata
Panch.et Sebert (New Caledonia), In all these the rays are dark
coloured, but in Agathis Palmerstoni they are light coloured. The
timber of the various species of Araucaria is usually light coi-
®ured. The rays are slightly darker, due to the presence of some
resin. The timber of A. Bidzcilli , however, is very dark, and in
this respect approaches a kauri.

Transverse.— The tracheides are thick walled and vary from
hexagonal to circular in shape. The tracheides are arranged in
fairly definite rows. The annual rings are not conspicuous, and
the summer wood may be reduced to a single row. Alongside the
medullary rays resinous tracheides are more or less common.
They are much more frequent in Agathis than in Araucaria.
However, Kanehira (4) notes that in Agathis alba (Lamarck)
Foxworthy, “ Resin cells normally absent.” In the only specimen
of Agathis Palmerstoni obtainable, the resin tracheides were also
absent. In Agathis microstachys they are scattered, not radially
arranged along the medullary rays. Medullary rays usually resin-
ous, particularly in Agathis .

Radial. — As is common in other Australian genera, the ray cells
are all of one kind. The conspicuous feature of the rays is their
irregular shape. They are more or less conspicuously contracted
at the ends, and at times the end wall is obliterated. This cliarac-
tei was observed by Penhallow (2) in the species of both genera
which he studied. The character is. apparently, common to all
species. It is not commented on by either Jeffry (6) or Baker
and Smith (1). It was noted by Jones (3). The walls of the
ray cells are thin, and do not show any local thickening. The
lateral walls bear a varying number of bordered pits, where in con-
tact with a tracheide. They range from 2 to 12 in number.

A conspicuous feature of the resin tracheides is the presence of
resin plates across the cells. These are more abundant near the
medullary rays. _ These plates are fully discussed by Penhallow
(2), and they give a pseudo-parenchymatous appearance to the
tissue. They vary in thickness, some are uniformly thin, others
are very thick where in contact with the cell wall and narrow to-
wards the centre. Usually the plate is entire, but it may have a
small perforation at the centre. These plates were also noted by
Jones (3), but they are not mentioned by Jeffry (6), who, how-

A ustral i. an Con if ei 'ous Timbei 's.

ever, does mention parenchyma. In no species studied has any
parenchyma been found. The bordered pits on the radial walls
range from one to four rows, the pits in each row alternating, and
when crowded have an hexagonal outline. In A gat his robusta and
A gat his Paimerstoni, where extra broad tracheides occur, there
is an irregularity of the four rows, and a fifth row is possible.
The opening of the pit is usually oblique.

Tangential. — Medullary uniseriate, rather low, averaging from
three to twelve cells high. Cells oval, resinous. Resin plates con-
spicuous, when present. Pitting on the tangential walls common
to all species. The pits may be either large or small.

Agathis robusta (C. Moore) F.v.M.

Transverse. — Rings poorly defined, summer wood 2-4 cells wide.
Resin tracheides bordering the medullary rays plentiful, a few
scattered. Rays resinous.

Radial. — Bordered pits 2-4 seriate, four rows abundant ; aper-
ture circular. Aledullary rays resinous; 2-8 pits per tracheide;
aperture oblique, narrow. Resin plates abundant.

Tangential. — Rays uniseriate. 3-15 cells high, maximum 25.
Pits present on the tangential walls.

The two conspicuous features of this timber are the presence
of four rows of bordered pits, and the abundance of resinous
tracheides. Baker and Smith ( 1 ) state in regard to this timber,
“ . . . xlyem tracheides are also devoid of this substance (i.e.,
resin) v . one that differentiates the timber from Araucaria.”
The resinous tracheides are, however, abundantly present. Again
(p. 3 77) they state, “ The large number (up to twelve) of simple
cells between the walls of the Jumina is also a good diagnostic
character of the genus A Note the curious use of the word “ cell,”
where evidently pit is intended. The large number of pits is also
found in A. micro stachys .

Authenticated material from Sydney Technological Museum and
from trade sources.

Agathis microstachys Bailey et White.

1 ransverse. — Rings poorly defined, 2-4 rows of summer
tracheides. Resin tracheides common, scattered, not occurring
along the rays. Medullary rays resinous.

Radial. — Bordered pits 1-2 seriate, occasionally three, aperture
narrow, oblique. Ray cells conspicuously contracted, very resin-
ous. Pits on the medullary rays 2-5 per tracheide, rarely more;
bordered, aperture oblique.

Tangential. — Rays uniseriate, 4-12 cells high, maximum 19.
Cells thin walled, round. Resin plates plentiful. Pits on tangen-
tial walls. Authenticated material from Government Botanist
Queensland.

R T. Patton:

Agathis palmerstoni F.v.M.

Transverse. — Medullary rays only slightly resinous. Resinous
tracheides absent.

Radial. — Bordered pits 1-4 rows, 3-4 rows frequent, aperture
circular. Resin in the medullary rays very infrequent ; pits on
lateral walls of rays 3-12 per tracheide, aperture oblique. Resin
plates absent.

Tangential. — Rays uniseriate, 3-12 cells high, maximum 25.
Bordered pits on the tangential walls.

Authenticated material from Queensland Forest Service.

Araucaria cunxinghami Sweet.

This timber has been described by Penhallow (2), and there is
nothing to add to his description. Baker and Smith (1) state in
respect to the resin in the medullary rays, “ . . . almost entire ab-
sence,” but Penhallow remarks, on the other hand, ** somewhat
resinous." The resin in the rays is a very variable quantity. Tim-
ber grown in the grounds of the University of Melbourne shows
no resin in the rays at all. Resin tracheides are also very variable
in their occurrence. Some specimens show large numbers, others
none. In the timber grown at the University, the resin tracheides
are zoned, and none occurs adjacent to and parallel to the rays In
regard to the numbers of rows of bordered pits on the radial walls
of the tracheides, Baker and Smith say 2-3, but Penhallow says
1-2. There is very occasionally a third row. A single row 'is
common. Penhallow says that pits on the tangential walls are
wanting, but they do occur as noted by Baker and Smith. Pen-
hallow remarks that the rays may be two seriate, but the author
has not noted this, and apparently neither did Baker and Smith.

Authenticated material from Queensland Forest Service, Tech-
nological Museum, Sydney. Trade Sources, and from known
trees.

Araucaria Bidwilli Hook.

This has been described bv Penhallow, who separates this tim-
ber microscopically from A. Cunninghamii by the bordered pits
being uniseriate, Both species have biseriate rows of pits. Baker
and Smith, speaking of the end walls of the ray cells, state that
they are right angled, while Penhallow remarks that the cells are
“ conspicuously contracted at the ends.” The latter character
frequently interferes with the direction of the end walls. Baker
and Smith give the number of pits on the ray cells per tracheide as
4, vvhile Penhallow gives 3-7. The latter number is correct.

Specimens of this timber grown locally show no resin tra-
cheides at all and there is very little resin m the ray cells. It is
apparent that the presence or absence of resin from either the tra-
cheides or the rays is very variable, and that its absence from any
particular specimen cannot be taken as a definite character.

Authenticated material from Sydney Technological Museum
and from known trees.

Australian Coniferous Timbers.

Athrotaxis.

This genus is endemic to Tasmania. Of the three species, the
only well-known one is A. selaginoides t)on, which provides the
King William Pine of commerce. This is one of the lightest of
timbers. The other two are small, and are very restricted in dis-
tribution. All three have reddish timber, and in structure they
resemble the genus Sequoia.

Transverse. — Rings very definite, the summer wood having
very thick walls, and very narrow lumen, which is very often less
than the thickness of the walls. Spring cells more or less defi-
nitely rectangular, several times longer than those of the summer
wood. Resin cells present or absent, when present scattered.

Radial— Border pits uniseriate or biseriate; when biseriate,
then opposed, not alternate. Bars of Sanio present. Medullary
rays narrow, irregular hastate cells often present when the rays
are only one cell high. End walls vertical or oblique, often curved.
Lateral walls of rays 1-6 pits per tracheide, bordered. Paren-
chyma present. Resin in parenchyma present or absent.

Tangential. — Medullary ravs variable, in height, short rays
common. Cells elliptical, thick walled. Pits on tangential walls of
spring wood, common, very small.

A. SELAGINOIDES, Doll.

Transverse. — Rings very narrow, up to 60 rings to an inch,
usually less than 10 cells wide. Summer wood 2-4 cells, spring-
wood up to 8 cells. Resin cells abundant.

Radial. — Bordered pits frequently two seriate. Bars of Sanio
conspicuous. Pits on the medullary rays 2-5 per tracheide. Resin
cells prominent.

Tangential. — Medullary rays short, commonly 2-4 cells high,
maximum 9, resin cells prominent.

Authenticated material from G. Weindorfer, Esq., and from
trade sources.

A. cupressoides Don.

Transverse. — Ring broad, comparable to Sequoia se mp ermrens.
Cells not as rectangular as in A. selaginoides. Resin cells abun-
dant, either in or towards the summer wood.

Radial. — Bordered pits mostly uniseriate. Medullary rays 1-2
pits per tracheide. Resin cells abundant.

Tangential. — Medullary slightly higher than in A. selaginoides.
Authenticated material from G. Weindorfer, Esq.

A. LAXIFOLIA Hook.

Transverse.- — As in A. cupressoides , but resin cells wanting.
In young specimens isolated unthickened cells occur as in Diselma
Arc her i.

OOOOO a QOOfi oooooo

R. T. Patton :

Radial. — Bordered pits uniseriate. Parenchyma abundant, but
no resin present. Traumatic resin parenchyma present in some
specimens.

Tangential. — x\s in A. cupressoides.

Authenticated material from G. Weindorfer, Esq., and Govern-
ment Botanist, Tasmania.

Callitris.

This is the most widespread genus of the Coni ferae in Australia,
being found in every State, and it has also the greatest number of
species. If we exclude the related African forms, which are
placed under Tetraclinis and Widdringtonia, Callitris is purely an
Australian genus. The species range from shrubs to moderately
large trees. The number of species is at present a matter of dis-
pute. Bentham (7) recognised nine species under the synonym
Frenela , but Baker and Smith (1) recognise at least 17. The
respective lists are as follows: —

Baker and Smith.
Callitris rolmsta R.Br.

tube real a ta R.Br.
verrucosa R.Br.
propinqua R.Br.
glauca R.Br.
a renosa A. Cunn.
intratropica Bent 1 1 . et
Hook.

gracilis R-T.B.
calcar ata R.Br.
rhomboidea R.Br.
Tasmauiea Baker et
Smith

Drummondii Bentli. et
Hook.

Roei Encll.

Morrisoni R.T.B.

Muelleri Bentli. et Hook,
oblonga Rich.

Macleayana Bentli. et
Hook.

Bentham.

Frenela

robusta A. Cunn.

., var. verrucosa

V '» ••

tt it it

it ?» yj

var. microcarpa

*• •> tt

Endliclieri Parlat.

rhomboidea Endl.

var. Tasmauiea

Drummondii Parlat.

Roei Endl.

F-

Muelleri Parlat.

australis R.Br.

Macleayana Parlat.

Parlatorei F.v.M.

Baker and Smith suppressed one of Bentham *s species. F. Par -
latorci , and created three new ones. It will be noted that Ben-
tham’s F, rolmsta was a very comprehensive one, and included no
less than seven otherwise recognised species. Undoubtedly these
are all closely related. C. gracilis R.T.B. is a doubtful species,,
and is regarded by some as a synonym for C. propinqua . Of the
remaining nine species given by Baker and Smith, seven are
agreed upon by both Baker and Smith on the one hand, and by
Bentham on the other. The other two are new. To the above list
must be added C. Baileyi C. T. White, which is a Queensland
species.

Australian Coniferous Timbers.

Transverse. — Growth rings more or less distinct, but never
strongly marked ; spring wood passes gradually into the summer
wood. Cells thick-walled, almost circular, arranged in radial
rows. Resin cells more or less abundant, but never so plentiful
as in Podocarpus, scattered or somewhat zoned. Medullary rays
resinous.

Radial. — Bordered pits uniseriate, at times incompletely
biseriate, scattered. Medullary rays strongly resinous, horizontal
walls parallel, end walls vertical or oblique or slightly curved.
Lateral walls of the rays 2-4 pits per tracheide, aperture oblique.
Resin cells present.

Tangential. — Rays uniseriate, height variable. Cells elliptical,
resinous.

Callitris robusta.

Under this species, Benthani grouped among others the five
species of R. Brown, C. robusta, C. verrucosa , C. propinqua, C.
glauca , C. tuber calata , Much confusion at present exists as to
nomenclature. As originally understood by Brown, C. robusta
was a West Australian species, but Victorian trees with large
cones have been placed under it. Timber specimens of all the
foregoing, except C. tuber culata, have been obtained, and nume-
rous specimens from commercial sources have also been received.
These all show without exception, structures which are very defi-
nite and characteristic. In radial section, each bordered pit is
associated with a more or less rectangular plate of thickening, ex-
tending across the tracheide, beyond the margins of the pit, but
not reaching to the upper or lower margin, as seen in Fig. 1 a,
These bands are connected to the tracheidal walls. The aperture
of the pit lies within the band. In tangential section these bands
appear as projecting awns as seen in Fig. lb. The awns arise
at the top and bottom of the aperture of the pit. Another feature
seen in radial section is the peculiar compound pit (Fig. 1 c) con-
necting the medullary rays with the tracheides. The pit is sub-

LO J)

Fig. 1.

rectangular, the length being in the radial direction. The pit is
not always parallel to the length of the ray cell, but may be slightly
oblique. There are from one to four apertures connecting a
tracheide with a ray cell, but there are never more than two aper-

R. T. Patton :

tures in these peculiar compound pits. Whenever in the four
species mentioned above, the bands of thickening are found on the
bordered pits of the tracheides. the compound pits are found on
the ray cells.

In C. arenosa A. Cunn. and C. intratropica Benth. et Hook., as
understood by Baker and Smith, and which were placed under C.
robustci as var. microcarpa by Bentham, the bands across the bor-
dered pits on the radial tracheidal walls are sparsely present, but
the double pits on the medullary ray cells are absent. Although
Bentham reduced these two species to varieties of C. robust a,
some botanists to-day regard C. intratropica as a valid species, but
would suppress C. arenosa, making it a synonym of the former.
In both of these, the resin cells as seen in cross section, are con-
gregated in the autumn wood; so that here the resin cells may be
regarded as zoned as distinct from the scattering cells of the pre-
ceding group. The two species, C. arenosa and C, intratropica ,
differ from one another in the height of the medullary rays. In C.
arenosa the rays are shorter than in the former group, rays of 2.
3 and 4 cell high being common. On the other hand, the rays of
C. intratropica are very elongated, a fact noted by Baker and
Smith. The rays generally range from 7 to 25 cells high, but the
maximum is 33. In the first group the rays are usually from 4 to
10 cells high.

C, gracilis was not seen, but a tangential microphotograph
(rather imperfect). Fig. 127, given in Baker and Smith’s work,
suggests that the awns are present. This species is regarded by
some as a synonym of C. propinqua, and in this the awns do
occur. C. calcarata was recognised as a distinct species by Ben-
tham, and by Baker and Smith. The rays are slightly longer than
in the first five species discussed, although Baker and Smith state
that the reverse is the case. The rays generally range from 5 to
14 cells high.

C. Baileyi , the most recent species to be described, does not
differ in any material way from the robusta group.

The thickening band is absent from C. Muelleri, C. oblonga , C.
M act cay ana, C. rhomboid ea and C. Tasmanica . This character,
therefore, serves as a useful guide in the classification of the
genus.

C, Muelleri has very short medullary rays, a fact noted by
Baker and Smith. The rays are usually from 2 to 4 cells high,
the maximum being 10.

C. oblonga also shows short medullary rays. In this case they
are shorter than those of C. Muelleri , being from 1 to 4 cells
high.

C. rhomboidea has also short medullary rays. Its structure
does not in any way differ from the preceding. C. Tasmanica,
which some regard as a synonym for C, rhomboidea . has longer
medullary rays. They are usually from 3 to 7 cells high, maxi-
mum 13.

Australian Coniferous Timbers.

C . Macleayana is a most distinctive tree, but its timber does not
show any marked character.

The three West Australian species, C. Drurnmondii , C. Roei
and C. Morrisoni , were not studied by Baker and Smith, and no
material has been available for this study.

Material of C. robusia , C. verrucosa , C. propinqua , and C.
glauca from Sydney Technological Museum, and from the
author’s collection. C. arenosa, C. intratropica, C . gracilis, C.
calcarata , C. rhomboidca, C. Tasmanica and C. Maclcayana. from
Sydney Technological Museum. C. Muelleri from the Govern-
ment Botanist, N.S.W., C. Baileyi from the Government Botan-
ist, Queensland, C. oblonga from Tasmania, but not authenti-
cated.

Acti nostro bus.

This genus is very closely allied to Callitris , and Mueller sup-
pressed it, and placed its two species under that genus. The two
species are confined to West Australia, and both are shrubs.
Specimens of A. pyramidal is Miq. have been authenticated by the
Forestry Department, W.A., but no material of A. acuminatus
Parlat. has been available. In structure the timber is indistin-
guishable from Callitris. Resin cells are plentiful in the trans-
verse section, and are generally zoned. They are usually in the
summer wood. In radial section the medullary rays shows a
strong tendency to be contracted at the ends. There is a strong
indication of this in some species of Callitris , but it is never
marked. End walls frequently oblique, often curved. There are
from 2 to 4 pits per tracheide on the medullary rays. There is no
indication of the plate of thickening on the bordered pits.

Diselma.

This genus was established by Hooker in 1859. It is monotypic,
and is confined to Tasmania. Bentham (7) followed Hooker in
his determination, but subsequently in 1880 Bentham and Hooker
in their Genera Plantarum suppressed this genus, and placed the
solitary species under Fitzroya. Diselma Archer i Hook, is a
shrub.

Transverse. — Rings narrow, well defined. Cells in the spring
wood rounded, thick-walled ; in the summer wood almost closed ;
radially arranged. Some cells which are conspicuously rectan-
gular, appear to be larger than the others. Their apparent large
size is due to the fact that no secondary thickening has taken
place. Rays more or less resinous.

Radial. — Bordered pits uniseriate, aperture oblique. Medullary
rays often irregular, cells sometimes arrow shaped (Fig. 2a),
directed towards the exterior. Cells similar to this are figured by
Jones (3) for Sequoia sempervirens . Lateral pits 2-4 per tra-
cheide. End walls at times very strongly pitted. Lateral walls

R . T. Patton:

simply pitted. Parenchyma present, devoid of resin; transverse
walls often strongly thickened at the centre, vertical walls, simply
pitted when parenchyma cells are next to one another.

Tangential. — Medullary rays uni seriate, 1-4 cells high, one and
two cells predominating. Cells elliptical, thick-walled, conspicu-
ously pitted (Fig. 2b). Pits present on the tangential walls.

Tracheides. — Very short, 0*5 to 08 mm. long.

Material authenticated by G. Weindorfer, Esq., and Govern-
ment Botanist, Tasmania.

Bibliography.

1. R. T. Baker and H. G. Smith. A Research on the Pines of

Australia.

2. D. P. Pen hallow. North American Gymnosperms.

3. W. S. Jones. The Structure of Timbers. ,

4. R. Kaneiiira. Identification of Philippine Woods by Ana-

tomical Characters.

5. R. Kanehira. Anatomical Characters and Identification of

Formosan Woods.

6. E. C. Jeffry. Anatomy of Woody Plants.

7. G. Bentiiam. Flora Australiensis.

Addendum.

Since the above was written another work from the hand of R.
Kanehira has been received entitled, “ Anatomical Characters and
Identification of the Important Woods of the Japanese Empire.”
In this appear figures of radial sections, from the medullary ray
region, of the following endemic Australian species, Pherosphaera
Hookeriana, Dacrydium Franklinii , Microcachrys tetragona , Phyl-
locladus rhomb oidalis, Araucaria Cunninghanm, Athrotaxis sela-
ginoides, Fitsroya Archeri, Callitris rhomboidea and Callitris

Australian Coniferous Timbers .

glaum. There are also photomicrographs of Callitris glauca
(radial sections), and Agathis robusta (tangential and transverse
sections). As the text is in Japanese characters the author is
unable to say how far these species have been described. The
figures of the two Tasmanian species, Pherosphaera Hookeriana
and Fitzroya Archeri do not agree with the material I have worked
with. Since receiving the above publication I have received from
Tasmania more material of Fitzroya Archeri , and it agrees with
the specimens previously obtained. The figure of Pherosphaera
Hookeriana given by Kanehira is that of F iter ova Archeri . The
figure given for Fitsroya Archeri is probably that of Pherosphaera
Hookeriana. The radial sections of Dacrydium Franklwii and
Phyllocladus rhomboidalis are very similar/a fact which has been
discussed in the preceding pages.

In the figure of Callitris glauca the compound pit on the medul-
lary ray has been shown, and the plates of thickening on the bor-
dered pits of the tracheides are also shown. Neither transverse
nor tangential sections are given of any of the species.

EXPLANATION OF PLATES.

Plate I.

Podocarpus data. — A, Transverse section showing the scatter-
ing resin cells and the indistinct boundarv of an annual ring-
x 80. " 5 ‘

B, Radial section, showing the boundary of an annual ring, and
resin in the medullary rays. x 80.

C, Tangential section showing the resin cells, x 80.

Plate II.

A, Athrotaxis cuprcssoides. Transverse section showing the
resin cells zoned in the autumn wood, and the very strongly
marked boundary of the annual ring, x 70

B, Athrotaxis selaginoides. Radial section showing the bordered
pits arranged in pairs and also Bars of Sanio. x 190.

C, A. cuprcssoides. Tangential section showing the resin cells,
x 70.

Plate III.

Diselma Archeri. — A, Transverse section showing the large,
unthickened cells, squarish in outline, x 230.

B, Radial section showing parenchyma and arrow-shaped cells
of the medullary rays, x 155.

C, Tangential section showing simple and bordered pits and the
short medullary rays, x 17u

Plate IV.

Agathis microstachys. — A, Transverse section showing the in-
definite boundary of an annual ring and an isolated resin tra-
cheide. x 100.

R. T. Patton : Australian Coniferous Timbers.

B, Radial section, showing the irregular nature of the cells of
the medullary rays, x 68.

C, Tangential section, showing the alternating hexagonal bor-
dered pits, x 100.

Plate V.

Callitris calcarata. — A, Transverse section showing scattering
resin cells and the rounded nature of the tracheides. X 70.

B, Radial section, showing the plates of thickening lying across
the bordered pits, x 1 56.

C, Tangential section, showing the awns arising from the bor-
dered pits projecting into the cavities of the cells, x 110.

Proo. R.S. Victoria, 1927. Plate 1

Podocarpus elata.

Proc. R.S. Victoria, 1927. Plate IT.

H. Marriott, photo.

A and C, Athrotaxis cupressoides.
B, A. selaginoides.

Proc. P.S. Victoria, 1927. Plate III

H. Marriott, photo.

Diselma Archeri.

Proc. R.S. Victoria. Plate IV.

H. Marriott, photo.

Agathis microstachys.

-> *

Proc. R.S. Victoria, 1927. Plate V.

H Marriott, photo.

Callitris calcarata.

[Proc. Roy. Soc. Victoria, 40 (N.S.), Pt. I.. 1927.]

Art. II. — Vitality of White Maces in Low Latitudes.

By C. H. WrCKENS.

[Read 7th April, 1927.]

One of the visitors to the Pan-Pacific Science Congress held in
Australia in 1923 was Professor Ellsworth Huntington, Research
Associate in Geography in Yale University. On his return to
America he published an account of his journey under the title
“ West of the Pacific,” in which he dealt with various aspects of
Chosen, Japan, China, Java, and Australia, and this paper is
concerned with that part of his book in which he uses Australian
statistical data, and especially data concerning Queensland, to
support a theory of his that persons of white race born in low
latitudes have less physical vitality than similar persons born in
temperate climates. Whether his theory is true or false, the
author will not undertake to say; all he wishes to do is to call
attention to the nature and extent of the evidence available in re-
spect of Queensland, and to suggest that this evidence does not
appear to support the theory.

Professor Huntington points out that during the eight years
immediately preceding the War, the crude death rate was lower in
Australia than in any country in the world, except New Zealand,
and that the Queensland rate was as low as that of any State in
Australia. 1 le also shows that even when allowance is made for
difference in age distribution, Queensland has still a lower death
rate than the healthiest country in Europe. lie says that when
the low death rate of Queensland was first brought to his atten-
tion, he thought there must be some mistake, but that careful in-
quiry had convinced him that the records are essentially reliable.
He also refers to the evidence of good health in Queensland fur-
nished by the investigations of the Institute of Tropical Medicine
at Townsville, and by the Bulletin on Tropical Australia issued
by the Commonwealth Bureau of Census and Statistics.

Having thus examined and found satisfactory the evidence of
vitality furnished in respect of persons living in Queensland, he
next investigates data concerning those born in that State, and
comes to the conclusion that “ although the people who go to
Queensland are so healthy that they reduce the general death
rate to a very low level, their children for some reason or other
are less healthy than are those born in the more southerly parts
of Australia or in the Old Country.” On the question of fertility
in Queensland, he notices the relatively high birth rate of that
State as compared with the rest of Australia, and says that as
in the case of deaths, the favourable condition is not due to the
people who are born in Queensland, but to those who come hither

C. H. Wickevs :

from other regions.” He thus considers it as established that birth
in Queensland of persons of white race increases the rate of mor-
tality and decreases the fertility of the race. The evidence which
he adduces to prove this twofold conclusion are two statistical
tables, the one on page 364, the other on page 366 of the book
under review, both based on data obtained from the Common-
wealth Statistician.

The first of these is an interesting table in which he shows,
from calculations that he has made on the basis of figures sup-
plied at his request, that for ages last birthday 15 to 49 inclusive,
the rate of mortality for each sex of persons born in Queensland
was sensibly higher than for Australian residents born in Victoria,
New South Wales, England or Scotland, and that such is the
case whether the State of residence is Victoria, New South
Wales or Queensland. The figures supplied to the author
gave death rates for quinqennial ages which he has sum-
marised for the range 15-49, by using a “standard popula-
tion.” Exactly what standard he has used he does not indicate.
His published results, however, appear to be in reasonable accord
with the data supplied to him. Covering as they do deaths during
the three years 1920, 1921, and 1922, of persons aged last birth-
day 15 to 49 inclusive, they are derived from births which oc-
curred in the period from 1870 to 1907 inclusive, and consequently
take no account of the remarkable improvement in infant and
child mortality which has taken place in Queensland during the
past fifteen years. In only one year in that period of fifteen has
the rate of infant mortality — the number of deaths under one year
per 1000 births — been higher in Queensland than the Australian
average. That was in 1919, a year in which Queensland experi-
enced a drought much severer than that experienced in other
States. For the whole period of 15 years the Australian average
rate was 7% higher than that for Queensland, and the Queens-
land rate for 1925 of 45 per 1000 births is the lowest ever
recorded for an Australian State, and not much in excess of the
remarkable rate of 40 per 1000 births recorded by New Zealand
for the same year. These figures show that for the first year of
life at all events it cannot be said that under modern conditions
the Queensland born are less healthy than those born in the more
southerly parts of Australia.

Fortunately, it is not necessary to stop here, for it is possible to
analyse the death rates to age 9 last birthday, inclusive of persons
born in the several States. This investigation was based on the
deaths which occurred at these ages during the three years 1920,
1921, and 1922. The mortality for these three years indicated
that for Australia as a whole out of every 1000 males born, 106
would fail to reach the age of 10, the corresponding number for
females being 87 out of every 1000 born. The figures for
Queensland for the same period were 102 failures out of every
1000 males born, and 84 failures out of every 1000 females born,
about 3J% better than the average Australian rate in both cases.

Vitality oj White Races .

The only State showing a better result than that for Queensland
is Tasmania, which out of 1000 male births had only 101 failures
to reach the age of 10 compared with 1Q2 for Queensland. For
females, however, the Tasmanian failures numbered 86, compared
with only 84 in Queensland.

For the purposes of this paper a series of triennial results for
Queensland, covering the five triennia ending with the year 1925
have been taken out. These give the failures to reach age 10 out
of 1000 males born as 116 for 1911-13. 112 for 1914-16, 106 for
1917-19, 102 for 1920-22, as already quoted, and only 87 for 1923-
25. In the case of females, the results are even more striking, the
failures to reach age 10 out of 1000 females born being 101 for
1911-13, 96 for 1914-16, 88 for 1917-19, 84 as already mentioned
for 1920-22. and only 70 for 1923-25. The author is at present
engaged in taking out a similar series of triennial rates for the
other States, but these are not yet completed. The heavy part of
the mortality under age 10 is, however, that under age 1, and data
in respect of this are available, indicating for Queensland for the
triennium 1923-25 an average for the sexes combined of 50
failures to reach age 1 per 1000 births, compared with 57 for Aus-
tralia as a whole ; that is, the Queensland rate was more than 12%
better than that for all Australia.

While on this subject of progressive improvement in mortality
rates, reference may be made here to the marked improvement in
rates of mortality at all ages that has taken place in Queensland
since the 'eighties, when that State was quoted as the shocking
example in the matter of high mortality in Australia. An examin-
ation of the crude rates of mortality for that decade discloses the
fact that in each year the male crude rate for Queensland was
consistently higher than the Australian rate for the same year, and
in one year (1884) there was an excess of as much as 50%. An
examination of the Queensland migration records shows that in
that decade there was a larger net immigration of males into
Queensland than in any similar period in the history of the State.
This suggests that the process of acclimatisation was expensive in
terms of human lives, and does not bear out Professor Hunting-
ton's view that the new arrival was the select of the select. In
fact, many of the deaths which he quotes in his table would repre-
sent first generation Australians, the offspring of these immi-
grants of the 'eighties, whereas the progressively improving re-
sults that have been given for recent years are increasingly second
or third generation Queenslanders. Concerning the population resi-
dent in Queensland, it may be mentioned that the expectation of
life at date of birth for 1881-90 was 41*3 years for males, and
49-8 years for females; for 1891-1900 it was 49-5 years for males
and 55 - 8 for females, for 1901-1910 it was 54-2 years for males
and 59-3 years for females. For the three years 1920-22 com-
plete life tables for Queensland have not yet been compiled, but
there is evidence of a corresponding improvement in both sexes.

C. H. W tokens ;

This consistent and rapid increase in the expectation of life at
date of birth was of course associated with an increasing propor-
tion of Queensland born in the population. A comparison of
these expectations of life with those for Australia as a whole in-
dicates that whereas for 1881-90 the Queensland male expectation
at date of birth fell short of that for Australia by more than 12%,
for 1901-10 the excess was less than 2%, and the indications for
1920-22 are that the Queensland expectation will exceed that for
Australia. There has been a marked improvement in mortality
rates throughout Australia, but the improvement has been more
marked in Queensland than in the rest of Australia, although
Queensland has been the only part of Australia that has had any
serious addition of Queensland-born persons to its numbers. In
all the circumstances it may be claimed that the weight of evi-
dence is against Professor Huntington’s verdict concerning the
vitality of the Queensland born.

We now come to the question of fertility, which Professor
Huntington claims decreases with birth in Queensland. The only
evidence on this point which he gives in his book is the

table mentioned as being on page 366, and the con-

clusion which he draws from the figures there quoted is
palpably fallacious. He has there a statement showing
the total issue at time of death of persons of various birth-
places, who died in Australia during the year 1921, and be-
cause those born in Queensland who died in that year had
smaller average issue than any of the others which he records, he
draws the conclusion that the Queensland-born were less fertile
than the others. What he has failed to remember is that the issue
of a person at date of death is a function of age, and that in the
case of deaths in Australia of persons who had been born in Ger-
many or Ireland or Scotland or England, the proportion of ad-

vanced age and, consequently, with maximum families will be much
larger than in the case of those born in Australia, and that in con-
sequence of the rapid comparatively recent growth in the number
of Queensland born the proportion of deceased Queensland born
with maximum issue will be smaller than in some of the older
States.

The case for the fertility of the Queensland-horn, however,
does not depend solely on the negative process of proving the
invalidity of Professor Huntington’s evidence. There is ample
positive evidence that the fertility of the Queensland-horn females
if not high, is as least as high as that of the females of corre-
sponding age born elsewhere than in Queensland, and resident in
Australia at the Census of 1921. The following table, which
deals with place of residence, not place of birth, is of interest,
and is included partly because the data in respect of the issue of
males according to birthplace is not available, and partly because
it furnishes some interesting comparisons with birthplace data
deduced later.

Vitality of White Races.

Average Issue at Census of 1921,

Age

Husbands

resident in

' Wives

resident, in

Queensland

Australia

Queensland

Australia

25 - 29

1*38

1-32

1-94

1-78

30 - 34

2*17

2-06

2-82

2-59

35 - 39

2-95

2-77

3-63

3-32

40 - 44

3-67

3-40

4-29

3-84

45 - 49

4-35

3-91

4-70

4-19

50 - 54

4-89

4-34

5-28

4-57

55 - 59

5-45

4-86

5-79

5-12

60 - 64

5*95

5-42

6-32

5-74

65 - 69

6*55

6-04

6-58

6*25

These figures indicate that the average issue at every age was
markedly higher in Queensland than in the rest of Australia at the
Census of 1921. There are, however, interesting supplementary
figures relating to average issue of wives according to age and
birthplace. These are as follows : —

Average Issue of Wives Resident in Australia at Census of 1921 .

Wives born in

Age

Queensland

Australia

British Isles

Europe

All Birth Places

25 - 29

- 1-98 -

1*84

- 1*32 -

1*36

- 1*78

30 - 34

- 2*86 -

266

- 2*04 -

209

- 259

35 - 39

- 3-69 -

344

2’70

2*76

- 3*32

40 - 44

- 436 -

397

- 3*30 -

3*36

- 3*84

45 - 49

- 4-74 -

4-30

- 3*75 -

3*84

4*19

50 - 54

- 530 -

4-64

4*28 -

4*36

- 4*57

55 - 59

- 5-92 -

521

4*79 -

4*88

- 5*12

60 - 64

- 671 -

593

5*28

5 37

- 5*74

65 - 69

- 7*38 -

6*61

- 5*80 -

5*88

6*25

We have here all the important ages, and at every one of
them there is a larger issue for the Queensland-born wife than
for the wife born from any other of the quoted birthplaces. A
comparison of the issue of Queensland-born wives shown in the
last table with the issue of wives of all birthplaces resident in
Queensland, as shown in the table before that indicates that birth
in Queensland connotes in respect of a wife a somewhat higher
issue than mere residence in Queensland, which appears to be con-
trary to Professor Huntington’s theory.

It will be convenient here to call attention again to Professor
Huntington’s evidence in respect of Queensland's fertility. He
has there committed that statistical fallacy sometimes known as
the “ fallacy of aggregates,” an error by no means uncommon,
and not always immediately evident. It may be illustrated by

C. II. Wiclcevs :

an example mentioned by an English Registrar-General, who said
that statistics showed that the occupation of farmer had a death
rate, over all, higher than that for the general population, but that
when death rates for successive age groups were compared, the
farmer’s rate in each age group was less than that of the general
population. The reason for the farmer’s higher death-rate when
age was not considered, was that there were so many more far-
mers at the advanced ages at which the death rates were high.
In other words, the farmer’s apparently high death rate was due to
his really low death rate. In the case of fertility, the results of
the 1921 Census indicate that, although the issue of Queensland-
born wives is high in each age group, yet, if the results are taken
irrespective of age, Queensland-born wives have an average issue
over all of 3T8; Australian-born wives, 3-30; wives born in the
British Isles. 3 -46; wives born in Europe, including the British
Isles, 3*53; and wives of all birthplaces, 3-34. As already ex-
plained, this is due to the larger proportion of Queensland-born
wives in the lower age groups, where the families are in all cases
smaller.

The data so far dealt with concern the whole of Queensland,
but not more than 25% of the population of Queensland is actually
within the tropics. The southern boundary of Queensland is
approximately 29° South latitude, so that the whole State can be
classed as being in low latitudes, if not altogether within the
tropics, A special Census Bulletin dealing with Tropical Aus-
tralia was issued in connection with the Census of 1921, and this
was supplemented later by the issue of a special Part of the
Census Report (Part XXVI. ) on the same subject as well as by
tropical sections of Part XXVIII.* Families. It is unnecessary to
refer to the details contained therein, but some figures compar-
ing the issue of Oueensland-born wives resident in tropical Aus-
tralia with the corresponding issue of wives born elsewhere, but
also resident in tropical Australia, may be quoted.

Average Issue of Wives resident in Tropical Australia at

Census

of 1921.

A »•

Wives born in

Queensland

Australi;

:i British Isles

Europe

All Birth Places

25 -

- 219

- 214

1-62 -

175

- 209

30 -

- 316

3’08

262

279

304

35 -

- 401

3*87

- 361

364

- 3-81

40 -

- 479

- 4-62

- 413

421

4 45

45 -

- oil

- 4-89

- 4-68 -

4-71

- 479

50 -

- 538

5 26

- 5*23 -

527

5*24

55 -

- 598

582

- 569 -

5-73

5 75

60 -

5 ‘76

- 592

- 61)9 -

6-20

- 611

65 -

677

7 05

- 5-71 -

5*90

- 6-01

Vitality of White Races. 2:8

Owing to the relative smallness of the numbers involved, the
rates here are somewhat less regular than those previously quoted,
but up to age 60 the total issue of Queensland born wives pre-
dominates. and for all birthplaces the total issue is in general
up to age 65 higher for tropical Australia than for all Australia.
Data concerning wives born in tropical Australia are, unfortu-
nately, not available.

There is a further small point in Professor Huntington’s critic-
ism of Queensland to which reference may be made. This is a
statement by him that there is a tendency for numbers of
Queesland females of adult age to get out of Queensland, with
the view apparently of so avoiding the climatic disabilities of that
State. This statement appears to be based mainly on the fact that
when the Census data for the Queensland population according to
sex and age are examined, it is seen that for early ages the num-
bers in each sex are fairly equal, but that at later ages there tends
to be a preponderance of males. His conclusion, however, is quite
wide of the mark. Like all new and progressive countries,
Queensland has an excess of males, an excess which with an in-
creasing number of births per annum is rapidly disappearing.
This disappearance is of course most marked in the younger ages,
which are mainly recruited from the local births in which there is
little difference in the proportion of the sexes. With the lapse
of time the ages having a marked male preponderance become
higher and higher, and in the absence of heavy immigration even-
tually disappear. The statistical peculiarity to which he refers is
thus due, in the main, not to a marked exit of adult females, but
to the influx by birth of infant females and their subsequent re-
tention in the State. This is indicated by the following table,
which shows at the Census of 1921 the numbers of each sex of
Queensland-born persons who were resident in Queensland at the
date of the Census. At all ages the numbers approximate equal-
ity, the excess of males being most marked at the younger ages
• — not, as suggested by Professor Huntington, at the older.

The total number of Queensland-born residents of Queensland
at the Census of 1921 was thus almost equally divided as regards
sex, and approximate equality was in evidence in each age group,
the most marked deviation being the excess of males under 5
years, due mainly to the normal excess of males at birth. Ifata
concerning the birthplaces of residents of other parts of Austra-
lia at the Census of 1921 indicate, however, that there is a slightly
higher migration of Queensland-born females than of Queens-
land-born males to the other States. The number of Queensland-
born persons recorded at the Census of 1921 in States other than
Queensland was 42,953, of which 20,142 were males and 22,811
females. In other words, at the Census of 1921 about 9}% of the
Queensland-born females resident in Australia were living out-
side their State of birth, and about 8-J% of Queensland-born
males. These proportions for residence outside the State of birth
are less than for any State of birth except New South Wales,

C. H. Widows:

Queensland-born Population Recorded in Queensland at Census

of 4th April , 1921

A g;e Group

Queensland born population

recorded in Queensland.

Males

Females

0 - 4

40,106

38,162

5 - 9

36,972

36,175

JO - 14

29,755

28,695

15 - 19

24,430

24,370

20 - 24

22,703

24,407

25 - 29

21,201

22,719

30 - 34

19,092

19,835

35 - 39

13,229

13,478

9,887

9,751

45 - 49

7,292

7,262

50 - 54

5,424

5,140

55 - 59

2,645

2,554

60 - 64

1,086

1.103

65 and over

491

577

Not stated

325

319

Total

234,638

234,547

the average for all Australia being 1H% for males and 10f% for
females. That is to say, there is a smaller rate of migration from
the State of birth among the Queensland-born of either sex than
is the case with those born in any other State of Australia, except
New South Wales. These are the figures for 1921. In 1911 the
position was very similar, Queensland’s figures being 8% for
males and 9% for females, and ranking still second to New
South Wales, whereas the figures for all Australia were 12^%
for males and 10f% for females. A slight female preponderance
in the migration of the native-born is not peculiar to Queensland,
but is also in evidence in Western Australia, and, to a larger ex-
tent, in Tasmania.

Summing up the position, it would appear that in depending on
Queensland to help in the establishment of his theory. Professor
Huntington has put his money on the wrong horse. If his theory
is true, there would appear to be some remarkable counterbalanc-
inf advantages, geographical, climatic or other, in the case of
Queensland, which must be counted to that State as a most valu-
able asset. As stated earlier, the immediate object of this paper is
not that of proving either the truth or the falsity of Professor
Huntington’s general theory, but of showing that available
Queensland data really furnish no evidence in its favour under
modern conditions. Whether the further extension of white
population into the tropical portions of the State will give equally
favourable results has yet to be ascertained, but the data avail-
able in this connection are of such a nature as to warrant expecta-
tions of satisfactory progress.

[Proc. Roy. Soc. Victoria, 40 (N.S.), Pt. I., 1927.

Art. III . — The Stony Greek Basin and the Covinella Dyke.

By D. ORR, B.Sc.

[Read 4th June, 1927.]

Contents.

1. Introduction.

2. Previous Literature.

3. Nature of the Basin.

4 Origin of the Basin.

5. The Corinella Dyke and its Relation to the Stony Creek

Basin.

6. Summary.

I. — Introduction.

The subject of this paper is a curious, ainphi theatrical depres-
sion about 50 acres in area lying immediately south of Jubilee
Park, 1-J miles south of the Daylesford Post Office. The Stony
Creek enters it in the south-west corner by a narrow gorge, flows
along the western side, and leaves by a similar gorge in the north-
west corner. Exposed by sluicing channels in the basin are lig-
neous shales with numerous fossil Eucalvpt leaves, remains of
diatoms, and abundant fresh water sponge spicules. The deposits
have been penetrated in a shaft for over 100 feet.

The map showing the relation of the Stony Creek basin to the
Corinella dyke is compiled from Quarter-Sheet No. 16 SE. of the
Geological Survey. A correction due to Mr. Whitelaw (1) is
made to the boundaries of the basalt and Ordovician in the neigh-
bourhood of the basin.

Much of the information about the shafts and deep leads in the
Eganstown district was obtained from Mr. Rehir, of the Victoria
Hotel, Daylesford, who was digging there in the '90\s.

II.— Previous Literature.

In his notes on Quarter-Sheet No. 16 SE., Mr. Hunter (2)
refers to the basin as “ a deep hole without an outlet,” and con-
siders it a point of Pliocene volcanic eruption. In a recent con-
versation Mr. Hunter informed me that he is now inclined to
regard the basin as a faulted block.

T. S. Hart (3) has discussed the origin of the basin in some
detail. His view of the sequence of events is .summarised as
follows: — The black clays were deposited prior to volcanic ac-
tivity, or as a result of the first modifications of the drainage
systems by the volcanic action. A portion of these was pre-
served by subsidence on a well defined line of weakness, namely,
a j* ne through the Corinella dyke, the zone of fracture in
Sailor’s Creek, and Wheeler’s Hill. The streams flowing on the

D. Orr :

west and north sides of the basin may have flooded it. The lava
flows from Leonard’s Hill then buried the old river valleys and at
least part of the basin. The drainage towards the pre-basaltic
stream resulted in the formation of Sailor’s Creek on the western,
and Stony Creek on the eastern, side. The basaltic barrier across
the exit from the basin formed a bar, which was only cut through
slowly. While the bar was being cut through, the stream cut out
a plain in the easily eroded black shales. A considerable amount
of basalt could have been removed at the same time by under-
cutting.

While these latter conclusions are most probably correct, the
earlier sequence leading to the deposition of the ligneous shales
seems to be capable of a more satisfactory explanation.

Mr. Whitelaw (1) states that “the area is a foundered block
at the intersection of the Ajax group of thrust faults, and a
younger cross fault.' 7

III. — Nature of Basin.

The floor of this crateriform depression lies about 100 feet
lower than the rim. It is surrounded on the west and the north-
ern sides by a basaltic plateau, in which is entrenched the Stony
Creek Gorge. Deep leads occur under the basalt, but about 50
feet higher than the level of the basin. The southern and eastern
banks are formed of Ordovician slates and shales, which rise to
somewhat higher levels than the basaltic plateau. In the north-
east. corner. Pliocene alluvial gravels outcrop at the surface. The
floor of the basin, which consists of recent alluvium, is fairly
le\ eh rising to the north and the east. It is dissected by numerous
sluicing channels. In the most easterly of these, black ligneous
shales are exposed at several places. When dry, they change to a
drab colour. They are very fine grained, and contain nume-
rous wood fragments and fossil Eucalypt leaves. Diatoms and
fresh water sponge spicules have been also recorded (1) from
them. These deposits have evidently accumulated in the still
waters of a lake, or some allied formation. Mr. Hart claims
that the shales dip from 45° to vertical, and that they therefore
have been much disturbed since their deposition. In the sections
examined in the basin the bedding was generally obscure, but no
evidence of a steep dip was obtainable. In 1864 a shaft was sunk
near the southern wall through these ligneous shales for over
100 feet. Alluvial wash was struck at 111 feet. It is doubtful
whether the shaft ever reached bedrock.

Thin seams of similar black ligneous clays occur under the basalt
at Sailors and Stony Creek Falls. These contain a fair propor-
tion of coarse grit, and no traces of either diatoms or sponge
spicules, as would be expected from the nature of their occur-
rence.

Similar deposits have been recorded (4) from the Exchequer
Co.’s shaft on the Royal Oak lead from Wombat Hill, where a

Stony Greek Basin.

■27

thickness of 85 feet was passed through. The deposit contained
wood in all stages of transformation into lignite, intermixed with
-leaves in all stages of preservation, identical with those of the
present day. The wood was frequently replaced by pyrite. At
Eganstown the shafts of the Great Extended, whose claim adjoins
the eastern boundary of the Corinella pre-emptive right, and the
shaft of the New National, whose claim joins the eastern boun-
dary of the Great Extended claim, both passed through 100 feet
of tripoli (diatomaceous earth).

Conditions at the time of the Newer Volcanic activity were
evidently favourable to the development of diatomaceous life, as
the deposits formed by the accumulation of their remains are asso-
ciated with the Newer Basalt in many Victorian localities (5).

As previously stated, Mr. Hart considers that the black ligneous
shales and clays of the Stony Creek basin are remnants of a much
more extensive deposit preserved by faulting. He suggests their
correlation with the deposits met with in the Exchequer shaft, and
in the Great Extended and New National shafts at Eganstown.
This would imply the existence, immediately prior to the Newer
Basaltic eruptions, of lake conditions necessary for the accumu-
lation of a deposit 100 feet in thickness over the greater part of
the Daylesford area. If such conditions had existed, one would
expect to find many other remnants of the deposit. These are
absent, and it is more probable that these three areas of ligneous
clays are of quite local and restricted occurrence, and that the
Stony Creek Basin deposits have accumulated there, filling in an
originally much deeper basin.

The most probable direction of the deep leads is shown in the
map. At the places where the lead is shown to enter and leave
the basin, the basalt extends down to a lower level than else-
where. On the western side it is difficult to determine how much
the relations have been interfered with by landslips. The
steep slopes are covered with a dense growth of blackberries,
which adds to the difficulty of locating the rock boundaries. Land-
slips have also obscured relations where the lead is shown to cross
on the east bank of the Stony Creek gorge. It is however fairly
certain that the basalt reached a much lower level here. River
gravels which outcrop beneath the basalt in this neighbourhood
also indicate that the lead crosses in that region. The exit of the
lead exposed in the cutting of the Ballarat Road can be readily
and definitely located.

Prior to entering the basin, the bed of the Stony Creek, down-
stream from the falls, consists of Ordovician slates and sand-
stones. Immediately it enters the basin all signs of Ordovician
in the bed of the creek disappear. The stream then flows over
basaltic boulders until the north-west corner is reached, where
Ordovician reappears and, from there down-stream, continues
to form the bedrock of the creek. Besides forming the high east
and south banks, the Ordovician slates and sandstones outcrop

JJ. Orr:

beneath the basalt on the north and west sides. The shaft near
the south bank, previously mentioned as passing through more
than 100 feet of black ligneous clays, shows that the junction
between clays and Ordovician must continue very steeply below
the surface here.

IV. — Origin of the Basin.

To account for such a formation in which a floor of Ordo-
vician slates and sandstones is enclosed by walls of Ordovician
and Ordovician capped by basalt, which rise to a height of 200
feet above the Ordovician floor, two modes of origin may be sug-
gested. These are (a) volcanic explosive activity, and (t>) sub-
sidence by faulting.

The first method was suggested by Mr. Hunter (1). The evi-
dence rather indicates that such was not the case, for although
the basin is surrounded on the north and west sides by basalt,
this had its source at Leonard’s Hill, some six miles to the south.
Nor is there any sign of volcanic fragmental rocks or accumula-
tions of broken Ordovician material around the basin, as would
be the case if it were due to a volcanic explosion.

This leaves the second method of origin, namely, faulting.
As will be seen in the next section, the basin lies at the extremity
of a well defined line of weakness and Assuring, at the time of the
Newer Basaltic eruptions. It is, nevertheless, rather difficult to
picture the cause and manner by which a cylindrical block could
be depressed vertically 200 feet, by faulting'

Consequent on the formation of the basin by faulting, thq
sequence of events was probably as follows : — It was flooded by
the pre-basalt ic stream whose valley was later filled by the basalt
flow from Leonard’s Hill. The course taken by this stream is
indicated in the map showing the direction of the present deep
leads in the neighbourhood of the basin. In the lake thus formed,
especially in the still backwaters, conditions would be favourable
to the accumulation of the black ligneous shales. That the mate-
rial brought into the basin by the stream was of a suitable nature
to form such a deposit is proved by the presence of similar black
ligneous clays beneath the basalt at Stony Creek falls, and fur-
ther south at Sailor’s Creek falls. It is probable that the basin
was entirely filled by these ligneous clays, as there is no indication
that much of the area occupied by the present basin was covered
by basalt, and certainly not by any great thickness of basalt. Next
came the pouring out of lava from Leonard’s Hill, filling the old
creek valley, its channel through the basin, and covering possible
part of the present basin. The concentration of the drainage
down the old river slopes towards the basalt flow gave rise
to the present Stony Creek on the east side and Sailor’s Creek on
the west side of this flow. While cutting through the basalt bar
at the north-west end of the basin, the Stony Creek cut a plain in
the easily eroded black shales, undermining any basalt that might

Stony Creek Basin.

29-

have overlain these. Since the extrusion of the Newer Basalt, the
streams in the Daylesford area have been rejuvenated, and the
present stream level is now generally some 50 feet below the level
of the old deep leads. Once the bar was cut through, the basin
was drained and deepened as the Stony Creek deepened its bed,
so that the floor is now well below the level of the old deep leads.
Recent alluvium derived from the slopes has covered the ligneous
clays, which are at present only exposed in the sluicing channels.

V The Corinella Dyke and its relation to the

Stony Creek Basin.

The boundaries of the feature marked in the map as the Corin-
ella Dyke are largely hypothetical. They have been taken from
Quarter-Sheet No. 16 SE. of the Geological Survey. In the
Corinella pre-emptive right paddock, where it is shown with
a width of 14 chains, it is questionable how much of the basalt
between these boundaries can be regarded as actual dyke and how
much as surface rock. The basaltic hill shown on the dyke in the
Corinella paddock is characterised by the vesicular nature of the
basalt there as a point of eruption. Burnt Hill immediately to the
SW. is very curious in structure. It is marked on the survey
map as Ordovician, and a note is appended to the effect that it is
a “ made hill of deposit.” It seems to consist on the upper slopes
of a mixture of Ordovician and basaltic boulders, mostly scoria-
ceous in nature. One, when broken, was found to be holocrys-
talline, probably a type allied to essexite. It was too incoherent
to permit a thin section being made for microscopical examina-
tion. This hill is probably the result of explosive volcanic
activity.

A microscopical examination of sections cut from samples
taken at various places along the dykes, shows the material to be
a rather fine grained olivine basalt. Much of it is, on a small scale,
of a spheroidal character, which gives outcrops of the un-
weathered rock a characteristic appearance.

Much of the surface rock in the Corinella paddock is extrusive
basalt that has filled in pre-basaltic river valleys. The deep leads
have been worked and found to be very rich. At a point on the S.
margin of the Corinella dyke (opposite the letter “ D v in “ Dyke 77
on the map) a lead, which was being worked in 1895, ran against a
a wall of basalt and was lost. It had proved to be very rich at this
locality, and its disappearance gave rise to considerable discussion.
On the advice of Mr. Hunter, who was surveying the area at the time,
it was sought on the north side of the dyke, where it was subse-
quently picked up. To try and locate the lead a shaft was sunk
between these two points to a depth of 120 feet. It was aban-
doned still in basalt.

Further east two other shafts have been sunk in the dyke. The
Great Extended shaft (4) bottomed at 240 feet with a “ dip of 16-

OF CORINELLA DYKE & STONY CREEK BASIN.

D. Orr :

-j

£>—4

Stony Creek Basin.

feet ” in the shaft. It is supposed to have passed through infu-
sorial (diatomaceous) earth. The shaft is on the northern mar-
gin of the dyke, and only passed through, several feet of basalt.
Levels opened towards the south met a wall of basalt. No sign
of diatomaceous earth was found in the dump heap around the
shaft. This was found to consist largely of a volcanic agglom-
erate composed of fragments of Ordovician slate and sandstone,
and fragments of volcanic rock set in a fine greyish matrix that
is possibly largely volcanic. Professor Skeats (6) has recorded
a somewhat similar occurrence of a monchiquite agglomerate
from Kangaroo Gully near Bendigo. He concludes that it had
originated through explosive volcanic activity. It is possible that it
was this material that was mistaken for the diatomaceous earth
through which the shaft was supposed to have passed. It is sug-
gested that this agglomerate fills a fractured zone in the Ordovician
bedrock, and owing to a southerly dip in this zone the shaft passed
into normal Ordovician at a depth of 246 feet.

Further east the dyke narrows to a width of 2 to 3 chains. In
1865 the New National Co. sank two shafts here. The first, which
reached a depth of 250 feet, is recorded (4) as passing through
50 feet of clay, 100 feet of infusorial remains, and 100 feet of
basaltic boulders and drift. It was abandoned at 250 feet in
drift. The other shaft bottomed at 273 feet in the north-east corner
with a “ dip of 10 feet ” to the south-east in the shaft. Ordo-
vician bedrock is proved by shafts to extend up to the dyke on
both sides at this part. As the dyke is only 2 to 3 chains in width
the accumulation of over 200 feet of drift and diatomaceous earth
beneath it is a remarkable occurrence, especially since the lower
100 feet consists of “ basaltic boulders and drift.” As the deposit
is sub-basaltic and older than the basaltic dyke, no source for
these basaltic boulders suggests itself. It is most probable that
here again, as in the Extended shaft, the “ tripoli ” is really a vol-
canic agglomerate, a large proportion being a very line volcanic
paste decomposed to a greyish clay. Lower down, the pieces of
undecomposed volcanic material may be more numerous and
larger in size, and the agglomerate was then mistaken for
44 basaltic boulders and drift.”

From these shafts, the course of the dyke eastwards is marked
by, a red basaltic soil. The most easterly outcrop is in a quarry.
Here the dyke has split into two walls separated by Ordovician.
It evidently ends at this point, for it does not outcrop further
east. But between the quarry and the Stony Creek basin, a wide
fractured zone in the Ordovician rocks is exposed in the bed of
Sailor’s Creek. The rocks here have been shattered and twisted.
Forming a matrix of this broken material is a greyish white paste
which has no perceptible gritty feel. It is probably decomposed
volcanic matter. According to Mr. Whitelaw (1) this fault dis-
places the country to the north side 5 chains to the right. There
is no trace of this fractured zone between Sailor’s Creek and the

D. Orr :

Basin, though it undoubtedly extends further eastward. The
nature of the country here makes it very unlikely that any sur-
face outcrops could be detected.

To account for this Corinella Dyke two hypotheses are avail-
able. The first is not very satisfactory, hut would account for the
presence of diatomaceous earth in the Great Extended and the
New National shafts. According to this hypothesis, the first
event to take place was the fracturing and Assuring of the Ordo-
vician rocks, developing a depressed trough-faulted block in the
Corinella paddock, and narrowing into a deep fissure towards the
Great Extended and New National Co.’s shafts. Further east it
is represented by the fractured zone in the bed of Sailor’s Creek
and terminates in the foundering of the Stony Creek basin
block. The later development of this basin has already been
traced. The Corinella down faulted area was probably flooded by
the streams in the area and partly filled by drift. The presence
of the basaltic boulders in this drift cannot be explained, and
remains an important argument against this mode of origin. The
amount of alluvial material washed into this lake decreased, and
in the clearer waters, diatoms flourished and the accumulation of
their remains, mixed with finer sediment, formed a deposit about
100 feet in thickness. In Newer Volcanic times activity was renewed
along this zone of fracturing, and basalt was intruded in places
as a dyke, and extruded from a localised centre of eruption, filling
in the trough fault and fissure, and covering over the diatom-
aceous deposits. In the quarry the basalt is probably intruded
as a dyke. The volcanic hill in the Corinella paddock is the centre
of the effusive type of eruption. In the neighbourhood of the
Great Extended shaft, volcanic material was explosively injected
into the Ordovician along the old fault zone, forming the volcanic
agglomerate.

The second hypothesis pictures a much more probable series of
events. It assumes that the term “ tripoli ” has been mistakenly
applied to a fine volcanic agglomerate while a coarser agglomerate
has been mistaken for a drift containing basaltic boulders. Along
the Corinella fault line a shattered zone was developed in the
Ordovician, and into it was explosively injected volcanic material,
forming a dyke of volcanic agglomerate. In places, as at the
upper parts of the Extended and New National shafts, a fine vol-
canic matrix forms the predominant part of the agglomerate,
while at the Sailor’s Creek zone it is mainly composed of shat-
tered blocks of Ordovician. In the eastern extremity the sagging
of the beds formed the Stony Creek basin. Contemporaneously
with the formation of this volcanic agglomerate, or at a slightly
later date, came the injection of basalt along this line of weakness.
The intrusion in the eastern part was confined to a dyke, but in
the west became effusive, culminating in the central type of erup-
tion denoted by the volcanic hill in the Corinella paddock.

Stony Greek Basin.

VI. — Summary.

A peculiar depressed basin of about 50 acres in area occurs
immediately south of Jubilee Park, Daylesford. It is surrounded
on all sides bv Ordovician slates and shales, which rise to a height
of 100 feet above the floor of the basin. The north and west sides
are capped by basalt, under which are deep leads about 50 feet
higher than the floor. It is open to the Stony Creek, which flows
along the west side. Exposed in sluicing channels, are ligneous
shales and clays containing diatoms and fresh water sponge
spicules. These have been penetrated by a shaft for over 100
feet.

This basin is considered to have originated by the depression
of a block of Ordovician, and to have been flooded by a pre-
basaltic stream. It was filled by ligneous clays. The stream enter-
ing the basin was carrying similar material. In the quiet back-
waters, conditions would be favourable for the growth of diatoms
and fresh water sponges. The pre-basaltic stream, and perhaps
part of the basin, was then filled by basalt. The twin streams.
Sailor’s Creek and Stony Creek, developed. While cutting through
the basalt bar at the north side of the basin, Stony Creek cut out
a wide plain in the easily eroded black shales, undermining any
basalt that covered them. The basin was subsequently deepened
as the stream entrenched itself deeper into bedrock.

The basin lies at the east extremity of a line of fracture and
intrusion in late Kainozoic times. At Eganstown this is known
as the Corinella dyke. Shafts through this dyke are supposed to
have passed through diatomaceous earth. No trace of this is
found around the dump heaps, but one consists largely of a vol-
canic agglomerate with a large proportion of fine greyish matrix.
This was possibly mistaken for diatomaceous earth. Two alter-
native explanations are considered to account for the known facts.
The more probable is that a fractured and shattered zone was
developed in the Ordovician rocks. Accompanying this was the
explosive injection of volcanic material forming a volcanic
agglomerate. The proportion of volcanic material to shattered
Ordovician varies at different parts. Contemporaneously, or at
a slightly later date, came the intrusion of basalt along the same
line of fracture as a dyke, culminating in the west in extrusive
volcanic action.

REFERENCES.

1. H. S. Whitelaw. Bull Geol Surv. V-ic., No. 42, p. 11, 1923.

2. S. B. Hunter. Prog. Rept. Geol. Surv. Vic., No. 9, p. 71,

1898.

3. T. S. Hart. Proc. Roy, Soc. Vic., n.s., xvii. (2), p. 366,

1905.

4. Dicker’s Mining Record, Melbourne, 1865.

5. D. J. Mahony. Bull. Geol. Surv. Vic., No. 26, p. 9, 1912.

6. E. W. Skeats. Proc. Roy. Soc. Vic., n.s., xxvi. (2), p. 375,

1914.

Proc. Roy. Soc. Victoria, 40 (N.S.), Pt. I., 1927.]

Art. IV. — An Olivine Anortkoclase Basalt

from Daylesford.

By D. ORR, B.Sc.
[Read 9th June, 1927.]

Contents.

1. Introduction.

2. General Geology.

3. Physiography.

4. Olivine Anorthoclase Basalt.

(a) Distribution and General Character.

(b) Chemical Character.

(d) Age.

(e) Comparison with Newer Basalt and with other Alkaline

Rocks of Victoria.

(f) Origin.

5. Summary.

I.— Introduction.

This paper is mainly concerned with the chemical and petrologi-
cal character of an alkaline basalt that has been recorded from the
Daylesford district (1). It is shown to be an olivine anorthoclase
basalt. The recent dissection characteristic of the area is also
briefly discussed.

The area covered by the anorthoclase basalt is contained in
Ouarter-Sheets No. 16 NE. and No. 16 SE., of the Geological
Survey of Victoria. Notes on these quarter-sheets appear in Pro-
gress Reports No. 8 and No. 9 of the survey. The central portion
of the area was later surveyed and mapped on the larger scale of
1 inch to 16 chains by H. S. Whitelaw (2). In all these publica-
tions of the survey no distinction is made between the olivine
anorthoclase basalt and the normal newer basalts of the area. Its
special nature was first noticed during a geological excursion from
the University. It was described by Professor Skeats (1) as an
olivine anorthoclase trachyte.

II.— General Geology.

The rock formations outcropping in the area are in decreasing
order of age: — Ordovician, Older Pliocene, Newer Volcanic,
Newer Pliocene and Recent. All have been fully described in the
publications of the Mines Department already referred to.

The newer volcanic series which forms a large part of the sur-
face rock of the district is part of the northern outskirts of the
great basaltic plain of south-west Victoria. The lava is in the

Olivine Anorthoclase Basalt.

form of separate flows, which have filled up pre-basaltic river
valleys, sealing up valuable auriferous gravels. These flows
-originated from numerous volcanic vents, the volcanic hills built
up around which form one of the characteristic topographical
features of the district. The mode of eruption in this area differs
from that of the main plain, which is considered to have origin-
ated by the fissure type of eruption.

III. — Physiography.

The area dealt with in this paper lies on the northern slopes of
the Main Divide. The Divide crosses the south-east corner of
'Quarter-Sheet 16 SE. of the Geological Survey. Leonard's Hill,
near the southern extremity of the anorthoclase basalt flow, lies
about 20 chains north of the Divide.

The area is one of great diversity of surface, being very deeply
dissected by numerous streams which ultimately find their way to
the Loddon River. The slopes of the valleys and the Ordovician
hills are generally densely timbered, while the elevated basaltic
plains have been cleared for agricultural purposes. The numerous
volcanic flows have been largely instrumental in the development
of the present topography. The volcanic hills around the points
of eruption rise well above the general level, and form prominent
landmarks. The most prominent of these are Mount Franklin,
Bald Hill, Fern Hill, Wombat Hill, Wheeler’s Hill, and Leonard's
Hill. Whereas the basalt formerly occupied the valleys of the
pre-basaltic streams, it is now found on tbe higher ridges and
plains. Frequently twin streams have deeply entrenched them-
selves on either side of a basaltic flow, which now forms a high flat-
topped ridge between the two streams. Sailor's Creek and Stony
Creek are twin streams that have entrenched themselves on either
side of the anorthoclase basalt flow. The head-waters of both
these streams occupy only small surface gutters, which suddenly
develop into deep gorges by steep falls over the basalt. Sailor’s
Falls has a vertical drop of about 100 feet, Stony Creek Falls one
•of about 50 feet.

All the streams of the area occupy recent deeply dissected
valleys. The level of the pre-basaltic streams, as denoted by the
present level of the deep leads, is generally 50 feet or more above
the present stream level. The most recent flow from Mount
Franklin is an exception. It was evidently extruded when present
stream level was reached, since the Jim Crow Creek, which flows
along the eastern edge of the flow, has much of its bed in
basalt. If, as is generally considered to be the case, these pre-
basaltic streams had reached base level, the area must have been
relatively uplifted in post-basaltic times. To form any reliable
opinion as to the manner by which these streams had their erosive
power increased an intimate knowledge of the country to the
north of Daylesford would be required. The tinif necessary to
acquire this information was not available.

I). Orr:

A trip was made along the Jim Crow Creek as far as its junc-
tion with the Loddon River at Newstead, about 16 miles north of
Daylesford. This creek is formed by the junction of Sailor’s
and Wombat Creeks. As far as Franklinford, six miles north
of Daylesford, it occupies a very steep and deeply dissected valley.
Past Franklinford to the junction with the Loddon. the character
of the valley changes. The stream flows in an open mature
valley, rising on the west and east to low rounded Ordovician hills.
The floor is occupied by the most recent basalt flow from Mount
Franklin. It was first thought that the cause of this recent dis-
section, most probably faulting, might be located in the neighbour-
hood where the change in the nature of this valley took place. An
excellent panoramic view over this part of the country to the Lod-
don is obtained from the summit of Mount Franklin. No indica-
tion of any fault escarpment could be noticed.

It is most probable that this open character of the valley of Jim
Crow Creek past Franklinford is due to the greater ease of
weathering of the Ordovician sediments compared with the basalts,,
which have confined the streams in the Daylesford district to
narrower and steeper valleys. The fact that auriferous gravels,
of apparently the same age as the deep lead gravels, outcrop above
the present stream level towards Newstead indicates that the cause
of the dissection is to be sought still further downstream.

IV. — Olivine Anorthoclase Basalt.

(a) Distribution and General Character.

The source of this basalt is Leonard’s Hill, a big volcanic hill
about 20 chains east of Leonard’s Hill station, six miles south
of Daylesford on the Daylesford-Ballarat railway line. The
upper part of this hill shows all the characteristics of a point of
eruption, consisting of very vesicular, scoriaceous basalt mixed
with finer volcanic material. From this hill the lava flowed north,
filling an old pre-basaltic river valley and sealing up the river
gravels. These have proved auriferous in the northern part of
the area, where the Llewellyn lead has been extensively worked.

The basalt is continuous until the Stony Creek Basin is reached,
past which, due to the deep dissection of the Stony and Wombat
Creeks, it has been isolated as cappings to flat topped hills;
namely, Table Hill, Hard Hill, and the Jubilee Park Hill. On the
Quarter-Sheet (No. 16 SE.) the basalt is shown as continuous on
the west bank of Stony Creek as far as the Ballarat Road, thus
differing from Mr. Whitelaw’s map, which shows a ridge of
Ordovician, of the same level as the basaltic plateau, breaking the
continuity immediately north of the outlet from the basin. Exam-
ination showed the latter mapping to be correct. The pre-
basaltic stream evidently meandered at this locality.

South of Leonard’s Hill, but not connected with it, two parallel
basaltic flows extend in a south-easterly direction for a distance

Olivine Anorthoclase Basalt

of about four miles, where they junction at the Werribee River.
From here basalt outcrops intermittently on either side of this
stream as it is followed southwards On the northern extremity
of each of these flows is a volcanic hill. The hill in the northern
extremity of the westerly flow has been isolated from the main
flow by stream dissection. Except in places on the surface of the
basalt flows, where the land has been cleared, the country here is
extremely rugged and thickly timbered. Time did not permit of
any detailed examination of these flows or of their relation to the
anorthoclase basalt flow. Specimens collected from the two hills
at the northern end of each flow, and from an outcrop on the
Werribee River, about a mile below their junction, were sectioned.
The first two (2200, 220 1 ) 1 were very similar, and resembled a
fine grained variety found at the base of Leonard’s Hill, though
they differed in the absence of anorthoclase. The third specimen
was a typical olivine basalt, the phenocrysts of olivine being
altered to brown iddingsite. It is therefore probable that these
flows are distinct from the anorthoclase basalt flow, and were ex-
truded from the hills at their northern extremities. If this is so,
the flows would be older than the anorthoclase basalt flow, as all
traces of scoria and ash have been removed from these hills by
the processes of weathering, and now only basaltic plugs remain
standing above the general level of the flow. Leonard’s Hill,
however, is in a much better state of preservation, and is therefore
younger.

On the surface, the anorthoclase basalt has weathered to a rich
red soil similar to that derived from the other basalt of the area.
Exposed in a rock face at Table Hill Quarry there is a large lenti-
cular patch, surrounded by un weathered basalt, that has been
weathered to a whitish clay. The junction between the two types
is quite abrupt. It is rather difficult to picture this as due to the
•concentration of ordinary meteoric weathering processes; more
probably it is due to chemical action of enclosed magmatic waters
and gases. That there were abundant is suggested by the common
occurrence of calcite filling the vesicles in the rock.

Excellent sections showing the junction between basalt and
Ordovician bedrock occur at Stony Creek and Sailor’s Creek falls.
On the under surface, for a thickness of about 6 inches, the
basalt is very vesicular in character.

A number of specific gravity determinations were made on
specimens collected from different parts of the flow. They varied
from 2-881 to 2-791, with an average of 2-838.

(b) Chemical Character.

The specimen selected for analysis was collected at Sailor’s
Creek Falls. At this locality there are numerous large blocks of
basalt which split fairly readily, allowing access to fresh rock.

numbers refer to rock sections in the collection of the Geological

Department, - University of Melbourne.

D. Orr:

Microscopical examination of a rock section of the specimen
analysed showed it to be a typical specimen. _

The analysis shows a silica percentage of 47-71%. This places
the rock in the basalt group. Chemically it is distinctly higher
in alkalies and lower in lime and magnesia than the normal
basaltic type, typified by R. A. Daly’s average of 161 basalts. It
therefore belongs to the alkaline suite of rocks.

The high percentage of CO, would seem to indicate that the
sample taken for analysis was not fresh. Microscopical examina-
tion, however, showed that this was not the case. None of the
minerals, including the felspars, was decomposed to any extent.
Some aragonite occurred in the vesicles. This would indicate a
primary origin for the carbon dioxide in the magmatic gases.

According to the Quantitative Classification, the rock would be
classed as • — Class 1 , Dosalane ; Order 5, Germanare ; Rang 3, An-
dase ; Sub-Kang 4, Andose.

Megascopically it is a dark bluish grey rock, in which small
phenocrysts of augite, olivine, and shiny felspar may be distin-
guished. In appearance it is very similar to the anorthoclase
basalts of the Macedon district. Large clear phenocrysts of fel-
spar, some over 1 cm. in length, showing a well-developed
cleavage and a lenticular outline through reaction with the molten
magmatic liquid, are occasionally found. They are most nume-
rous at Sailor's Creek and Stony Creek Falls, where an abundance
of broken material is available for examination. Vesicles lined
with zeolites and aragonite are fairly numerous.

The vesicular blocks at Leonard’s Hill contain, though not
plentifully, large phenocrysts of a clear felspar. Similar crystals
are found in the scoriaceous material at Mount Franklin, where
they have been identified as anorthoclase. Near the foot of
Leonard's Hill is found a very dense, fine-grained type of basalt,
in which few, if any, phenocrysts can be recognised with a lens.
It occurs in boulders which under the hammer break up into con-
centric shells.

The microscopical character of the rock has been determined by
the examination of a large number of rock sections made from
specimens gathered from different parts of the flow. There is a
marked uniformity in all these sections, with the exception of
those of material from the point of eruption.

Microscopically it consists of pheifocrvsts of idiomorphic
plagioclase, showing distinct though rather fine lamellar twinning,
allotriomorphic anorthoclase, granular olivine and hypidiomorphic
augite and hypersthene. The ground mass is composed mainly of
lath-shaped plagioclase, showing a well marked trachytic arrange-
ment of crystals, with grains of olivine, augite and magnetite. A
clear felspar sometimes occurs as an interstitial allotriomorphic
mineral in the ground mass. Its refractive index, found by the

Olivine Anorthoclase Basalt.

Becke method, is less than that of the felspar laths. This is
probably sanidine or untwinned anorthoclase. Interstitial glass is
sometimes present in small amounts. Aragonite is frequently
found in the vesicles [2192, 2202, 2208],

The reaction border surrounding the allotriomorphic crystals
of anorthoclase and the corroded outline of the crystals show
that they were not in equilibrium with the molten liquid. Their
distribution is rather irregular, as, though usually present, they do
not occur in all the sections. Twinning cannot be distinguished,
but they have the characteristic wavy extinction of anorthoclase.
The refractive index is lower than that of the plagioclase felspar,
whose refractive index, determined by the Becke method, is dis-
tinctly higher than, while that of the anorthoclase is practically the;
same as that of the balsam. The size varies 'from large phenocrysts
[2204, 2205, 2206, 2207] to small corroded remnants [2191, 2193,
2195] The smaller crystals are crowded with inclusions, but the
interior of the large ones, although surrounded by altered reaction
border with inclusions, is quite clear and free from inclusions.

The predominant pyroxene is a violet-grey titauiferous variety
of augite. Hvpersthene occurs in a subordinate amount [2191,
2193, 2203, 2195], being distinguished by its straight extinction
and faint pleochroism.

The felspar laths were determined, wherever possible, by the
Michel Levy statistical method, and were found to vary from an
acid labradorite to a basic andesine.

Generally the rock is fairly fresh. Calcite is developed at the
expense of the plagioclase [2191, 2194, 2206], frequently form-
ing central inclusions in the larger phenocrysts. The olivine is
generally altered to brown iddingsite [2195, 2199, 2206], and
more rarely to serpentine [2192]. Serpentine, not directly asso-
ciated with olivine,. is frequently present [2203, 2208, 2195].

This microscopical examination shows the rock to be an olivine
anorthoclase basalt.

Sections of the sCoriaceous type from the summit of Leonard's
Hill show a few porphyritic crystals o;f magnetite, olivine, augite
and, in one case [2197] hornblende, set in a decomposed glassy
ground mass. Two sections were prepared containing the large
felspar phenocrysts. In one case [2196] the phenocryst had a
well-developed cleavage and a wavy extinction inclined at 3-7° to
the cleavage. The refractive index \Vas practically the satiie as
that of the balsam. It is therefore anorthoclase. In the other
section [2198] the felspar showed brdad lamellar twinning and
a higher’ refractive" index characteristic of the plagioclase felspars.

In the dense fine-grained variety found at the foot of Leonard’s
Hill [2199], the trachytic structure, sp characteristic of the .f el-
spar laths in the ground mass of the other sections, is absent, fjdie
phenocrysts are smaller ancf-less numerous, anorthoclase : peing
represented by one small corroded remnant.

D. Orr :

( d ) Age.

The anorthoclase basalt flow covers auriferous drifts of the
same age as those preserved beneath the normal basalt flows of
the district. Leonard’s Hill, compared with other hills, such as
Fern Hill, Bald Hill, Wombat Hill, points of eruption of normal
basalt types, is of the same order of preservation. The deep lead,
like the other deep leads, lies about 50 feet above present stream
level, and therefore was formed before the rejuvenation of the
streams, causing the present deep dissection, took place. The flow
therefore belongs to the newer basaltic period of volcanic action,
and is late Kainozoic in age.

The anorthoclase basalts of the Macedon district are consid-
ered (3) to have been extruded immediately prior to the nor-
mal newer basalts. In this area, however, if the age assigned to
the basalt flows immediately south of Leonard’s Hill is correct,
the extrusion of the normal types had already commenced when
the anorthoclase basalt was poured out.

( e ) Comparison with Newer Basalts and with other Alkaline
Rocks in Victoria.

An average of six analyses of newer basalt from the Camper-
down district is given with the chemical analyses. This may be
taken as typical of the newer basalts of Victoria. Compared with
R. A. Daly’s average, it shows a marked similarity throughout.
The normal newer basalts are therefore typical olivine basalts
from which the anorthoclase basalt differs both chemically and
mineralogically.

Chemical Analyses.

Si0 2

4771

48-83

5152

48-78

48 00

Al a 0g -

15*66

1669

16-58

15-85

1411

Fe,0, -

247

2-66

2 35

5-37

5 61

FeO

843

8*40

768

6-34

6 11

MgO

445

5"56

403

•

603

8-81

CaO

7o3

795

610

891

8-68

Na a O

369

2-92

4-11

318

3 01

K a 0

223

2T0

■ -

2.99

1-63

1-25

H a O -f-

1-93

066

022

1*03

0-73

H ,0 - .

0-48"

1-34

1-39

►

0-73

- -

0-80

CO*

2 67

tr.

—

TiO t

1-96

2-85

215

•

1-39

2 20

089

0-74

0-82

0-47

0 50

MnO

tr.

0-25

013

0-29

013

Li a O

n.d.

—

ei, -

n.d.

0*04

0-05

—

so t

n.d.

—

CoO,NiO -

n.d.

—

006

—

003

SrO

n.d.

•

—

tr.

—

Total -

100-10

100-99

100-18

100 00

99-97

Olivine Anorthoclase Basalt.

Norms and Classification.

1 .

Orthoclase

1279

1223

17-79

Albite

28*82

24-63

34-58

A nortliite

1974

26-41

18-07

Nepheline

1-42

—

Diopwide -

1046

723

5-94

Hypersthene

—

13-12

4T8

Olivine

11 89

4-32

8-59

Magnetite

371

3 94

3-48

llnienite -

3-65

547

4-10

Apatite

1 86

1-68

202

Class

Order

Rang

Sub-rang -

Magmatic Name -

Andose

1. Anorthoclase Olivine Basalt, Stony Ck. Falls, Daylesford. Analyst,

D. Orr.

2. Anorthoclase Basalt, Sugnrloaf Hill, N.N.E. of Wooden d. Bull.

Geol. Surv. Yic. No. 24, p. 33.

3. Olivine Anorthoclase Trachyte, allot, iv., Parish of Cobaw. Bull.

Geol. Surv. Vic. No. 24, p. 25.

4. Average of 161 typical basalts, mostly olivine-bearing. R. A. Daly,

Journ. Geol. xvi., p. 409, 1908.

5. Average of 6 basalts, Gamperdown district. Mem. Geol. Surv.

Vic. No. 9, p. 22, 1910.

In Victoria there are three Kainozoic centres of alkaline vol-
canic activity, viz. : N.E. Victoria, in the Omeo, St. Bernard Hos-
pice and Mt. Leinster districts; Central Victoria at Macedon; and
Western Victoria around Coleraine. Besides the main Macedon
area in the central Victoria province, there are two other volcanic
foci of alkaline nature recorded by Professor Skeats (1). These
are Mount Wilson, about six miles south east of Leonard’s Hill,
which is composed of trachy-phonolite, and Blue Mountain, four
miles north of Blackwood township, which is tentatively described
as anorthoclase-olivine-trachyte. In the same paper Professor
Skeats describes the Daylesford anorthoclase basalt as an olivine-
anorthoclase-trachyte from which it cannot be distinguished either
megascopically or microscopically. It is therefore probable that
the Blue Mountain occurrence is very similar to the Daylesford
type.

The alkaline rocks of the Macedon district (3) include solvs-
bergites, anorthoclase trachytes, anorthoclase basalts, limburgites,
Macedonite, and Woodendite. Analyses of an anorthoclase basalt
from Sugarloaf Hill and an olivine anorthoclase trachyte from
the Parish of Cobaw are recorded with that of the Daylesford

R Orr :

type. The Daylesford rock shows marked chemical affinities
with both types. This is reflected in the similar positions which
the three rocks occupy in the Quantitative Classification. Though
the chemical similarity in the case of the Sugarloaf type and the
Daylesford type is very marked, the comparison then ceases.
Structurally the Sugarloaf type is of the basalt variety, while the
Daylesford type has a well marked characteristic trachytic struc-
ture, and might be called an anorthoclase trachy-basalt. Although
the Daylesford rock has a distinctly lower Si0 2 content than the
olivine anorthoclase trachyte from the Parish of Cobaw, the
similarity in the chemical analyses of the two types is very
marked, and as both types structurally belong to the trachyte type
they are therefore more closely related than the Daylesford type
and the Sugarloaf type. The silica content ( 51-52%)* of the latter
rock would, according to Hatch’s classification, place it in the
basalt group with the Daylesford rock.

(/) Origin.

The origin of the Kainozoic alkaline rocks of Victoria has been
discussed by E. W. Skeats and H. S. Summers (3). They con-
sider that at the beginning of the Kainozoic period, Victoria was
invaded by a basaltic magma and formed a basic sub-alkaline pro-
vince. In early Kainozoic times “ came the separation and pour-
ing out of the older basalts of the eastern and central portions of
Victoria. This left a magma moderately rich in alkalies, and by
some process of differentiation alkali magmas separated out into
at least three lesser magma basins, viz. : at Omeo, Macedon and
Coleraine. On the exhaustion of these lesser magma basins, ex-
trusion once more took place from the main reservoir, giving the
newer basalt series.” The alkaline lavas of Blue Mountain and
Mount Wilson together with the Daylesford anorthoclase basalt
may be considered as offshoots from this lesser alkaline magma
basin of central Victoria.

In all these other Victorian occurrences of alkaline lavas,
whenever evidence of age is obtainable, it points to a slightly ear-
lier period of eruption than the newer basalt. As previously indi-
cated, however, it seems probable that the extrusion of newer
basalt types had already commenced in the Daylesford locality
when the anorthoclase basalt was extruded. Considering the
close intimacy, both in space and time, of the anorthoclase basalt,
and the newer basalts, it cannot be doubted that both have been,
derived from a common magma, and are therefore genetically,
related.

Mineralogically the anorthoclase basalts differ from the normal
basalt only in the presence of anorthoclase. This mineral accounts
for the higher alkali percentage, and if absent would bring the
rock chemically into line with the normal newer basalts, with which
it is so intimately associated. Anorthoclase has been recorded in>

Olivine Anorthoclase Basalt .

several instances from the newer basalts, and careful micro-
scopical work would no doubt increase the number of occurrences.
H. J. Grayson and D. J. Mahony have recorded anorthoclase from
the ejected material from Mount Noorat (4), and it is fairly plen-
tiful in the vesicular blocks of scoria at Mount Franklin. Dr.
Stillwell has shown it to be present in both the newer and older
basalts of Broadmeadows (5). Professor Skeats and Dr. Sum-
mers have recorded it in the newer basalt of Ballarat and Mace-
don (3). These latter two occurrences were as intensely cor-
roded phenocrysts. It is therefore evident that the anorthoclase
molecule is present in very minute amounts in the newer basalt
magma, and that it separated out at an early stage in the crystalli-
zation to be later wholly or partially re-absorbed by the magma.

It is possible that the portion of the magma from which the
anorthoclase basalt was derived had been enriched in these anor-
thoclase crystals. Concentration of the alkalies by resurgent
gases has been suggested by R. A. Daly and C. H. Smith. The for-
mer (6) regards the source of these gases to be assimilated basic
sediment, the most efficient being limestone. The latter (7) re-
gards magmatic or “ juvenile ” gases to be the more likely effec-
tive concentrating agent.

In the rock under discussion, aragonite and zeolites are found
in the vesicles, and would indicate the association of magmatic
gases with the lava. None of the minerals associated with the
more active “ mineralisers, ,, however, was detected. In this
area the evidence is not sufficient to support one or other of the
views as to the source of the magmatic gases. Seepages of carbon
dioxide, and the fact that the mineral springs of the district are
highly charged with this gas, show that the Ordovician sediments
would be competent to supply this gas if assimilated on a suffi-
ciently large scale.

Summary.

Among the newer basalts of the Davies ford district is an
alkaline lava which was extruded from Leonard’s Hill, six miles
south of Daylesford. The area is one of great diversity of surface
due to the recent rejuvenation and consequent dissection by the
streams. This dissection is discussed, but no conclusive evidence
as to its cause could be advanced. A chemical analysis shows that
the rock has alkaline affinities. This is borne out by micro-
scopical examination which shows it to be an olivine anorthoclase
basalt. The age of the rock is discussed, it being considered to
have been extruded subsequently to the commencement of newer
basaltic activity. It is compared with normal newer basalt types,
and its relation to other Kainozoic alkaline basalts is discussed.
It is considered to be genetically related to the newer basalts, and
a possible method of concentration of anorthoclase crystals or
molecules by resurgent gases, either magmatic or consequent on
assimilation, is suggested.

D. Orr : Olivine Anorthoclase Basalt.

REFERENCES.

1. E. W. Skeats. Rept. Aust. Assoc. Adv. Sci., xv., Hobart

Meeting, p. 305, 1921.

2. H. S. Whitelaw. Bull. GeoL Surv. Vic., No. 42, 1923.

3. E. W. Skeats and H. S. Summers. Ibid., No. 24, 1912.

4. H. J. Grayson and D. J. Mahony. Mem. Geol. Surv .

Vic., No. 9, p. 17, 1910.

5. F. L. Stillwell. Proc. Roy. Soc. Vic., n.s., xxiv. (1), p.

156, 1911.

6. R. A. Daly. Igneous Rocks and their Origin, p. 410. 8vo,

McGraw-Hill, New York, 1914.

7. C. H. Smith. Am. Journ. Sci., xxxvi., 1913.

[Proc. Roy. Soc. Victoria, 40 (N.S.), Pt. I., 1927.]

Art. V. — Notes on the Coastal Physiography of Port Campbell ,

Victoria.

By J. T. JUTSON, B.Sc., LL.B.
(With Plates VI., VII.)

[Read 14 th July, 1927 .]

Introduction.

The Port Campbell area lies between Cape Otway and Warr-
nambool, and its coast is justly famed for its magnificent scenery.
Inland, the country forms a gently undulating plain of sedimenta-
tion, rising at the coast in places to a height above sea-level of
over 200 feet, and being destitute of timber near the ocean. The
rocks composing this plain are Tertiary sediments, which have
been so slightly disturbed that they lie very close to the horizon-
tal. The country has been comparatively little eroded. There are
no large streams, the most prominent being the Port Campbell
Creek (at the mouth of which the small township of Port Camp-
bell is situated) and the Sherbrooke River, a few miles to the
east of Port Campbell. Both these streams, and some smaller
ones, enter the ocean at sea-level; but numerous small water-
courses and other channels that are mere gulches in the cliffs,
occupy hanging valleys. By reason of the non-resistant character
of most of the rocks, the valleys are fairly wide, open ones, with
rather gently-sloping sides. In some valleys, as in parts of that
of the Port Campbell Creek, the valley-sides tend to be steep,
owing to the intercalation, high up, of bands of indurated lime-
stone. The Port Campbell Creek, in its lower portion, is sluggish,
and its valley flat-floored and marshy, features no doubt at least
partly due to the formation of the small bar to be referred to
immediately.

The coast-line consists of rugged cliffs, and its general outline
is so simple that there are practically no harbours, with the excep-
tion of the tiny one of Port Campbell, which lies at the mouth of
the creek of the same name, and is about 150 yards wide by about
250 yards long. The head of that “ harbour ” has been determined
by the formation of a crescentic sand bar, capped by a low sand
dune across the valley, which it almost closes, the exit of the
stream being by a narrow channel on the western side. The bar
provides an excellent bathing beach. The western and eastern
sides of the “ harbour ” are formed of steep cliffs in process of
marine abrasion.

J. T . Jutson

Fig. 1 . — Sketch map of the Port Campbell coast, comprising the area referred to in the paper. The map was copied from
one available at Port Campbell and appears to be approximately correct. Its authorship is unknown to the writer.

Physiography of Port Campbell .

The site of the township on the left bank of the stream has
been determined by the fact that there the left side of the valley
possesses a long gentle slope, whilst the right side rises precipi-
tously from the valley floor.

The impure limestone, referred to below, is used for road-
making in the district.

Previous Literature.

The only previous physiographic references that the writer is
aware of are those of J. W. Gregory (1, p. 40) and A. V. G.
James (2, pp. 144 and 194) by each of whom Port Campbell (i.e.
the “ harbour ”) is referred to as a drowned river valley, and
by Gregory also as an example of a ria (1, p. 56).

Physiographic Features of the Coast-Line.

In its general features, the coast-line may be described as con-
sisting of a series of high cliffs, practically vertical, and in places
overhanging, which frequently follow a sinuous line (which may
be described as “ crenulate,” although here not necessarily indicat-
ing a youthful submerged shoreline (3, p. 278)), owing to the
occurrence of numerous small bays and narrow gorges, the latter
of which in some instances are continued some distance under-
ground as sea caves, as shown by the u Thunder Cave ” at the
Survey Gorge ; by “ the Blowhole ” at a point perhaps 200 yards
inland, where the roof of the long horizontal cave has fallen in;
and by Tom Pierce’s and Miss Carmichael’s caves in the Loch
Ard Gorge. Tom Pierce’s cave is about 40 or 50 feet long, 15 to
20 feet wide, and 10 to 15 feet high at the entrance, but becom-
ing so low that one cannot stand erect in it; whilst Miss Car-
michael’s cave is about 50 feet or more long, 15 to 20 feet wide,
and in places 20 feet high. Many of the bays, however, are
slight open indentations. The gorges themselves are but a few
yards wide and, as a rule, would be about 100 yards in length.

Arches also are formed by marine erosion of opposite faces of
small peninsulas or islands. Examples of such arches occur at
the peninsula known as “ London Bridge,” about five miles west
of Port Campbell; at a small projecting tongue of land to the
north-east of Point Hesse, where it is known as the “ Marble
Arch ” ; and in an island in the small “ bay ” immediately to the
east of the Loch Ard Gorge.

Islets, usually with vertical cliffs, are fairly common — a rather
unusual feature of the Victorian coast-line.

The rocks forming the cliffs are chiefly Tertiary shales, fine-
grained sandstones, and impure limestones. The lower beds
largely consist of easily eroded shales, with an occasional harder
band, whilst those towards the top of the cliff have been made
more resistant to erosion by the deposition of ferruginous or cal-
careous material in the form of thin bands along the bedding

J. T. Jutson :

planes, and also as nodules and irregular masses. 1 These features,
therefore, by aiding marine abrasion, and by retarding atmos-
pheric erosion, facilitate the formation and maintenance of prac-
tically vertical cliffs, without (under the present stage of develop-
ment) materially checking the rapid recession of the cliffs. The
attack of the waves on a soft band of rock is greatly aided by its
horizontal character. This is shown by the undermining along
such a stratum, both when the direction of attack is approximately
at right angles, and when, in consequence of a projecting strip of
land, the direction of attack is parallel to the trend of the coast.
In the latter case, it is very instructive to watch the end of a wave
working along the stratum, and to note the even groove made by
such action.

The hanging valleys indicate that the cliffs are receding faster
than these valleys can deepen themselves, and so the latter do not
reach the sea-coast at sea-level.

Cliff erosion is also assisted by the presence of pronounced,
usually vertical joints, fairly widely spaced, often forming two
sets, more or less at right angles to each other, sometimes very
curving and sometimes dipping into the cliff. The undermining
of the cliffs, their steep character, and the percolation and evapo-
ration of water along these joints, bring about the fall of huge
masses usually rectangular in cross-section; and the new face of
the cliff consequently presents a somewhat buttressed appearance.

A striking feature of the coast-line is the number of taluses
formed as a result of the processes just described, many of the
component blocks of which are of great size. These masses,
being particularly exposed to the action of the sea and of the
weather, must be removed relatively rapidly.

Broad well-developed beaches and water-worn pebbles and
boulders are scarce. Beaches at the foot of the cliffs in the bays
do occur ; but usually they are narrow and the sand scanty. Many
are accessible at low tide only, and in some parts they are prob-
ably always inaccessible. One of the best beaches is that in the
bay immediately to the west of London Bridge. The waterworn
boulders, when naturally occurring, may have been largely re-
moved for the purpose of roadmaking, as several heaps, gathered
together for this purpose, of the fairly hard, impure limestone
were noticed at the time of the writer’s visit in January, 1925. 2
If, however, boulders and shingle are generally absent from the
beaches, it is a matter of some surprise, considering the quantity
of resistant rocks in the cliff beds.

1. — One result of the occurrence and irregular distribution of these harder

bands is the formation by differential weathering of pillars of com-
paratively soft rock with projecting hard caps. Where a considerable
area of the rocks has been exposed, as at the tops of the cliffs just
to the west of Point Hesse and near the 41 Blowhole,” close to the Loch
Ard Gorge, the surface becomes honeycombed and extremely uneven
and ragged, with varied and often grotesque forms.

2. — These collected waterworn rocks may possibly be due to stream action;

but if so, whence they came is unknown to the writer.

Physiography of Port Campbell . 49

The comparatively short and few streams entering the sea in
this area cause little fluviatile detritus to be carried into the sea ;
hence the waves are freed from the task of .removing the material
that would otherwise be deposited near the shore, and are thus
enabled to make a more continuous attack on the coast.

Wave-cut rock platforms lie at the base of the cliffs and around
the stacks, which are comparatively close to the shore; but
whether or not these platforms stretch seaward to any extent was
not observable, although waves frequently break far out, thus
suggesting the continuance of the platforms to those areas.

One obstacle to marine abrasion is the occurrence of the great
seaweed commonly known as “ kelp, 5 ' which in many places is so
firmly attached to the rocks at sea-level, and is so abundant, that
rock removal must be considerably checked.

As a result of the powerful marine abrasion, a coast-line has
been produced which in certain parts is very broken on a small
scale. These features are observable in the vicinity of London
Bridge, and especially at and adjacent to the Loch Ard Gorge-
Diminutive bays and "gorges succeed one another, and a moderate
number of islets fringe the shore in places. A physiographet
without personal knowledge of the area, looking at a map of the
coast-line drawn on a fairly large scale, would probably infer that
the physiographic features referred to are due to a recent sub-
mergence of the land. This question is subsequently discussed,
but here it may be pointed out that although slight submergence
may have .taken place, and although the “ harbour ’ of Port
Campbell 3 may possibly be due to submergence, yet practically no
traces of submergence remain ; and the small bays and gorges are
clearly due to erosion, as they certainly are not the drowned ends
of normal valleys. Moreover, their actual formation can be seen
now going on.

Reflection might suggest that the gorges occur only at the
mouths of valleys, and that consequently, even though they are
not drowned valleys, their erosion by the waves has been hastened,
owing to the formation of the valley above, and the action of
waterfalls (where the valleys are “ hanging ' ones) in bringing
about a recession of the cliffs upstream. But although gorges do
occur as indicated, yet there are others quite as long, which have
been formed without these aids, and therefore the ocean is respon-
sible for the whole gorge in certain cases. Doubtless where strong
seas — as at Port Campbell — are available, they gather force by
concentration in a narrow area, despite the increased friction they
suffer. Hence if breaches are made in a cliff face, the concen-
trated power of the water will tend to erode gorges — at least in
soft rocks — at the head of which extensive sea-caves may be
formed, as for example, in the Lord Ard and Survey Gorges.

3. — The mouth of the Sherbrooke River was not examined by the writer,
so that no expression of opinion can be given as to whether or not it
represents a drowned river valley.

■SO

J . T. Jutson :

One would imagine, however, that there must at least be a tem-
porary limit to the extension of the gorges, owing to the continu-
ally increasing friction and consequently reduced power of the
waves as the gorges lengthen; but the activity of the sea may be
revived as the outer cliffs are cut back.

The formation of such gorges might be expected if there were
much variation in the power of resistance to marine abrasion of
the rocks of the coast, but such variation at Port Campbell is not
apparent on a general inspection; hence some surprise is excited
by their occurrence. There must, however, be differences in resist-
ing-power among the rocks, although difficult to detect ; and per-
haps wave-attack varies in strength from point to point.

The gorges, however, may be primarily or largely determined
by structural features, such as strong joints or small faults
highly developed in certain areas; since erosion would advance
more rapidly along these lines of weakness than in areas where
they were absent, other things being equal. This aspect was sug-
gested to the writer by Professor Skeats, but it must be reserved
for future investigation.

The numerous islets are due to marine abrasion, and not to
submergence of the land; their shape and distribution, and their
standing, as far as can be seen, upon a wave-cut platform nega-
tiving the latter idea. In a mainland so little dissected, such
islets could not result from subsidence. They are, therefore, un-
doubtedly stacks. None of these stacks is joined to the mainland
or to another stack by a sand bar. this fact being no doubt due to
the rapid removal by the waves of most of the detritus from the
cliffs.

The scarcity of beaches is apparently due to the tremendous
scour of the waves, and to the fineness of the constituents of the
rocks attacked. Such rocks yield little sand, and the fine materials
are carried well out to sea. Moreover, the absence of much river-
borne detritus must tend to restrict the formation of beaches.
This limitation of beaches hastens in turn the wearing away of
the cliffs owing to their lack of protection by the beaches. Two
localities are of interest as minor examples of the prevention of
wave attack. In the western arm of the Loch Ard Gorge a high
bank of blown sand is preventing — but probably only temporarily
* — further marine abrasion of the cliffs around this arm. In the
small bay just west of “ London Bridge,” where there is a well-
developed beach, a small dune at one point acts in the same way.
In each case, rough thick stalactites have formed on the inwardly-
inclined cliff, so that the cessation of wave attack is not quite
recent. Incidentally, it may be noticed that if man removes sand
and gravel from the shore, such action will hasten the wearing
away of the cliffs by the sea.

A little to the west of the large sand ridge described below,
there is on the beach a large bed of rounded pebbles and boulders,
covered with vegetation, in front of a talus, which thus at present
protects the cliffs from marine abrasion.

Physiography of Port Campbell.

A striking feature observable in some localities is the small
serrations in the Tertiary rocks at the top of the cliffs. The serra-
tions have been filled with red soil containing ironstone fragments,
which deposit in many places forms a thin surface cover on the
marine Tertiary sediments. The origin and age of this deposit
.are not obvious, although it appears to be of quite recent age, and
to have been laid down subaerially.

Reclamation of Land from the Sea.

The physiographic features noted above are so characteristic of
the coast that much surprise is felt when a long sand ridge is found
abutting the coast. It commences a few hundred yards west of

Fig. 2. — Diagrammatic plan and section of the sand ridge area west
of Port Campbell.

Port Campbell “ harbour ” and extends in a gently curving line
westerly for perhaps three-quarters of a mile. It follows the
coast-line so closely that at high tide the water reaches to its base.
The ridge varies in height, the highest point being about 80 feet
above sea-level. Its crest usually is not more than a few feet
wide, and its width at the base varies from a few yards to about
25 yards. It is clothed, but not densely, with vegetation, mostly
small shrubs, but there is ample opportunity for movement of the
sand by the wind. The sand is fine-grained, and, so far as cur-
sorily examined, consists chiefly of quartz. At its western end,
the sand ridge throws off inland a short minor ridge, which
divides into two, but reunites. Near the eastern end, the sand

J . T. Jutson :

rests for a short distance on a rock platform six to eight feet
above the beach.

At each end of the main ridge the coast-line consists of precipi-
tous cliffs up to 200 feet or more high, of the kind already de-
scribed as characteristic of the coast as a whole. These sea cliffs
are being abraded strongly by the ocean. Although they give
place to the sand ridge along the coast, yet they are connected
behind the ridge by a line of cliffs (which for convenience may
be referred to as the “ inland cliffs ”) which forms an arc of which
the sand ridge is the chord.

The inland cliffs present a striking contrast with the present
sea cliffs in that they are clothed with vegetation; show the
double curve of atmospheric denudation in profile ; are not ver-
tical, although steep; are not undermined; and have alluvial fans
at the mouths of the small creeks or gullies that dissect them.
These features are absent in the sea cliffs, as will be seen from the
description given above of the coast generally. What are appar-
ently old taluses occur at the foot of the inland cliffs, but these
may have been formed either under marine or atmospheric condi-
tions. Another contrast is that the gullies and short creeks which
cut the inland cliffs are not hanging like those of the sea cliffs.
This points to rapid recession of the latter, but to very slow back-
ward cutting of the former.

Between the inland cliffs and the sand ridge^ is a narrow belt of
low-lying land possessing a somewhat uneven floor, owing to the
formation of the alluvial fans mentioned above, to the probable
drifting of sand from the sand ridge, and to some slight stream
action. The greatest width of this belt is about 70 or 80 yards,
and in its higher portions its surface is probably 25 feet or more
above sea-level.

The obvious interpretation of the phenomena described is that
the line of inland cliffs was originally bounded by the sea; that
the sand ridge was thrown up by the waves and wind, forming a
typical bay dune-covered bar, which either altogether or almost
entirely cut off the sea behind it from the ocean; that silting (or
perhaps silting combined with uplift) has taken place until the
water has given place to land; that subsequently thereto, by the
accumulation of wind-driven sand from the ridge and of detritus
from the gullies and face of the inland cliffs, the land surface of
this reclaimed area has been raised to its present height ; and that
by atmospheric erosion the face of the inland cliffs has been
changed from their typical sea-front form in this locality to the
equally typical double curve of atmospheric denudation.

This interpretation appears to meet the facts of the case best,
although others are possible. It is, however, uncertain, on the
facts known to the writer, whether uplift has played any part in
bringing about the present conditions.

It may be contended that the bay which has been cut off pene-
trated far enough into the coast-line for the water to become rela-

Physiography of Port Campbell.

tively quiet, and silting to occur — resulting in the formation of
land — without necessarily the formation of a bar, and that the
land gradually encroached on the sea, until the present coast-line
was approximately fixed, after which the sand ridge was formed
by wind action. Against this idea is the fact that the coast-line in
the neighbourhood is much indented on a small scale by the abra-
sive action of the waves, and in several instances the indentations
are quite as pronounced and some are more sheltered than the bay
referred to, and yet in almost all these localities wave abrasion is
still actively proceeding.

Another point requiring consideration is that at the eastern end
of the sand ridge, the rock platform already referred to, stands
six to eight feet above the present beach. This fact suggests that
if that platform were originally continuous across the bay to the
inland cliffs, and had been formed as a platform of marine abra-
sion, it could hardly have been formed at its present height, and
hence a recent uplift of the land relative to the sea may have
occurred ; and in that event the change from sea to land would be
effected merely by the uplift and not by silting. No direct evi-
dence of such an uplift in the immediate neighbourhood of the bay
was noticed (except possibly just to the west of the sand ridge,
where some waterworn pebbles and boulders fill some hollows: in
the cliffs at a height of about eight feet above the beach, but these
plight also be regarded as the remains of a storm beach) although
such evidence, if it originally existed, would no doubt usually be
rapidly destroyed, except that, in other indentations, remnants of
similar platforms should be observable. Whether or no such a
platform exists in the old bay behind the sand ridge cannot at
present be stated, as its surface, if it exist, is covered with recent
debris — the wash from the inland cliffs and the sand from the
sand ridge. On the whole, however, doubt is thrown on the exist-
ence of a wide platform by reason of the fact that no evidence
of its occurrence is found along the greater part of the sea margin
of the ridge, and on account of the absence of such platforms in
neighbouring indentations. Moreover, the rock platform under
the sand ridge can be explained on the assumption that it is an old
low stack.

A short bore or some shallow excavations in the low-lying
ground might yield some interesting results.

The sand ridge cannot be regarded as quite fixed, as the vege-
tation is sufficiently scattered to allow the sand to be carried by
the wind over the crest landward. In this way, if there be no
erosion of the sand ridge, the ridge must be increasing in width.
There is evidence however, in some places, of the erosion by the
sea (which would be expected, seeing that the cliffs at each end of
the ridge are being cut back) with the result that the vegetation
has in places been destroyed, and the loose sand is partly being
removed by the sea and partly being carried by the wind over the
ridge landward. As the strength of the marine abrasion on this
coast is not likely to abate for some time to come, the sand ridge

J. T. Jut son :

must either be removed or migrate. The fact that the reclamation
of the old bay and the modification noticed above of the forms of
the inland cliffs have apparently taken place since the formation
of the sand ridge, indicates some degree of antiquity for the sand
ridge; and the most feasible explanation of why the ridge has
been able to maintain its identity for so long a period, despite
marine abrasion, is that the latter process is counter-balanced by
the deposition of wind-blown sand on the landward face of the ridge.
Consequently the ridge may be migrating towards the inland cliffs.
If this be so, an interesting point is, assuming these conditions to
continue, whether or no the ridge will reach the inland cliffs, and
be banked against, and temporarily protect from erosion the
latter. This in turn will depend on the rate of migration of the
sand ridge and the rate of erosion of the inland cliffs, points
about which at present we know nothing.

The Stage of Development of the Coast.

D. W. Johnson (3, p. 249) distinguishes between the shore pro-
file and shoreline development. They may be in the same or in
different stages, so that if one be young or mature, the other is
not necessarily in that particular stage.

In the Port Campbell district, the vertical high cliffs, their rapid
erosion, the stacks and the scanty beaches indicate that the shore
profile is in the youthful stage ; whilst on the other hand the
simply curved coast-line (apart from the minute irregularities
caused by the small bays and gorges) the absence of spits, bay-
bars (with the exception of that referred to in detail above),
tombolos and offshore bars, indicate that the stage of shoreline
development is mature.

Whether the coast-line was originally one of submergence or
of emergence, or is due primarily to faulting, is difficult at present
to say. The existing forms of the shore profile and of the shore-
line could result from any of the types mentioned. If it were one
of submergence, then the drowned valleys (except the M harbour ”
of Port Campbell and possibly the mouth of Sherbrooke River,
assuming them for the moment to be such) together with any
associated phenomena, such as spits and bay-bars, have been re-
moved by marine abrasion. Similarly, if it were one of emer-
gence, the offshore bar, the lagoon and possible marsh (if these
features or any of them occurred) and the low-lying land of the
uplifted coastal plain have been removed. In the same way, a
fault-coast would have been retrograded.

The absence of any direct evidence of recent uplift has been
referred to in the immediately preceding section of this paper.
It may here, however, be pointed out that if a small uplift could
be proved to have taken place, such would almost certainly estab-
lish the fact of a prior submergence, since the bed of the Port
Campbell Creek is now so close to sea-level that, if it has

Physiography of Port Campbell.

•recently 'been raised several feet, it must have been an estuary
prior thereto for some distance from its mouth.

As regards direct evidence of submergence, there is none with
the possible exception of Port Campbell " harbour ” and the
mouth (not seen by the writer) of the Sherbrooke River. But
so far as concerns the “ harbour/’ in view of the evidence adduced
above as to the power of the ocean to excavate small bays and
gorges; of the fact that the stream has provided a well-defined
opening in the cliffs which the sea could easily further widen;
of the fact that wave-cut platforms and cliffs do exist inside the
“ harbour and of the short distance the sea penetrates, it cannot
be assumed from the mere form of the “ harbour ” that it repre-
sents a drowned portion of a valley. Of course the shortness of
the “ harbour ” could be accounted for either by the building, in
the old estuary, of the dune-covered bar already described, owing
to which — above the bar — siltation has taken place and marshes
have resulted 4 , or by the retrogradation of the shoreline, or by
both these operations, the second of which we may be sure has
been in force.

The writer has been permitted, bv the courtesy of Captain J.
K. Davis, Director of Navigation, to inspect the charts of the
ocean in the vicinity of Port Campbell, but no definite conclusions
regarding the questions discussed in this paper can be drawn from
them, although such might be possible, were greater detail
shown.

Judgment on the original character of the shoreline must there-
fore be suspended pending an examination of a wider area.

It may be noted that G. S. Griffiths (5, pp. 76-79) states there
is evidence of late submergence and of subsequent emergence at
the Portland Promontory. Similar observations have been made
in so many places around Australia that one naturally leans
towards those ideas at any similar spot under examination.
Each locality must, however, be independently judged.

Summary.

The Port Campbell area lies between Cape Otway and Warr-
nambool. Inland, the country forms a plain of Tertiary marine
sedimentation, not much dissected. The coast consists of rugged
high vertical cliffs of the same class of rocks.

The coast in places follows a sinuous line owing to the occur-
rence of small bays and narrow gorges. Arches and long sea-
caves are prominent features, and numerous islets occur. These
features are due to marine erosion, and not { as regards the bays,
gorges and islets) to submergence of pre-existing valleys.

s * (b P. 31) refers to the blocking of the mouths of various

streams east of Warrnambool, of which Curdie’s and the Gellibrand
rtivers are examples, by sand dunes, without apparently believing
tftat their estuaries are drowned valleys. The "harbour" of Port
Campbell may be of similar origin.

J. T. JutsOV:

• The upper beds of the coastal cliffs, being somewhat more re-
sistant to erosion than the lower, facilitate the formation and
maintenance of practically vertical cliffs. A series of vertical
joints also tends to the same result.

Hanging valleys occur in the shorter watercourses, indicating
that the sea-cliffs are retreating faster than these valleys can
deepen themselves.

Taluses are abundant; but well-developed beaches and water-
worn pebbles and boulders are scarce.

Wave-cut platforms occur at the base of the cliffs, but their
seaward extent is unknown to the writer.

A prominent sand-ridge runs along the coast to the west of
Port Campbell. Behind it lies a belt of low-lying ground which
is bounded by a line of old sea-cliffs, which have lost their typical
marine form, and have taken on the double curve of atmospheric
denudation. I he sand ridge is regarded as having been formed as
a bar across the mouth of the old bay, and then built up by wind-
blown sand. The bay became practically closed to the ocean.
Silting occurred, and the low-lying belt at the rear was formed,
thus reclaiming a moderate area of land. The question of uplift
is considered in this connection, and the conclusion is arrived at
that as there is no definite evidence of recent uplift, the proba-
bilities are that the low-lying belt was formed entirely by silting
and subsequent atmospheric accumulation. If that be so, the sand
ridge must he of some antiquity.

. The sand ridge is notfixed. It is being eroded on the seaward
side by the ocean, and is widening on the landward side by the
deposition of sand blown over it from the seaward face. Conse-
quently the ridge may be migrating landward.

The shore profile is youthful, whilst the stage of shoreline de-
velopment is mature.

Whether the shoreline was originally one of submergence or
emergence, or is due primarily to faulting, has not been deter-
mined, although some slight submergence has perhaps taken
place.

REFERENCES.

1. J. W. Gregory. The Geography of Victoria. 2 Ed., Mel-

bourne, 1912.

2. A. V. G. James. The Intermediate Geography Text Book.

2 Ed,. Melbourne, 1925.

3. D. W. Johnson. Shore Processes and Shoreline Develop-

ment. London and New York, 1919.

4. T. S. Hall. Some Notes on the Gippsland Lakes. Viet.

Nat., Nxxi., pp. 31-35, with map, 1914.

5. G. S. Griffiths. The Geology of the Portland Promon-

tory. Trans. Roy. Soc. Vic., xxiv. (1), pp. 61-80, 3
pis., 1887.

Proc. K.S. Victoria, 1927. Plate VI,

3 .

Proc. R.S. Victoria, 1927. Plate VII.

Fig. 3.

Physiography of Port Campbell .

EXPLANATION OF PLATES.

Plate VI.

Fig. 1, — The old sea-cliffs, and the low-lying ground between
these* cliffs and the sand ridge. A small portion of the latter is
shown in the foreground at the left-hand corner.

Fig 2. — The sand ridge bounding the ocean, and the low-lying
ground between the ridge and the old sea-cliffs. A part of
these cliffs is shown at the right hand side of the photograph.

Fig. 3. — The present and the old sea-cliffs at the western end
of the sand ridge, which is shown tapering out in the fore-
ground. The existing cliffs are shown between the letters A and
B, and the old cliffs between the letters B and C.

Plate VII.

Fig. 1. — Near the Survey Gorge, illustrating the formation
of vertical cliffs, and of small bays and gorges, by marine
abrasion.

Fig. 2. — Stacks at the Bay of Islands. Note the rectangular
forms due to two sets of master joints at right angles to each
other, and also the cones of red soil, with ironstone fragments,
capping the ordinary Tertiary rocks in the foreground. Vic-
torian Railways photo.

Fig. 3. — Sea-cliffs west of the sand ridge showing great
taluses, composed of huge vegetation-covered blocks.

END OF VOLUME XL, PART I.
[Published December, 1927].

[Pkoc. Roy. Soc. Victoria, 40 (N.S.), Pt. II., 1928. J

Aut. VI . — A Revision of the Genus Pultenaea. Part V.

By H. B. WILLIAMSON, F.L.S.

[Read 13th October, 1927; issued separately 21st April, 1928.1
Pultenaea mollis Lindley.

An attempt is here made to clear up the difficulty involved in
this species, a difficulty which seems to have arisen from an error
in the determination of Robertson’s Mt. Sturgeon specimens as
P. viscosa R.Br., FI. Aust., II., 127. Referring to Revision, Part
III., Proc. Roy. Soc. Vic., p. 107, it may be taken as certain that
the specimens collected (a) by Robertson, FI. Aust., II., 127;.
( h ) by Mueller about 1855; and (c) by the author were the
same species. The leaves are scarcely to be distinguished from
those of the Clyde Mt. and Parramatta specimens of P. viscosa
R.Br., and Mueller determined (a) and (c) as viscosa. Speci-
mens (&) and (c) have been determined at Kew as P. mollis
LindL, as they agree with those distributed by Lindley, and which
are considered at Kew as portions of the plant gathered by
Mitchell and given to Cunningham. Mueller determined his own
specimens as P. mollis , and they agree exactly with those of the
author, who in conversation with the distinguished botanist in
1895 learned that 40 years before the locality had been the site
of his camp on the Wan non.

It seems, therefore, that P. viscosa has been wrongly recorded
for Victoria. In view of the fact that Bentham’s description was
framed to include the Mount Macedon specimens of Mueller,
which now must be kept distinct, and the true mollis of Lindley,
which is so close to the more recently described P. viscosa, it must
for a time be doubtful whether this species should have been set
up at all. The plant which has been accepted by Victorian collec-
tors as P. mollis (Grampians, Mt. Macedon, Gcmbrook, and re-
cently Bairnsdale, T. S. Hart), differs materially from the type
which appears to be confined to the southern Grampians, the
Wannon River and Portland, and as all these specimens agree on
the whole with the Mount Macedon specimen collected by
Mueller, it had better be kept distinct as P. angustifolia, Mueller’s
MS. name on the label. A description is here given.

Pultenaea angustifolia (F.v.M. Herb.), sp. nov.

Frutex circiter 2 m. (At us, ramulis puhescentibus , foliis tenui-
bus fere teretibus 10-20 mm. longis glabris vel puhescentibus,
floribus breves ramos terminantibus in capitula ( 5-7 fl.) congrc-
gaiis pedicellis 2-3 mm. longis, stipulis minimis nigris recurvatis ,
bractcis latis pedicellis brevioribus plerumque bifidis , bracteolis
latis saepc viseosis nonmimquam carinatis arete appressis tulw

II. B. Williamson:

calycis aequilongis , calyce fere glabra lobis subaequalibus obtusis
vel subacutis superioribus paulo latioribus quam inferior a, ovario
villoso .

Mt. Macedon, Mueller ; Gembrook Ranges ; Grampians ;
Bairnsdale (T. S. Hart, Sept., 1927).

Var. viscosa, var. nov.

Calyce bract eis et bract eolis viscosis.

The form from the Grampians, Vic., with its calyx, bracts and
bracteoles viscous and its leaves rather longer.

The relation of the foregoing species to each other is shown
thus : —

A. Leaves almost flat.

B. Leaves incurved, flowers densely crowded in a head (7 to
10) on very short pedicels, bracteoles half the length of
the calyx. p. mollis.

B. Leaves almost straight, flowers in heads (5-6) not crowded,
on pedicels 2-3 mm. long, bracteoles nearly as long as the
calyx. P. viscosa.

A. Leaves thin, terete, channelled above, flowers not densely crowded,
on pedicels 2mm. long, bracteoles short and broad.

P. angustifolia.

Calyx viscous. var. viscosa.

PULTENAEA HIBBERTIOIDES Hk. f.
var. prostrata, var. nov.

Frutex pro strains, foliis ct floribus valde confertis.

A prostrate form with much crowded leaves and flowers.
Waterloo Bay, Wilson’s Prom., Nat. Park, J. W. Audas, Nov.,
1908. The flowers of var. conferta Bth., Cobden, S.W. Vic., are
much crowded, but its leaves are not distinctly so as in this form.
This is evidently the plant that has been recorded from George-
town, North Coast of Tasmania.

PULTENAEA KeNNYI H.BAV.

=P. micro phylla Sieber var. cun eat a Bth.

This agrees with the var. c uneat a Bth. in Ann. Wien. Mus.,
ii., 83 (FI. Aust., II., 117). As from this form to the normal P.
microphylla a series of intermediates as regards width of leaves
has been examined, the. plant should still retain varietal rank
only.

Pultenaea foliolosa Cunn.

Benth. in Ann Wien. Mus., ii., 83.

This plant has very small leaves, often only 1 mm. long, almost
orbicular, crowded on short branchlets of 1-2 cm. long, at the ends
of which are a few flowers in the axils. Stipules are hairlike,
nearly as long as the leaves, and the bracteoles are similar to the
leaves, and provided with hair-like stipules. The calyx is about
4 mm. long, with lobes not longer than the tube; the two upper

Revision of the Genus Pultenaea . Part V.

ones much falcate and united to the middle. The ovary is glab-
rous except for a tuft of white hairs.

The type of this species came from westward of the Welling-
ton Valley, N.S.W., A. Cunn. There are specimens from Lach-
lan River, Fraser; Darling Downs, Q., Mrs. Ford; Texas, Q.,
Boorman; Eastern Downs, H. Law. The form from Chiltern,
“ Mayday Hills,” “ Between Meadow Creek and King River,”
N.E. Victoria, differs from the normal in size of flowers and
leaves, the latter reaching 4 nun. in some specimens. It shows a
transition towards P. stypheliaides Cunn. through its intermediate
var. mutica F.v.M., having often lanceolate bracteoles fixed high
upon the calyx. From New England, N.S.W., there is a plant
which so much resembles P. foliolosa in general appearance that it
is little wonder that it has been placed under that species. Its re-
markable calyx, however, renders it quite distinct, and with the
concurrence of the authorities at Kew it is now described, the
species name being in honour of the collector of the Tenterfield
specimen.

Pultenaea Stuartiana, sp. nov.

Frutex pawns, ramulis numerosis brevibus , foliis minimis raro
-2 mm. longis ova to- orb i c ularib us supra concavis inf ra <scabridis
a pice recurvatis, stipulis parvis, floribus fere sessilibus in ax ill is
superior ibus, bracteolis foliaceis quasi stipulatis , calyce 4 mm.
longo villoso lobis l at is tubo longioribus oblongis inter se aequi-
longis et forntd similibus superior ibus non falcatis, ovario glabro
a pice contain g create.

New South Wales: Tenterfield, C. Stuart; Torrington, J. L.
Boorman.

This differs from P. foliolosa Cunn. in having the long lobes
of the calyx similar in size and shape.

Pultenaea accrosa R.Br.

A record for this species in Victoria has been made, the plant
having been gathered near Mt. William in the Grampians, J. W.
Audas, Nov., 1923.

Pultenaea graveolens Tate.

This plant has apparently not been gathered in Victoria since
Mr S. Johnson sent it to Mueller from Meredith many years ago,
until with Mr. E. Cooper, senr., the author found it at Steiglitz in
October, 1925.

H. B . Williamson :

Addendum.

[Read 8th December, 1927.]

PULTENAEA PATELLIFOLIA, Sp. tlOV.

F rut ex fere glaber 0*5 — i m. alius, ramulis pubescentibus, foliis
fere sessilibtis alternatis late obovatis vel orbicularibus circiter
3 mm. longis latisque mar gin e paululum incurvatis ad apiceni re-
curvo-mucronatis supra glabris infra sparsim pubescentibus,
stipulis parvis, floribns subumbellatis 3-6 ramulos terimnantibus,
pcdicellis pilosis 5-6 win. longis, bracteis parvis , bracteolis viscosis
orbicularibus appressis vix 2 mm. longis subter tubum calycis
affixis, calyce circiter 5 nun. longo fere glabro lobis fere aequalibus
tiibo brevioribus, vex HI 0 luteo lineis atro-rubris instructo, alts
luteis, carina atro-rubra, ovario villoso in stylum subulatum ex-
tenuato, legumine ovato-oblongo breviter acmninato, seminibus ?
distinct e strophiolatis .

Fig. ].

1. P. patellifolia — A, leaf, under side; B, leaf, upper side; C, upper calyx

lobes ; D, lower calyx lobes ; E, bracteole.

2. P. foliolosci, calyx and bracteoles.

3. P. Stuartiana- } calyx and bracteoles.

Revision of the Genus Pultenaea . Part V.

«1

Mt. Byron, Black Range, Western Grampians, Vic., Mr. Harold
Smith, October, 1927.

An almost glabrous shrub, ^ to 1 rm high, with pubescent
branchlets. Leaves almost sessile, alternate, broadly ovate to
orbicular, many dish-shaped, about 3 mm. in diameter, with
slightly incurved margins, and ending in a recurved mucrone,
glabrous above, slightly hairy below. Stipules small. Flowers
almost umbellate, 3 to 6, terminating the short branchlets, with
hairy pedicels 5 to 6 mm. long. Bracts small. Bracteoles viscid,
orbicular, appressed, scarcely 2 mm. long, and fixed below the
tube of the calyx. Calyx about 5 mm. long, almost glabrous, with
nearly equal lobes shorter than the tube. Standard yellow with
dark-red lines. Wings yellow. Keel dark red. Ovary villous
tapering into a subulate style. Pod ovate-oblong, shortly acumi-
nate. Seeds two, with a distinct strophiole.

This graceful and very distinct species has some resemblance
to Bossmea cor dig era, although it is not a scrambling shrub, but
distinctly erect. It belongs to the Section CoelophyUum, and its
nearest ally is P. Vrolandii Maiden, which it resembles in having
viscid bracteoles forming a complete cup fixed just under the
calyx; but in the case of P. Vrolandii this cup is inflated, and
almost conceals the calyx, while in the new species it is very small,
and tightly appressed to the calyx. The leaves are quite unlike
those of any other Pultenaea , their dish-like appearance suggest-
ing the specific name, and the umbel-like inflorescence with flowers
on long pedicels is unusual in the genus. It may be remarked that
although the flowers are at first terminal, the ends of the branch-
lets are somewhat produced after flowering, so that the flowers
are not then strictly terminal.

The discoverer, Mr. Smith, handed the plant in at the National
Herbarium, and the Government Botanist, Mr. Rae, noting that it
was worthy of further investigation, instructed his senior assis-
tant, Mr. Audas, to proceed to Horsham to obtain additional
material. The author was invited to accompany him, and through
the good offices of Mr. Smith a visit was paid to the locality, 38
miles S.W. of Horsham, where the plant was found in abundance
both in the sandy soil at the foot and among the rocks at the sum-
mit of Mt. Byron. As a vernacular name “ Mt. Byron Bush-pea ”
is suggested.

Pultenaea D’Altonii H.B.W.

New Locality: Mt. Talbot, towards Mt. Byron, J. W. Audas,
October 30th, 1927. Previously recorded only from the Little
Desert near Nhill (type locality), and from near Steiglitz.

[Proc. Rot. Soc. Victoria, 40 (N.S.), Pt. II., 1928.]

Art. VII .—Some Tertiary Volcanic Minerals and their
Parent Magma.

By D. J. MAHONY, M.Sc., F.G.S.

[Read 13th October, 1927 ; issued separately 21st April, 1928.]

In many Victorian scoria cones and bedded tuffs of Tertiary
age, minerals much larger than the normal constituents of the
associated volcanic rocks are found. These minerals may be free
from other matter or embedded in vesicular basalt or may form
the nuclei of volcanic bombs. Those examined are felspars, horn-
blende, augite and olivine.

Felspars.

Analyses of Felspars.

Si0 2

- 65*09

66*20

- 65*64 -

65*14

64*62

65 37

61*93

66*23

A1 2 0 3

- 21*86

22*60

- 21*18 -

2092

22*90

21*56

22*08

19*97

Fe 2 0 3

- 0.30

tr.

- 0*05 -

0*31

0*40

—

tr.

049

FeO

- 0*08

—

- nil. -

—

nil.

MgO

- 0*05

0*38

- 0*21 -

018

0*14

—

tr.

0*17

CaO

- 1*93

1*92

- 1*55 -

1*42

1*30

1*67

nil.

0 50

Na 2 0

- 8*66

805

- 7*45 -

7*43

7 29

7*35

8*84

8*07

k 2 o

- 1*68

1*03

3*66 -

3 93

3*07

3*74

6*34

3 36

h 2 o+

- 012

0*60

- nil. -

0*07

005

0*25

h 2 0-

- 0*06

—

- 0 06 -

0*07

—

007

0*02

CO,

nil.

—

- — .

—

nil.

TiO a

- 001

—

- nil. -

003

tr.

nil.

P.o s

nil.

—

- nil. -

—

nil.

MnO

- 002

—

- _ -

nil.

tr.

—

Li 0

■ strong fcr.

—

-present -

tr.

nil.

—

nil.

so 3

nil.

—

- — ,

—

0*32

- tr.

—

nil.

NiO,CoO

nil.

—

nil.

BaO

- 0*02

—

nil.

SrO

- 016

—

nil.

ZrO

- —

—

- — -

—

tr.

10007

100*80

- 99*80 -

99*50

100*09

10001

99*19

98*81

Rating

- Al.I

A2.II

- Al.I -

Al.I

A2.1I

B3.V

Volcanic Minerals .

Felspars : Norms by Percentage Weight.

1-2

- 7

- 9*99 - 6*20

21-71

- 23*34

- 18*19

- 22*16

- 37*57

- 19*93

- 73*48 - 68*27

63*22

- 63*06

- 61*86

- 62*33

- 75-00

- 68*43

- 10*10 - 9*55

7*72

- 703

6*46

- 8*31

- 8*30

- 2*48

Si0 2

- 3*67 - 10*53

4*42

- 3*32

- 7*27

4*54

—

- 4 50

A] s O s

- 2*08 - 4*71

208

■ 1*81

- 5 18

- 2*35

—

- 212

Enst.

0*12 - 0*94

0*52

- 0*45

- 0*35

. —

—

- 0*42

Ity

- — • —

—

- —

. _

- —

- 0*90

- 0 08 - —

‘ —

- 0*10

Alt

- 0*21 - —

—

- 0*24

- 0*40

008

- —

—

H 2 0, etc.

- 0*18 - 0*60

0*06

- 0*14

- 0*37

- 0*32

- —

- 0*02

10007 - 100*80

99.81

- 99*49

- 100*08

- 100*01

- 120*87

- 98*80

Felspars

: Molecular

Percentages.

1 - 2

Or -

- 10*24 - 7*03

22*49

- 23*98

- 20*15

- 22*80

- 22*94

- 20*97

Ab -

- 79*9t - 82*16

69*50

- 68*77

- 72*68

- 68*64

- 68*46

- 76*42

- 9*81 - 10*82

8*01

- 7'25

- 7*17

- 8*56

- 8*61

- 2*61

99*99 - 100*01

100*00

- 100*00

- 100*01

- 100*00

1. Mount Lookout, Aberfelclv, Gippsland. Large crystal from

vesicular basalt. Sp.gr. 2*618. Analyst, A. G. Hall.

2. Mount Lookout, Aberfeldy, Gippsland. Only about one

gramme available for analysis. A duplicate determina-
tion gave alkalis 9* 18%. Analyst, P. G. W. Bayly.

3. Mount Anakie, about 16 miles NNW. of Geelong. Sp.gr.
2*62. Analyst, F. F. Field.

4. Mount Franklin, near Davies ford. Sp.gr. 2*613. Analyst,

A. G. Flail.

5. Mount Franklin, near Daylesford. Fine, cross-hatched twin

lameilae. Analysts, Miss J. M. Robertson and C. E.
Eales.

6. Mount Franklin, near Daylesford. Fine cross-hatched twin

lamellae. Analyst, F. W. J. Clendinnen.

7. Magorra, near Jumbunna, S. Gippsland. Glassy tabular

crystals from dense basalt. Alkalis evidently too high.
Analysis, Geol. Surv. Vic. Lab. (1901).

8. Mount Noorat, near Terang. Alkalis unsatisfactory on ac-

count of fusion and insufficient material for duplicate
determinations in their case; duplicate determinations
of the other oxides agreed closely. Sp.gr. 2*608.
Analyst, A. G. Hall.

D. J. Mahony :

I am indebted to Professor H. S. Summers for analyses Nos. 5
and 6. The relative values of the analyses are indicated by using
Washington's method of rating (1), the degree of accuracy
being expressed by the letters A, B, C and D, and the degree
of completeness by the figures, 1, 2, 3 and 4. Analyses rated
as A1 are excellent; A2 or Bl, good; A3, B2 or Cl, fair; and the
others poor.

Felspar is abundant at Mt. Franklin and at Mt. Anakie in loose
basalt scoria, and is also found at many other points of eruption
of the New Volcanic basalts. On the shores of Lake Purrumbete
and other lakes of the Western District it is fairly common, and
lias been derived from bedded tuffs. It also occurs at Mt. Look-
out (Aberfeldy) and other volcanic centres in Gippsland asso-
ciated with the Older Volcanic basalt. It is found as colourless
and transparent cleavage fragments, generally small but ranging
up to about two inches long, as rounded lumps evidently partly
absorbed by molten igneous rock, and occasionally as crystals
more or less rounded. Ilowitt (2) described some specimens
from Mt. Anakie as having albite and peri cline twin lamellae, and
he measured extinction angles as follows : —

P. M. Perpendicular to P & M.

+3°7' +12° 30' +10°0'

+3°3' +12° 19' +11°5'

Other examples since examined often resemble microcline in
having twin lamellae in two directions, but extinction angles vary

Numbers correspond with Analyses.

Volcanic M inerals.

somewhat in different parts of the same specimen. A section
parallel to (001) from an Aberfeldy specimen extinguished at 6° ;
this felspar is twinned on the albite law. It contains a few minute
vermiform intergrowths, probably quartz, and some minute
specks too small for determination. It is quite clear and unde-
composed.

The analyses and the triangular diagram show that all these
felspars are anorthoclases ranging from potash albite to potash
oligoclase, and that they are very different from the felspars
entering into the composition of the associated basaltic rocks.
Every analysis (except one which is rated as poor) gives an
excess of Si0 2 and Al 2 O a above the calculated amount required
to combine with the alkalis. These substances may be in solid
solution with the felspar, or the quartz may be in ter grown as
indicated in the Aberfeldy felspar mentioned above.

Analyses of Ferro-Magnesian Minerals.

°/

/ o

mol. propn.

% mol. propn.

mol propn.

Si0 2

4-900

0-8127

- 39*09 -

06483

4153

0-6889

A1 2 0 3

8*66

0-0847

- 15*34 -

0-1500

095

0-0093

Fe,,0 3

2.78

00174

- 12*43 -

0 0778

nil.

—

FeO

6*52

0*0908

- 6-79 -

00945

902

01256

MgO

14*53

0*3604

- 7-67 -

01902

48-02

1*1909

CaO

15*64

0-2790

935 -

0-1668

0-31

00055

Na 2 0

112

0018L

2-32 -

0-0374

0!4

0-0020

k 2 0

0*05

0-0005

1-51 -

00160

003

0-0003

h 2 o+

006

0*0028

- 091 -

0-0506

038

00211

h 2 0-

0*14

—

- 009 -

—

010

—

-co 2

nil.

—

nil. -

—

nil.

—

Ti0 2

1*27

00158

4'76 -

00593

tr.

—

**o 5

nil.

—

strong tr.

—

nil.

—

MnO

0*24

00034

tr.

—

• —

—

Li,0

nil.

—

nil. -

—

S0 3

tr.

—

tr.

—

nil.

—

NiO

0*04

0-0005

—

CoO

tr.

—

BaO

nil.

—

— -

—

013

0-0041

—

Cr 2 0 3

0 03

00002

_ -

—

nil.

—

100*21

100-26

10048

Bating

Al.I

9. Augite, Mount Noorat, near Terang, Sp.gr. 3*342. Analyst,
A. G. Hall.

66 D. J. Mahony :

10. Hornblende, Mount Anakie. Sp.gr. 3-32. Analyst, F. F.

Field.

11. Olivine, Mount Anakie. Granular material from an “ olivine-

bomb.” Analyst, F. F. Field.

Augite.

Only one analysis of augite has been made, and unfortunately
none of the material (which consisted of a black cleavage frag-
ment from Mt. Noorat) was kept for microscopical investigation.
It is a type rich in alumina and alkalis and poor in lime. This is
best shown by taking the maximum, minimum and average figures
from the analyses of augites quoted by Iddings (3) and compar-
ing them with those for the Noorat specimen.

AI 2 O 3 (36 anal.)

CaO (36 anal.)

Na 20 (18 anal.)

Maximum

915

23*46

1-47

Minimum

2 82

15-98

tr.

A verage

5-86

20-89

081

Noorat augite

8‘66

15‘64

Augites are considered to consist of various proportions
of the molecules R0.R0 2 , R0.R 2 0 3 .R0 2 and sometimes
R 2 0.R 2 0 3 . (RO a ) 4 . On this basis the composition of the Noorat
specimen works out as follows : —

Radicle

Mol. propn.

ro 2

8285

R 2 0 3

1023

7341

R 2 O
214

R 2 0.R 2 0 3 .4R0 2

856

214

—

214

ro.r 2 o 3 .ro 2 -

809

—

RO.R0 2

6532

—

6532

—

Surplus

—

The molecular formula is therefore

65 RO . R0 2 +8R0.R 2 0 3 . R0 2 +2R 2 0 . R 2 0 3 . ( R0 2 ) 4
and there is an excess by weight of 0-53% Si0 2 or 0-70% Ti0 2 .

Hornblende.

One good analysis of hornblende is also available. The material
consisted of black cleavage fragments from Mt. Anakie. Exam-
ined microscopically the crushed mineral is a monoclinic amphi-
bole, brown (burnt umber) in colour in very small fragments, the
large being opaque. It is moderately pleochroic (brown to buff)
and the extinction angle measured from the c-axis on cleavage
fragments parallel to the a-axis is 9°. The analysis shows that
it is rich in Al 2 O s and alkalis, and poor in lime for this type of

Volcanic Minerals.

mineral, as indicated by comparison with figures taken from Idd-
ings’s quoted analyses of hornblendes from igneous rocks.

A1 2 0 3 (30 anal.)

CaO (30 anal.)

Na 2 0 (26 anal.)

Maximum

Minimum

Average

17-36

1*50

1069

13 03

925

11-62

318

0- 37

1- 67

Anakie hornblende

15*34

9-35

2 32

The calculation of the formula offers some difficulties. The
hornblendes consist of some or all of the molecules R0.R0 2 ,
R 2 0.R0.2R0 2 , R 2 0.R 2 0 3 .2R0 2 and R0.R 2 0 3 .R0 2 . The simple
method used above in the case of augite is not applicable since no-
radicle is confined to a single molceule. Let w, x, y and z be the
amounts of the above molecules, and a the amount of R0 2 in the
whole of them. Then

w-\-2x-\-2y-\-z=a (total R0 2 required)
x-\-y=1040 (total R 2 0)
y-\-z=227& (total R 2 0 3 )
w-{-x-\-z=4575 (total RO)

and from these equations

zv—2a — 9973
a'=6655 — a
y=a — 5615
-=7893 — a

The maximum value of a is therefore 6655 and the minimum 1
5616. It is reasonable to choose the maximum since this figure-
will give the least excess of R0 2 . Then ,r— 0. The composition
now works out as follows: —

Radicle

Mol. propn.

R 0 2

7076

R 2 0 3

2278

4575

ro 2

1040

wRO.R0 2

3337

—

3337

—

y R a 0.R 2 0 3 .2R0 2 -

2080

1010

—

1040’

»R0.R 2 0 s R0 2

1238

—

Surplus

421

—

The molecular formula is therefore

33RO.RO 2 +12RO.R 2 O 3 .RO 2 +10R 2 O.R 2 O 3 .(RO 2 ) 2

and there is an excess by weight of 2-54% Si0 2 or 3-37% TiOo..

The ejected felspar, augite and hornblende are, therefore, all
types rich in alkalis, alumina, and silica, and poor in lime.

Olivine.

Olivine from points of eruption either forms small, well-defined':
crystals, or more commonly granular masses up to a foot or more

■68

1). J. Mahony :

in diameter containing an admixture of other minerals. Greenish-
yellow crystals about an eighth of an inch long, sp. gr. 3*486, were
•collected from the scoria on top of Mt. Terang. The crystal
forms developed are (010), (001), (110), (120), (101), (Oil)
and (021). The mineral was found to contain 15*80% FeO,
from which its composition is estimated to be

Forsterite (Mg 2 Si0 4 ) 77*57% by weight

Fayalite (Fe 2 Si0 4 ) 22*43%.

The analysed olivine from Mt. Anakie was picked out from
the crushed granular nucleus of a volcanic bomb, and is yellow-
ish-green in colour. Its calculated composition is

Forsterite, 86*76% by weight.

Fayalite, 13*24%.

How the other oxides in this analysis are combined is difficult
to picture, but possibly they form a basic felspar.

Both olivines are rich in magnesia, and contain only about 10%
of the fayalite molecule.

Magma.

Bowen (4) has shown experimentally that minerals rich in
magnesia are the first to crystallise from a cooling magma con-
taining the elements of diopside and the plagioclases, and that
alkaline minerals appear at a late stage of cooling; also that in
systems involving such mix-crystals as the olivines, the earlier
crystals are enriched in magnesia. It therefore seems highly prob-
able that the olivine was formed at an earlier stage in the history
of the magma than the other minerals considered above and that
its association with them is more or less fortuitous. The infer-
ence is that the original magma during the progress of cooling
separated by some process of differentiation into two types, one a
normal basalt and the other an alkali-gabbro ; and that the felspar,
augite and hornblende discussed above come from the alkali-
gabbro differentiate. This inference is supported by the facts
that ejected blocks of essexite type occur in the basaltic tuffs of
Lake Bullenmerri (5); that alkaline volcanic rocks of Tertiary
age are found at Macedon (6 and 7) ; and that analcite basalts
are plentiful in Gippsland.

If the alkaline magma be admitted, the distribution in Victoria
•of the minerals here considered indicates that it is widespread, and
is at least as old as the Older Basalt. In contrast is the fact that
typical alkaline volcanic rocks are comparatively rare. The expla-
nation may be that the alkaline magma was relatively too viscous
to flow to the surface with the same ease as the basalt. The
fine-grained texture of the basalt, the infrequency of porphyritic
crystals in it and the glassy nature of its quickly cooled portions
show that it arrived at the surface in a completely fluid state; and
the extent and thinness of the flows show that it was very mobile.
On the other hand, the comparatively large size of the 'minerals

Volcanic Minerals.

69 >

considered in this paper indicates long continued crystallization of
the magma in which they originated ; the alkaline volcanic rocks-
of Macedon are typically porphyritic; and it is known that
alumina increases the viscosity of a “ melt.” It would therefore
appear that the original magma separated into more and less
alkaline portions before Tertiary volcanic action began. The
basaltic portion remained highly mobile, and rose to the surface
more easily than the alkaline, partly crystallized, more viscid por-
tion, which in consequence seldom formed lava flows, though
some of its constituents together with molten basalt were hurled
by explosions from points of eruption. The anorthoclase basalts
of the Macedon district and the analcite basalts of Gippsland
may represent a less advanced stage of differentiation or the
mixing of the two types of magma.

The magmatic differentiation suggested by the study of the
ejected minerals agrees in general with the deductions made by
Professors Skeats and Summers (7) as a result of their exhaus-
tive study of the Macedon area; but it appears that the process
began before the ejection of the Older Basalts of Gippsland and
continued or was repeated until the end of the volcanic period.

REFERENCES.

1. H. S. Washington. Chemical Analyses of Igneous Rocks.

US. A. Geol. Sum. Professional Paper 99, 1917.

2. A. W. Howitt. On Oligoclase Felspar from Mt.' Anakie,

m Victoria. Rept. Aust. Assoc. Adv. Sci., vii., pp 375-7
1898.

3. J. P. Iodings. Rock Minerals. New York: 1906.

4. N. L. Bowen. The Later Stages of the Evolution of the

Igneous Rocks. Journ. Geol., xxiii., Supplement, pp.
33-39, 1915. 1

5. II. J. Graason and D. J. Maiiony. The Geology of the

Camperdown and Mount Elephant Districts. Geol
Sum. Vic., Mem. 9, 1910.

6. J. W. Gregory. The Geology of Mount Macedon, Victoria.

» _ Rro f- R °y- Soc - n.s.. xiv. (2), pp. 185-251, 1902.

7. E. W. Skeats and H. S. Summers. The Geology and

Petrology of the Macedon District. Geol. Surv. Vic
Bull. 24, 1912.

[Proc. Roy. Soc. Victoria, 40 (N.S.), Pt. II., 1928.]

Art. VIII . — Experimental Error of Field Trials in
Australia.

By H. C. FORSTER, B.Ag.Sc., and A. J. VASEY, B.Ag.Sc.
(Department of Agriculture, Melbourne).

[Read 13th October, 1927; issued separately 28th April, 1928.]
Introduction.

The question of the “ Probable Error ” in field trials is one
which has in recent years come to the fore in connection with the
work of experimental stations in Europe and America. It has
sometimes been suggested that in Australia the water supply
available in the soil for the crop is often the limiting factor to
growth and seed production. This might lead to a more uniform
growth, and thus the experimental error of plot observations
might be thereby diminished. It seemed, therefore, worth while
to investigate the matter fully, and with this end in view the
classic experiment of Hall and Mercer was repeated at the State
Research Farm, Werribec, Victoria, it being felt that such
an investigation should lead to valuable results which would be a
guide for the future in the “ lay out ” of the trial plots.

The experiment was undertaken to determine —

(1) the variation in an apparently uniform acre of wheat as
measured by the “ Standard Deviation/’ and the “ Prob-
able Error ” of 1/1 60th acre plots,

(2) the optimum

(j* ) size,

(/') shape,

Method.

During the season 1926-27, the North Railway Field at Werri-
bee was planted with “ Free Gallipoli ” wheat, and it produced a
fair average crop, which was, before harvest, expected to yield
about 24 bushels to the acre. An acre of this was selected for
the experiment, and many casual observers were agreed that as
far as the eye could judge, it was an even area of wheat.

A preliminary survey was made on the 29th November, 1926,
when it was observed that the drilling was somewhat irregular.
There was one double-sown row in every stroke of the drill, and
therefore it was decided to include two of these double-sown rows
in each plot. Accordingly each plot was made 30 x 20 links, and
the dimensions of the whole acre, 300 x 320 links, excluding

Error of Field Trials in A ustralia.

•paths. Further it was found that near the western boundary of
the acre, a strip a few yards wide had been damaged by cart-
tracks. This was consequently excluded/

A straight row on the western side of the acre was taken as a
base-line, and from this all measurements were set off. These
allowed for the division of the acre into four quarters by means
of two intersecting paths.

On account of the danger of shaking-out by storms before the
harvesting of the whole area had been completed, an occurrence
which would have wrecked the whole experiment, it was deemed
advisable to mark out only one quarter-acre at a time. This was
then harvested immediately. Owing to extremely favourable
weather conditions during the harvesting period, such precautions
proved unnecessary.

Along the boundaries of each quarter-acre pegs were put in
corresponding to the corners of the outside plots. The boun-
daries of each 1 /160th acre plot were then defined by stakes
whose positions were obtained by sighting from the outside pegs.
Paths were then cut in a N.-S. direction, dividing the quarter-acre
into 5 strips of 8 plots each. These paths were 4 drill-rows wide,
and were made by hand-cutting 2 rows on each side of the
actual boundaries of the plot as defined by line and plumb-bob.
As the bags for the reception of the produce from each plot had
previously been marked, the crop cut in the formation of the
paths was transferred immediately to the corresponding sack.

Cutting was commenced on the 22nd December with a single-
horse mower fitted with a carrier-arrangement. The mower was
driven in an E.-W. direction across the paths, thus cutting five
plots. It was stopped in each pathway — specially cut for this pur-
pose — to enable the crop cut from each plot to be bagged straight
from the carrier. After four swathes of the mower, a strip "of
about one foot was left along the northern boundary of each plot.
This was cut by hand, the exact boundary being defined by line
and plumb-bob as before. Plots were then thoroughly gleaned
for any heads that had been broken off, as well as anv loose
straws.

Before cutting, the plots were examined for the number of
rows they contained, and for the presence of any disturbing fac-
tors. There were very few weeds. In a similar manner the other
three quarter-acres were harvested, the bagged produce of the
plots being carted and stored as the harvesting of each quarter-
acre was completed. Field work was finished on the 7th Januarv,

Thrashing was commenced on the 18th January. This was per-
formed by means of a motor-stripper, which consisted of the
drum and beaters of a typical Australian harvester, driven by a
stationery engine mounted on the same under-carriage. After
thrashing the wheat fed into the beaters, both straw and grain
were delivered into a bin at the rear. Here the straw was col-
lected, and later re-thrashed separately from the grain. The

Forster and Vasey :

grain was winnowed to an even sample, and weighed to the
nearest ounce, which was considered the limit of the overall-
accuracy of the experiment. Thrashing was completed on the
27th January.

Results and Discussion.

Table 1 . — Plan and weights in ounces of grain harvested from

160 wheat plots.

125 -

117 -

125 -

135 -

159

| 139 -

150 -

161 -

143

- 157

121 -

135 -

123 -

147 -

147

130 -

141 -

154 -

144

- 161

124 -

137 -

126 -

135 -

142

j 142 -

138 -

135 -

139

- 136

125 -

134 -

127 -

130 -

144

| 133 -

141 -

147 -

142

- 147

128 -

135 -

121 -

136 -

138

140 -

135 -

149 -

139

- 139

132 -

127 -

118 -

134 -

135

1 137 -

137 -

138 -

127

- 145

137 -

133 -

122 -

131 -

120

137 -

139 -

142 -

133

- 148

141 -

128 -

130 -

126 -

1 19

149 -

143 -

132 -

133

- 144

123 -

119 -

120 -

1 1 1

139 -

152 -

142 -

140

-p

- 143

115 -

125 -

135 -

121 -

125

143 -

135 -

143

- o

- 146

121 -

120 -

133 -

132 -

128

144 -

133 -

127 -

137

- 137

137 -

123 -

141 -

124 -

128

135 -

139 -

137

- 142

127 -

131 -

136 -

121 -

135

135 -

133 -

335 -

133

- 131

122 -

124 -

144 -

132 -

128

136 -

143 -

145 -

143

- 142

122 -

108 -

141 -

143 -

125

139 -

148 -

150 -

152

- 154

130 -

135 -

161 -

154 -

154

153 -

141 -

130 -

152

- 161

Table 1 shows the yields of plots together with their position
in the field. The yields varied from 108 to 164 ozs., the variation
being 20-6% on either side of the mean. The frequency curve
as shown in Figure 1 was obtained by grouping the yields into
periods of 5 ounces each.

With the curve from the actual results is shown the normal
curve of error calculated to fit the results. Owing to the small
number of observations, the approximation of the actual curve
( vide Figure 1) to the above is considered close enough to justify
the conclusion that the material was homogeneous, and that the
formulae applicable to such, may be used in this case.

A study of Table 1 shows that there is a definite rise in yield
from East to West, while the variations from North to South are
apparently irregular. The graph (Fig. 2) of the sum totals of
the rows of plots, as set out in Table 2, verifies these con-
clusions.

Error of Field Trials in Australia .

Fig. 1. — Frequency curves Jor 130 wheat plots .
(Actual and 'Theoretical).

It is necessary to consider briefly these disturbing elements
before proceeding to the main discussion and conclusions.

The presence of such a regular rise in the field under observa-
tion is a factor which has appeared in most investigations of this
character. In their Mangold experiment, Hall and Mercer had a
similar experience in a variation from North to South of 7*3%,
which, after being observed and noted, was subsequently disre-
garded in the calculation of results. In this case, there is a varia-
tion from E.-W. of 6-9% on either side of the mean. The
irregular variation from South to North is similarly 5*9%. Since

YIELD IN OZ S PER ROW

Forster and Vasey :

Table 2. — Fin ing weight of rows of plot yields.

South-North

Eas>t-V\ r est

1414 ozs.

2030 ozs.

1403 „

2035

1354 „

2102 „

1370 „

2121 ,,

1360 „

2138 „

1330 „

2231 „

1342 „

2256

1345 „

2264 „

1312 „

2237 „

1331 „

2333 „

1312 „

IQ/IC

loTO ,,

1317 t .

1359 „

1382 „

1471 „

these variations, viz. 6*9/ from E.-W., and 5-9% from S.-N., are
of approximately the same order, it is possible in a similar manner
to disregard this regular variation from side to side.

Error of Field Trials in Australia. 75

Hall and Mercer in their experiment measured each plot as a
definite distance along a certain number of rows, thus taking
area of crop as their unit. On account of'the irregular drilling, it
was impossible in this experiment to include a definite number of
rows in each plot ; therefore area of land was taken as the unit.

The examination of the number of drill rows showed a varia-
tion of from 33 to 35 rows per plot. This variation, 3-til, is of
the order of 3%. On taking only those plots containing 34 rows,
the yields varied from 115 to 164 ounces, a range of approxi-
mately 18% on either side of the mean. Thus the normal varia-
tion due to chance is far greater than the difference that could be
produced by such variation in the number of rows, and this may
therefore be grouped with these chance errors. A more accurate
comparison may be drawn between the Standard Deviation of all
the plots (S.D. = 10-9=fc0-41 ozs.), and that from those containing
the same number of rows (34), S.D.=11 -9 ±0-58 ozs.). These
two figures are of the same order. Now, since this S.D. is a
measure of the variance of the plot yields, the above assumption
is confirmed.

The Variation in an Apparently Uniform Acre of Wheat
as measured by the Standard Deviation and the Probable
Error of l/160th acre plots.

Table 3. — Calculation of the Standard Deviation.

Group

Frequency

Deviation from
Arbitrary Mean

fx2

107-111

- —5

-10

112-116

—4

-4

117-121

-3

-36

108

122-126

-2

-36

127-131

-1

-18

132-136

£2

137-141

142-146

147-151

152-156

112

157-161

125

162-166

Totals

160

—

761

It may be calculated by the usual formulae that the mean yield
of the 1/ 160th acre plots is 136’5 7*3 ozs., i.e., there is an even

chance that the yield from any one plot will be between 143 8 ozs.

7ti

Forster and Vasty •

and 129 2 ozs. Further that if a comparison were made between
a pair of 1 /160th acre plots of two different varieties of wheat
on similar land to that found here, any differences between yields
of less than 23*3 ozs. (177% of the mean), would not be
significant.

Optimum Size of Plot.

In order to determine the optimum size of plot for purposes of
yield trials, i.e. that size of plot which will give the least variation
from the mean, it was necessary to compare the S.D. of different
sizes of plots. By the grouping of adjacent plots, the yields from
areas of different sizes have been obtained. The method of
grouping is indicated by the accompanying dimensions in Table 3.

Table 4. — The Standard Deviation (%) of Plots of Various

Sizes.

Size of
Plot

No. of

Plots

Dimensions

Standard ( B ,
Deviation

1/ 160th

160

30 x 20 Iks.

8-0%

l/80th

30x40 „

7-0

l/40th

60 x 40 „

5'8

1 /20th

80x60 „

1 / 10th

80 x 120 „

- 4’6

N.B. — The small number of results in the two latter cases
detracts somewhat from the reliability of the figures 5-2 and
4' 6% respectively.

From the above table and the following graph, it will be noted
that the S.D. (%) falls rapidly from 8-0% in the case of the
1 /160th acre plots to 5-8% at the l/40th acre plots. Further in-
crease in size up to 1 /10th acre only reduces this quantity to
4*6%. Now, since the larger the area, the greater the difficulty
in obtaining an “ apparently uniform ” area of soil, it follows that
little is to be gained by increasing the size of plot for yield trials
above 1 /40th of an acre.

STANDARD DEVIATION PER CENT

Error of Field Trials in Australia.

Fig. 3. — Actual and Theoretical Curves of the Standard Deviation
of Plots of Various Sizes.

N.B. — The theoretical curve is obtained by the division of the
S.D. of the 1/1 60th acre plots by the square root of the number
of the original number of small plots combined in each grouping.

Optimum Shape of Plot.

It is generally considered that a long narrow plot is more
desirable for field-scale work than a short square plot, and the
following table tends to establish this belief.

Forster and Vasey :

Table 5. — Standard Deviation of Plots of Various Shapes.

Size of

Plot

No. of

Plots

Dimensions

Standard */.
Deviation

l/40th ac.

60 x 40 Iks.

5-8%

l/40th „

20 x 120 „

l/20tli „

80x60 „

l/20th „

20 x 120 „

3*7

It is important to note that on account of the gradual increase
in yield from east to west, plots with their axis in a north to south
direction cannot be used in the above comparison.

Optimum Number of Replications desirable.

Having determined the size and shape most desirable from a
practical standpoint, it was necessary to find the number of
replications required for a working minimum of error. The S.D.
was then calculated for two scattered l/20th acre plots, four
scattered l/40th acre plots, etc. Maximum scattering was ob-
tained by entering the yields of the various sized plots on slips of
paper, which were later drawn from a bowl, and thus the various
sets of pairs, fours, eights, etc., were made up.

Table 6. — Standard Deviation of 1/10 th acre plots obtained by
random grouping of various numbers of units .

No. of Units
in 1/10 ac. plot

No. of
observations

Standard Deviation

4*6%

4*0

3-18

246

2*3

N.B. — Only a low reliability can be placed on the figure 4 6,
due to the small number of results.

From this it would appear that a greater number of replications
than four or five is not warranted, as the small increase in
accuracy so obtained would entail a great amount of extra work.

Error of Field Trials in Australia .

HO. OF REPLICATIONS

Fig. 4. — Relation between the Standard Deviation and the
Number of Replications.

(Actual and Theoretical).

N.B. — The theoretical value is obtained by the division of the
S.D. of 1 /10th acre plots by the square root of the number of
units into which it was divided.

Conclusions.

While the small number of observations necessarily detracts
from the accuracy of some of the results, the following conclu-
sions seem to be justified, supporting, as they do, most of the
previous work overseas.

80 Forster and Vasey: Field Trials in Australia.

( 1 ) That in this field experiment, there are two types of error —

(a) casual, due to small chance errors in harvesting tech-
nique, uneven seeding, manuring, hare-tracks, etc.
These may be so gradual as to be inappreciable to the
eye.

(b) more regular errors, due to marked soil variations,
climate, etc.

(2) That the casual error attaching to a single plot decreases
with the increasing size of plot, but the more systematic
error of soil variation becomes more important as the plot
increases in size.

(3) The optimum size for field trials for cereals under con-
ditions such as these, is l/40th acre.

(4) That there would appear to be grounds for the belief that
a long narrow plot is the more desirable for field trials.

(5) That the error attaching to a l/40th acre plot is diminished
to a working minimum by a replication of five times in any
one series.

It is absolutely essential that these results be applied with
caution. They are only of value for the conditions which pre-
vailed during the period of the experiment, on the particular soil
on which the experiment was conducted. Thus in the first place
they will apply only to areas of crop in which the eye is unable to
detect any serious lack of uniformity. If a field, used for yield
trials, contained areas in which the crop was locally affected owing
to disease, extra-heavy rain or some other exceptional circum-
stance, there would be no reason for expecting that the statistical
results obtained in the Werribee work would hold good in such
an area.

In the second place, with different climatic conditions the re-
sults might be different, but the marked similarity between the
results at Werribee and at Rothamsted suggests that this is not
likely to be a very serious source of trouble.

Finally, the authors wish gratefully to acknowledge all assist-
ance received. The experiment itself was undertaken under the
direction of the Department of Agriculture, Melbourne, at the
suggestion of Mr, H. A. Mullett, Superintendent of Agriculture.

They are also greatly indebted to Professor S. M. Wadham for
his many suggestions and helpful criticism of this report.

The facility and accuracy obtained in the field work would have
been impossible but for the assistance of Mr. A. Morgan,
B.Ag.Sc., mid Messrs. Pescott and Skene, students in Agriculture
in the University of Melbourne.

[Proc. Roy. Soc. Victoria,. 40 (N.S.), P'r. II., 1928.]

Art. IX . — Contributions to the Flora of Australia , No. 3f*

Additions to the Flora of the Northern Territory and
Locality Records.

ALFRED J. EWART, D.Sc., Ph.D., F.L.S., F.R.S.

(Professor of Botany and Plant Physiology),
and

PHYLLIS H. JARRETT, B.Sc.

(Caroline Kaye Scholar in Botany, University oi Melbourne).

[Read 8th December, 1927; issued separately 28th April, 1928.]

The records made in this paper are mainly the result of a
revision of the collection of Northern Territory plants made by
the Horn Expedition in 1894. This collection is housed in the
Tate Herbarium at the University of Adelaide, and we have here
to acknowledge our gratitude to Professor Osborn for making
this material available to us.

Many of the species have not been previously recorded either
in the report of the Horn Expedition or in the Flora of the North-
ern Territory, and other records add considerably to the range
of species already known without definite locality from Northern
Australia, and recorded in the Melbourne National Herbarium
Census.

This paper is a further step in the collection of material for a
complete Flora of the Northern Territory.

GRAMINEAE.

Eriochloa punctata Hamilton.

Swallow Creek, Gidia Creek, and Opossum Water hole, R.
Tate, May, 1894, (labelled E. polystachya H. B. et K).
Panicum cffnsum R.Br.

Tennant’s Creek, R. Tate, 1894.

P. r ever sum F.v.M.

Finke River, R. Tate, 1894.

Setaria glauca (L.) Beauv.

Palmerston, ArnheinTs Land, M. Holtze, 1882.

A. verticillata (L.) Beauv.

Opossum Waterhole, R. Tate, 1894.

Alopecurns geniculatus L.

Barrow Creek, R. Tate, 1894.

*No. 33 in Proc. Roy. Soc. Vic., n.s., xxxix. (2),p. 154.

Ewart and J arret t :

Eriachne ovata Nees var. pallida Benth.

MacDonnell Ranges, R. Tate, 1894.

Triraphis danthonioides F.v.M.

Gill’s Range, R. Tate, 1894.

This is the first definite locality recorded for this species in
the Northern Territory.

Chloris divaricata R.Br.

Finke River, R. Tate, 1894 (labelled C. acicnlaris ).
Fimbristylis Neilsonii F.v.M.

MacDonnell Ranges, Rev. Kempe, 1883.

Scirpus americanus Pers.

Illara Water, R. Tate, 1894 (labelled S. pangens ).

This species had not been previously recorded for the North-
ern Territory.

CENTROLEPIDACEAE.

Centrolepis polygyna (R.Br.) Hieron.

South of the MacDonnell Ranges, R. Tate, 1894.

POLYGONACEAE.

Polygonum serrulatum Lag.

Finke River, R. Tate, 1894.

CHENOPODIACEAE.

Chenopodium cristatmn F.v.M.

Ilpilla Gorge, R. Tate, 1894.

Atriplex campanulatum Benth.

On the sandhills south of the MacDonnell Ranges, R. Tate,-
1894. This species has not been previously recorded for
the Northern Territory.

A. fissivalve F.v.M.

Near Elizabeth Creek, — Giles.

This species has not been previously recorded for the North-
ern Territory.

A. liinbatum Benth.

Finke River, R. Tate, 1894.

Bassia convexula R. FI. Anders. (=B. echinopsila F.v.M.).

The Goyder and Ilpilla Gorge, R. Tate, 1894.

B. criacantha (F.v.M.) R. H. Anders.

This species has not been previously recorded for the North-
ern Territory.

Kochia Georgei Diels.

Mt. Olga, W. H. Tietkens, 1899.

K. villosa L. var. enchylaenoides J. M. Black.

Charlotte Waters, R. Tate. 1894.

Flora of Australia.

Threlkeldia inchoata J. M. Black.

Adminga Creek, R. Tate, 1894.
Arthrocnemum halocnemoides Nees.
Finke River, R. Tate, 1894.

AMARANTACEAE.

Trichinum arthrolasium F.v.M.

MacDonnell Ranges, R. Tate, 1894.

T. helipteroides F.v.M.

Upilla, Barrow Creek, R. Tate, 1894.

T. helipteroides F.v.M. var. minor J. M. Black.

Finke River, R. Tate, 1894.

T. nobile L. (—Ptilotus nobilis (L.) F.v.M.).

Mt. Sonder, MacDonnell Ranges, R. Tate, 1894.

T. parvifolimn F.v.M. (—Ptilotus parznfolius F.v.M.).

Crown Point, Finke River, R. Tate, 1894.

This is the first definite locality recorded for this species in
the Northern Territory.

Amarantus Mitchelli Benth.

On the sandhills south of the MacDonnell Ranges, R. Tate,
1894.

This species has not been previously recorded for the North-
ern Territory.

Gomphrema affinis F.v.M.

Croker Island, R. Tate (No. 60), March, 1883; Pine Creek,
MacDonnell Ranges, R. Tate, 1894.

G. Brozunii Moq.

Finke River, Rev. Kempe, 1882 ; Mt. Sonder, R. Tate, 1894.
G . parvifolia Benth.

Mt. Norris Bay, R. Tate (No, 79), March, 1883; Pine Creek
and Barrow Creek, R. Tate, 1894.

AIZOACEAE.

Trianthema crystallina Vahl. var. clavata J. M. Black.
Throughout the MacDonnell Ranges, R. Tate, 1894.

PORTULACEAE.

Portulaca filifolia F.v.M.

Stuart’s Pass, MacDonnell Ranges, R. Tate, 1894.

P. oleracea L. var. grandifolia Benth.

On the sandhills of the MacDonnell Ranges, R. Tate, 1894.
Calandrinia polyandra (Hook.) Benth.

Finke River, R. Tate, 1894.

This is the first definite locality recorded for this species in
the Northern Territory.

£4

Ewart and Jarrett:

C. pusilla L. (=C. volubilis Benth.).

MacDonnell Ranges, Rev. Kempe, 1883.

This species has not been previously recorded for the North-
ern Territory.

C. remota J. M. Black.

Charlotte Waters, Baron von Mueller.

CRUCIFERAE.

Menkea australis Lehm.

South of the MacDonnell Ranges, R. Tate, 1894.

M. sphaerocarpa, F.v.M.

Mt. Olga, Rev. Kempe, 1883.

CRASSULACEAE.

Crassula bonariensis (D.C.) Cambess.

Finke River, Rev. Kempe, 1883; (labelled Tillaea purpurata
Hook.). '

C. colorata (Nees) Ostenf.

MacDonnell Ranges, R. Tate, 1894.

LEGUMINOSAE.

Acacia Bynocana Benth. (=A. Wilhelmiana F.v.M.).

Along the west end of Lake Amadeus, W. H. Tietkens, 1889.
A. Cambagei Baker.

The Goyder, Swallow Creek, and on the slopes of Mt.
Daniel, R. Tate 1894 (labelled A. homaloplxylla Cunn.).
A. coriacea D.C.

MacDonnell Ranges, R. Tate, 1894.

A. ligulata A. Cunn.

West of the MacDonnell Ranges, W. H. Tietkens, 1889.
Cassia Sturtii R.Br. var. involucrata J. M. Black.

Stuart’s Pass, MacDonnell Ranges, R. Tate, 1894.

Bauhinia Leichardtii F.v.M. var. cinerascens.

MacDonnell Ranges, C. A. Winnecke, 1883.

Isotropis Winneckiana F.v.M.

MacDonnell Ranges, C. A. Winnecke, 1883.

Crotalaria unifoliata Benth.

, Cameron’s Well, Central Australia, R. Tate, 1894.

Indigofera saxicola F.v.M.

Yam Creek, R. Tate (No. 7), 1883.

Swainsona Burkei F.v.M. var. parviflora.

Near Mt. Sonder, R. Tate, 1894.

N. canescens F.v.M. var. Horniana Tate.

Glen Helen Gorge, R. Tate, 1894 (labelled Y. Horniana ).

S. stipularis F.v.M.

Common in the scrub near the Goyder, W. H. Tietkens,

Flora of Australia .

ZYGOPHYLLACEAE.

Zygophyllum Billardierii D.C. (=Z. ammo phy Hum F.v.M.).

Finke River at Idracowrie, R. Tate, 1894.

Z. compression J. M. Black.

Sonder’s Range, R. Tate, 1894.

Tribnlus mi nut us Leich.

Ayer's Rock, MacDonnell Ranges, W. H. Tietkens, 1889.

POLYGALACEAE.

Poly gal a orbicularis Benth.

Port Darwin, R. Tate (No. 95), 1883.

STACKHOUSIACEAE.

Macgregoria racimigera F.v.M.

MacDonnell Ranges, C. A. Winnecke, 1883; Mt. Gillen and'
North of Alice Springs, R. Tate, 1894.

This is the first definite locality recorded for this species in
the Northern Territory.

TILIACEAE.

Corchorus vermicularis F.v.M. ( =Scorpia simplicifolia Ewart
and Petrie, 1926).

Wycliffe, A. J. Ewart, 1924.

This curious plant is only recorded from one locality in
BenthanTs Flora, and it is sparsely distributed in the
Northern Territory, usually near to river banks or on
flood-plains. In North-West Australia it has recently
appeared in many localities from Derby to Fitzroy
Crossing and Leopold, mostly on grazed river areas
after floods, in some cases being now the dominant vege-
tation over acres of ground. Apparently it is a native
plant whose spread is favoured by grazing; sheep, cattle
and. horses usually avoid it, and even goats appear to
eat it only sparingly.

STERCULIACEAE.

Melhania incana Heyne ( =Sideria revert a Ewart and Petrie,
1926).

Tayloi s Well, A. J. Ewart, 1924. This is the first record
of this plant in the interior of Northern Australia. The
suppressed genus “ Sideria ” was placed under the Mal-
vaceae.

Ruelingia hermanniae folia Steetz.

Near the Finke River, Rev. Kempe, 1882; Watson Hills, W.
H. Tietkens, 1889. (Labelled Commersonia Kempeana
F.v.M.).

'86

Ewart and Jarrett :

DILLEN I ACEAE.

Pachynema sphenandrum F.v.M.

Near Yam Creek, R. Tate, 1894.

This is the first definite locality recorded for this species in
the Northern Territory.

LYTHRACEAE.

Rotala occultifolia Koch van Leichardtii Koch.

West of the MacDonnell Ranges, W. H. Tietkens, 1889;

Deering Creek, R. Tate, 1894.

This is the first definite locality recorded for this species in
the Northern Territory.

MYRTACEAE.

Micromyrtns ciliata J. M. Black (=Thryptomene flavifolia
F.v.M. ).

Along the south side of Gill’s Range, R. Tate, 1894.

HALORRI-IAGIDACEAE.

Loudonia Roei Schlechtd.

On the sandhills south of Gill’s Range, R. Tate, 1894.

BORAGINACEAE.

Hcliotropinm he ter ant hum F.v.M.

Near Lake MacDonald, Central Australia, W. H. Tietkens,
June, 1889.

H. tenuifolium R.Br.

MacDonnell Ranges, R. Tate, 1894.

CONVOLVULACEAE.

Ipomoea heterophylla Schrank.

Port Darwin, R. Tate, (No. 50), 1883.

I. lonchophylla J. M. Black.

Swallow Creek, R. Tate, 1894 (labelled I. heterophylla ).

SCROPHULARIACEAE.

Striga curviflora Benth.

MacDonnell Ranges, W. H. Tietkens, 1889.

S. hirsiita, Benth.

Port Darwin, R. Tate (No. 99), 1883.

This species has not been previously recorded for the North-
ern Territory.

ACANTHACEAE.

Ruellia bracteata R.Br.

Yam Creek, R. Tate (No. 29), 1883.

Flora of Australia.

MYOPORACEAE.

Myopornm deserti A. Cunn.

South of the MacDonnell Ranges, W. H. Tietkens, 1889.
M. montanum R.Br.

MacDonnell Ranges, R. Tate, 1894 (labelled M. Dampieri ),
Eremophila Elderi F.v.M,

James’s Range, R. Tate, 1894.

E. Latrobei F.v.M. var. Tictkcnsii (=E. Tietkensii F.v.M.).

On the south side of Gill’s Range, R. Tate, 1894.

E. neglecta J. M. Black.

At Yellow Cliffs, near Charlotte Waters, R. Tate, 1894.

VERBENACEAE.

N ewcastlia cephalanthci F.v.M.

Finke River, R. Tate, 1894.

RUBIACEAE.

Oldenlandia elaiinoides F.v.M.

Deering Creek and Haast’s Bluff, MacDonnell Ranges, R.
Tate, 1894.

This species has not been previously recorded for the North-
ern Territory.

GOODENIACEAE.

Scaveola ovalifolia R.Br. var. parviflora.

Mt. Sonder, Illawarta and Idracowra, R. Tate, 1894. This is
the first record of this variety in the Northern Territory.

COMPOSITAE.

Vittadinia bracJiycomoides F.v.M.

Throughout the MacDonnell Ranges, R. Tate, 1894.

Calotis cuneifolia R.Br.

In the mulga scrub at Glen Edith, R. Tate, 1894 (labelled
C. dentrix).

This species has not been previously recorded for the North-
ern Territory.

C . scabio si folia F.v.M.

Finke River. R. Tate, 1894.

H die hr y sum bracteatum Willd.

Finke River. Rev. Kempe, 1883 (labelled H. lucidum
Henck.).

REFERENCE.

Ewart and Petrie, 1926. Contributions to the Flora of Austra-
lia, No. 31. Proc. Roy. Soc. Vic.;, n.s., xxxviii.

[Proc. Roy. Soc. Victoria, 40 (N.S.), Pt. II., 1928.]

Art. X. — Fossil Plants of the Stony Greek Basin .

By REUBEN T. PATTON, B.Sc., M.F.

(With Plate VIII.)

[Read 8th December, 1927 ; issued separately 7tli June, 1928].

The geology of the Stony Creek Basin, Daylesford, has been
the subject of many papers, the last of which being that by
Orr (1). In this basin is a thick deposit of black ligneous clay,
the origin of which is a matter of doubt. Although in places the
deposit contains a large amount of plant material, yet owing to its
lack of any definite lamination it is very difficult to secure un-
broken specimens. This applies particularly to the leaves of the
genus Eucalyptus, which occur abundantly. Small fragments of
what is apparently fern material are present, but the identification
is difficult. One fern appears to be Pteridium aquilinum , which
is at present world wide. Another specimen has large broad
frond segments with large orbicular sori, characters which are
identical with the living species Polypodium piistulatum. The
veining of the leaves of the Eucalypt leaves can be very distinctly
made out in fresh material. No complete leaves were obtained.
The veining is of two distinct types : one has the veins very
oblique and the other has the veins set at an angle of about 45°.
The oblique veining occurs among others in the living species E.
aniygdalina, and the other type is seen in the living species E. vim -
inalis, Both these species occur living in the area under discus-
sion. The veining of E. aniygdalina is very variable, so that it is
quite possible that the leaves all belong to the same species. The
leaves are all comparatively narrow and falcate, and about 4 to 6
inches long. It is quite probable that the fossil leaves belong to
the existing species. Besides the leaves, however, there are
woody masses which are very soft and cheesy in consistency. The
material is very soft and, therefore, difficult to section, but when
dry it is very brittle and fractures like coal.

Microscopically it is seen that the cell walls have been enor-
mously swollen, so much so that in most parts the cell cavity has
been obliterated. This swelling of the walls has also caused the
bordered pits to a very large extent to disappear, and other
characters are also very much affected. This makes the identifi-
cation very difficult. However, it is easily seen that the wood is
of gymnospermous origin, The annual rings are very distinct, and
are also very broad. Approximately the spring and the autumn
wood are about equal in breadth. The summer wood is very
dense, and owing to the swelling of the walls the lumen is com-
pletely obliterated. The spring wood is very open, and compara-
tively thin walled. This portion of the ring is very much dis-
torted. At first sight it would appear that the wood had been
subject to strong pressure in a radial direction, but the nature of

Fossil Plants.

the deposit in which it occurs does not favour this suggestion.
The distortion is entirely due to the swelling of the walls. In
cross section no resin canals nor resin cells' can be observed, but
it is quite possible that even if the latter were present in the sum-
mer wood they would not be observed. In longitudinal radial sec-
tion it is seen that the bordered pits, which are but rarely pre-
served, were arranged in single rows. The medullary rays are
homogenous. The pits connecting the medullary rays with the
tracheides are large, broad, elliptical and simple. These, too, have
been largely obliterated by the swelling of the walls. This charac-
ter had been observed in some fossil wood sent by Baron von
Mueller to Schenk (2, pp. 872-4), and named by the latter Phyl-
locladus Muelleri . These large pits had already been noted in
the living species Phyllocladus trichomanoides, which is endemic
to New Zealand. These pits also occur in the Tasmanian species,
P. rhomboidalis. This is also an endemic species. These pits are,

Fig. 1. — A, B and C leaves of Eucalyptus spp.

0 Radial section of woody material shewing uniseriafce
bordered pits of the medullary rays.

however, not confined to the genus Phyllocladus , for they also
occur in the two endemic Tasmanian species, Dacrydinm Frank -
Uni and Microcachrys tetragona. The last genus is endemic to
Tasmania, and is monotypic. The other two genera in which
the large elliptical pits occur, are, however, very widely distri-
buted. Phyllocladus occurs in Tasmania, New Zealand, New
Guinea, Borneo and the Philippine Islands. Dacrydinm occurs in
Tasmania, New Zealand, Fiji Is., New Caledonia, New Guinea,
Borneo, Philippine Is., Malay and Chile. From the distribution
of the existing species it is seen that these two genera range over a
very wide area, and it is therefore very surprising that, while these
two genera are found on the south, east and north of Australia,

Reuben T. Patton : Fossil Plants.

they are nowhere found at present on the mainland itself. The
two genera, Phyllocladus and Dacrydiitm , as far as their Tas-
manian species are concerned, are so very similar as regards their
wood anatomy that it is impossible to separate them. It is there-
fore impossible to say in which genus this fossil wood should be
placed. The longitudinal tangential section does not show any
definite characters, and this is also a feature of the Tasmanian
species mentioned above. Similar fossil wood has been obtained
from the Malakoff Reef at Ballarat, and also from the Langi
Logan Mine at Ararat. From the distribution of the fossil wood,
therefore, it is apparent that somewhere about the Newer Basaltic
period at least one of the above genera was present in Australia
itself. The disappearance of the genus from the mainland has been
probably due to secular changes of climate since basaltic times.
The three genera mentioned are found in the wetter areas of Tas-
mania, and therefore it is most probable that similar conditions
previously existed in those parts of Victoria where the fossils
have been found. The presence of Polypodium sp. also indicates
a wet habitat. Polypodium to-day exists as an epiphyte on tree-
ferns, and other arboreal vegetation in the moist gullies of the
State. This further supports the suggestion that the climate
was formerly moister than it is to-day. The distribution of the
species Eucalyptus amyc/dalina, E. viminalis and Ptcridium aqui -
liniitn is not controlled by climatic but by soil conditions. All
these three are found in the wettest areas as well as in the com-
paratively dry regions. The area where the fossils have been
found has, therefore, apparently become progressively drier, and
therefore moisture loving species have, so to speak, been driven
out.

REFERENCES.

1. D. Orr. The Stony Creek Basin and the Corinella Dyke.

Proc. Roy. Soc. Vic., n.s., xl. (1), pp. 25-33, 1927.

2. A. F. W. Schimper and A. Schenk. ZitteTs Handbuch der

Palaeontologie. Part II. — Palaeophytologie. Pp. xi,

958. 433 text figs. 8vo, Munich and Leipzig, 1890.

EXPLANATION OF PLATE VIII.

Distribution of the genera Phyllocladus and Dacrydium.

Proc. K.S. Victoria, 1928. Plate VIII.

Distribution of Phyllocladus and Dacrydium.

[Proc. Roy. Soc. Victoria, 40 (N.S.), Pt. II., L928.J

Art. XI . — The Staurograptus Bed of Victoria.

By W. J. HARRIS, M.A., and R. A. KEBLE.

(With Plate IX.)

[Read 8th December, 1927 ; issued separately 7th June, 1928.]

Messrs. W. J. Harris and W. Crawford recently found some
dendroid graptolites of considerable importance to Victorian
Ordovician stratigraphy, in a band of slate on the bank of a
creek 3 miles north east of Romsey. The band contains the genus
Staurograptus , and we regard it as being very near the base of
the Lower Ordovician. The band (approximately 27 chains,
N.18°W. from the south-west corner of Allot. 26, Parish of
Springfield, on a water reserve), is of hard, black slate interca-
lated with bands of chert, quartzite, and altered sandstone. A
note on Quarter Sheet 5 SE. refers to the outcrop as “ black
shales.” The strike is almost north and south and dip 86° west.
Easterly up the creek is an outcrop of quartz-porphyry. 1 Still
further east shales, mudstones and sandstones occur, lithologically
similar to those outcropping in typical Silurian areas. 2

Only two graptolite genera, viz. Staurograptus and Dictyoncma,
have been recognised, in both cases preserved as films on the bed-
ding planes of the slate, but in profusion. Staurograptus is a
new record for Victoria; Dictyouema has been found at several
localities, more particularly at Taylor's Quarry, 5 miles to the
north. There D. macgillivrayi T. S. Hall, D. pulchellum T. S.
Hall, and D. grande T. S. Hall, occur with Bryograptus, Clono -
graptus, Tetragraptus , etc.

Genus Staurograptus Emmons.

.Staurograptus diffissus, sp. nov.

(Plate IX., Figs. 1-5.)

Polypary broadly conical to saucer shaped ; usually vertically
compressed. Sicula long, conical, suspended by a long nema ; no
primary disc observed.

The primary theca grows beyond the aperture of the sicula;
the polypary begins with four or more branches.

Polypary small, in evathiform aspect less than T5 cm. wide
distally, in vertically compressed aspect (apparently less mature
forms) not exceeding 1 cm. It develops by dichotomy to approxi-

1- — This is shown on the Quarter Sheet as M greenstone,” and was probably
presumed to belong to the diabases of which the hills to the north are
largely composed. It is an acid dyke similar to those found further
south.

2. — Professor Skeats, however, has described these as Heatlicotian eastwards up to
the Basalt junction. Pan-Pacific Scl. Congress, Australia, Melb. Handbook,
1 ). 134 ; 1923 ; reprinted in Proc. Pan*Pac. Sci.Cong., Aust. 1923, ii., p. 16G7, 1923.

Harris and Keble .*

mately 16 branches of the fourth order; none of our forms seems
to have developed further.

Branches slender, about 0-25 mm. wide, all the branches slightly
flexuous, branching at irregular intervals. In the horizontally
compressed polypary the branches of the third order diverge at an
average angle of 45°.

Thecae number from 20 to 25 in 10 mm., in contact for one-
third of their length, outer wall straight or slightly concave, aper-
tural margin slightly concave. Ventral margin makes, with the
axis of branch, an angle of about 40°.

Remarks. — The nema of the mature forms is about 7- 0 mm.
in length, and is often split, giving the appearance of a double
nema, bifurcating at different distances from the sicula. In one
instance the strands of the nema are twisted around each other
below the sicula, but reunite and apparently form a single tube
at a still lower level. Except as regards size, the vertically com-
pressed polypary bears a considerable resemblance to X. dichoto-
mous Emmons It differs, however, from that species in the
angles of bifurcation and the details of its thecae. In the cyathi-
form aspect the typical nema is readily recognised.

Associates . — Dictyonema scitulum, sp. now, D. cmipanulatum r
sp. nov., and Crustaceae.

Genus Dictyonema Eichwald.

Dictyonema campanulatum, sp. nov.

(Plate X., Figs. 6-13.)

Polypary cyathiform, flabelliformly compressed in mature
specimens, attaining a length of approximately 1*2 cm. and a
width of 1*5 cm. Branches irregularly disposed, somewhat
flexuous; outside branches convex to the' axis of the polypary
proximally, approximately straight distally, inside branches
flexuous throughout. Bifurcations fairly frequent. Branches
from 0 3 to 0-4 mm. wide of increasing width, 10 with interspaces
in a width of 10 mm., space between the branches more than the
width of the branches. Stout transverse dissepiments P0 mm.
to 2 0 mm. apart, which, with the adjacent branches enclose an
irregularly shaped interspace.

Thecae 12-14 in 10 mm., acutely dentiform.

Sicula about 0-7 mm. long with long attenuated nema.

Remarks. — The material on which this description is based can-
not be regarded as ideal. Nevertheless there is little doubt regard-
ing the distinctness of D campanulatum from any other form
known to us.

Some specimens (PI. IX , Figs. 7, 8, 10, 12, 13) show curious
double or triple nemas, hair-like filaments, one of which occasion-
ally ends in a small triangular body suggestive of a peduncular
attachment.

Associates . — Staurograptus diffissus , sp. nov., and D. scitulum,.
sp, nov. Crustaceae.

The Staurograptus Bed of Victoria.

Dictyonema scitulum, sp. 110V.

(Plate IX., Figs. 14-19.)

Polypary cyathiform, flabelliformly compressed, in mature
specimens attaining a length of 2 cm., a width distally of about
2 cm. (included in an angle of 85°).

Branches nearly parallel, regularly disposed, outer ones slightly
concave to axis of polvpary proximally, and straight distally, inner
ones straight throughout. Bifurcations infrequent. Branches
0-4 mm. (04-0-5 mm.) wide, of constant width, 13-14 occupying
(with interspaces) a width of 10 mm. The spaces between the
branches is somewhat less than the width of the branches. Com-
paratively stout transverse dissepiments, from 0 -7 mm. to 1-5
mm. wide, connect the branches and these with the branches en-
close a subrectangular interspace. Thecae 14-17 in 10 mm. dis-
tally. Thecal apertures thickened and acutely dentiform.

Sicula 1 mm. long.

Remarks. — The type specimen, although preserved as a film,
exhibits some of the characteristics revealed by Wiman (1),
Bulman (2) and others in their work of isolation of specimens in
relief from matrices with dilute acids.

Two types of thecae may be recognised, the thecae and gon-
angia ” of Wiman (T). The latter arise from opposite sides of
the former and throughout their short length appear to be dis-
posed in alternately right and left hand spirals, their apertures
being opposed. The apertures are visibly thickened. An attempt
was made to trace some plan of arrangement of the cell groups
about the branches, but, other than that indicated, unsuccessfully.

The dissepiments are straight bars connecting adjacent
branches and show no evidence of fusion midway. An apertural
process, very similar to that described by Ruedemann in regard
to the thecae of D. furcifernm (3, p. 607), extends from the flat-
tened aperture of the “ gonangium ” and impinges on the dorsal
part of the adjoining branch.

There is little doubt that D. scitiditm , sp. now is closely related
to D. furcifernm , but unfortunately the thecae of the type speci-
men are not clearly enough shown to ascertain whether the differ-
ence is varietal or specific. On the other hand, Ruedemann (3,
pi. iii., f. 11) has only figured a portion of a polypary, and until
better material is forthcoming, it has been thought desirable, on
account of its stratigraphical importance in Victoria, to give
D. scit ulum specific rank.

Associates . — Staurograptus diffissus, sp. now and D. cam pa -
nnlatum , sp. now Crustaceae.

Correlation of Fauna.

The importance of the Springfield association to the Victorian
Ordovician sequence lies in the facts that it is the oldest graptolite

Harris and Keble:

fauna yet discovered in Australia, and is comparable with the
oldest but one of the graptolite associations of America and
Europe. The graptolite succession is generally alike in all parts
of the world and the forms described in this contribution are so
closely related to those found in similar associations elsewhere
that there is little doubt that the Springfield slates are very near
to the base of the Ordovician. Making world-wide comparisons,
stratigraphically above them should occur a fauna equivalent to
that of the American Deep Kill Zone III, containing Clonograptus
flex il is and Tetragraptus (4. p. 130) ; such a position and associa-
tion is held by the Taylor's Quarry slates east of Lancefield
(5, P . 175).

If conditions were suitable to its preservation and it is acces-
sible, a bed containing exclusively a Dictyonema allied to D. f label-
lifonne Etch, should yet be found in Victoria stratigraphically
below the Springfield slates. This bed in other parts was for-
merly regarded as marking the closing stage of the Cambrian,
but latterly both in American and Europe, it has been recognised
as introducing the extensive Ordovician transgression. Such is
probably the case in Victoria, for stratigraphically above the
Springfield slates we have a very comprehensive suite of Lower
Ordovician graptolites which have been zoned and subzoned, while,
apparently, stratigraphically below them a little east of their strike
some distance north we have the Cambrian Dinesus trilobite
fauna. It is probably in this direction that the missing bed will
be found.

Bibliography.

1. C. Wiman. Ueber die Graptoliten. Bull. Geol. Inst. Upsala ,

ii., Art. No. 6, 1895.

2. O. M. B. Bulman. Notes on the Structure of an Early

Dictyonema. Geol. Mag., lxii. (728).

3. R. Ruedemann. Graptolites of New York. N.Y. State

Mus. Mem . 7, 1904.

4. R. Ruedemann. Paleontologic Contributions from the

New York State. Mus. and Sci. Dept. Bulls. 227 , 228 r
1919.

5. T. S. Hall. The Graptolites of the Lancefield Beds. Proc.

Roy. Soc. Vic., n s., xi. (2), 1899.

EXPLANATION OF PLATE IX.

All Figures X 2-6.

Figs. 1-5. — Staurograptus diffissus , sp. nov.

1. Young polypary showing sicula and netna. Paratype*

2. Polypary vertically compressed. Paratype.

3. Polvpary laterally compressed. Paratype.

4. Polypary vertically compressed. Paratype*

5. Polypary vertically compressed. Holotype.

Proc. R.S. Victoria, 1928. Plate IX.

The Staurograptus Bed of Victoria.

Figs. 6-13 . — Dictyonema campanulatiim, sp. nov.

6. Young polypary, with single nema. Paratype.

7. Young polypary with divided nema. Holotype.

8. Proximal portion of polypary, showing divided nema.

Paratype.

9. Polypary showing typical form of polypary. Paratype.

10. Polypary, distorted, showing divided and attachment

suggestive of peduncular appendage. Paratype.

11. Portion of polvpary showing dissepiments. Paratype.

12. Proximal portion of polypary showing divided nema.

Paratype.

13. Complete polypary with tripartite nema. Paratype.

Figs. 14-19. — D . scituhm, sp. nov.

14. Portion of polypary showing interspaces and dissepi-

ments. Paratype.

15. Imperfect polypary showing interspaces. Paratype.

16. Portion of mature polypary showing transverse dissepi-
ments. Paratype.

17. Distorted polypary. Paratype.

18. Polypary. Paratype.

19. Complete polypary. Holotype.

1 927 .

A tl-Oll :

HIS EXCELLENCY THE RIGHT HON. BARON SOMERS, K.C.M.G., D.S.O., M.C.

{h'Cdtbent :

Prof. W. E. AGAR, M. A., D.Sc., F.R.S.

^ice-Jprcdibcnts :

K. CHAPMAN, A.Ii.S.

D. K. PICKEN, M. A.

Jljon. ^ t ea sure v :

N. A. ESSERMAN, B.Sc., A. Inst. P.

gi) 011. Ji tinman :

F. A. CUDMORE.

$jon. £ecretarg :

F. L. STILLWELL, D.Sc. (resigned).
R. T. PATTON, B.Sc., D.I.C., M.F.

3i)on. -Secretary), ^lailumntinil anb jLlhtjssical Section :

E. KIDSON, O.B.E., D.Sc., F.Inst.P. (resigned).

J. S. ROGERS, B.A., M.Sc.

Council :

Prof. E. W. SKEATS, D.Sc., A ll.C.Sc.,

F.G.S.

Prof. A. J. EWART. D.Sc., Pll.D., F.R.S. ,
F.L.S.

Assoc. Prof. H. S. SUMMERS, D.Sc.
Prof. T. H. LABY, M.A., Ph.D., Sc.D.,
F.Inst.P.

J. M. BALDWIN, D.Sc.

Catt. J. K. DAVIS.

Prof. W. A. OSBORNE, M.B., B.Cil.,
D.Sc.

E. J. DUNN, F.G.S.

J. SHEPHARD.

W. RUSSELL GRIMWADE, B.Sc.

W. IIEBER GREEN, D.Sc.

Dr. A. C. D. RIVETT, M.A., D.Sc.

Committees of tlje Council

publication (Committee :

THE PRESIDENT.

THE HON. TREASURER.
THE HON. SECRETARY.

&}onovarj> JhiMtor* .*

C. A. LAMBERT.

J. SHEPHARD.

Honorary) Architect :

\V. A. M. BLACKETT.

TErustce* :

PROF. SIR W. BALDWIN SPENCER, K.C.M.G., F.R.S.
J. A. KERSHAW, F.E.S

1928.

LIST OF MEMBERS

WITH THEIR YEAR OF JOINING.

[Members and Associates are requested to send immediate notiee of any
change of address to the Hon. Secretary.]

Patron.

His Excellency. The Right Hon. Baron Somers, K.C.M.G.,,
D.S.O., M.C.

Life Members.

Fowler, Thos. Walker. M.C.E., “ Fernhill ” 8 Fitzwilliam- 1879
street Kew, E.4.

Gregory, Prof. J. W., D.Sc., F.R.S., F.G.S., University, 1900
Glasgow.

Love, E, F. J., M.A , D.Sc., F.R.A S„ The Grove. Coburg, 1888
N.13.

Selby, G. W., Glenbrook-avenue, Malvern, S.E.5 1889

Ordinary Members.

Agar, Prof. W. E., M.A., D.Sc., F.R.S., University, Carl- 1920
ton, N.3.

Anderson, George, M.A., LL.M., B.Com., 222 Beacons- 1924
field-parade, Middle Park, S.C.6.

Aston, R. L., B.E., M.Sc., Trinity College, Parkville, N.3 1927

Austin, E. G., Boeri Yallock, Skipton 1922

Baker, Thomas, Bond-street, Abbotsford, N.9 1889

Baldwin, ]. M., ALA., D.Sc., Observatorv, South Yarra, 1915
S.E.l.

Bale, W. M.. F.R.M.S., 83 Walpole-street, Kew, E.4 .... 1887

Balfour, Lewis J.. B.A., M.B., B.S., Burwood-road, Haw- 1892
thorn, E.3.

Baragwanath, W.. Geological Survey Department, Trea- 1922
sury Gardens, Melbourne, C.2.

Barrett, A. O., 25 Orrong-road, Armadale, S.E.3 1908

Barrett, Sir J. W., K.B.E., C.M.G., M.D., M.S„ 105 Col- 1910
lins-street, Melbourne, C.l.

Brittlebank, C. C., 48 York-street, Caulfield, S.E.8 .... 1898

Chapman, F., A.L.S., F.R.M.S., F.G.S., National 1902

Museum, Melbourne, C.L

Cudmore, F. A., 12 Valley View-road, East Malvern, S.E.6 1920

100

List of Members.

Davis, Captain John King, “ St. Carols,’’ Caroline-street, 1920
South Yarra, S.E.l.

Dunn, E. F.G.S., “ Roseneath,” Pakington-street, Kew, 1893
E.4.

Dyason, E. C., B.Sc.. B.M.E., 92 Oueen-street, Melbourne, 1913
C.l.

Elliott, R. D., 395 Collins-street, Melbourne, C.l 1927

Esserman, N. A., B.Sc., A. Inst. P., Research Laboratories, 1925
Maribyrnong, W.3.

Ewart, Prof. A. J., D.Sc., Ph.D., F.R.S., F.L.S., Univer- 1906
sity, Carlton, N.3.

Gault, E. L., M.A., M.B., B.S., 4 Collins-street, Mel- 1899
bourne, C.l.

Gepp, H. VC, Greensborough-road, Heidelberg, N.22 .. 1926

Gilruth. J. A., D.V.Sc., M.R.C.V.S., F.R.S.E., 7 Clowes- 1909
street, South Yarra, S.E.L

Green, W, Heber. D.Sc., University, Carlton, N.3 1896

Greenwood, Prof. J. X., D.Sc.. University, Carlton, N.3 . . 1927

Grimwade, W. Russell, B.Sc., 420 Flinders-lane, Mel- 1912
bourne, C.l,

Herman, H., D.Sc., B.C.E., M.M.E., F.G.S., u Albany,” 8 1897
Redan-street, St. Kilda, S.2.

Hills, Loftus, D.Sc., “Mount Royal,” Upper Ferntree 1925
Gully.

Hurst, W. W., B.Sc., Ph.D., Urquhart-street, Hawthorn, 1927
E.2.

Janssens, Eugene, B.Sc., 2 Argyle-street, St. Kilda, S.2 . . 1923

Kelly, Bowes, Glenferrie-road, Malvern. S.E.4 1919

Kenyon, A. S., C.E., Lower Plentv-road, Heidelberg, N.22 1901

Kernot, Assoc. Prof. W. N., B C E M.Mech.E , M Inst.C.E., 1906
University, Carlton, N.3.

Kershaw, J. A., F.E.S., National Museum, Melbourne, 1900

C. l.

Laby, Prof. T. H.. M.A., Ph.D., Sc.D., F.Inst.P., Univer- 1915
sity, Carlton, N.3.

Lewis, J. M., D.D.Sc., “Whitethorns,” Boundarv-road, 1921
Bur wood, E.13.

Littlejohn, \\'. S., M.A., Scotch College, Hawthorn, E.2 . . 1920

Llewelyn, Miss Sybil, M.A., M.Sc., “ Cleveden,” 34 Barklv- 1924
street, St. Kilda, S.2.

Lyle, Prof. Sir 1 hos. R., M.A., D.Sc., F.R.S., Irving-road, 1889
Toorak, S.E.2.

MacCallum, Prof. Peter, M.C., M.A., M.Sc., M.B., Ch.B., 1925

D. P.H., University, Carlton, N.3.

Mahony, D. J., M.Sc., “ Lister House,” Collins-street, Mel- 1904
bourne, C.l.

List of Members.

101

Mann, S. F., Caramut, Victoria 1922’

Masson, Prof. Sir David Orme, K.B.E., M.A., D.Sc., 1887
F.R.S.E., F.R.S., 14 William-street, Sth. Yarra,

S.E.l.

McCallum, Dr. Gavin, 127 Collins-street, Melbourne, C.l 1925
Merfield, C. J., F.R.A.S., Observatory, South Yarra, S.E.l 1913
Merfield, Z. A., F.R.A.S., University, Carlton, N.3 . . . . 1923

Michell, Prof. j. H., M.A., F.R.S., 52 Prospect Hill-road, 1900'
Camberwell. E.6.

Milieu, Senator J. D., 90 William-street, Melbourne, C.l 1920
Miller, Leo F., “ Moonga,” Power-avenue, Malvern, S.E.4 1920
Miller, E. Studlev, 396 Flinders-lane, Melbourne, C.l .. 1921

Monash, Lieut. -General Sir John, G.C.M.G., K.C.B., Doc. 1913
Eng., LL.D., State Electricity Commission, 22
William-street, Melbourne, C.L

Mullett, H. A., B.Ag.Sc., Dept, of Agriculture, Melbourne, 1923
C.2.

Osborne, Prof. W. A., M.B., B.Ch., D.Sc., University, 1910
Carlton, N.3.

Patton, R. T.. B.Sc., M.F., Hartley-ave., Caulfield, S.E.8 1922
Payne, Prof. H., M.Inst.C.E., M.I.MechE., University, 1910
Carlton, N.3.

Penfold, Dr. W. J., M.B., Alfred Hospital, Commercial- 1923

road, Prahran, S.l.

Petrie, A. H. K., B.Sc., University, Carlton, N.3 1925

Picken, D. K., M.A., Ormond College, Parkville, N.3 .... 1916

Piesse, E. L., 43 Sackville-street, Kew, E.4 1921

Pratt, Ambrose, M.A., 376 Flinders-lane. Melbourne, C.l 1918

Quayle, E. T., B.A., 27 Collins-street, Essendon, W.5 . . 1920

Rae, F. J., B.Sc., B.Ag.Sc.. Botanic Gardens, South Yarra, 1927
S.E.L

Reid, J. S., 498 Punt-road, South Yarra, S.E.l 1924

Rivett, Dr. A. C. D.. M.A., D.Sc., Council for Scientific and 1911
Industrial Research, 314 Albert-street, East Mel-
bourne, C.2.

Rogers, J. Stanley, U.A.. M.Sc., University, Carlton, N.3 . 1924

Ryan, Rev. Wilfrid, S.J.. M.A., F.G.S., Newman College, 1926
Carlton, N.3.

Schlapp, H. H., 31 Queen-street, Melbourne, C.l 1906-

Shephard, John, “ Norwood,” South-road, Brighton Beach, 1894
S.5.

Shillinglaw, Godfrey V., 64 Dandenong-road, Caulfield, 1925
S.E.7.

Singleton, F. A., M.Sc., University, Carlton, N.3 1927

Skeats, Prof. E. W., D.Sc., A.R.C.Sc ,F.G.S., University, 1905
Carlton, N.3.

102

List of Members.

Smith, B. A , M.C.E., Mutual Building, 395 Collins-
street, Melbourne, C.l.

Spencer, Prof. Sir W. Baldwin, K.C.M.G., M.A., D.Sc.,
F.R.S., National Museum, Melbourne, C.l.

Stillwell, F. L., D.Sc., 44 Elphin- grove, Hawthorn, E.2 . .

Summers, Associate Prof. IF S., D.Sc., University, Carl-
ton, N.3.

Sweet, Georgina, D.Sc., Cliveden Mansions, Wellington-
parade, East Melbourne, C.2.

Tbirkell, Geo. Lancelot, B.Sc., 4 Grace-street, Malvern,
S.E.4.

Thomas, Dr. D. J., M.D., 12 Collins-street, Melbourne,
C.l.

Tiegs, O. W., D.Sc., University, Carlton, N.3

Trinder, E. E., M.I.H.V.E., “ Ruzilma,” Orrong -grove,
Caulfield, S.E.7.

Wadham, Prof. S. M., M.A., Agr.Dip., University, Carl-
' ton. N.3.

Walcott, R. H., Technological Museum, Melbourne, C.l . .

Weber, E. K.. 49 Armadale-street, Armadale, S.E.3 ....

\\ ickens, C. H., F.I.A., F.S.S., Commonwealth Statistician,
315 Post Office Place, Melbourne, C.l.

Woodruff, Prof. H. A., M.R.C.S., L.R.C.P., M.R.C.V.S.,
Veterinary School, Parkyille, N.2.

Young, Assoc. Prof. W. J., D.Sc., University, Carlton, N.3

1924

1887

1927

1902

1906

1922

1924

1925

1922

1927

1897

1927

1923

1913

1923

Country Members.

Caddy, Dr. Arnold, “ Chandpara,” Tylden, Vic 1924

Crawford, W., Gisborne, Vic 1920

Drevermann, A. C., Dookie Agricultural College, Dookie, 1914
Vic.

Easton, J. G., “ Kiewa,” Murphy-street, Bairnsdale, Vic. 1913

Harris, W. J., B.A., High School, Echuca, Vic 1914

Hart, T. S., M.A., B.C.E., School of Mines, Bairnsdale, 1894
Vic.

Hope, G. B., B.M.E., 14 Carrical,” Hermitage-road, New- 1918
town, Geelong, Vic.

Janies, A., B.A., M.Sc., High School, Colac, Vic 1917

Kitson, Sir Albert E., C.M.G., C.B.E., F.G.S.. 29 Alfred- 1894
place, Kensington, London, S.W.7, England.

Langford, W. G., M.Sc., B.M.E., 44 Vailala,” Elizabeth- 1918
street, Gordon, Sydney, N.S.W.

Lea, A. M., F.E.S., 241 Young-street, N. Unley, S. Aus. 1909

103

List of Members .

Mackenzie, H. P., Engr. Commr. R.N.(Ret), Trawalla, 1924
Vic.

Oliver, C. E., M.C.E., c/o J. E. Minifie, 12' Martin-street, 1878

* El wood, S.3.

Parker, L. C., B.Sc., High School, Ballarat 1927

Sutton, T. W., 127 Doncaster-avenue, Kensington, Sydney, 1924
N.S.W.

Trebilcock, Captain R. E., M.C., Wellington-street, Kerang, 1921
Vic.

White, R. A., B.Sc., School of Mines, Bendigo, Vic 1918

Corresponding Member.

Lucas, A. H. S., M.A., B.Sc., Sydney Grammar School, 1895
Sydney, N.S.W.

Associates.

Albiston, H. E., M.V.Sc., Veterinary School, Parkville, 1925
N.2,

Allen, J. M., B.A., 41 -Nirvana-avenue, East Malvern, 1924
S.E.5.

Allen, Miss N. C. B., B.Sc., University, Carlton, N.3 .... 1918

Archer, Howard R., B.Sc., c/o J. M. Moffatt, Faulkner- 1921
street, Armidale, N.S.W.

Ashton, H., ‘'The Sun,” Castlereagh-street, Sydney, 1911
N.S.W.

Bage, Mrs. Edward, “ Cranford,” 7 Gellibrand-street, 1906
Kew, E.4.

Bage, Miss F., M.Sc., Women’s College, Kangaroo Point, 1906
Brisbane, Old.

Baker, F. H., 167 Hoddle-street, Richmond, E.l 1911

Barkley, H., Meteorological Bureau, Melbourne, N.3. . . 1910

Barnard, R. J. A., M.A., University, Carlton, N.3 1926

Bordeaux, E. F. J., G.M.V.C., B. es. L„ Mangalore-street, 1913
Flemington, W.2.

Breidahl, H., M.Sc., M.B., B.S., 23 Chatsworth-avenue, 1911
North Brighton, S.5.

Brodribb, N. K. S., Ordnance Factories, Maribyrnong, 1911
W.3.

Brookes, Leslie R., B.A., 3 Fern-avenue. Windsor, S.l. . . 1922

Brvce, Miss L. M.. B.Sc., 22 Victoria-avenue, Canterbury, 191S
E.7.

Buchanan, G., D.Sc., University, Carlton, N.3 1921

Carter, A. A. C., “ Fairholm,” Threadneedle-street, Bal- 1927
wyn, E.8.

104

List of Members.

Chapman. W. D., M.C.E., “ Hellas,” Heidelberg-road, 1927
Clifton Hill, N.8.

Collins, A. C., Public Works Department, Treasury Gar- 1928
dens, Melbourne, C.2. *.

Chappie, Rev. E. If., The Manse, Warrigal-road, Oakleigh, 1919
S.E.12.

Clinton, H. F., Produce Office, 605 Flindcrs-street, Mel- 1920
bourne, C.l.

Cook, G. A., M.Sc.. B.M.E., 18 Elphin-grove, Hawthorn, 1919
E.2.

Cookson, Miss I. C., B.Sc., 154 Power-street, Hawthorn, 1916
E.2.

Coulson, A. L., M.Sc., D.I.C.. F.G.S., “ Finchley,” King- 1919
street, Elsternwick, S. 1.

Cox, E. H., Literary Staff, “ The Argus,” Elizabeth-street, 1924
Melbourne, C.l.

Crespin, Miss I., B.A., 67 Studley Park- road, Ivew, E.4 . . 1919

Banks, Sir Aaron T., 391 Bourke-street, Melbourne, C.l. 1883

Dare, J. H., B.Sc., State School, Brunswick, N.10 . . . . 1925

Deane. Cedric, “ Cloyne.” State-street, Malvern, S.E.4 . . 1923

Feely, J. A., Observatory, South Yarra, S.E.l 1924

Fenner, C, D.Sc., Education Department, Flinders-street, 1913
Adelaide, S.A.

Ferguson, W. H., 37 Brinsley-road, E. Camberwell, E.6 . . 1894

Finney, J. M., 36 Toorak-road, Malvern, S.E.4 1925

Flecker, Dr. H., 71 Collins-street, Melbourne, C.l 1922

Gabriel, C. J., 293 Victoria-street, Abbotsford, N.9 . . . . 1908

Hardy, A. D., F.L.S., Forests Department, Melbourne, 1903
C.2.

Hartung, Prof. E. J., D.Sc., University, Carlton, N.3. 1923

Hauser, H. B., M.Sc., Geology School, University, Carl- 1919
ton, N.3.

Hercits, E. O., M.Sc., A. Inst. P., University, Carlton, N.3 1923

Heslop, G. G., D.V.Sc.., 7 Hudson-street, Caulfield, S.E.7. 1923

Hill, Gerald F., Council for Scientific and Industrial Re- 1924
search, 314 Albert-street, East Melbourne, C.2.

Hills, E. S., B.Sc., Geology School, University, Carlton, 1928
N.3.

Holmes, W. M., M.A., B.Sc., Observatory, South Yarra, 1913
S.E.l.

Horning, Eric, Newman College, Carlton, N.3 1924

Howitt, A. M., Department of Mines, Melbourne, C.2 ... 1910

Jack, A. K., M.Sc., 49 Aroona-road, Caulfield, S.E.7 .... 1913

Jessep, A. W., B.Sc., M.Ag.Sc., Dip. Ed., Horticultural 1927
Gardens, Burnley, E.l.

List of Members.

105

Jona, J. Leon. M.D., B.S., D.Sc., “ Hazelmere,” 104 1914

Wattle Tree-road, Malvern, S.E.4.

Jones, Miss K. A. Gilman, Church of England Girls' Gram- 1922
mar School, Anderson- street, S. Yarra. S.E.l.

Jutson, J. T., B.Sc., LL.B., “ Darlington,” 9 Ivanhoe-par- 1902
ade, Ivanhoe, N.21.

Kannuluik, W. G., M.Sc., Natural Philosophy Dept., Uni- 1927
versity, Carlton, N.3.

Keartland, Miss B., M.Sc.. Cramer-street, Preston, N.18 1919

Keble, R. A., National Museum, Melbourne, C.l 1911

Lambert, C. A., Bank of N.S.W., Melbourne, C.l 1919

Leslie, J. R., 99 Toorak-road, South Yarra. S.E.l .. .. 1923

Lewis, Miss R. M., 52 Campbell-road, East Kew, E.4 .... 1925

Long, Miss M. E., Physiology School, University, Carlton, 1924
N.3.

Lilly, W. H., Department of Lands, Public Offices, Mel- 1896
bourne, C.2. *

Macdonald, B. E.. Metorological Offices, Melbourne, C.l. 1920

Mackenzie, G., 1 High-street, Prahran, S.l 1907

Maclean, C. W., 56 Cole-street, Elsternwick, S.4 1879

Mclnerny, Miss K., M.Sc., University, Carlton, N.3. . . 1918

McLennan, Ethel, D.Sc., University, Carlton, N. 3. .. 1915

Melhuish, T. D'A., M.Sc., Adelaide Chemical and Fertilizer 1919
Co., Currie- street. Adelaide

Mollison, Miss E., M.Sc., Roval-crescent, Camberwell, 1915
E.6.

Moore, F. E.. O.B.E., Chief Electrical Engineer's Branch, 1920
P.M.G/s Department, Treasury Gardens, Mel-
bourne, C.2.

Morris, P. F.. National Herbarium, South Yarra, S.E.l. . . 1922

Nelson, Miss E. A.. M.A., M.Sc., University, Carlton, N.3 1924
Newman, B. W., Meteorological Offices, Melbourne, C.l 1927
Nicholson, Miss Margaret G., 59 Murray-street, Elstern- 1920
wick, S.4.

Oke, C., 56 Chaucer-street. St. Kilda. S.2 1922

Orr, D.. B.Sc.. 860 Mount Alexander-road, Essendon, W.5 1927

Parr, W. J., 17 Bokhara-road, Caulfield, S.E.8 1927

Pern, Dr.* Sydney, M.R.C.S., L.R.C.P., 16 Collins-street, 1920
Melbourne, C.l.

Petersen, Miss K., B.Sc., 56 Berkeley-street, Hawthorn, 1919
E.2.

Pretty, R. B., M.Sc., Technical School, Wonthaggi, Vic. 1922

Raff, Miss J. W., M.Sc., F.E.S., University, Carlton, N.3 1910

Richardson, Sidney C.. 2 Geelong-road, Footscra}?’. W.ll 1923
C.l.

List of Members .

106

Rosenthal, Newman H., B.A., B.Sc., 49 Odessa-street, St. 1921
Kilda, S.2.

Ross, Miss D. J., M.Sc., Merton Hall, Anderson-street, 1924
South Yarra, S.E.l.

Salier, D. G., B.Sc., Queen’s College, Carlton, N .3 1924

Sayce, E. L., B.Sc., Research Laboratories, Maribyrnong, 1924
W.3.

Sharman, P. )., M.Sc., “ Glenalvie,” 9 Daphne-street, Can- 1916
terbury, E.7.

Shearer, J., B.Sc., Queen’s College, Carlton, N.3 1924

Shiels, D. O., M.Sc., Ph.D., Chemistry School, Univer- 1927
sity, Carlton, N.3.

Showers, Allan F.. B.Sc., Brewster-street, Essendon, W.5 1922

Simpson, H. \V. C., 64 Dandenong-road, Caulfield North, 1927

S.E.7.

Smith, J. A., 25 Collins-place, Melbourne, C.l 1905

Stickland, John, 433 Brunswick-street, Fitzroy, N.6 1922

Stillman, Miss E. G., B.Sc., “ Taiyuan,” 5 Grange-road, 1919
East Kew, E.4.

Sutton, C. S., M.B., B.S., Education Department, Mel- 1908
bourne, C.2.

Thomas, R. G., B.Ag.Sc., C/o Dr. Thomas, Northam, W.A. 1922
Thompson, Mrs. G. R., 26 Fawkner-street, St. Kilda, S.2 1922

Thorn, Wm., 37 Chrystobel-crescent, Hawthorn, E.2 .... 1907

Traill, J. C., B.A., B.C.E., 630 St. Kilda-road, Melbourne, 1903
S.C.3.

Treloar, H. M., Meteorological Offices. Melbourne, Cl .. 1922

Trii dinger, W., 27 Gerald-street, Murrumbeena, S.E.9 . . 1918

Turner, A. H., M.Sc., Natural Philosophy Dept., Univer- 1927
sity, Carlton, N.3.

Turner, A, W., M.V.Sc., Veterinary School, Parkville, 1925
N.2.

Wilcock, E. L., B.Sc., Dookie College, Dookie, Vic 1925

Williamson. H. B., F.L.S., “The Grange,” 231 Waverley- 1919
road, East Malvern, S.E.5.

Wilson, F. E., F.E.S., 22 Ferncroft-a venue, E. Malvern, 1921
S.E.5.

Wilson, Major PI. W., O.B.E., M.C., C.deG., B.Sc., 630 1923
Inkerman-road, Caulfield, S.E.7.

Withers, R. B., 10 Nicholson-street, Coburg, N.13 1926

Woodward, J. H., Queen’s Buildings, 1 Rathdown street, 1903
Carlton, N.3.

INDEX.

The names of new genera and species are printed in italics.

Actinostrobus, 13 ; pyramidalis, 13.

Agathis, 5 ; microstachys, 7 ; palm-
erstoni, 8 ; robusta, 7.

Anatomy of Australian Coniferous
Timbers, 1.

Araucaria, 5 ; Bidwilli, 8 ; Cun-
ningham i, 8.

Athrotaxis, 9; cupressoides, 9;
laxifolia, 9 ; selaginoides, 9.

Augite, 6G.

Australia, Field Trials in, 70 ;
Flora of, 81.

Australian Coniferous Timbers, 1.

Basalt from Daylesford, 34.

Callitris, 10 ; arenosa, 12 ; Baileyi,
12 ; calcarata, 12 ; glauca, 15 ;
gracilis, 12 ; intra tropica, 12 ;
Macleayana, 13 : Muelleri, 12 ;
oblonga, 12 ; rhomboidea, 12 ;
robusta, 11.

Coastal Physiography of Port
Campbell, 45.

Coniferous Timbers. Australian, 1.

Contributions to Flora of Austra-
lia, 81.

Corinella Dyke, 25 ; Relation to
Stony Creek Basin, 29.

Dacrydium. 2 ; Franklinii, 2.

Daylesford, Basalt from, 34.

Dictyonema, 92 ; campanulatum ,
92 ; scitulum, 93.

Diselma, 13; Archeri, 13.

Error of Field Trials, 70.

Ewart, Alfred J., 81.

Experimental Error of Field
Trials, 70.

Felspars, 02.

Field Trials, Experimental Error,
70.

Fitzroya Archeri, 15.

Flora of Australia, 81 ; Northern
Territory, 81.

Forster, H. C., 70.

Fossil Plants of the Stony Creek
Basin, 88.

Harris, W. J., 91.

Hornblende, 66.

Jarrett, Phyllis H., SI.

Jutson, J. T., 45.

Keble, R. A., 91.

List of Members, 98

Low Latitudes, Vitality of White
Races in, 17.

Magma, 08.

Mahony, D. J., 02.

Members. List of, 98

Microcachrys, 2 ; tetragona, 2.

Minerals, Volcanic, 62.

Northern Territory, Flora of, 81.

Olivine, 67.

Olivine Anorthoclase Basalt from
Daylesford, 34 ; Chemical Char-
acter, 37 ; Comparison with
Newer Basalts, 40 ; Distribution,
36 ; Origin , 42 : Petrological
Character, 38.

Orr, D., 25, 34.

Patton, II. T., 1, 88.

Pherosphaera, 1 ; Fitzgeraldi, 1 ;
Ilookeriana, 2, 15.

Phylloeladus, 4 : rhomboklalis, 4.

Physiography. Coastal, 45.

Podocarpus, 3 ; alpina, 4 ; data,
4 ; ladei, 4 ; pedunculata, 4 ;
spinulosa, 4.

Port Campbell, Coastal Physio-
graphy of, 45 ; Physiographic
Features of Coast Line, 47 ; Re-
clamation of Land from Sea, 51 ;
Stage of Development of Coast,
54.

Pultenaea accrosa, 59 ; angnsfi-
folia , 57 ; angustifoUa var.

ri&cosa, 5S : D' Alton! i, 61; folio-
losa, 58 ; graven lens, 59 : bibber-
tioides var. prostrata, 58 : Kenn-
yi, 58 ; microphylla var. cuneata,
58 : mollis, 57 ; p at elli folia, 60 ;
Revision of the Genus, 57 ;
8 '1 uartiana, 59.

Revision of the Genus Pultenaea,
57.

Scorpia simplicifolia, 85.

Sideria reverta, 85.

Stau rograptus, 91 ; cliffissus, 91.

Staurograptus Bed of Victoria,
91 ; Correlation of Fauna, 93.

Stony Creek Basin, 25, 88 ; Nature,
26 ; Origin, 28.

Tertiary Volcanic Minerals, 62.

Timbers, Australian Coniferous, 1.

Vasey. A. J., 70.

Victoria, Staurograptus Bed of,
91.

Vitality of White Races, 17.

Volcanic Minerals, 62.

White Races, Vitality of, 17.

Wiekens, C. H., 17.

Williamson, H. B., 57.

END OF VOLUME XL.

[Part II. Published 14th June, 1928. J

5 a

NGS

fixtoria.

VOL. XLI. (New Series).

PARTS I. AND II.

Edited under the Authority of the Council.

PROCEEDI

OF THE

opl ^orixtl) of

ISSUED 27th SEPTEMBER , IQ28 and 11th APRIL , ig2g.

( Coutninnt # Papers read before the Society during the months of
September to December , ig2g).

THE AUTHORS OK THK 8KVKKAL TAKERS ARK INDIVIDUALLY RESPONSIBLE FOR THK
SOUNDNESS OF THK OPINIONS GIVKN AND FOR TJ1K ACCURACY OF THK
STATEMENTS MADE THEREIN

ROYAL SOCIETY’S HALL,

VICTORIA STREET, MELBOURNE, 0. 1.

FORD & SON PRESS PTY. LTD., DRUMMOND ST., CARLTON, MELBOURNE.

1929 .

' :

CONTENTS OF VOLUME XLI.

Part I.

PAGE

Art. I. — Notes on New Zealand Hydroida. By R. E. Treb-

ilcock. Plates (I. -VII.) ... ... ... l

II. — The Diurnal and Annual Fluctuations of Tempera-
ture in the Interior of a large Tree. By A. O.
Barrett. ... ... ... .. ... 32

III— Some Trematode Parasites on the Gills of Victorian
Fishes. By Winifred Kent Hughes, B.Sc.

(Plates VIII.— XI.) 45

IV. — Note on the Reflection of X-rays from Glass and
Quartz. By R. Bingham, M.Sc., T. H. Laby,

Sc. D., and J. Shearer, M.Sc. , t . ... 55

V. — Contributions to the Flora of Australia, No. 35.

The Naturalized Aliens of Victoria. By Alfred
J. Ewart, D.Se., Ph.D., F.L.S., F.R.S. ... 59

Part II.

VI. — On Grooved, Pitted and Miniature Pedestal Rocks
at Lake Goongarrie, Western Australia. By
J. T. Jutson, B.Sc., LL.B. (Plates XII., XIII.) 63

VII. — Notes on Australian Termites (Isoptera). Descrip-
tions of new species. By Gerald F. Hill.

(Plate XIV.) ... ... ... ... 85

VIII. — The Devonian and Older Palaeozoic Rocks of the
Tabberabbera District, North Gippsland, Vic-
toria. By Professor Ernest W. Skeats, D.Sc.,
A.R.C.S., F.G.S. (Plate XV.) 97

IX. — The Building Stones of Victoria, Part II. The
Igneous Rocks. By Kathleen McInerny, M.Sc.

(Plate XVI.) 121

X. — Longe Range Rainfall Forecasting from Tropical

(Darwin) Air Pressures. By E. T. Quayle, B.A. ICO

XI. — Solitary Waves at the Common Boundary of two

Liquids. By Frances E. Allan, M.A. ... 1G5

XII. — The Geology and Palaeontography of the Cathedral
Range and the Blue Hills, in North-Western
Gippsland. By Edwin S. Hills, B.Sc. (Plates
XVII., XVIII.) 176

PAGE

Art. XIII. — On the Flanged Cowry, Palliocypraea gastroplar.

By Frederick Chapman, A.L.S., F.G.S., etc.
(Plates XIX., XX.) 202

XIV. — On some Trilobites and Brachiopods from the
Mount Isa District, N.W. Queensland. By
Frederick Chapman. A.L.S., F.G.S. (Plates
XXI., XXII.) 206

XV. — On a New Species of Capulus found attached to a
Pterygotus Carapace. By Frederick Chapman,
A.L.S., F.G.S. (Plate XXIII.) ... ... 217

XVI. — Notes on and Additions to Australian Fossil Poly-
placophora (Chitons). By Edwin Ashby,
F.L.S., etc. (Plate XXIV.) 220

List of Members .. ... ... ••• ••• ••• 234

Index ... ... .. ••• ••• ••• • •• 241

[Proc. Roy. Soc. Victoria, 41 (N.S.), Pt. I., 1928.]

Art. I . — Notes on New Zealand Hydroida ,

By R. E. TREBILCOCK.

(With Plates I.-VII.)

[Read 8th March, 1928; issued separately 27th Se ptember, 1928]

During a visit to New Zealand in April and May, 1923, I spent
a few days collecting Hydroids at Auckland, Island Bay (Wel-
lington), New Brighton (Christchurch), St. Clair (Dunedin),
and Bluff. The collecting was done solely in rock pools at low
tide, and among algae Washed ashore, but it resulted in several
new species, a number of species not hitherto recorded from New
Zealand, and numerous other and interesting forms.

The present paper deals with this collection, and also includes
references to a few specimens collected by my late colleague, Mr.
J. F. Mulder, from Stewart Island “ oyster ” shells many years
ago, and to some specimens kindly sent to me by Mr. W. M. Bale,
F.R.M.S., for examination and comparison.

I have to thank Mr. W. M. Bale for the very great assistance
he has rendered me by identifying species in respect of which I
had doubts, by sending me literature and specimens from his own
extensive collection for examination, and by valuable advice
throughout the preparation of this paper.

Fam. CLAVIDAE.

Endocrypta huntsmani (Fraser).

Cryptci huntsmani Fraser, 1911, p. 19.

Endocrypta huntsmani Fraser, 1912, p. 216; 1913, p. 149; 1914,
p. 109.

Ascidioclava parasitica Kirk, 1915, p. 146.

I have compared a specimen of Ascidioclava parasitica from
New Zealand (kindly sent to me for examination by Mr. Bale)
with specimens of Endocrypta huntsmani from Departure Bay,
west coast of Canada (received from Dr. C. McLean Fraser),
and there is not the slightest doubt that not only do they belong
to the same genus, but also that they are not specifically distinct.

Mr. Bale’s specimen is from the peripharyngeal groove of a
Poly car pa, and was collected at Wellington, N.Z.

Fam. PENNARIIDAE.

Penn aria australis Bale.

Numerous specimens, growing in a shallow rock -pool at the en-
trance to Auckland Harbour.

This species has not hitherto been recorded from New Zealand.

R. E . Trebilcock :

INCERTAE SEDIS.

SAABA ( ?) SCAN DENS, n. Sp.

(Plate I., Figs. 1, la.)

Specimens from Island Bay no doubt belong to the same genus
as Saaba arenosa (Bale),

The hvdrocaulus is fascicled in its proximal part, unjoin ted, and
consists of string-like stems, with a few ascending branches. The
perisarc has a tough and cartilaginous appearance. The polyp-
tubes are few, and spring irregularly from all sides. Near the
end of the polyp-tubes there is an annular thickening, and often
another just below it, probably due to regeneration. Beyond the
distal thickening the edge of the polyp-tube is sometimes quite
sharp, sometimes rather ragged. Inside the annular thickening is
often a narrow septum. Unfortunately no hydranths are present.

Unlike Bale's species, the present form is climbing in habit, the
hvdrocaulus adhering loosely to other hydroids.

The height of the largest specimen, which is incomplete, is
about 35 mm.

Springing from several of the polyp-tubes are large sac-like
bodies, which are possibly gonangia. As in Saaba arenosa they are
formed of very thin colourless perisarc which does not stain
readily, and have the whole exterior surface coated with closely
adhering grains of sand, calcareous particles and Foraminifera.

In the absence of hydranths the position of this genus is very
uncertain, and it can only provisionally be referred to the
Hydroida.

Fam. CAMPANULARIIDAE.

Obelia geniculata (Linn.).

Loc. — Island Bay, and New Brighton (Christchurch).

Obelia australis v. Lendenfeld.

I found this species, with numerous gonothecae, growing in
tide-pools below the swimming pool at St. Clair (Dunedin).

Orthopyxis Formosa, n. sp.

(Plate I., Figs. 2-2e.)

Hydrorhiza broad, flattened, with flanged margin, forming an
irregular network.

Hydrosoma varying from 1-2 mm. to 2-3 mm. in height.

Peduncles somewhat flattened, usually strongly and spirally
undulated ; a single spherule below each hydrotheca.

Hydrothecae large, compressed; in broad aspect with wide
base and only slightly expanding upwards, and with thickening of

Notes on New Zealand Hydroida.

walls mostly extending from the base up to just below the rim,
where there is an additional thickening on the outside forming
a stout band which completely surrounds the upper part of the
hydrotheca ; in narrow aspect narrower at the base and with little
thickening except at the base and near the rim.

Margin rising slightly above the thickened rim, very thin, fur-
nished with about 12 rounded teeth.

Gonothecae not present.

Hab. — At entrance to Auckland Harbour, on floating seaweed.

Occasionally hydrothecae are found without any thickening,
but such have the walls more or less wrinkled, and are obviously
abnormal. The hydrothecae vary considerably in size, as may be
seen by reference to my figures which are all drawn to the same
scale.

Orthopyxis delicata, n. sp.

(Plate II., Figs. 1-1/.)

Hydrorhiza broad, only slightly flattened, with a slightly flanged
margin.

Hydrosoma varying from 0*75 mm. to 1-5 mm. in height;
peduncles somewhat flattened, strongly and spirally undulated
throughout their entire length; a single spherule below each
hydrotheca.

Hydrothecae large, slightly compressed ; in broad aspect with a
wide base, but expanding considerably upwards ; in narrow aspect
narrow at the base and expanding still more upwards so as to make
the aperture more nearly circular ; walls in broad aspect thickened
near the base and becoming only slightly thinner towards the rim,
in narrow aspect much thinner throughout ; no annular thickening
around the rim.

Margin of hydrothecae rather thin, furnished with from about
10 to 15 rounded teeth.

Gonothecae about 1-2 mm. in length, springing from short ex-
panding peduncles, somewhat pear-shaped, more rotund on one
side than the other, thus making them curved, with a large, cir-
cular, terminal opening.

Loc. — St. Clair (Dunedin), growing over the surface of deli-
cate algae and polyzoa. In the latter case the hydrorhiza forms
an anastomosing network roughly corresponding with the cells of
the polyzoa.

The hydrothecae vary considerably in size, as will be seen by
reference to my figures, which are all drawn to the same scale.

Orthopyxis crenata (Hartlaub).

Numerous specimens with gonothecae, from Island Bay.

Silicularia bilabiata (Coughtrey).

I found this species growing profusely over Laminaria washed
ashore at Island Bay.

R. E. Trebilcock :

SlLICULARIA CAMPANULARIA (v. Lend.).

I found this species in large numbers at St. Clair (Dunedin)
and Bluff.

Fam. LAFOEIDAE.

Hebella calcarata (L. Agassiz).

I found numerous examples of this widely spread species grow-
ing over Sertularella subarticulata from Stewart Island.

Hebella corrugata (Thornely).

I found specimens of this widely spread species growing on
Thccocarpns forniosus var. imrmatus , n. var., at Island Bay.

In my specimens the aperture of the hydrothecae is not as
oblique as figured by most authors.

Measurements. — Length of pedicel 0-20 — 0-45 mm.; length of
hydrotheca 0-72 — 0-90 mm.; diameter of hydrotheca 0-37 — 0-55
mm.

Not previously recorded from New Zealand.

Filellum serratum (Clarke).

Specimens of this species occur on several other hydroids from
Bluff, Stewart Island, and Island Bay. Not previously recorded
from New Zealand.

Perisiphonia quadriseriata, n. sp.

(Plate II., Figs. 2-2 d.)

Hydrorhiza consisting of a dense mass of tubes forming a
parchment-like disc.

Hydrocaulus stout, straight, attaining a height of 14 cm.,
fascicled, consisting of a single axial tube, bearing hydrothecae,
surrounded bv a large number of peripheral tubes bearing sarco-
thecae only.

Hydrocladia sub-opposite, forming an angle of about 60°
with the hydrocaulus, attaining a length of about 22 mm.,
stout, flattened, oval in section, about 0-5 mm. in broader (ver-
tical) diameter, and about 0-4 mm. in narrower; fascicled, con-
sisting of a single axial tube bearing hydrothecae, surrounded by
a large number of peripheral tubes, none of which spring from
the axial tubes of the hydrocladia; hydrocladia not of smaller
diameter at base than elsewhere.

Hydrothecae borne on axial tubes of hydrocaulus and hydro-
cladia, close-set, each usually overlapping the proximal portion of
its successor in the same series, cylindrical, with a slightly bulging
profile, lying closely adpressed to the axial tube for a considerable
part of their length, the terminal portion curving away from the
axial tube and projecting for a short distance (about 0-1 mm. in
the longest example) through the fascicle of peripheral tubes.
Aperture of hydrothecae circular, margin smooth.

Notes on Neiv Zealand Hydroida.

Base of hydrothecae passing* into a projection of the axial tube
corresponding to a hydrothecal peduncle, and containing a
strong diaphragm which slants from the 'outside inwards and
slightly downwards.

Hydrothecae on hydrocaulus in two series on opposite sides of
the peripheral tube, both series lying in the same plane, alternate
though not always regularly spaced, varying in length with the
thickness of the mass of peripheral tubes through which they pro-
ject.

Hvdrothecae on the hydrocladia in four series, two series close
together on the abcanline side, and two series close together on
the adcauline side, the two abcauline series being widely separated
from the two adcauline. Hydrothecae, in each series, regularly
alternating with those of adjacent series on both sides of it, but
opposite to those of the series diagonally opposite.

Sarcothecae numerous, scattered on hydrorhiza, hydrocaulus
and hydrocladia, short, cylindrical, diameter wide compared with
their length, borne on slight projections from the peripheral tubes,
and separated from the projections by a constriction.

Gonosome, not present.

Locality. — Island Bay, washed ashore.

The hydrocladia would appear from casual examination to be
opposite, but, on dissecting away the mass of the peripheral tubes,
it is found that the axial tubes of the hydrocladia spring from the
axial tube of the hydrocaulus at points that would otherwise be
occupied by alternate hydrothecae. The hydrocladia are thus sub-
opposite.

The peripheral tubes communicate freely with each other, and
with the axial tube of the hydrocaulus by numerous circular or
oval apertures. They do not however directly communicate with
the axial tubes of the hydrocladia except by one or two apertures
near the base of each hydrocladium and within the bundle of peri-
pheral tubes of the hydrocaulus.

The axial tubes of the hydrocladia are not thickened to any
great extent, but that of the hydrocaulus has a massive wall, built
up of numerous layers.

In the axil of each hydrocladium is a hydrotheca which differs
from the others. It springs from the hydrocladium immediately
above the point of origin of the latter, but lies within the peri-
pheral tubes of the hydrocaulus through which it projects. It
curves in the same direction as the hydrothecae of the hydro-
caulus, and in the opposite direction to the other hydrothecae on
the adcauline side of the hydrocladium. Perhaps it would be
more correct to say that the axial tube of the hydrocladium
springs from the base of this hydrotheca, for the latter certainly
belongs more to the hydrocaulus than to the hydrocladium. The
wall of the lower part of this axial hydrotheca is much thickened,
and from an aperture just below the diaphragm springs a peri-
pheral tube. This tube immediately divides, one branch running
up the hydrocaulus and the other along the hydrocladium.

R. E. TrebilcocTc :

Other peripheral tubes running up the hydrocaulus from below
also divide near the point of origin of the hydrocladium, one
branch continuing upwards and the other going along the hydro-
cladium.

The proximal hydrotheca on each hydrocladium on the abcau-
line side is separated from the others by a short distance, the first
hydrotheca of the adjoining series being missing.

The present species is represented in my collection by a single
specimen.

Fam. HALECIIDAE.

Halecium flexile Allman.

I found a few small colonies of this species growing in a rock
pool at St. Clair (Dunedin), also at Bluff.

Halecium lenticulare, n. sp.

(Plate III., Figs. 3-3 d; Plate IV., Figs. 1-1 b.)

Colonies small, attaining a height of a little more than 1 cm.,
not fascicled, regularly and markedly sympodial (an example of
cincinnal monopodia), the stems usually having' a marked zig-zag
appearance, sparingly bipinnate.

Primary hydrothecae low. Secondary hydrothecae with a large
basal cavity, somewhat symmetrically ’ developed. Hydrothecae
usually with very thin walls, open margin curved outwards and
usually everted, particularly on the adcauline side. Diaphragm
well developed, but extremely thin ; aperture narrow, often ter-
minating in a very delicate membranous tube, which stretches a
considerable distance into the basal cavity. Below the diaphragm
a well developed adcauline thickening of the wall (“pseudo-
diaphragm ”) often extending completely round the hydrotheca
but always thicker on the adcauline side. Sometimes another
thickening, similar, but not so pronounced, near the base of the
basal cavity.

Regenerations of hydrothecae markedly active, unregenerated
examples being an exception to the general rule.

Gonothecae borne on the apophyse immediately below the
primary hydrothecae. Male gonothecae borne mainly on the
distal parts of the colony, small, ovate, much flattened. Female
gonothecae usually confined to the proximal parts of the colony,
large, lenticular, with a circular opening in the distal part of the
abcauline side ; margin of opening thickened.

Localities.— St. Clair (Dunedin), Bluff, Island Bay (Welling-
ton).

The type specimen is from Bluff Harbour.

The specimens from St. Clair differ from the type in that most
of the perisarc is enormously strengthened by internal annular
thickenings, arranged more or less diagonally around stem and
branches.

Notes on New Zealand Hydroida.

The specimens from Island Bay resemble those from St. Clair
in the above respect, but differ from those from the other two
localities. In the latter the hydrothecae, primary and secondary,
are usually almost, if not quite, in a straight line with the inter-
node from the apophyse of which they spring, thus giving the
colony a distinct zig-zag appearance. In the Island Bay speci-
mens, however, the apophyse and hydrothecae curve away from
the line of the internode, and the succeeding internode is in a
straight line with its predecessor, the whole stem thus being
straight. (PI. IV., Fig. 1.) Specimens from all three localities
bear gonothecae.

Halecium corrugatissimum, n. sp.

(Plate III., Figs. 1-1/.)

Colonies small, attaining a height of about 5 cm. Stem not fas-
cicled, strongly annulated, and divided into very short internodes.
Growth irregularly sympodial, the main axis not being produced
beyond the primary hydrotheca, immediately below the diaphragm
of which spring one, two or three branches, the branching thus
being sometimes falsely dichotomous or falsely trichotomous.

The branches do not lie in one plane, but are quite irregular in
this respect.

Pedicels expanding upwards, much and deeply corrugated.
Hydrothecae large, margins broadly expanded and slightly
everted, with a well marked row of dots. Margins or the whole
hydrotheca sometimes reduplicated.

Diaphragm fairly well developed, but thin; aperture wide.

Gonothecae borne on short pedicels below the hydrothecae,
mainly on the upper part of the colony, ovoid, with from 5 to 7
deep and even annulations ; upper part usually devoid of annula-
tions, and more or less hemispherical.

Locality. — St. Clair (Dunedin).

The young hydranth is enclosed by an almost spherical, thin,
chitinous ball, which eventually splits away from the hydrophore.
Fig. 1 d shows an example with the splitting process just com-
menced.

This species is closely allied to H. speciosiim Nutting (1901, p.
181, pi. xxii., figs. 1, 2).

Halecium expansum, n. sp.

(Plate III., Figs. 2-2c; Plate IV., Figs 2-2 b.)

Colonies small, attaining a height of about 6 mm., main stem
either not fascicled or sometimes strengthened by one or two irre-
gular peripheral tubes, irregularly sympodial, the prolongation of
the axis often developing quite as strongly as the branch, false
dichotomy occasionally met with.

Primary hydrothecae sessile. Secondary hydrothecae with a
large basal cavity. Hydrothecae wide, shallow and expanding,

RE. Trebilcock:

with very thick walls, margin never curved outwards. Diaphragm
well developed and massive, aperture large. Walls of basal cavity
without any marked thickening. Between each hydrotheca and
the basal cavity of the hydrotheca above it, a more or less
spherical intemode, the base of which springs from the inner
edge of the diaphragm.

Regenerations of hydro thecae fairly active.

Gonothecae not present.

Locality. — Growing on roots of algae in rock-pools at St. Clair
(Dunedin).

The branches are not confined to any particular plane, but are
quite irregular in this respect. In the distal parts of some of the
colonies the branching is very active, resulting in a bushy mass.

The ring of chitinous “ dots ” around the hydrothecae are not
conspicuous; in fact, I could find them only in a few young
hydrothecae that happened to be empty.

Campanulina humilis Bale.

(Plate IV., Figs. 3-3rf.)

I found numerous specimens of the creeping form of this species
growing on other hydroids, polyzoa, and the roots of algae, at St.
Clair (Dunedin). After a long search among the material with
a view to finding a gonosome. I discovered several branched speci-
mens. These in their mode of branching agree with Campanulina
turrita Hineks, except that the branching is more active and the
hydrothecae consequently more crowded. My specimens are all
small, the largest being only 1-35 mm. in height. The stems and
branches are all closely annulated, or, sometimes, spirally con-
stricted, throughout.

The specimens are in excellent state of preservation, with many
of the polyps fully expanded. There is a web between the ten-
tacles (see Fig. 3c), but it extends only a very short distance up.

Unfortunately, no gonothecae were present, so the assignment
of the species to the genus Campanulina is provisional. Hitherto
the species has been known by its creeping form only.

Thyroscyphus simplex (Lamouroux).

Specimens, with gonothecae, from Island Ray. The gonothecae
are as described by Bale (1915), except that they are somewhat
shorter and. broader, their length being from 1-1 mm. to 12 mm.,
and their diameter at the widest part 0-7 mm.

Fam. LINEOLARIIDAE.

Lineolaria flexuosa Bale.

There is a specimen of this small and little known species in the
collection of the late Mr. Mulder. It has not hitherto been re-
corded from New Zealand.

Locality. — Stewart Island.

Notes on New Zealand Hydroida.

Fam. SYNTHECIIDAE.

Synthecium patulum (Busk.)

Billard (1925) has united 5\ patulum Busk and 5\ ortho gonium
Busk under the name S. patulum. concluding from his exam-
ination of specimens collected by the “ Siboga ” Expedition that
the distinctions pointed out by Bale (1914), namely, the arrange-
ment of hydrothecae, pinnae, and nodes on the hydrocaulus, can-
not be relied on to separate the two specifically.

In the absence of gonothecae, I am assigning to this species
specimens collected by me at Auckland Harbour and from
Stewart Island.

They both differ from Billard’s figures in that the orifice of the
hydrothecae is very little, if at all, everted. Reduplication of the
margin is not common. The measurements of the hydrothecae,
however, generally agree with those given by Billard.

Length adnate -
Length free -
Diameter at mouth
Interval between pairs of hydro-
thecae (measured from ad-
nate part of one to the base
of the next above it)

Auckland

Specimens.

0*46 — O’ 60 mm.
O’ 16 — O’ 35 mm.
0’19— 0’21 mm.

Stewart Is.
Specimens.

O’ 50 — O’ 69 mm.
0 -20— 0-43 mm.
0* 25 — 0*32 mm.

O’ 18 — O’ 50 mm.

0 ’ll— 0’34 mm.

Fam. SERTULARIIDAE.

Selaginopsis monilifera (Hutton).

(Plate IV., Figs. 4, 4a.)

Bale (1915, p. 266) in his description of this species, under the
name of Selaginopsis dichotoma, states that the hydrophyton is
monosiphonic. All my New Zealand specimens are fascicled.

There is considerable variation in the distance between hydro-
thecae even in a single specimen; the space occupied by four
hydrothecae in the four series (measured from the base of the first
to the base of the fifth) varies in my specimens from 06 mm.
to 1-45 mm. Occasionally, when widely separated, the hydro-
thecae lose their quadriserial arrangement and all lie in the same
plane.

Localities. — St. Clair (Dunedin), Bluff, and Stewart Island.

This species has not hitherto been correctly figured.

Genus Sertularella.

Characters. — Hydrothecae with operculum of three or four
valves; hydranths with abcaulinary caecum; hydrothecae alter-
nate, opposite, verticillate, or spirally arranged.

ii. E. Trebilcock :

The addition of the last three words to the above is rendered
necessary by the discovery of Sertularella irregularis, n. sp.

One of the characteristics of the genus Sertularella is the
presence of a more or less developed abcaulinary caecum, or
blindsac. From the circumference of this, most markedly at the
abcauline side, spring a number of threads a split external ecto-
dermal lamella, according to Kiihn which stretch to the wall of the
hydrotheca, and hold the caecum in position. In many
species, especially, curiously enough, in those in which the caecum
is well developed, these threads have no definite points of attach-
ment to the chitin, but appear to be attached only to the layer of
ectoderm cells which is left adhering to the inside of the hydro-
theca when the young hydranth shrinks away from it. This layer
often breaks away, and lies loosely within the hydrotheca, but
attached by the threads to the caecum.

In other species these threads have definite points of attach-
ment to the chitin itself, the points of attachment being marked
by a number of scars, or, sometimes, chitinous denticles which are
usually minute. Occasionally, however, there is a comparatively
large denticle at the abcauline side, and sometimes this is the only
one that can be detected.

This line of denticles stretches obliquely backwards and down-
wards around the lower part of the hydrotheca. This line of
attachment is the “ apparent oblique septum ” referred to by Bale
(1914, p. 21) in his description of S. divaricata.

I have found that the angle at which this ring encircles the
hydrotheca is fairly constant in each of the species I have exam-
ined, but often differs in closely allied species, and thus gives con-
siderable assistance when in doubt as to which of two species a
specimen belongs. For example, in S. johnstoni the highest point
reached by the ring is about half way up the abcauline side of the
hydrotheca ; in N. pygmaca, which is so closely allied as to be con-
sidered by some authorities to belong to the same species, without
doubt erroneously, the ring reaches to about one-third of the way
up; while in 5\ rentoni it is quite close to the base, and only
slightly diagonal.

Sertularella pygmaea Bale.
Specimens, growing on other hydroids, from Bluff.

Sertularella rentoni Bartlett.

(Plate I., Fig. 3.)

I found this species growing in a rock-pool at St. Clair (Dun-
edin). No gonothecae were present. It has not hitherto been
recorded from any other locality than Victoria.

In S. pygmaea the line of points of attachment of the caecum is
less conspicuous than in S. rentoni, and at the front of the hydro-
theca is usually much higher up, and runs around it in a more

Notes on New Zealand Hydroida.

diagonal position. This enables the two species to be separated
with ease, even when no gonothecae are present.

Usually the walls of the hydrothecae are thin, but the margin
is sometimes slightly thickened, but never to the extent found in
the species next described.

Sertularella macrogona, n. sp.

(Plate I., Figs. 4-4 d.)

Hydrocaulus attaining a height of about 1 cm., simple or irre-
gularly branched, branches usually springing from inside a hydro-
theca. Hydrocaulus and branches divided by slightly oblique joints
into short internodes, each (with an occasional exception) bearing
a hydrotheca.

Hydrothecae adnate for about one-third of their height, both
series springing from the front, the two planes in which they lie
being at about right angles to each other ; hydrothecae tubular,
smooth, slightly concave in outline on the abcauline side, the
adcauline side being ventricose; the margin slightly thickened,,
especially at the adcauline side ; aperture with three broad rounded
marginal teeth, one superior, short and comparatively incon-
spicuous, and two lateral, much longer; no internal submarginal
denticles. Operculum of three triangular pieces.

The hydrothecae lie at, approximately, an angle of 45° with
the stem, and the aperture is, approximately, at an angle of 45°
with the length of the hydrotheca, thus making the aperture at
about right angles with the stem.

Gonothecae large, about 4 times as long and times as broad
as the length of a hydrotheca, usually springing from inside
hydrothecae, but sometimes borne on the front of the stem just
below a hydrotlieca, obovate, flattened and slightly depressed just
above its widest part, and from the centre of the depression rises
a narrow, converging tube with a circular orifice.

Locality. — St. Clair (Dunedin).

Though the branches usually spring from inside a hydrotheca,
generally broken, they sometimes arise from the front of the
hydrocaulus, just beneath a hydrotheca. One specimen shows a
branch springing from inside a broken gonotheca.

The lower part of the hydrocaulus, below the proximal hydro-
theca, is deeply and closely annulated or spirally thickened.

This species is allied to rentoni Bartlett, but differs in many
important respects, as will be seen by a comparison of the descrip-
tion of the two species.

The points of attachment of the caecum of the hydranth are in
the same position as in .S', rentoni , and marked in the same way by
a faint ridge with minute chitinous processes.

Sertularella procera, n. sp.

(Plate I., Figs. 5-5 d.)

Colony tall and slender, attaining a height of 22 cm., tapering
very gradually towards the apex. Main stem giving rise to a few

11. E. Trebilcock :

principal branches of the same structure as itself. Main stem and
principal branches fascicled, except in distal parts. Main stem
and branches giving rise on all sides to irregularly placed pinnae
(the longest of which attain a length of about 1-3 cm.), except at
the distal part where the pinnae all lie in the same plane and are
more or less regularly alternate.

Pinnae sometimes strengthened in their proximal part by a tube
running out from the fascicled stem or branch, but otherwise not
fascicled, themselves pinnate or bipinnate, or with somewhat irre-
gular sub-dichotomous ramification; the more regular forms with
the sub-pinnae alternate; usually three hydrothecae (including
one in the axil) between every two sub-pinnae or sub-dichotomous
branches.

Intel nodes of stem and main branches usually bearing three
hydrothecae ; those of the pinnae and sub-pinnae bearing only one
each. Pinnae not terminating in stolons.

Hydrothecae tubular or sub-conical, both series directed
strongly to the front, curved outwards, sometimes somewhat
abruptly; margin with three conspicuous teeth, one superior and
two lateral, and an operculum of three pieces; no internal sub-
marginal denticles. Points of attachment of caecum of hvdranths
extending very obliquely across the hydrothecae, reaching to
nearly half way up in the front.

Gonothecae borne on the pinnae, large, obovate, surrounded by
a number of prominent annular ridges, except on the proximal
part of back which is smooth and adpressed to the pinna; distal
portion of gonotheca not projecting forward ; rising from the
centre of the flattened end surrounded by the distal ridge a
narrow, long, expanding tube; aperture usually central, rarely
slightly eccentric.

The annular ridges on the gonothecae are similar to those of
*S. johnstoni and S. pygmaca except that they are somewhat more
pronounced and the flange rather wider.

A fenestra is sometimes present at the base of the hydrothecae
but more often absent. Similar fenestrae are numerous on the
axillary tubes of the stem and branches, often on the abcauline
side. Inter-communication between the tubes is frequent Occa-
sionally bridges like the letter H are met with between tubes not
m contact with one another, though how they originate is not clear.

This species is closely allied to S . divaricctta var. subdiclwtoma.

A single pinna or two could not be distinguished from that species*
but the manner of growth is altogether different.

Locaiities.— Bluff, several fine specimens; Stewart Island (T.

F. Mulder), fragments.

Sertularella subarticulata (Coughtrey).

(Plate VII., Figs. 7 -7b.)

. Locality. — Bluff, numerous specimens; Stewart Island, grow-
ing on “ oyster ” shells. This species has not hitherto been cor-
rectly figured.

Notes on Neiv Zealand Hydroida.

13 ’

Sertularella irregularis, n. sp.

(Plate V., Figs. 1-lb.)

Hydrorhiza somewhat flattened, giving rise to shoots at close

intervals. . .

Hydrocaulus unbranched or sparingly branched, attaining a
height of about 5 mm., divided into short internodes each of
which bears a hydrotheca.

Hydrothecae not arranged in two series, but in an irregular
spiral, every fourth one approximately completing one turn, so
that there are three irregular, longitudinal series. Hydrothecae
adnate for about one-third of their height, tubular, slightly con-
verging towards the mouth, slightly curved, the adcauline side
being convex and the abcauline concave; aperture with three
pointed, prominent teeth, one superior and two lateral, and . an
operculum of three triangular flaps. No internal sub-marginal
denticles. Hydranth with about 16 to 20 tentacles, abcaulinary
caecum weakly developed.

Gonothecae usually springing from the interior of a hydrotheca,
but sometimes from the ^stem, ovate, marked with about 8 to 10
conspicuous transverse annulations which are not provided with
a flange, terminating at the summit in a short tube which is some-
times slightly expanded at the orifice.

Locality. — St. Clair (Dunedin), growing on roots and stems of
algae in rock-pools below the swimming pool.

Sertularella crassiuscula Bale.

I collected specimens of this species, with gonothecae, at Bluff
and New Brighton. In most of the hydrothecae in my specimens
the marginal teeth are very shallow indeed, and the margin prac-
tically entire. Only in a comparatively few cases can 1 make out
the four teeth, which are little more than undulations of the
margin.

Sertularella fusca, n. sp.

(Plate V., Figs. 2-2 b.)

Hydrocaulus pinnate, or, occasionally, slightly bipinnate, attain-
ing a height of about 15 mm., divided by oblique joints into inter-
nodes each bearing a hydrotheca.

Pinnae alternate, often crowded, one springing immediately
below almost every hydrotheca of the hydrocaulus, both series
directed in a marked degree to the front, the angle between them
usually being 45° or less.

Hydrothecae alternate, both series directed in a marked degree
to the distal side, the angle between the two series being usually a
little less than 90°, adnate for a third, or less, of their height, con-
stricted and somewhat thickened immediately below the mouth,
especially on the adcauline side, where the margin is slightly

R. E . Trebilcock:

everted ; abcauline side almost straight, adcauline side ventricose ;
margin with three prominent, rounded teeth, and an operculum of
three pieces; three small and often inconspicuous internal sub-
marginal denticles.

Gonothecae not present.

Locality. — St. Clair (Dunedin), in rock-pools below the swim-
ming pool.

In several of the specimens there spring, from immediately
below the base of one or more of the proximal hydrothecae,

S. simplex

St. Clair

Dunedin

S. simplex
Auckland

Harbour

S. robusta

var.

quasiplana

Hydrotheca

length -

0-40 - 0-52

0 42 - 0 54

0 50 - 0'55

greatest diameter

0 20 - 0 25

0 25 - 0*32

0-29 - 0 33

diameter at mouth (c) -

0-14 - 0 20

017 - 0 20

015 - 0 20

proportion adnate

i i

3 " 2

about 4

Internode

length -

0 36 - 0 62

0*45 - 0 94

0 55 - 0 82

diameter (d)

01 5 - 0*20

013 - 0*16

015 - 016

Gonotheca

length -

1-24 - J 70

1-36 - 172

diameter -

0-60 - 0-83

0-77 - 0 90

S. robusta
forma typica
(a)

S. angulosa
Bale
(b)

S. robusta

var.

flucticulata

Hydrotheca

length -

0 -45 - 0 60

0 57 - 0-60

0 65 - 0 75

greatest diameter

0-23 - 0-37

029

0 34 - 0 37

diameter at mouth (c) -

016 - 0 22

0-16 - 0-19

022-0-25

proportion adnate

i i

3 " 2

about 4

Internode

length -

0-45 - 1-00

0 50 - 0-60

0-45 - 0-65 (e)

diameter (d)

0-15 - 0-21

017

0-18-0-30

Gonotheca

length -

1-95

1-60-1-92 (f)

diameter -

1-00

1-20-1-35 (/)

(a) Measurements taken from specimens from Bluff and Stewart Island.

(b) Measurements taken from Bale’s figure.

(d) Half way between base of hydrotheca and proximal end of internode.

0) Proximal inter node of branch often attains I 10 mm.

if) From robust specimen from Bluff. Other specimens vary from 1*30 mm.
x 0 85 nun. to 1*80 imn. x l - 0 mm.

stolons which turn downwards, but not in contact with the stem.
In one example one of these reaches the hydrorhiza, and anas-
tomoses with it.

Notes on New Zealand Hy droid a .

Sertularella simplex-robusta Group.

The New Zealand forms belonging to this group are very diffi-
cult to separate from one another. Though S . simplex (Hutton),
at the one end of the group, and 5'. robust a var. flucticulata, n. var.,
at the other, are so different even to the naked eye that no one
could have any difficulty in separating them, there are numerous
intermediate forms connecting them, which merge gradually into
one another.

One great difference between the two above named is in the
size of the hydrothecae, but the table of measurements on the
opposite page shows how completely the gaps between them are
filled up by other forms. The measurements are in millimetres.

Sertularella simplex (Hutton).

(Plate VI., Figs. 1-1 d, 2-2e.)

Hutton’s description (1872) of this species under the name of
.Sertularia simplex is not very full, and could easily include several
forms which undoubtedly belong to a distinct species, though
where to draw the line between them is not easy to decide.
Coughtrey (1874) adds to Hutton’s description sufficient further
details to fix the typical form more satisfactorily. He, however,
includes under this name two other forms, which he figures ( 1874,
pi. xx., figs. 9 and 10), the latter of which he afterwards
(1875) describes as a distinct species under the name of Sertula-
rella robusta . The second form he describes as having faint,
shallow grooves, generally three in number, that cross the hydro-
theca.

It is very difficult indeed to decide to which species some of the
forms in my collection belong, as there are so many intermediate
forms, differing from one another in small details, but all of them
having the general characteristics of the typical form.

The figure and description given by Bale (1924) may be taken
as illustrating the typical form of the species. The hydrothecae
of the specimens collected by me at the entrance to Auckland
Harbour agree with his figure and description, but, though it was
the most common species on the beach, I failed to find a gono-
theca.

Hutton (1872) says “ Hydrothecae distant,” and many of my
specimens agree with this. There is, however, considerable varia-
tion in this respect even in a single colony. Figs 1 and la are
drawn from two shoots springing from the same hydrorhiza, and
illustrate how considerable the variation may be.

The hydrorhiza anastomoses very freely; and my Auckland
specimens show a considerable number of single hydrothecae
growing from the hydrorhiza. No doubt they are the beginnings
of new shoots which would ultimately develop, but in their present
form they remind one forcibly of Allman’s untenable genus

R E. Trebilcock:

Calamphora. The hydrorhiza in these specimens is not flattened
to any appreciable extent.

Specimens growing in tide pools at St. Clair (Dunedin) differ
somewhat from the former, the hydrothecae being smaller, and
usually directed more to the front, and slightly less of it being
adnate. The gonothecae are also different from the typical form
as figured by Bale, inasmuch as the tubular neck is absent. They
have four conical projections at the summit, but these are not well
developed. (See Fig. 2c.) The difference between the two forms
is not sufficient to warrant the constitution of a new species in the
case of such a variable hydroid as the present.

Stechow (1923) has figured quite a number of species belong-
ing to this group, all of which have four external teeth and three
internal submarginal denticles, as having one superior, one in-
ferior, and one lateral internal denticle. I feel sure, however, that
he has done this inadvertently. Bale describes the denticles as
being “ two within the two upper emarginations of border, and
the third below inferior marginal tooth/' and, in all probability,
this is the arrangement of the internal denticles in all species hav-
ing four teeth and three internal denticles. Plate VI., Fig. Id,
shows the appearance, when looking straight into the aperture, of
the hydrotheca. The external teeth are not apparent in this figure,
as they face directly towards the observer, but their position is
indicated by the angles at the bases of the triangular flaps of the
operculum, a and b representing the positions of the superior and
inferior marginal teeth respectively, and c and d those of the
lateral.

Hutton says in his original description that this species is
“ simple or rarely branched." I have a small branched specimen
from St. Clair that I consider belongs to this species. It shows
considerable variation both in the length of the internodes and in
the manner in which the hydrothecae lie. In some shoots both
series project to the front in a marked degree; in others the two
series lie almost in the same plane. In this specimen the gono-
thecae differ somewhat from those already described, being, as
a rule, narrower. None of them possesses more than three pro-
jections at the summit. (See Figs. 2b-2d.)

All the above described forms have the hydrothecae entirely
destitute of any transverse undulations, and I consider that the
forms with undulated hydrothecae should not be assigned to this
species.

Sertularella robusta Coughtrey.

(Plate VI., Figs. 3-3c.)

Of the three forms originally assigned by Coughtrey to Sertu -
laria simplex I include under the present species those figured by
him (1874, pi. xx., figs. 9, 10). He included only Fig. 10 under
the name Sertularia robusta, apparently leaving Fig. 9 as Sertu -
la?ia simplex. These are what he refers to as the 4t several pigmy

Notes on New Zealand Hydroida.

varieties in which the hydrothecae are transversely wrinkled.” I
take it that the word ‘ pigmy ’ here refers to the height of the
colony, and not to the size of the hydrotheeae, because reference
to his figures shows that the hydrotheeae of Fig. 9 do not differ
appreciably in size from those he figures as the typical form (Fig.
8). In his Description of Plate he says, “All objects magnified
50 diameters except where otherwise specified.” This, as Mr.
Bale has pointed out to me, is obviously incorrect. The length of
a hydrotheca of the typical form, in my collection, of S. simplex ,
which varies very little from that figured by Bale, is about 0*5
mm. Assuming that the hydrotheeae of the typical form of S.
simplex as figured by Coughtrey are about the same length, his
figures of S. simplex and its variety S. robusta cannot be magnified
more than about 18 diameters.

I have endeavoured to separate the forms with undulated
hydrotheeae into more than one species, but, although the large
and small forms differ very considerably from one another in size
and general appearance, I have so many intermediate forms that
I am unable to do so satisfactorily, and must be content for the
present in describing as varieties certain forms that differ some-
what from the typical form. At first it seemed that there was a
decided gap between the largest of the small forms and the small-
est of the large, but this is filled by Sertularella angulosa Bale (1893,
p. 102, pi. iv., fig. 6), which, I think, must be regarded as a variety
of S. robusta ( Sertularella robusta var. angulosa ). In the smaller
forms the hydrotheeae are about 0-5 mm. in length, and differ
from simplex in having transverse ridges. In 5\ angulosa the
hydrotheeae are about 0 6 mm. in length. In this form the stem
is zig-zag. In many of my New Zealand specimens also the stem
is zig-zag (though most of them are almost, if not quite, straight),
and one specimen so closely resembles Bale’s figure as to leave no
doubt whatever in my mind concerning the identity of the species.

As it is quite impossible to say definitely which form Cough-
trey had before him and used as his type of .S', robusta , I am
assuming that it was the more common form, and not either of
the varieties I am describing under the names of var. qitasiplana
and var. flucticulata. Both of these are comparatively rare, while
the common form is so plentiful that Coughtrey could not pos-
sibly have missed it.

The full description of a typical specimen of Sertularella robusta
is as follows: — Shoots simple, attaining a height of 10 mm.,
divided by slightly oblique joints into internodes which vary con-
siderably in length, each bearing a hydrotheca on its upper part.
Hydrotheeae adnate for about one-third of their height, large,
divergent, barrel-shaped, but smaller towards the summit, usually
more ventricose on the adcauline than the abcauline side, with
about six distinct sharp, transverse ridges, completely surrounding
them, but usually becoming less distinct on the abcauline side;
aperture expanding, with four well-defined teeth, aperture some-

R. E. Trebilcoclc :

times at about right angles to the length of the hydrotheca but
often with the inferior tooth projecting further than the others;
three internal, compressed, vertical, submarginal denticles, two of
which are within the two upper emarginations of the border, and
the third opposite the inferior marginal tooth.

Gonothecae large, borne sometimes on the hydrocaulus and
sometimes on the hydrorhiza, ovate, with several' distinct cross
undulations, upper part sometimes in the form of a tubular neck,
which, however, is not always distinctly present; summit usually
with about four conical projections.

Locality— Bluff, and on “ oyster ” shells from Stewart Island.
I have already referred to the single hydrothecae that are found
on the hydrorhiza of .S’, simplex. In the present species single
hydrothecae are also met with quite commonly, and remind one
even more forcibly of Allman’s genus Calamphora.

Sertularella robusta var. quasiplana, n. var.

(Plate VI., Figs. 4, 4a.)

I have separated 5". robusta from 5". simplex on account of the
presence in the latter and the absence in the former of transverse
undulations on the hydrothecae. The present form seems to be on
the border line. At first sight it looks like a rather robust form of
A. simplex, but closer examination reveals the presence of three or
four transverse rugae completely surrounding the hvdrotheca. It
is true they arc often rather faint; in fact, in some hydrothecae,
such as the one figured, they would be likely to be overlooked but
for the presence of minute diatoms which grow thickly along the
shallow depressions between the ridges. This variety differs
from the others not only in the above respect, but also in having
the hydrothecae usually broader at the base, in proportion to their
height, than the others. They are often adnate for as much as
one-half of their height. In some specimens I have found the
superior tooth projecting slightly more than the others, but usually
the mouth is at right angles to the length of the hydrotheca. The
teeth are, as in the typical form, well defined. My specimens
attain a height of 13 mm.

The gonothecae do not differ from the typical form in any
important particular. They are, however, somewhat larger.

Locality. — Island Bay.

Sertularella robusta var. flucticulata, n. var.

(Plate VI., Figs. 5, 5a.)

This variety differs from the others mainly in its much greater
size in all its parts, and only for the existence of V angulosa Bale
(1893, p. 102, pi. iv., fig. 6), I would have no hesitation in ranking
it as a distinct species. The rugae appear like little waves on the
adcauline side (hence the proposed name of the variety), but rarely

Notes on New Zealand Hydroida.

extend more than half way round the hydrotheca, the abcauline
side being almost, if not altogether, free from undulations. The
internal submarginal denticles are very large and well developed,
but form very thin vertical plates. Usually the mouth of the hydro-
theca is not at right angles to its length, the inferior tooth project-
ing considerably more than the others. The teeth are rarely well
developed, and are often no more than a slight wave in the other-
wise entire but oblique peristome.

The gonothecae are broader than those found on the other vari-
ties, and are borne on stem and branches.

Unlike the other forms belonging to this species the hydrophy-
ton branches rather freely, but the branching is quite irregular.
There is considerable variation in the length of the internodes,
even in the same colony. In one specimen they vary from 0-45
mm. to 0*85 mm.

Locality. — Bluff.

Thuiaria farquhari Bale.

(Plate VII., Fig. 4.)

A fine specimen of this species collected at Bluff densely clothes
the stem of an ascidian for several inches.

Thuiaria buski Allman.

(Plate VII., Figs 1-1 c.)

Hydrocaulus not fascicled, attaining a height of about 3 inches
(fide Allman), straight or almost so, unbranched, pinnate. Stem
usually thick, divided by slightly oblique nodes into inter nodes of
variable length, each bearing from 1 to 5 pairs of hydrothecae.
Pinnae irregular, usually with a tendency to alternate disposition,
rarely opposite, stout, divergent at nearly right angles, borne on
slender apophyses from which they are separated by a rather
oblique, conspicuous node ; usually divided into 2 or 3 long inter-
nodes each bearing from 4 to 8 (sometimes up to 11) pairs of
hydrothecae; nodes oblique, very rarely transverse.

Hydrothecae in pairs, strictly opposite both on hydrocaulus and
pinnae, adnate in front, widely separated behind, most of their
length vertical, upper portion turned outward and narrowed, aper-
ture vertical, widened laterally, with two lateral lobes, facing out-
ward and forward ; edge of peristome thin, especially on the ad-
cauline side, where the sinus between the lateral lobes is filled up
by a very thin prolongation of the wall of the hydrotheca; some-
times a slight broad internal thickening of the perisarc just inside
the peristome on the abcauline side, but no well developed internal
denticle.

Pairs of hydrothecae usually closely approximated, sometimes
actually touching, on the pinnae; more separated on the hydro-
caulus.

B. E. Trebilcock :

Gonothecae borne on the front of the pinnae, near the base
of same, ovoid, about 2-5 mm. in length, aperture round, entire,,
on a very short neck.

Colour of perisarc, dark brown.

Locality. — Island Bay and Bluff.

Allman’s (1876) description and figures of this species are
faulty. He describes and figures the pinnae as being divided into
short" internodes, each of which bears a single pair of hydrothecae
only, and his figure shows the joints as being transverse. At Mr.
Bale’s request Captain A. K. Totton, M.C., of the British
Museum (Natural History), has examined Allman’s type, and in
a letter, which the former has kindly placed at my disposal, writes
as follows: — “The successive pairs of hydrothecae on the type
of D. buskii are closely approximated though not quite touching,
but there is not a node between each pair. It would be unwise to
say more about the nodes than this, because the type specimen
is a poor one, very imperfect and much overgrown.”

In my experience nodes in the pinnae of Sertulariidae are never
found in the position shown in Allman’s figure, immediately above
the base of the hydrothecae. The explanation of his mistake is
doubtless that he examined a pinna lying approximately in the
position shown in my Fig. 1 c, and mistook for a node the base of
the hydrotheca on the further side of the pinna. In his Fig. 4 an
oblique view of the pinna is shown. His Fig. 7 is a lateral view,
not “ oblique,” as he calls it. This is borne out by the relative dis-
tances between the hydrothecac and the back of the pinnae. The
importance of this is that if his Fig. 4 showed a true lateral view
of the pinna, the distance by which the hydrothecae are separated
at the back would be greater than it really is. I have searched my
material for a transverse node such as he shows, but find that
transverse nodes in this species are very rare, and when present
occur, as one would expect, above the adnate part of the hydro-
thecae. Allman’s figure would make them occur behind the adnate
part.

In all my specimens the colour of the perisarc is very pale horn.

This species is allied to T. bicalycula , but differs from it in
several respects, the most striking of which are that in the latter
the hydrocaulus is much stouter and consequently the hydrothecae
on it are far more widely separated at the back, these hydrothecae
are not spaced regularly and not always in pairs, the apophyses
are much stouter and are not separated from the pinnae by a con-
spicuous node, and on the pinnae of the latter species the pairs of
hydrothecae are not placed so closely together.

Thuiaria buski var. tenuissima, n. var.

(Plate VII., Fig. 2.)

Specimens from Island Bay and Bluff, attaining a height of 35
mm., differ sufficiently from the typical form to be ranked as a
distinct variety. The whole hydrophyton is more slender, the

Notes on New Zealand Hydroida.

hydrocaulus being little, if at all, thicker than the pinnae and
Scarcely distinguishable from it in arrangement of the hydro-
thecae. The hydrothecae are somewhat smaller and the pairs are
not so closely approximated. The hydrocaulus and pinnae being
narrower, the hydrothecae are not "so widely separated at the
back. The pinnae are much shorter than those of the average
specimen of the typical form, and the node between pinna and
apophyse is more oblique. The apophyses are much longer and
more slender. To the eye this variety closely resembles Allman's
natural size figure of T: buski (Allman, 1876, pi. xiv., fig. 3).

Gonothecae not present.

Thuiaria spiralis, n. sp.

(Plate VII., Figs. 3- 3e.)

Hydrocaulus attaining a height of 16 cm., arranged in a loose
but fairly regular spiral, sparingly branched, pinnate. Stem thick,
fistulous, divided by slightly oblique joints into internodes, each
bearing from 1 to 5 pairs of hydrothecae. Pinnae quite irre-
gular, occasionally opposite, stout, divergent usually at an angle
of 60° or more, borne on short stout apophyses from which they
are separated by a conspicuous oblique node ; pinnae themselves
giving rise to secondary pinnae borne on similar but somewhat
more slender apophyses ; pinnae and secondary pinnae stout,
usually divided by slightly oblique nodes into internodes, each
bearing from 1 to 7 (sometimes up to 11) pairs of hydrothecae,
but sometimes undivided.

Hydrothecae in pairs, strictly opposite, in contact in front,
widely separated at the back, especially those on the hydrocaulus,
most of their length vertical, upper portion turned outward and
forward, and narrowed; aperture vertical, widened laterally, with
two lateral lobes; edge of peristome thin, especially on the adcau-
line side in the sinus between the lateral lobes, slightly thickened
just inside the abcauline side, but with no well developed internal
denticle.

Pairs of hydrothecae fairly closely approximated on the pinnae,
but widely separated on the hydrocaulus.

Colour of perisarc, dark brown.

Gonothecae borne on the front of the pinnae, near the base of
same, large, ovoid, about 2*5 to 2-7 mm. in length, and 1*1 to 1-3
mm. in diameter; aperture round, entire, on a very short neck,
scattered, vertically flattened, irregular denticles sometimes pro-
jecting into the interior round the neck, but not always present.

The spiral habit and dark colour of this species at once makes
it easily distinguished from T. buski and T. bicalycula , to which
it is allied. It also differs from the latter in the regular arrange-
ment of the hydrothecae on the hydrocaulus, and the oblique joints
between the pinnae and the apophyses. The hydrothecae on the
pinnae are not so closely approximated as in T. buski.

R. E. Trebilcock :

The pinnae, following the twisting of the hydrocaulus, are given
off in all directions, and do not lie in a single plane. The cauline
hydrothecae also spirally follow the twisting of the hydrocaulus.

Sertularia episcopus (Allman).

I found specimens of this species growing profusely over algae
washed ashore at Island Bay.

Sertularia fasciculata (Kirchenpauer).

I collected a specimen of this species at Island Bay, and another
at Bluff.

Sertularia bispinosa (Gray).

A specimen of this species is in the collection of the late Mr. J.
F. Mulder, but the locality is not stated. The gonothecae contain
a ring of tiny internal denticles. One of the gonothecae is totally
destitute of “ shoulders.”

Sertularia trispinosa Coughtrey.

(Plate V., Fig. 3.)

I found numerous specimens, with gonothecae, at St. Clair
(Dunedin), and Bluff.

Attention has been drawn (Mulder and Trebilcock, 1914, p. 38)
to the presence of a tiny aperture, from which sometimes pro-
trudes a short and delicate tube, in the perisarc of the infrathecal
chamber of S. minima, S. minuta, and allied species. This aper-
ture is also found in a similar position in S', trispinosa, but in no
case can I find any trace of a tube. In this species I have noticed
protruding from the aperture a small mass of ( ?) protoplasm,
but whether it is a sarcostyle or not 1 am unable to determine, as
the soft parts of my specimens of this species are not in suffi-
ciently good state of preservation.

In most cases in .S', trispinosa these apertures are missing, and,
when present, I have not found them paired.

Stechow treats these structures as nematophores, and creates
a^new genus Ncmclla for the reception of the species possessing

Sertularia trispinosa var. inarmata, nov.

(Plate V„ Fig. 4.)

A specimen, collected by me at Island Bay, having a large num-
ber of shoots, differs from the typical form in its gonothecae
which are totally destitute of “ horns ” or even “ shoulders. 1 ’ At
first I felt disposed to treat it as merely an accidental variation,
especially as in some instances there is a slight irregularity in
outline of the gonothecae at the spot usually occupied by ’ the
“ horns.”

Notes on New Zealand Hydroida .

However, after examining the whole of the gonothecae, of
which there are a considerable number, and among which I find
no exception, I have come to the conclusion that the difference is
sufficient to warrant this form being named as a distinct variety.

In the trophosome the variety does not differ in any respect
from the typical form.

Sertularia minima Bale.

(Plate VII., Figs. 5, 5a.)

I have specimens of this species from Island Bay (Welling-
ton), St. Clair (Dunedin), and Bluff. In specimens from all
three localities the tiny apertures and tubes are found springing
from the infrathecal chambers. In many of the specimens from
St. Clair two apertures and tubes are found instead of one, and
similar structures are also found on the hydrorhiza, but in the
latter case the tubes are much longer.

Two apertures, with tubes, are also found in some of the Bluff
specimens. In the latter the gonothecae sometimes have and
sometimes are without the internal submarginal denticles.

The Island Bay specimens belong to the variety pumiloides.

Sertularia divergens Busk.

A few specimens of this species were growing on an " oyster ”
shell from Stewart Island. “ Tridentata xantha ” Stechow (1923a,
p. 64; 1925, p. 236, fig.) does not belong to this species, but is
probably a young form of Sertularia unguiculata Busk.

Sertularia unguiculata Busk.

I collected specimens of this species at St. Clair and Bluff,
those from the latter locality having gonothecae. They do not
differ in any respect from the average Victoria specimen.

Dynamena quadridentata (Ellis and Solander).

A few fragmentary specimens of this species from “ oyster ”
shells from Stewart Island are in the collection of the late Mr. J.
F. Mulder. Not hitherto recorded from New Zealand.

Stereotiieca elongata (Lamouroux).

Bale (1924) states that “specimens from Lyttelton, in Profes-
sor Chilton’s collection, do not differ in any respect from the
small form abundant on the southern Australian coast.” My
New Zealand specimens, on the contrary, which I collected at
Island Bay, St. Clair (Dunedin) and Bluff, belong to the larger
variety, and most of them are more robust than the average large
southern Australian specimens, and usually branch more freely.

li. E . Trebilcoclc :

Fam. PLUMULARIIDAE.

Plumularia pulchella Bale.

I collected a single specimen of this small species, with gono-
thecae, growing on Stereotheca elongata , at Bluff. It differs in no
respect from the form usually found on the Victorian coast. As
in the Victorian specimens, the gonothecae are of two sizes, one
about twice the length of the other. Possibly they are of different
sexes, but in the absence of gonangial contents f am unable defi-
nitely to decide that point. This species has not hitherto been
recorded from New Zealand.

Plumularia setacea (Linn.).

I collected numerous specimens of this species at St. Clair and
Island Bay.

Plumularia setaceoides Bale.

I collected numerous specimens of this species, with gonothecae,
at Island Bay, St. Clair (Dunedin), and Bluff. It has not
hitherto been recorded from New Zealand. The specimens do
not differ materially from the average specimen from Victoria.

I have considerable doubt whether Plumularia unlsoni Bale
(1926), (=P. delicatula Bale, not Busk, not Quelch) is specific-
ally distinct from P. setaceoides, but must examine further speci-
mens of P. wttsoni in a well preserved condition before coming
to a definite conclusion. Some of my specimens from Island Bav,
which had been washed ashore and dried, cannot be distinguished
from the last named species, though others were undoubtedly P.
setaceoides.

Plumularia iiyalina Bale.

(Plate VI., Fig. 6.)

I collected specimens of this species at St. Clair, Island Bay,
and Bluff. It has not hitherto been recorded from New Zealand.

This species has always been looked upon as possessing pinnae
each beai ing essentially a single hydrotheca only, and would thus
be placed in Nutting’s genus Monotheca. The better opinion
seems to be that the retention of this genus is not warranted. In
specimens collected at St Clair and Island Bay I find an addi-
tional argument in favour of this view. Several of the pinnae
bear two hydrothecae each, and are divided, like a typical Plumu-
laria, into alternate long and short internodes, the former each
bearing a hydrotheca, one median inferior and two lateral superior
sarcothecae, and the latter each bearing a single median sarcotheca
only.

1 he retention of the genus Monotheca would doubtless be very
convenient, but the existence of forms such as the above is a
strong argument against it.

Notes on New Zealand Hydroida .

Thecocaulus minutus, n. sp.

(Plate VII.,, Figs. 6, 6a.)

Hydrocaulus attaining a height of about 5 mm., not fascicled,
unbranched, lower part usually destitute of appendages, remainder
divided into alternate hvdrothecate and non-hydrothecate inter-
nodes, the latter usually short. Pinnae, the proximal two usually
opposite, the remainder alternate, bearing from one to three
hydrothecae, divided into alternately long and short internodes of
which only the former bear hydrothecae.

Hydrothecae free for two-thirds of their length, campanulate,
longer than broad, broad at base, slightly constricted at the rear
near the margin; margin smooth, circular slightly everted at the
rear.

Sarcothecae bithalmic, canaliculate, narrow at base, one median
below each hydrotheca, and a pair of laterals above it, one median
on each intermediate internode of stem and pinnae, and sometimes
one median above the caulinary hydrothecae, on the upper part of
the hydrothecate internodes.

The pinnae are each borne on a prominent apophyse, which
springs from beside each caulinary hydrotheca.

The first internode, and sometimes the second, are short, and
bear no appendages. The intermediate internodes, both on the
stem and pinnae, vary in length, but are usually short. Some-
times on the stem two intermediate internodes are found in suc-
cession.

The joint above each intermediate internode, both on stem and
pinnae, is oblique.

The lateral sarcothecae rise to about the level of the margin of
the hydrotheca. Gonosome, not present.

Locality. — St. Clair (Dunedin).

Thecocaulus heterogona Bale.

Mr. Bale has kindly sent me a specimen of this interesting
species. In the axil at the back of each hvdrotheca there is a sar-
costyle protected by an extremely delicate, monothalmic, rudimen-
tary, bract-like sarcotheca, shaped something like the terminal
half of the bowl of a spoon. These sarcothecae are difficult to
detect anywhere, but particularly so on the pinnae; in fact, I
could distinguish them there in only a few instances.

Aglaophenia acanthostoma Allman.

I collected several specimens of this species at Bluff and St.
Clair (Dunedin).

Aglaophenia lax a Allman.

(Plate V., Figs. 5-5 b.)

I collected numerous specimens of this species with corbulae at
Island Bay. The largest specimen attains a height of nearly 60

R. E . Trebilcock:

mm. In my specimens the teeth of the hydrothecae are more
rounded than shown by Bale (1924, p. 260, fig. 15).

There are two forms of this species represented in my collec-
tion, one with the hydrocladia lax, there being about 26 to the
cm., the other with close-set hydrocladia, as many as 46 to the cm.
In general appearance these two forms are so different as to lead
one at first to the conclusion that they belong to different species,
but, apart from the distance between hydrocladia I can detect no
difference between them, and, as the length of the hydrocladial
internodes varies in some specimens, there is little doubt that
examination of a large number of specimens would reveal inter-
mediate forms connecting these two extremes.

Bale's conjecture that the gonosome, when found, would prove
to be of the same character as in A. acanthocarpa and A. divari-
cata I find to be correct.

Thecocarpus formosus (Busk) var. inarmatus, nov.

(Plate V., Figs. 6, 6a.)

I was fortunate in collecting a number of specimens of this
apparently rare species at Island Bay. M. Billard, to whom I sub-
mitted specimens, has kindly compared them with his specimens
from Madagascar, and writes me as follows : —

Malgre des differences, je crois qiPH s’agit de la merae espece.
Dans vos eclianti lions, seule la dent laterale adcaulinaire est bifur-
quee a tel point meme quelle npparait com me en formant deux; lea
deux autres situ^es du cote de la median© ne le sont pas du tout.
Jo dois dire quo men dessin represente un cas extreme et que
dans tout-es les hvdrotli&ques les dents laterales voisines de la
median© ne sont pas toujours aussi francliement bifurquees; le
dessin de Marlctanner donne un cas intermedia ire entre ce qui
exist© cliez mes exemplaires africains et vos exemplaires neo-
y. ela-ndais . Dans ceux-oi j’ai note le plus faible developpment du
repli intrathecal et du processus sp ini forme median. Dans les
echantillons que vous possedez , les liydroclades sont-ils terminea
par une epine ayant a sa base une dactylotlieque? Je n’ai pas
observe ce detail dans les specimens que j J ai re<jus. Il-y^aurait
lieu je crois de fa ire de la forme neo-zelandaise une variete dis-
tincte."

Billard's letter leaves little for me to add in the description of
this variety. In none of my New Zealand specimens are the
hydrocladia terminated in a spine.

The median spiniform processes of the hydrothecae are hollow,,
and vary considerably in size.

Gonosome, not present.

REFERENCES.

Allman, G. J., 1876. Diagnoses of New Genera and Species
of Hydroida. Journ. Linn. Soc. } Zool., xii.

Bale, W. M., 1893. Further Notes on Australian Hydroids,
with Descriptions of Some New Species. Proc. Roy ...
Soc. Vic., n.s., vi.

Notes on New Zealand Hydroida .

, 1914. Further Notes on Australian Hydroids,.

III. Ibid., xxvii.

, 1915. Report on the Hydroida collected in the

Great Australian Bight, and other Localities, III. Biol.
Results “ Endeavour ” iii.

, 1924. Report on some Hydroids from the New

Zealand Coast, with Notes on New Zealand Hydroids
generally, supplementing Farquhar’s List. Trans. N.Z ..
Inst., lv.

, 1926. Further Notes on Australian Hydroids, V..

Proc. Roy. Soc. Vic., n.s., xxxviii.

Billard, A., 1925. Les Hydro’ides de FExpedition du Siboga, II.,,
Synthecidae et Sertularidae.

Coughtrey, M., 1874. Notes on the New Zealand Hydroideae.
Trans . N.Z. Inst., vii.

• , 1875. Critical Notes on the New Zealand:

Hydroida. Ibid., viii.

Fraser, C. McLean, 1911. The Hydroids of the West Coast of
North America. Bull . Lab. Nat. Hist., State Univ .
Iowa.

- , 1912. Endocrypta huntsmani. Science, xxv. (New

York).

, 1913. Hydroids from Vancouver Island. Canada

Gcol. Survey, Victoria Memorial Mus., Bull. No. I, pt ...
xv.

* , 1914. Some Hydroids of the Vancouver Island

Region. Trans. Roy. Soc. Canada , [3], viii.

Hutton, F. W., 1872. On the New Zealand Sertularians. Trans.
N.Z. Inst., v.

Kirk, H. B., 1915. On Ascidioclava, a new genus of Gymnoblastic
Hydroids. Ibid., xlvii.

Mulder, J. F., and Trebilcock, R. E., 1914. Victorian Hydroida,
with Description of New Species: Part III. Geelong
Naturalist , [2] , vi. (1).

Stechow. E., 1923. Zur Kenntnis der Plydroidenfauna des Mit-
telmeeres, Amerikas und anderer Gebiete. II. Tiel, Zool.
Jahrb., 1923,

, 1923a. Diagnosen neuer Hydroiden aus Austra-

lien. Zool. Anz., lix.

EXPLANATION OF PLATES.

Plate I.

Fig. 1. — Saaba (?) wardens, n. sp. Xl5.

la. Gonangium (?) of same. Xl5.

Fig. 2. — Ortho pyxis formosa, n. sp.

2. Showing flanged hydrorhiza, renovation of stem, and
broad view of hydrotheca. X20.

2a. Another hydrotheca, broad view. X20.

i?. E. Trebilcock :

2b. The same, narrow view. X20.

2c. A large hydrotheca, with renovated margin. X20.
2d. A thin walled specimen, showing hydranth. X20.

2e . Transverse section through perisarc of stem, show-
ing flattening and thickening. X20.

Fig. 3. — Sertul arella rentoni Bartlett. X70.

Fig. 4. — Sertularella macrogona, n. sp.

4. Typical specimen. X20.

4a. Front view. X70.

4b. Side view. X70.

4c. Hydrotheca, showing operculum. X70.

4 d. Gonotheca. X15. ;

Fig. 5. — Sertularella pro c era, n. sp.

5. Part of typical colony. Xi«

5a. Part of pinna. X35.

5b. The same showing fenestrae. X35.

5c. A hydrotheca from the hydrocaulus. X35.

5 d. Gonotheca. X15.

Plate II.

Fig. 1. — Ortho pyxis delicata , n. sp.

1-1 d. Hvdrothecae, showing variation in size. X50.
la. Showing renovation of margin. X50.

le. Gonangium. X15.

l f. Hydrotheca, viewed from above. X50.

Fig. 2. — Perisiphonia quadriseriata, n. sp.

2. Complete specimen. Xi.

2a. Part of hydrocaulus and of one hydrocladium and
the base of another, dissected after maceration in
liquor potassi, showing part of three peripheral
tubes. X35.

2b. Part of hydrocladium, broad view. X35.

2c. The same, narrow view. X35.

2d. Three sarcothecae. XI 10.

Plate III.

Fig. 1. — Halecium corm gatissimum, n. sp.

1. Complete colony, showing method of branching.
XI 5.

la. Two hydrothecae. X70.

lb, c. Hydrothecae, showing reduplication of hvdrothecae.

X70.

l d. Young hydrotheca. X70.

le. Hydrotheca and gonotheca. X70.

1/. Hydrotheca, XI 10, to show thinness of perisarc.
Fig. 2. — Halecium expansum, n. sp.

2. Complete colony, showing method of branching, with
hydranths, one of them young. X15.

Notes on New Zealand Hydroida.

2a. The same species, showing normal method of:

branching. X50.

2b, c. Hydrothecae. X HO.

Fig. 3 . — Halecium lenticular e, n. sp.

3. Complete colony, showing method of branching and
position of gonothecae, the proximal one of which
has been twisted on its stalk to show broad aspect.

X15.

3a. An example of vigorous reduplication of hydrotheca.
X15.

3b, c. Examples of irregular branching. X35.

3d. Male gonotheca. X15.

Plate IV.

Fig. 1 . — Halecium lenticular e, n. sp.

1. Specimen from Island Bay. X35.

la. Specimen from St. Clair. X50.

lb. Specimen from Bluff, showing delicate diaphragm,,
and reduplication of margin of hydrotheca. XllO.

Fig. 2 . — Halecium expansum, n. sp.

2. Distal part of a colony, with hydranths, showing
crowded state of hydrothecae. X50.

2a, Another example of active branching. X50.

2b. An example of unusually active renovation of hydro-
thecae. X 50.

Fig. 3 . — Campanulina humilis Bale.

3. Simple form. X50.

3a, b. Branched form, showing hydranths. X50.

3c. Base of four tentacles, showing web. X360.

3c/. Tip of tentacle, showing nematocysts. X360.

Fig. 4 . — Selaginopsis monilifera (Hutton). X15.

4a. A young hydrotheca of same, showing three-toothed
margin. X35.

Plate V.

Fig. 1 . — Sertidarella irregularis, n. sp. X50.

1 a,b. Gonothecae of same. X35.

Fig. 2 . — Sertidarella fusca , n. sp. X 35.

2a. Another view of same. X35.

2b. Hydrotheca. X50.

Fig. 3. Sertularia trispinosa Coughtrey, showing pore (a) in
infrathecal chamber. X50.

Fig. 4. Sertularia trispinosa var. inarmata, n. var. Gonothecae
X15.

Fig. 5 .—Aglaophenia laxa. Allman. XHO.

5a. Hydrotheca, viewed from the front. XllO.

5b. Adcauline part of top of hydrotheca. X200.

R. E. Trebilcock:

Fig. 6. — Thecocarpus formosus (Busk) var. inarmatus , nov.
XI 10.

6a. Front view of hydrotheca. XHO.

Plate VI.

Fig 1. — Sertularella simplex Hutton. (Loc. Auckland.)

1. la. Specimens taken from same hydrorhiza. Xl5.
lb. Typical hydrotheca. X35.

1 c. Single hydrotheca growing from hydrorhiza. X35.
Id. Mouth of hydrotheca viewed from above, showing
operculum, internal denticles, and position of (a)
superior, (6) inferior, and ( c and d) lateral teeth.
X50.

Fig. 2. — Sertularella simplex Hutton. (Loc. Dunedin.)

2. Part of hydrocaulus, showing unusual anastomosis.
X35.

2a. Hydrotheca from specimen with unusually massive
walls. X35.

2 b-e. Gonothecae, showing extent of variation. X35.
Fig. 3. — Sertularella robust a Coughtrey.

3. Typical hydrotheca from Island Bay. X35.

3 a. Typical hydrotheca from Bluff. X35.

3b. Single hydrotheca growing from hydrorhiza, front
view. (Loc. Bluff.) X35.

3c. Gonotheca. (Loc. Island Bay.) X15.

Fig. 4. — Sertularella robust a var. quasiplana, n. var. (Loc. Island
Bay.) X15.

4. Typical hydrotheca. X35.

4a. Gonotheca. X15.

Fig. 5. — Sertularella robust a var. flucticalata, n. var.

5. Branched form with gonothecae. X20.

5a. Typical hydrothecae. X42.

Fig. 6. — Plumularia hyalina Bale, showing hydrocladium bearing
more than one hydrotheca. X50.

Plate VII.

Fig. 1. — Thuiaria buski (Allman). (Loc. Island Bay.)

1. Part of average pinna. X15.

la. Pinna without joints, side view. Xl5.

lb. Part of pinna with two kinds of nodes, side view.

X15.

lc. Part of pinna showing both series of hydrothecae.

X15.

Fig. 2. — Thuiaria buski v ar. tenuissima , n. var. X 15.

Fig. 3. — Thuiaria spiralis , n. sp.

3. Distal half of type specimen. Nat. size.

3a. Part of hydrocaulus showing spiral arrangement of
hydrotliecae. (The hydrocaulus has been straight-
ened in the drawing to economize space.) X15.

Proc. E.S. Victoria, 1928. Plate I.

RK.T. del.

Proc. R.S. Victoria, 1928. Plate II.

R.E.T. del.

■

Proc. R.S. Victoria, 1928. Plate III.

R.E.T. del.

Proc. K-.S. Victoria, 1928. Plate IV.

R.E.T. del.

Proc. R.S. Victoria, 1928. Plate V.

E.E.T. del.

Proc. R.S. Victoria, 1928. Plate VI.

R.E.T. del.

Proc. R.S. Victoria, 1928. Plate VII.

R.E.T. del.

Notes on New Zealand Hydroida.

Fig. 4. — Thuiaria farquhari Bale, gonotheca. X15.

Fig. 5. — Sertularia minima Bale.

5. Infrathecal chamber, showing a pair of pores, with

tubes. XllO.

5a. Side of another specimen, showing two pores on one
side. XllO.

Fig. 6. — Thecocaulus minutus , n. sp.

6. Hydrocladium. X50.

6 a. Front view of part of hydrocladium. X50.

Fig. 7. — Sertularella snbarticulata (Coughtrey).

7. Small complete colony. Xi-
7a. Hydrothecae. X50.

7b. Gonotheca. X15.

[Proc. Roy. Soc. Victoria, 41 (N.S.), Pt. I., 1928.]

Art. II . — The Diurnal and Annual Fluctuations of
Temperature in the Interior of a large Tree.

By A. O. BARRETT.

(Read 12th April, 1928; issued separately 27th September, 1928.)

In the autumn of 1926 it became necessary for me to find out
the temperature of the earth in basements of stone, concrete,,
asphaltum, wood, and the like.

It occurred to me as to whether the trunk of a living tree has a
temperature different from that of the surrounding air. Is the
temperature of the heart wood different from that of the half-
formed surface timber where growth and respiration are active?
Does the translocation of food-materials produce heat? Is this
heat (if any is found) neutralised by the ascending soil water?

The changes in the internal temperature of an inanimate object
always lag behind those of the air outside, but if one continues
recording the daily rise and fall long enough — for twelve months
— the temperature of a column of iron or stone will average that
of its surrounding air.

The choice of a tree fell on a specimen of Pinus canariensis in
my garden at “ Lalbert,” Armadale, Victoria, which grew on a
lawn amidst other trees, but whose trunk was surrounded by a
dense hedge of Coprosma 6 ft. high and 3 ft. thick, which left
inside a space 3 ft. wide where one could walk. The shade of the
other trees, the branches above, and this hedge, constituted an
effective screen between the sun and the trunk; therefore the
sun's rays could not shine directly on to the trunk at any time,
and only air of shade temperature could ever reach the trunk,
conditions which one usually finds in a forest or wood. The girth
of the trunk was 12 feet 6 inches at three feet from the ground
on 10/7/26, and a year later had increased by an inch. The old
dead corrugated bark is about three inches thick. The spread of
the branches above is about 70 feet, and its height is 50 feet.

On 10th July, 1926, I took an auger 5/8 in. diameter and 2 ft.
long, and having cut a circular cavity in the dead bark 3 in.
diameter just down to the living wood, "bored a hole 23 in. long to
the centre of the trunk, parallel to the earth, from north to south,
at a height of 3 ft. from the soil ; also another similar hole at a
tangent to the circumference of the living sap wood, so that the
thermometer would be totally enclosed, and so that its bulb would
be some 4 in. inside from the surface of the dead corky bark, and
about 1 in. into the living outside ring. This hole ended 3 in.
from the beginning of the core hole, and was at the same level,,
and was bored from the S.E. to the N.W.

Fluctuations of Temperature in a large tree.

Now when one bores a hole into a living tree, heat is engen-
dered by the act ; also the cells of wood arotmd the hole begin to
flow with sap, consequently the temprrafure of the first few days
recorded by a thermometer is not normal, until the effects of the
friction and injury have faded away.

The holes being prepared, thermometers (fitted with rubber
corks) which recorded exactly similar temperatures as did my
maximum and minimum thermometer at 53°F., at 60° and at 72°,
were inserted — one in each hole. The thermometer for the core
hole was wired to a skewer of hard wood, whose outer end pro-
truded 1 in. from the rubber cork, in order to facilitate removal
for observation ; both corks exactly fitted the holes, and by cutting
a niche in the corks one could always pull them out so that the
column of mercury was uppermost, and put them back the same
way. This enables one to see the position of the column instantly
on withdrawal, and to read the temperature accurately, even if
the column moves. The whole operation of withdrawal, reading
and replacement takes only four to five seconds, after one becomes
accustomed to it. The maximum and minimum thermometer was
hung 1 in. away from the bark at the spot where the bulb of the
bark thermometer was, and over the hole of the core thermometer.
It is useless to take the temperature of a tree trunk in one part
of a forest unless one records the temperature of the air at the
same spot.

The temperatures were taken at sundown, but in the warm
weather the minimum was read in the morning, and the maximum
in the evening of the same day. In the winter the temperatures
in my garden were very similar to those issued daily in the
Argus, but in the summer the temperatures in the garden were
much lower than those of the bureau. Having proceeded thus
for about one month, and shown the idea to Professor Ewart, he
encouraged me to continue for at least twelve months. With few
exceptions, due to absence, the recording of temperatures pro-
ceeded daily for twelve months, and to facilitate the summarising
of the results they are presented in graph form. From perusal of
the results it will be seen that: —

(a) The mean annual temperature of the heart wood was
1°F. lower than that of the air.

(b) The mean annual temperature of the alburnum was
1T°F. higher than that of the air.

( c ) The mean annual temperature of the duramen and
alburnum combined was the same as the mean tempera-
ture of the air.

Further observations showed that although the trunk was
shaded, the average temperature of the alburnum on the north side
was one degree higher than that on the south side, and hence the
average temperature of the centre of the tree was not more than
half a degree lower than the mean of the two sides. 1 his differ-
ence 1 suggest is due to the fact that the average temperature of

A . 0. Barrett :

the air on the south side of this tree was 1° lower than that on the
north side. The flow of heated air in Australia is from the north
and the flow of cold air is fi^un the south, though it has also been
suggested that the ascending water stream cuts off a small fraction
of the external heat of the air from the duramen, which otherwise

DECREES

FAHRENHEIT

Pki. 1. I emperature of Air and of a Pin us cana-rien&is throughout yoar IG1G-17.

Average mean temperature of air for year = 58*7 Fahr.

Average mean temperature of bark and core = 58 7 Falir.

I runk 12 -C girth at 3 from ground, completely shaded by dense hedge of Coprosma

behaves in regard to external temperature variations as an inert
mass having no appreciable production of heat of its own.

During t lie whole period of observation there did not occur at
any time any positive indication of any alteration in temperature
due to pollination, or the sudden bursting into growth of the
needle buds or the growth of branches, nor did sudden drenching
with rain produce the fall of temperature one would expect. On
March 18th, 192/, the bark on the trunk became soaked with rain,
and the temperatures rose 0-5°. This happened again on April
12th, 1927. I suggest this is due to the rushing of the water by

Fluctuations of Temperature in a large tree.

capillarity into the vesicular tissue of dry, dead bark, developing
heat by friction and chemical action. The rain was colder than
the bark.

As a rule the difference between the maximum and minimum
temperatures did not result in much variation in the heat of the
alburnum at the time of occurrence in this tree. In June, 1927,
owing to a succession of frosts followed by warm afternoons, I
took the opportunity of recording the temperature of the albur-
num frequently, and from this it will be seen that there was only a
difference of 0-5°F. between the temperature at about 6-8 a.m.,
when the ground was white with frost, and the temperature about
4-6 p.m., when by contrast the air was warm and the afternoon
delightfully sunny. 1 suggest this is due in this tree to the fact
that the alburnum is insulated from the air on the outside by a
thick layer of dead bark, which is a bad conductor of heat. The
slowness of the change in temperature in the core of this tree
trunk is illustrated by noting the fact that by the 12th June, 1927,
the average daily mean temperature of the air had fallen 7
degrees owing to frost, whereas two feet inside the tree it
required four days to reduce the temperature by 2°F.

At the same time I became possessed of the idea that the
temperature of this tree was more subject to change from atmos-
pheric causes in the first half of this year than it was in the last
half of last year. I suggest it is due to the fact that the tree is
drained of water due to its spring growth, and that consequently
its specific heat is lower than in the spring.

I have noticed that dead, dry timber seasoned, fluctuates more
than timber of the same dimensions does in a living tree, and con-
sider this is due to its low specific heat, owing to the absence of
“ free ” water. 1

On one occasion the temperature of the core fell faster than
that of the alburnum. This was on and about the 21st to the
26th May, 1927, when the temperature of the core was reduced
by the falling temperature of the air, would have been the tem-
perature of the alburnum, but for the fact that again rain
drenched the bark, and either its condensation in the dead bark
or the fact that it was warmer than the air, caused a rise in tem-
perature, which warmed the alburnum. As soon as the rain
stopped the bark dried and the temperature of the outside of the
tree suffered a quick fall, and on the 26th became lower than the
core.

While taking these temperatures daily, I began to take the tem-
perature of many other varieties of trees and their parts, and
noticed many curious happenings which may be of interest. This
has resulted in the conviction in my mind —

(1) That all dicotyledonous trees average (over long periods)
almost the same temperature as the air of the forest or

1 It has been suggested by Professor Ewart, however, that it is due to the effect of the
transpiration current.

A. 0. Barrett:

locality where they live, although small trees have a
greater daily variation of temperature than large trees,
as they have more bark surface per unit of mass than
the latter.

(2) That trees with smooth bark have a greater daily varia-
tion in temperature than those with thick, corky, or
stringy bark.

(3) That the parts of trees upon which the sun shines have
a greater variation than those in permanent shade, and
that the thinner branches have a greater daily variation
than the trunk.

(4) That the twigs from which the leaves grow vary in tem-
perature hourly.

(5) That the temperature of the smooth barked part of a
branch on eucalypts varies more than that of the
part which — although the same thickness — is nearer the
trunk, and which is covered with stringy or hairy or cor-
rugated bark.

(6) That the average temperature of any part of the trunk of
a large tree shows no evidence of any material average
difference in temperature from that of the atmosphere.
There is always the “ lag,” but the temperature average
is practically the same over long periods.

(7) That all leaves in my garden, whether of Australian or
other origin, which admit of the bulb of a thermometer
being wrapped up in them, are of the same temperature
as that of the air with which they arc surrounded.

(8) That the ascending water current can only influence the
temperature of the trunk in the alburnum or water-con-
ducting wood.

In conclusion I wish to thank Dr. Ewart for his assistance and
for codifying my results; also Messrs, Lang and Mitchell, consult-
ing engineers, for the preparation of the graph.

Fluctuation of Temperature in a large tree ,

Temperature of Core and living sap cells of Pinus Canariensis.

Note — A ll tree temperatures were taken at sundown.

Air Shade
Temperatures

Trek Temperatures

1926

.July 10

Core

54 °F

Bark

54°

Max.

58°

Min .

43°

Remarks.
Auger heat

54.5

Rain

54.5

Av. 21 days 53

54.1

57.5

46.6

Mean air temp*

August 1

52°

Rain

54.5

Av. 16 days

54.1

Mean=51°

54.5

Buds loosening

54.5

Pollen cones

forming

Frosty

53.5

55.5

•y

Av. 15 days

53.4

55.2

58.7

42.5

Mean =50.6 °

A. 0. Barrettt:

1926

September 1

Core

Bari'

56.5

58.5

Av. 15 days

57.3

59.5

63.5

62.5

54.5

55.5

59.5

58.5

56.5

59.5

56.5

Av. 15 days

54.6

60.5

October 1

56.5

58.5

56.5

Av. 14 days

56.8

58.1

58.5

60.5

56.5

Min.

Remarks.

Buds bursting

freely

Pollen cones

ripe

Fruit cones ap-

pear

44.7

Mean=52.9 °

Hot

wind

Pollen cones

empty

Rain

50.3

Mean =60. 3 °

Re -bored hole

centre

Very wet and

windy

Wet and

windy

Fine, S. W.

wind

Gales, rain

Rain

49.5

Mean=57.6 *

Hot

»»

» J

Max.

61.2

70.4

65.8

Fluctuation of Temperature in a large tree .

1926

Core

Bark

October 24

56.5

- 60.5

57.5

- 59.5

57.5

- 62.5

57.5

- 63

57.5

- 62.5

57.5

- 62

- 60

Av. 15 days

56.8

- 60.9

November 1

57.5

- 59

57.5

- 59.5

57.5

- 58.5

57.5

- 57.5

- 56.5

- 60

- 65

- 65.5

- 63

- 60

57.5

- 65

57.5

- 63

57.5

- 63

- 62

Av. 15 days

57.3

- 60.9

- 62

- 61

58.5

- 59

- 61

- 62.5

- 64

- 63

- 61

58.5

- 59.5

58.5

59.5

- 60

59.5

- 67

- 64

- 61

Av. 15 days

- 61.7

December 1

- 62

- 70

60.5

- 67

- 65

60.5

- 63

- 64

61.5

66.5

- 68

61.5

- 64

- 62

- 61.5

- 66

Av. 14 days

60.8

- 64.5

Max.

Min.

Remarks.

- '48

>»

- 48

Cool

- 50

Hot

- 49

- 59

Gales N.W.

- 49

Bleak (rain)

- 40

99 99

73.4

- 49.8 -

Mean=61.6 °

- 55

- 47.5 -

- 50

Cold, Gale

- 45

„ „ hail

- 48

99 99 >,

- 45

99 99 »»

- 47

- 55

N.W. gale

- 55

6 p.m.

- 51

- 47

- 46

Very dry

- 54

- 55

71.4

- 50.3 -

Mean=60.8 °

- 49

- 45

Gales rainy W.

- 48

Showers passed

- 47

Dry conditions

- 52

- 59

- 55

- 54

Cold wind S.W

- 51

Needles ex-

- 53

panding

- 47

- 52

Hot, very dry

- 60

- 53

70.8

- 51.8 -

Mean=61.3

- 54

- 48

- 53

- 62

- 60

15 points rain

- 51

W. Gale

■ 74

- 47

- 57

- 56

Leaves small

- 58

and scanty.

- 48

too dry

- 52

Fine and cool

- 51

- 45

73.8

- 53

Mean=63.4 °

A. 0.

Barrett :

1926

Core

Bark

Max.

Min.

Remarks .

16 points rain

Rain

55 - )

40 points

60.5

Hot

61.5

No dew

61.5

64.5

Av. 15 days

75.6 -

55.6 -

Mean =65.6°

1927

Core

Bark

Max.

Min.

Remarks.

January 1 — 5 |

49 -For period of 6

absent

-days Rain, drizzle

64.5

62.5

►

Hot and dry

63.5

Old needles

102

dropping freely

Av. 9 days

62.7

63.4 -

Mean = 75.7°

64.5

73.5

Cool change

25 points rain

64.5

63.5

Av. 13 days

64.3

75.4 -

54.2 -

Mean = 64.8°

February 1

►

60 mile gale

63.5

64.5

Premature Au-

tumn

Fluctuation of Temperature in a large tree .

1927

Av. 14

March

Core

Baric

Max.

Min.

Remarks.

> 53

No moisture or

dew

No sap flow,

63.5

heat now felt

63.5

at once thro’

the bark, no

63.5

evaporation

to resist heat

: days

63.5

66.4

82.1

56.9

Mean = 69,5 c

63.5

65.5

66.5

63.5

-r+

63.5

A little rain

63.5

1 days

63.7

65.2

75.5

56.3

Mean = 65.9°

Very hot

63.5

Sudden cold

63.5

change at

63.5

noon

63.5

Lawns nearly

63.5

dead, 160

63.5

points since

January 1

days

63.5

65.4

77.2

56.4

Mean = 66.8°

63.5

Rain at last

Trunk of tree
soaking, yet
temp, of outer
rings rose 0.5 °

62.5

F. 70 pts. rain

60.5

Drizzle

61.5

•*.

60.5

61.5

days

61.5

61.3

66.4

52.8

Mean = 59.6°

A . O. Barrett:

1927

April

Av. 15

May

Core

Bark

Max.

Min.

Remarks.

59.5

61.5

59.5

60.5

Max. and min.

-i

for two days

A little rain

59.5

days

60.8

67.3

50.7

Mean =59.0°

59.5

Max. and min.

59.5

-i

for two days

59.5

Hole in bark

closing up, re-

bored

58.5

57.5

58.5

57.5

60.5

Many trees

dying, leaves
falling

> days

58.7

58.5

62.5

48.5

Mean =55.5°

59.5

Nice rain 30

Av. 16

points. Copi-
ous condensa-
tion of mois-
ture on ther-
mometer in

core lately.

57.5

This has not

57.5

occurred be-

fore. None on

the thermom-

59.5

eter in bark

60.5

Rain

58.5

Rain, 50 points

58.5

Dull

58.5

days

57.4

58.6

47.7

Mean =55.8°

Fluctuation of Temperature in a large tree

1927

Core

Bark

Max.

- 64

Min.

Remarks .

Rain, 25 points

57.5

Rain

50.5 -

57.5

Fall of temp.

50.5 -

in core not al-
ways preced-

56.5

ed by fall of

temp, in bark

55.5 -

Rain

55.5 -

53.5

Fine, 26° on

grass, frost
Fine, frost

Ay. 13 days

56.4 -

56.4

58.5

43.6

Mean =51.0°

June 1

54.5 -

Fine, frost

9t ft

J1 tt

53.5 -

Rain

53.5 -

54.5

53.5 -

52.5

51.5

Drizzle & rain

51.5

Frost, fine

52.5 -

»» »»

51.5

51.5 -

51.5

Ay. 15 days

53.1 -

52.8

40.2

Mean =48.6°

51.5 -

50.5

51.5 -

51.5

51.25 -

52.5

51.5 -

Record num-

51.5 -

52.5

ber of frosts

51.5 -

52.5

51.5 -

51.5

Fine

51.5 -

48.5

Fine, very cold

51.5 -

49.5

fog at 8 a.m„
Fine, warmer

51.5 -

50.5

Av. 15 days

51.4 -

51.4

55.3

Mean =48.1°

July 1

It took 4 days

for the succes-
sive waves of

cold to affect

the core

Rain all over

Vic. 50 to 100

points

51.5

Ay. 11 days

56.2

45.7

Mean =50.9°

Barrett : Temper atare in a large tree .

Summary of Average Temperatures.

July

10-31

August

1-16

17-31

Sept.

1-15

16-30

October

1-15

16-31

Nov.

1-15

16-30

Dec.

1-15

16-31

Jan.

1-14

15-31

Feb.

1-15

16-28

Mar.

1-16

17-31

Apl.

1-15

16-30

May

1-16

17-31

June

1-15

16-30

July

1-11

Average

Tree

Core

Bark

54.1

53.4

55.2

57.3

54.6

60.5

56.8

58.1

56.8

60.9

57.3

60.9

59 -

61.7

60.8

64.5

62.7

64.3

63.5

66.4

63.7

65.2

63.5

65.4

61.5

61.3

60.8

58.7

58.5

57.4

58.6

56.4

53.1

52.8

51.4

57.7

59.8

Air

Max.

Min.

57.5

46.6

58.7

42.5

61.2

44.7

70.4

50.3

65.8

49.5

73.4

49.8

71.4

50.3

70.8

51.8

73.8

75.6

55.6

63.4

75.4

54.2

82.1

56.9

75.5

56.3

77.2

56.4

66.4

52.8

67.3

50.7

62.5

48.5

47.7

58.5

43.6

40.2

55.3

56.2

45.7

67.5

49.8

Tree

Air

Mean

53.5

54.3

50.6

55.6

52.9

57.5

60.3

57.4

57.6

58.8

61.6

59.1

60.8

60.3

61.3

62.6

63.4

62.5

65.6

65.3

75.7

65.6

64.8

64.9

69.5

64.4

•

65.9

64.4

66.8

61.4

59.6

60.4

58.6

55.5

55.8

56.4

52.9

48.6

51.4

48.1

51.5

50.9

58.7

^Proc. Roy. ko £. Victoria, 41 (N.S.), Pt. I., 1928.]

Art. III. — Some Trematode Parasites on the Gilts of
Victorian Fishes.

By WINIFRED KENT HUGHES, B.Sc.

(Howitt Research Scholar in Zoology, University of Melbourne.)

(With Plates VIII.-X1.)

[Read 12th April, 1928 ; issued separately 27th September, 1928.]

This work was carried out under the guidance of Dr. O. W.
Tiegs, and my thanks are due to him for his assistance and in-
terest.

Methods. — The material was fixed in 1% formalin. Bouin,
Zenker, and corrosive sublimate were also tried, but the formalin
proved most satisfactory. Iron haematoxylin was used for all
sections, and Erhlich’s haematoxylin for whole specimens. In
cases where the iron haematoxylin overstained the vitellaria in
sections, Erhlich was used as an alternative.

Genus Anchylodiscus Johnston and Tiegs, 1922.

Anctiylodiscus gadopsis, n. sp.

(Plate VIII., Fig. 1 ; PI. X., Fig. 6.)

Found in great numbers on the gills of Gadopsis sp. (River
Black Fish).

Locality. — Campaspe River, Vic.

Formalinised animal measures about CL36 mm. in length,
breadth 0 08 mm.

External Features. — Small body, slightly oval in section, with
indication of head at anterior end; at posterior end is a well-
marked hooked disc. The hooks are arranged in nine pairs,
two consisting of large hooks, the bases of which are slightly
bifurcated, and are connected by a single crossbar, while seven
consist of minor hooks, one pair of which lies across one of the
crossbars. The large hooks are supported by a ring of chitinous
material (PI. VIII., Fig. 1). Immediately in front of the pharynx
there are two pairs of eyes, of which the anterior pair is the
smaller.

Three pairs of u head organs ” are present (PI. VIII., Fig. 1) ;
the cephalic glands are situated slightly anterior to the eyes.

The “ brain ” lies between the eyes, and is the only indication of
the nervous system.

Alimentary Canal. — The mouth is situated vent rally, and is
anterior to the eyes. The pharynx is large, intestine is bilobcd

46 Winifred Kent Hughes:

and devoid of caeca and ends blindly towards the posterior end
of the body.

Reproductive System. — The testis is slightly elongated, lies
dorsal to the ovary ( PI. X., Fig. 6), and extends posterior to it.
'Fhe vas deferens passes forwards dorsally as a very wide tube,
and opens posterior to the pharynx by a chitinous penis, which is
a straight-pointed structure (PI. VIII., Fig. 1).

The ovary is a large median structure, situated about half-way
along the length of the animal. The oviduct passes forwards and
opens to exterior close to the male opening. There is no vagina.
The vitelline system is very large, and occupies the larger part of
the body. The transverse yolk duct passes across anterior to the
ovary and opens into the oviduct.

The egg is enormous, the ripe egg in the oviduct displacing
the organs of the body.

Points by which A. gadopsis is distinguished from A. tandani
T. II. Johnston and O. W. Tiegs (8) :■ —

1. One pair of minor hooks lies across one of the crossbars in a
median position (PI. VIII.. Figs. 1, 2).

2. No vesicula seminalis could be determined. The time of the
season may account for this,

3. The penis is straight instead of curved.

Genus Squalonchocotyle Cerfontaine, 1898.

SQUALONCHOCOTYLE ANTARCTICA, 11. Sp.

(Plate IX., Figs, 4, 5; Text-Fig. 1.)

This marine^ parasite belongs to the sub-family Onchocotylinae
Cerfontaine (7), of the family Octocotylidae van Ben et Hesse
(3). It is very similar to X. vulgaris Cerfontaine (7), and X.
cjrisca Cerfontaine (7). Found on the gills of Mustclus antarcti-
cus.

Locality. — Port Phillip Bay.

Average length of formalinised animal 10 mm., breadth 1 mm.

External Features (PI. IX., Fig 4). — Body elongated and
Fattened dorso-ventrally. At the posterior end is a fixing organ
which is composed of a fixing disc with a caudal appendage. On
the former are three pairs of large suckers arranged in two
parallel rows and each provided with a single large hook. Two
smaller, unarmed suckers are present at the extremity of the
caudal appendage, and a pair of small minor hooks is situated
between them. The body is attached to this organ at the level of
the middle pair of large suckers.

Each large hook ends in a small pointed structure which is re-
curved at right angles, and sharply defined from the main body
of the hook by its smaller diameter. The minor hooks are Y-
shaped, the three arms being more or less equal, and the base of

Trematode Parasites .

the Y ends in a small hook which is recurved so as to point in a
direction parallel to the long axis of the hook (Text-fig. la, lb).

Fig. 1 . — S qu a 1 o n c h. ocoty l e antarctica , n. sp.. drawn with camera
lucida. A, major hook: B, minor hook: (', egg.

Alimentary Canal. — The mouth is situated ventrally, and is
surrounded by a large circular sucker ; a short muscular pharynx
opens into the oesophagus, which divides immediately into the
two main branches of the intestine. These unite at the posterior
end of the body, and pass into the fixing organ. Here the intes-
tine divides again into two single branches passing anteriorly and
posteriorly along the disc and caudal appendage respectively.
Small unbranched caeca are given off both medially and laterally
along the length of the intestine.

Reproductive System (PL IX., Fig, 5). — The reproductive
organs are typical of the sub-family. The common genital
opening is median, and is situated just posterior to the pharynx.
The two vaginal openings are lateral, and on nearly the same level
as the genital opening.

The eggs are oval and narrow, with two short polar filaments.
The length of the egg is approximately 150 /x, which is roughly
twice the length of the filaments (Text-fig. If.).

S. antarctica is distinguished from .S', vulgaris by the following
points : —

1. Shape of buccal sucker.

2. Shape of minor hooks.

3. Length of polar filaments of egg.

.S', antarctica is distinguished from S. grisca bv the following
points : —

1. Structure and shape of large hooks.

2. Position of vaginal openings.

Winifred Kent Hughes :

Macrophylla, n. gen.

MaCROPHYLLA ANTARCTICA, n. gen. et sp.

(Plate X., Figs. 7-9.)

Marine form from the gills of Mustelus antarcticus. Found on
only two specimens of about a hundred examined.

Locality. — Port Phillip Bay.

Length of formalinised animal 13-15 mm., breadth 1 -3-2-5 mm.

External Features (PL X., Fig. 7). — At the anterior end
on either side is a single pair of large Hat expansions, ridged on
their ventral surface. Sections of these structures, examined
under high magnifications, seem to reveal them as glandular in
nature. At the hinder end is a large disc., attached by its middle
to the body of the worm (PI. X., Fig. 7). This disc is provided
with a very large sticker, divided by radii into five secondary
suckers, of which the largest is incompletely divided into three
compartments.

Body Wall (PI. X., Fig. 9). — This consists, so far as could
be made out on the material available, of an epidermis with a
marked cuticle. The musculature consists of a circular layer
divided into three secondary layers, a well defined longitudinal
layer, and vertically running fibres.

Alimentary Canal. — The crescent-shaped mouth opens into an
enormous pharynx, which extends well over half the breadth of
the animal. The pharynx is extremely muscular, and is pro-
vided with large unicellular glands. The intestine is bifurcated,
long and narrow, and extends almost to the posterior end of the
worm. Along its length numerous branching caeca arc devel-
oped.

Reproductive System (PI. X., Fig. 7, 8). — There are two
compact testes situated behind the ovary, about a third of the
length of the animal from the anterior end. The left testis is situ-
ated a little in front of the right. The two vasa deferentia lead
into a common tube which travels to the left of and dorsal to the
ovary. It then passes forwards almost to the level of the repro-
ductive openings, crosses under the vagina, turns hack upon itself
and enters the penis. This is a pear-shaped muscular organ,
which passes 1o the exterior along a narrow duct whose opening
is situated on the side of the animal just behind the left glandular
expansion, and immediately anterior to the uterus.

The ovary is a well-marked, median, round body, considerably
larger than the testes. The oviduct passes forwards as a straight
tube and continues as the uterus to open just behind the male
opening. The vitelline glands extend along either side from the
anterior to almost the posterior end of the body. The two longi-
tudinal yolk ducts open into a transverse duct, which runs an-
terior to the ovary, and opens into a dilated short median yolk-
duct, which in turn opens into the oviduct. The vagina has a

Tt rmatod e Pa > asi tes.

49>

common opening to the exterior with the male duct. It passes
behind into a curious muscular organ (PL X., Fig. 8), which
in turn opens into a slightly convoluted tube, leading backwards
and emptying into the transverse yolk duct. The muscular organ
above referred to, and the proximal part of the convoluted tube
which leads away from it, lie suspended in a cavity, indicating that
they are distensible structures. It is probable that the tube is a
receptaculum seminis, though I have never observed spermatozoa
within it.

In two specimens an egg was present in the uterus ; it is oval in
shape, and measured about 220 /i in length, 107 /t in breadth.
The genus Macrophylla is distinguished from its nearest ally
Tristonmm by the following points; —

1. Two compact testes as contrasted with numerous testicles.

2. Only five instead of seven distinct radii in posterior sucker.

3. Glandular membranes at anterior end in the place of suckers*

Genus Octobothrium Leuckart, 1827.

OcTOBOTHRIUM THYRITES, n. Sp.

(Plate XI., Figs. 10-12 ; Text-Fig. 2.)

Marine form found on gills of Thyrites atun (Barracouta).

Locality. — San Remo, Vic.

Length of formalinised animal 7-8 mm., breadth 2-2-5 mm.

External Features (PL XI., Figs. 10, 11). — Body elongated
and flattened dorso-ventrally, tapering forwards each end; it is
more pointed anteriorly. A distinct 44 head ” region is marked off
anteriorly by the two lateral openings of the vagina which are
surrounded by tumid lips. Posterior end terminates in a fixing
disc which is not sharply defined from the body. Disc is provided
with eight suckers arranged in two rows converging posteriorly,
and each sucker has a complex armature (PL XI., Fig. 11, and
Text-fig. 2a). Two pairs of small . hooks are situated at the
extreme posterior end, the more anterior pair being the larger
(Text-fig. 2c). In the “ neck ” region the body wall is so folded
as to give a false appearance of segmentation. (PL XI., Fig. 10).

Body Wall. — This consists of a syncytium with a marked
cuticle. The musculature consists of a thick layer of well devel-
oped longitudinal fibres, the circular fibres being only poorly
developed.

Alimentary Canal. — The mouth is situated on the ventral side
of the extreme anterior end. On each side of the mouth is a
pair of small oral suckers, -which is characteristic of the group.
The mouth leads into a wide, but not muscular pharynx, opening
into the intestine, which forks immediately posterior to the trans-
verse vaginal duct. Numerous diverticula are given off along the
whole length of the intestine, both medially and laterally.

Winifred Kent Hughes:

Reproductive System (PI. XL, Fig. 12). — The male organ is
represented by numerous testicles occupying the middle of the
posterior 3/5ths of the body. The single vas deferens passes up
medially and dorsal ly as a coiled tube, and opens to the exterior
on the ventral surface of the “ head ” by a sucker which is sur-
rounded by small hooks (PI. XL, Fig. 12, and Text-fig. 2b).

The ovary is lobed and single, situated on the right side about
2/5ths of the total length from the anterior end. In stained speci-
mens a lobed structure (PL XL, Fig. 10, y."). which takes up
the stain very readily, appears in sections to be composed of
young ova. There is some doubt about this, as similar stained
cells extend from this structure to the posterior end of the body.

Fig. 2 . — Octotfothrium tkyrites, n. sp., drawn with camera lucida.
A, armature of posterior sucker ; B, hooks of genital apparatus ; C,
hooks at posterior extremity.

The ovary, however, appears to be continuous with the lobed por-
tion. The oviduct passes forwards dorsally, the shell gland opens
into it immediately posterior and dorsal to the yolk reservoir. It
continues forwards ventrally as the uterus to open posteriorly to
the male genital opening. The vitelline glands extend along
either side from the “ neck ” to the posterior end of the body.
The longitudinal ducts open into a large transverse yolk duct
which opens medially into the yolk reservoir, to the right of
which lies the ovary. A single yolk duct passes down and opens
into the oviduct. Laurer's canal is clearly seen as a narrow tube

Trematode Pa ras i tes .

running from the junction of the oviduct and vitelline duct to the
intestine. The two vaginal openings are connected by a single
transverse duct, from the centre of which a single duct passes
down on the right hand side to open into the double recep-
taculum seminis (Fig. 12). From this a narrow duct passes
down and opens into the transverse yolk duct on the right. By
this arrangement, sperm not used in fertilisation may be passed
into the intestine. In fact, I have observed in sections structures
which closely resemble sperm together with yolk granules in
Laurer’s canal.

History. — Hermann in 1782 was the first to describe a worm of
this group. Then in 1828, Leuckart, and later Kuhn, described
the same species. At first the posterior end was taken for the
anterior end, and vice-versa.

I have followed Bronn’s classification, which gives the follow-
ing synonyms for Octobotkrium : — Dactylocotyle, Choriocotyle,
Pterocotyle , Glossoctyle , Octocotylc, Ophiocotyle and Octosoma.

Of the previously described species I have not been able to ob-
tain the literature describing — 0 . lance datum, 0. sagittatum, and

O. arcuatum.

0. thyrites is distinguished from 0 . thunninae by the following
points : —

1. The arrangement of the suckers on the posterior disc.

2. The posterior disc is not divided from the body by a con-
striction.

3. The intestine is forked below, not above, the sexual open-
ings.

4. The genital armature is distinctive in both species.

Genus Ancyrocephalus Creplin, 1839.

(Syn. Diplectanum Diesing, 1858). (MacCallum 9.)

Ancyrocephalus bassensis, n. sp.

(Plate VIII., Fig. 3.)

Marine form found in great numbers on the gills of Platyce-
-phalus bassensis (Flathead).

Locality. — Port Phillip Bay.

Length varies considerably; formalinised animal, measuring
0 49 mm. to 0-95 mm.; average breadth 0-09 mm.

External Features. — Body elongate and circular in section. At
the anterior end there is a slight indication of a “ head ” region;
at the posterior end is a fixing disc which is not sharply marked
off from the body. The disc is armed with nine pairs of hooks,
consisting of two large pairs, the bases of which are bifurcated
and connected by two cross bars, and seven minor pairs (PL
VIII., Fig. 3).

Winif red Kent Hughes :

Musculature. — This consists of an inner and outer longitudinal
layer, with a circular layer between them. The longitudinal layer
at the posterior end is strongly developed to supply the disc.
Three pairs of head organs and numerous cephalic organs are
present, the latter being arranged laterally and extending from
the anterior pair of eyes to the region posterior to the pharynx.

Alimentary Canal. — The mouth is situated on the ventral sur-
face in the “ head ” region, and opens into a very muscular
pharynx. This opens into a short oesophagus, passing into the
simple forked intestine which ends blindly at the posterior end
of the body, and is devoid of caeca.

Reproductive System. — The position of the testis and ovary
varies according to the degree of contraction of the animal. The
testis is a single large rounded structure taking up the whole of
the diameter of the body and lies immediately posterior to the
ovary. A single vas deferens is given off anteriorly on the left,
and passes forwards ventrally to a large vesicula seminalis, opening
to the exterior by a chitinous penis. A well marked prostate
gland lies to the left of the penis, into the base of which it open9

(PI. VIII., Fig. 3).

The ovary is much smaller, and lies dorsally and anteriorly to
the test is. The oviduct is short, and passes forwards as the uterus
to open to the exterior with the vas deferens at the common
genital opening which is median and ventral in position. The
vagina is very short, and opens on the left just anterior to the
ovary; it connects with the receptaculum seminis which is globu-
lar in shape, before passing immediately into the genital junction.
The vitellaria are well developed, extending along the lateral
margins, and the transverse yolk duct passes across anterior to
the ovary.

I have found no reference to any other description of this
genus occurring on Australian fishes.

Bibliography.

1. Beneden, E. van. Le genre Dactylocotyle, son organisa-

tion, et quelques remarques sur la formation de l’ceuf des
Trematodes. Bull de VAcad. Roy. de Belg., [2], xxv.,
Bruxelles , 1868.

2. Beneden, P. J. van. Note sur V Octobothrium nierlangi,

Bull . de VAcad. Roy de Belg., xxiii. (2), Bruxelles ,
1856.

3. Beneden, P. J, van, et Hesse, C. E. Recherches sur les

Bdellodes et les Trematodes marins. Mem. Acad. Roy.
Belg., xxxiv., 1863.

4. Bronx, H. % G. Klassen und Ordnungen des Thierreich. Tre-

matodes iv. la.

5. Cerfontaine, P. — Contributions a l’etudedes Octocotylides,.

II., Le genre Dactylocotyle. Arch. Biol., xiv., 1896.

Trematode Parasites.

6. Cerfontaine, P. — Contributions a l’etude des Octocotylides,

IV. , Nouvelles observations sur Je genre Dactylocotyle
et description de Dactylocotyle luceae . Arch. Biol., xv.,
1898.

7. Cerfontaine, P. Contributions a l’etude des Octocotylides,

V. , Les Onchocotylinae, Arch. Biol., xvi., 1900.

8. Johnston; T. H., and Tiegs, O. W. New Gyrodactyloid

Trematodes from Australian Fishes, together with a re-
classification of the super-family Gyrodactyloidea. Proc.
Linn. Soc . NSW. , xlvii. (2), 1922.

9. MacCallum, G. A. Some new species of Ectoparasitic

Trematodes, Zoologica , i. (20), 1915.

10. Perona, C., e Perugia, A. Di alcuni tramatodi ectoparassiti

di pesci mar ini. Ann. Mus. civic odi storia natur di
Genova, [2\, vii. (xxvii.), 1889.

11. Pratt, H. S. North American Invertebrates, No. 13, Tre-

matodes, Pt. 1, The Heterocotylea or Monogenetic
forms. Amer. Nat., xxxiv., 1900.

EXPLANATION OF PLATES.

Plate VIII.

Fig. 1. — Anchylodiscus gadopsis , n. sp. Entire animal, ventral
view.

Fig. 2. — A. tandani Johnston and Tiegs. Disc in ventral view
(from T. H. Johnston and O. W. Tiegs (8) ).

Fig. 3 . — Ancyrocephalus bassensis , n. sp. Entire animal, ven-
tral view.

Plate IX.

Fig. 4 . — Squalonchocotyle antarctica , n. sp. Entire animal, ven-
tral view.

'Fig. 5. — X. antarctica , n. sp. Longitudinal section.

Plate X.

Fig. 6 . — Anchylodiscus gadopsis, n. sp. Transverse section,
showing the relative position of ovary and testis.

Fig. 7 . — Macrophylla antarctica, n. gen, et sp. Entire animal.
Fig. 8. — M. antarctica, n. gen. et sp. Reproductive organs,
slightly diagrammatic.

Fig. 9. — M. antarctica, n. gen. et sp. Section through body
wall, highly magnified.

Plate XI.

Fig. 10 . — Octobothrium thyrites, n. sp. Entire animal with ali-
mentary canal drawn on the right side only, and vitel-
laria on the left side.

Winifred Kent Hughes: Trematode Parasites.

Fig. ii. — 0. thyrites, n. sp. Posterior end, highly magnified.
Fig. 12. — O. thyrites, n. sp. Reproductive organs, slightly dia-
grammatic.

Explanation of Lettering.

a.s. Sucker with chitinous armature; b. “brain”; b.c. Buccal
sucker; c. cuticle; c.a. caudal appendage; c.b. cross-bar; c.g.
cephalic glands; c.g.o. common genital opening; c.m. circular
muscle layer; d.m. disc muscle; d.s. double sucker; e. eye; eg.
egg; ep. epidermis; g.o. genital openings; h.o. “head organ”;
i. intestine; i.c. intestinal caecum; l.c. Laurer’s Canal; l.h. major
hook; l.m. longitudinal muscle layer; m. mouth; o. oviduct;
oe. oesophagus; o.s. oral sucker; ov. ovary; p. penis; p.g.
prostate gland; ph. pharynx; p.i. part of intestine; p.s.
posterior sucker; r. radius; r.s. receptaculum seminis; s. sucker;
s.g. shell gland; t. testis; t.s. testicles; t.y.d. transverse yolk duct;
u. uterus ; v. vitellaria ; v.d. vas deferens ; vg. vagina ; vg.o. vagi-
nal opening: v.s. vesicula seminalis; y.g. yolk glands; y.o. young
ova; y.r. yolk reservoir.

Proe. R.S. Victoria, 1928. Plate VIII.

W.K.H. del.

Proc. R.S. Victoria, 1928. Plate IX

W.K.H. del.

Proc. R.S. Victoria, 1928.

Plate X

W.K.1JL del,

Proc. P.S. Victoria, 1928. Plate XT.

W.K.H. del.

[Proc. Roy. So-. Victoria, 41 (N.S.), Pt. I., 1928.J

Art. IV . — Note on ike Reflection of 'X-rays from Glass

and Quartz .

R. BINGHAM, M.Sc., T. H. LABY, Sc.D., and T. SHEARER.

M.Sc.

(Natural Philosophy Laboratory, University of Melbourne.)

(Read 12th April, 1928; issued separately 27th September, 1928.)

The reflection of X-rays has been observed by the authors for
glancing 1 angles up to 45° from glass and 40° from quartz, the
critical or limiting angle not having been observed. The X-rays
incident on the glass or quartz are, it is believed, heterogeneous

X-rays of wave-length in the region of about 50 A.U. The
most intense wave-length in the incident radiation is probably the

K« line of carbon of 45 A.U. wave-length.

That X-rays incident at angles of less than 1° are reflected
by glass and by metals is now a well-known phenomenon (1), but

that a wave-length of about 50 A.U. would be reflected when inci-
dent at about 40° is not to be expected from the observations
already made. Hoi week claims to have observed the reflection of
long X-rays from polished bronze at 11*7° and 16-2° (2).

Dauvillier (3) remarks that he observed in this region (50 A.U.)
optical reflection from glass and from a thin film of melisstc acid
on lead.

Assuming that we have correctly interpreted the preliminary
observations which we have made, it can be concluded that X-rays
can be reflected from spherical surfaces and brought to a focus,
which will make new methods for the study of long X-rays pos-
sible.

Experiments.

The apparatus used in our experiments is a vacuum spectro-
graph in which the crystal has been replaced by a piece of opti-
cally flat plate glass or quartz which acts as the reflector of the
X-rays. The target, Wehnelt cathode, the reflector, the camera
and photographic film are in the same vaccum. 2 The X-rays

J* angles of incidence and reflection stated are glancing angles.

“• 7k ,®P ec ^rograph in the use of a common vacuum for the “ X-ray
spectrograph is similar to that described bv Shearer, Phil.
Mag., Oct., 1927.

Bingham , Laby and Shearer:

emitted by the target pass through a circular hole in a shield and
then fall on the reflector. The reflected rays pass through a slit
into a camera and fall on a Schumann film. Copper and carbon
targets have been used. The difference of potential between the
cathode and the copper target was at first up to 10,000 volt recti-
fied A.C. In later experiments, the P.D. between the carbon
target and the Wehnelt cathode has been about 375 volt given by
a battery.

The photographic film on development shows a slit image of
the focal spot on the target. This slit image is the same whether
it is taken direct, without the glass or quartz reflector, or with the
reflector, except that in the latter case there is lateral inversion of
the image. The angle of incidence is accurately equal to the angle
of reflection. Using the copper target reflection up to an angle
of 29° from glass was observed, and with the carbon target rays
reflected at 45° from glass and 40° from quartz have been
photographed. In each case, this angle is the largest attainable
with the spectrometer. An exposure of 1200 milliampere second
and a potential difference of 375 volt gives a well-defined image
of the focal spot on Schumann film.

Estimates of intensity with the Schumann film are difficult to
make. The ratio of intensity of the reflected beam to that of the
incident beam in the case of the graphite target is of the order of
\ up to 30° glancing angle. At 40° incidence this ratio had con-
siderably diminished.

The evidence that the radiation is optically reflected appears to
be conclusive.

What is the nature of the radiation which is reflected?

As the photographic film is enclosed and placed opposite the
slit (0-05 mm. wide) in a metal box which is at the potential of
the negative end of the filament of the cathode, the rays cannot
be cathode rays.

The radiation —

(a) is emitted, as shown by the slit images, from the same
focal spot as that from which short wave X-rays 3 were
proved to be emitted in other experiments ;

( b ) penetrates aluminium foil 4 0-0006 mm. thick, about 1%
of the incident radiation being transmitted (this foil was
tested and found to absorb visible light) ;

( c ) is absorbed by glass and by fluorite;

( d ) is emitted by a carbon target on which 375 volt electrons
are incident;

3. — That these rays were X-rays was fully verified by wave-length measure-

ment.

4. — In a previous paper by one of the authors (Shearer, F\ M. Vol. IV.,

p. 7 47, 1927), two thicknesses of aluminium foil were not found to be
transparent to X-rays in this region. It should be noted that only
one thickness of foil is penetrated in the observation recorded above,
and all the conditions in these experiments tend to give an incident
beam of increased intensity.

Reflection of X-rays from Glass and Quarts.

(e) is not emitted when the filament is hot, and no potential
is applied between it and the anode;

(/) is in its action on Schumann film approximately propor-
tional to the exposure measured in milliampere second
at constant voltage.

If the radiation is not X-rays emitted according to the usual
laws 5 connecting wave-length with applied potential, it can be
longer in wave-length than is given by those laws; but it would
appear to be very improbable that it is shorter. The observation
(c) above excludes the assumption that the radiation is in the

range of about 8000 down to 1200 A.U. Observation ( b ) ex-
cludes the region longer than 8000 A.U. It remains to consider

the region from about 100 to 1200 A.U. All the evidence from
the observations of Schumann. Lyman, Millikan and Hohveck 6
show that radiation in this region is highly absorbed by all forms
of matter, and thus ( b ) above excludes the Millikan and Lyman
regions of the spectrum. Observation ( e ) confirms that the
radiation is not one emitted by a hot body at a temperature up
to 1200° C., the highest temperature of the filament. The absorp-
tion measurement in ( b ) is consistent with the radiation being of

wave-length from 50 to 80 A.U.. assuming the A 3 law of absorp-
tion. Observation (/) implies that the radiation is produced by
the incidence of electrons on the target.

Taken as. a whole, the evidence strongly supports the view that

X-rays of wave-length about 45 A.U. can be reflected at angles up
to 40° from glass and quartz.

The Lorentz dispersion formula in the form

3=1

= I! ( - 2 — 2 ) = 4-478. 10- 14 X ( '2 J

2tt7U ~ ' 1 ' — i'„ a / i - Vs

where e and m are the charge and mass of the electron, n s is the
number of electrons per unit volume of natural frequency v s
(wave-number iA) and v is the frequency of the incident
radiation (wave-number v), has been shown to give, in the case
of X-rays, positive values of 3 which are confirmed by experi-
ment (1). Total reflection is therefore to be expected for radia-
tion incident on substances for which /* is less than unity at a
glancing angle less than a certain critical value. This was first
observed by Compton (1) from surfaces of glass, silver, lacquer
and calcite.

5.- — These laws have been assumed by Holweck for the range 44 to 300 A.U.
“ De la Lumiere aux Rayons X ”, Chap. III.

— Holweck finds that ft/p for celluloid increases rapidly up to about 300
A.U., becomes a maximum at about 320 A.U., and then rapidly decreases
toward 1200 A.U.

Bingham , Lahy and Shearer ; X-rays.

In the figure the critical glancing angle for glass 0 C is plotted

against wave-length in A.U. of the incident radiation, where 0 G
is obtained from the relation cos 0 c ~i^ and /i is calculated from
the Lorentz formula. n s is obtained from the following data
(assumed) : —

Density of glass : 3 gm./cc.

Composition : 71 %Si0 2

13% CaO
14% Na.O.

Number of electrons per atom:

M N

Si(14) -

4 - —

Na(ll) -

1 - —

Ca(20) -

8 - 2

0(8)

— . —

The following values of X s (

= r) in

A.U. were adopted :

K L

Si(14) •

112

Caf-20) -

Na(ll) -

322

0(8)

717

The results obtained in this paper do not appear to be recon-
cilable with this curve.

We have to thank Messrs. H. Massey and C. Mohr for computing
the data shown in Fig. 1.

REFERENCES.

(1) A. FI. Compton. X-rays and Electrons, Chap. VII.

(2) Holweck. De la Lumiere aux Rayons X, p. 85, Paris*

1927.

(3) A. Dauvillier. Journal de Physique, viii. (1), 1927.

[Peoc. Roy. Soc.. Victoria, 41 (N.S.), Pt. 1928.]

Art, V. — Contributions to the Flora of Australia , No. 35*-
The Naturalized Aliens of Victoria .

ALFRED J. EWART, D.Sc, Ph.D., F.L.S., F.R.S.

{Read 12th July, 1928; issued separately 27th September. 1928.]

These now form a prominent part of the Flora of Victoria, and.
they are steadily increasing both in numbers of species and of
individuals. In 1909 the number of aliens recorded was 363, and
in 1928 it had risen to 461. This rate of increase represents
approximately one every two months, or slightly more than five a
year, and this rate of increase has been maintained with remark-
able uniformity during the past sixty years. The alien plants are
also more numerous in individuals than the native flora, although
the latter represents a much larger number of species, about 3000,
and the aliens occupy a greater area of the soil than does the
native flora outside of the forest areas.

Nevertheless, all the aliens are not obnoxious, since they include
all the clovers, trefoils and medicks,, most of the more valuable
pasture grasses, and some garden plants that have run wild. Less
than a hundred of the aliens are serious weeds, and few of them
represent as serious a menace as does our native bracken on newly
cleared forest land. In addition, two native plants, the Chinese
Scrub ( Cassinia arcuata ) and the Nut Grass ( Cyperus rotundus ),
have proved so troublesome as to be proclaimed for the whole
State, while .the native Prickly Acacia ( A . arm at a) has been pro-
claimed for eleven shires, and the Three-cornered Jack (Em ex
australis) for two.

The sources of origin of the aliens show several points of
interest. Naturally most are derived from plants with a wide
general distribution. Thus 140 are native to Europe, Asia and
Africa, and 66 to Europe and Asia. Of plants native to single
continents, Europe has contributed 57 aliens, America 30, North
America 18, South America 12, North and Central Africa 11,
South Africa 29, and Asia 2. From the Mediterranean region 31
aliens are derived, whereas only 11 are native to Europe, Asia
and America, 7 to Europe and Africa, 4 to Europe and America,
and 2 to Asia and Africa. Only 8 of our aliens arc general cos-
mopolitans exclusive of Australia, 13 are cosmopolitan to the tem-
perate regions, 7 to the warmer zones, and 2 come from the N.
Temperate and Arctic zones.

Not included in the above are the following single cases: —
Avelina Micheli is derived from Italy, Calycotome spinosa from
Spain, Centaur ea Picris from the Caspian region, Chi oris abys-

*No. 34 in Proe. Roy. Soc. Vic., n.8 , xl. 02), p. 81.

Alfred J. Ewart :

sinic a from Abyssinia, Leycesteria formosa from the Hima-
layas, Lychnis divaricata from Sicily, Lycium chinense from
China, Rubus phoenicolasius from Japan, and Verbena venosa
from the Argentine. Only one alien, Eleusine coracana, has its
home in Asia, Africa and America, but several are native to
other parts of Australia. Albkcia lophantha is native to West
Australia and Andropogon erianthoides to New South Wales and
Queensland, but both have become naturalised in Victoria.
Hibiscus Trionum , which is native to Europe, Africa, Asia and
Australia (with the exception of Victoria and Tasmania), has
reached Victoria as an alien with the advent of civilisation, and in
the same way Setaria macrostachya , which is native to Asia and
Tropical Australia, has established itself in the South, aided by
man. The activities of man, particularly through the transport of
fodder, are probably responsible for the relatively high proportion
of aliens contributed by South Africa, and these include some
of our worst weeds both here and still more in West Australia
(Cape weed, Onion grass, Stinking Roger, etc.).

1 lie native flora of Victoria, exposed as it is to the competition
of imported aliens and to the pressure of settlement, is in a con-
dition of rapid flux. It is probable that less than half of the
original flora will survive within 50 years, and that many plants
originally widely spread will be confined to special localities.
Were it not for the disturbing factors introduced by man the
spread of the introduced aliens might have been used as a test of
Willis s age and area hypothesis. As it is, although in a very
general way the older weeds are more widely spread than the
more recent introductions, the rule does not apply to hardly any
comparable pair of individual cases. Thus the Evening Primrose,
Oenothera biennis (1887), has covered less ground than the
Foxglove, Digitalis purpurea (191 7). The Musk Weed, Mya-
grum pcrfoUatmn (1916) has become more abundant than the
Horehound, Marrubimn vulgar e (1870), and Onion grass,
Romulea Bulbocodium (recorded in 1873, but abundant in Mel-
bourne in 1860), with twenty years’ start has hardly covered more
ground than St. John's Wort, Hypericum perforatum (recorded
in 1893), but introduced in 1880. A still more striking case is
that of the Stinkwort, Inula graveolens (1893), which rapidly
overtook the Stinkweed, G ilia squarrosa (1887), both in area and
in abundance.

One would expect the largest number of the naturalised aliens
to belong to the Compositae (70), and the disproportionately high
number derived from the Leguminosae (50), and from the Gram-
ineae (102 ) is an aftermath of the pastoral phase when the world
was searched for fodder plants to improve our pastures. The
native Gramineae comprise 125 species, and many of these are
dying out, so that in the near future the grass flora will be mainly
foreign. Another curious disproportion is shown among the
Monocotyledons. There are 13 alien Irids and only four of the
Lihaceae, while no Orchid or Amaryllid has become naturalized,

Flora of A ustralia .

611

and only one alien sedge has crept in among the 1 1 1 native species
of the Cyperaceae. As the native Irideae are only 8 in number,
this is the first native order in which the' aliens have widely out-
numbered the natives. Of the total of 461 naturalized aliens all
but twelve belong to natural orders already represented in the
flora. The new orders added are Aroidaceae (1), Cactaceae (2),
Dipsacaceae (2), Fumariaceae (1), Polemoniaceae (1), Ponte-
deriaceae (1), Resedaceae (2), Salicaceae (1), Valerianaceae (1),
but no member of the Myrtaceae, Sapindaceae or Rhamnaceae
has become naturalized. Aliens are relatively high in the
Labiatae, Solanaceae and Scrophulariaceae, nearly half the latter
order being now represented in Victoria by naturalized aliens (24
native species to 18 aliens).

Strictly speaking, the age and area hypothesis is held to apply
to closely related plants or to species of the same genus, although
if true at all there seems no reason why it should not apply gene-
rally or why, if it does not apply generally, it should be true in a
restricted form. Even taking species of the same genus, it appears
that the time factor is of far less importance in determining the
area covered by a species than its suitability to new habitats, its
means of distribution, its aggressiveness and its resistance to foes
and injurious agencies. In the case of the genus Poa , P. annua ,
P. pratensis and P. trivialis were recorded as naturalized in 1878,
1888 and 1888 respectively, but P. pratensis has taken the lead
because it is better suited generally to the local conditions, and
P. compressa (1908) is rapidly overtaking some of the earlier
introductions. Similarly, in the case of the clovers, taking those
which spread by natural means, Trifolium glomevatum (1892), is
more widely spread than T. arvense (1887), and of the red (T.
pratense ), yellow (T. procumbens ) , and white clovers (all 1864),
white clover ( T repens) has taken the lead mainly because of its
superior means of natural distribution and its greater staying
power.

According to Willis, however, comparisons cannot be made
between single pairs of species, but only between groups of not less
than 10 closely related species. As a matter of fact, if the age
and area hypothesis has any general value, any average of any 10
pairs selected at random should be as good as two groups of ten
each of related species. Even using 10 pairs of related species it is
easy to construct natural cases in which the “ law ” could not
apply. Thus, suppose a genus of five species is diverging through
subgenera B, C and D, each of five species, and that in groups
B and C, the size of the seed diminishes, and in group D that of
the pappus, so that groups B and C have twice the rate of dis-
persal of A and D.

Then, taking any descending order of age for the species
groups B and C will occupy double the area relatively to a given
age as compared to groups A and D. Beneath the areas are set
out proportionately to the ages and rate of spread in each group.

■62 Alfred J. Ewart: Flora of Australia.

The ages selected are immaterial if they are set out in descending
order.

Genus

Age

Area in

Sub-

Genus

Age

Area in

years.

10.000 acres.

years.

10.000 acres.

6.000

2,500

5,500

2,000

5.500

2,000

4.500

1.500

4.500

1,000

Sub-

Genus B.

3,500

Sub-

Genus D.
1

800

3.000

600

3.000

600

2.500

500

2.000

500

Proportion
total age to
total area

1 : 135

1 : 175

Hence the proportion between age and area may vary widely
•’even in comparisons between groups of ten, each containing equal
numbers of plants with the same rate of dispersal. If groups A
and C are compared with groups B and D, the relative propor-
tions between age and area are as 1 : 126 and as 1:61.

If groups B, C and D had all twice the rate of dispersal of A,
then a simple arithmetical calculation shows that in 6000 years
the area of group C and D would become equal to that of group A
and B, whereas the average group ages would be as 57 : 100. Sup-
pose that the dispersal of a plant is uniform, so that it spreads at
the rate of a mile a year; then the area covered is proportional to
the square of the distance of linear dispersal, i.e., to the square
of the time, so that in one, two and three years the areas are res-
pectively one, four and nine.

Hence if the groups A, B, C and D all had the same rate of
spread and their average ages were 10,000, 5,000, 2,500, and
1,000 years, then the areas covered would be proportional to the
squares of the ages, i.e., as (A) 100 : (B) 25 : (C) 9 : (D) 1.
Thus the accidental inclusion of a single A plant in a D series
because of an apparent close affinity would vitiate subsequent cal-
culations, and to avoid such inclusions it is necessary to assume
the age and area hypothesis, i.e., the very thing set out to be
proved. It seems probable that the age of a species is one of the
least important of the factors governing its distribution, and that
in only few cases can a relation be traced between the age of
species and the area they cover at the present day. The area of a
cosmopolitan is limited by that of the surface of the earth, and
during its existence a species like the common bracken or any
other cosmopolitan may have travelled several times round the
earth. Bracken certainly, and other cosmopolitans probably also,
have had sufficient time to cover the surface of a planet far larger
than Jupiter, and in such cases the present area of distribution
cannot bear any definite relation to the age of the species.

END OF VOLUME NLI., PART I.
[Published September, 1928 ].

[Pkoc. Roy. Soc. V ectori a , 41 (N.S.), Pt. II., 1929.]

Art. VI . — On Grooved , Pitted and Miniature Pedestal
Rocks at Lake Gomigarrie , If extern Australia.

By J. T. JUTSON, B.Sc., LL-B.

(With Plates XII., XIII.)

[Read 13th September, 1928; issued separately 30th January, 1929.]

Contents.

I. Introduction.

II. Previous Literature.

III. Climatic Conditions.

IV. Description of the Occurrences.

( A ) G r o o ve d Rock s.

(a) Greenstones.

(b) Quartz and Jasperoid Rocks.

(B) Pitted Rocks,

(a) Greenstones.

(b) Quartz and Jasperoid Rocks.

V. Tjie Origin of the Phenomena.

VI. Records of Pitting and Grooving Elsewhere by Wind
Action.

VII. Slum mary.

VIII. References.

I. Introduction.

Lake Goongarrie is a “ dry ” lake or playa in sub-arid south-
central Western Australia. It lies north of Kalgoorlie and just
to the east of the railway line from Kalgoorlie to Leonora, and
extends northwards from Goongarrie to Comet Vale. It is
situated on the Great Plateau of Western Australia, which in the
district referred to is about 1200 feet above sea-level.

In the course of geological work some years ago at the mining
centres of Goongarrie and Comet Vale, the writer made a sketch
survey of the lake, the map of which has been published on a re-
duced scale in this journal in connection with a paper (1) by the
writer on various physiographic phenomena observed during the
course of the survey and is here reproduced. The same map, but
on a larger scale, appears in an official report (2, Plate I.) by the
writer, published by the Geological Survey of Western Australia.

In addition to the phenomena described in the paper (1) just
referred to, certain rocks, some in situ and some fragmentary,
were observed to be pitted, grooved or undermined in a remark-
able manner, and under conditions with regard to adjacent rocks

J. T. Jutson :

" - . QEFtNEQ

H- L

Xx High "Lands
\ A A Lowlaa/DS
: S A IV D 5 ( ft'DGES

0 lY //V SOUTHERN

pom r/o/v)

///i\\ Hock Cl iff*

^ Isolated Hills

A\'

Sketch Map of Lake Coohcarr/e

i ^ i

Scale of Miles j.

Grooved , Fitted and Pedestal Pocks

-that indicated that the forces responsible for the production of
such features were working in a peculiarly restricted way. A
description of such occurrences, together with a discussion as to
the origin of such features, is therefore of interest; but this
interest is heightened by the possibility that light may thereby be
thrown upon the origin of the surface features of the interior of
Western Australia, concerning which — especially in regard to the
“ dry ” lakes or playas — no unanimity has yet been reached. This
paper is therefore submitted.

IL Previous Literature.

The writer is not aware of any similar phenomena in Western
Australia having been described, and he consequently believes that
this paper contains the first description of such phenomena occur-
ring in that State, or indeed in any other part of Australia; but
the literature of the other States, which may bear on the matter,
has not been searched.

Extra- Australian literature treats of some related occurrences,
and such will be referred to below.

III. Climatic Conditions.

In order that the facts and arguments submitted may be
better appreciated, a brief statement of the climatic conditions of
the area is advisable.

The Comet Vale-Goongarrie district has an average rainfall of
about ten inches per annum, most of which falls in fairly steady
rain during the winter months. In summer, the individual falls
are frequently heavier than those of winter, and consequently are
probably responsible for more erosion than the winter rains, not-
withstanding the greater abundance of the latter. There is great
variation in the annual quantity of rain, some years being as low
as four or five inches, whilst a wet year may have up to 19 or 20
inches, which, however, is exceptional.

The. range of temperature is considerable. In the summer the
temperature mav frequently rise above 100 F., and in the winter
it may fall below 50°F. in the daytime and may reach freezing-
point at night. In the summer there is great radiation of heat at
night, which thus often brings about a pronounced fall in the tem-
perature during that time. The nights are therefore almost always
cool.

The humidity is low, but the evaporation of watei is enot-
mous, as is shown by the records from Coolgardie, farther south,
where the amount reaches 87 inches annually.

Frosts occur in the winter, and hence are responsible for a cer-
tain amount of rock splitting. The variation in day and night
.temperatures also brings about exfoliation of the rocks.

J. T. Jut son :

From the writer's observations, the winds, taken as a whole,,
are not very strong, except the westerlies, which at times blow
with great force, and are apparently the dominant winds.

The lake floor is almost always free from water; hence the
name “ dry ” lake. When rain falls it spreads as a sheet a few
inches thick over the lake floor, but it rapidly evaporates. This
floor is destitute of vegetation, but at its margin, in those places
where the ground is but slightly higher, samphire and other salt-
loving vegetation grow, whilst the rest of the country carries small
trees and shrubs, forming a scrub, with much bare ground be-
tween the plants.

IV. Description of the Occurrences,

(A) Grooved Rocks.

(a) “ Greenstones A* — The western side of Lake Goongarrie at
Comet Vale consists, in places, of moderately high and steep cliffs
f tough, line-grained basic rocks (“greenstones”), of which
amphibolites are probably predominant. At the immediate foot
f the cliffs there is a rock floor of similar rocks.

Just south of a large “ natural quarry/’ almost due east of the
Id Sand Prince Lease/ a remarkable set of grooves may be
bserved in the lowest rocks of the cliff face. These grooves are
closely spaced, are usually in straight lines, and run in all direc-
tions in a horizontal plane, in that plane in some cases being
roughly parallel to and in other cases intersecting one another;
but their inclination to that plane is generally vertical or close to
the vertical. They may vary from an inch or two to about ten
inches in length, with a depth varying from a few lines to about
four inches, and a maximum width of an inch. In length, depth
and width, the grooves tend to taper into the solid rock.

On careful examination, these rectilinear grooves are found to
follow the small irregular divisional planes (due to jointing and
earth movements) which abundantly traverse the rocks, but the
grooves are not mere openings in the rocks caused by the two
sides of a divisional plane becoming forced apart. The rock
material has been actually removed by some natural agent so as
to leave a distinct groove of the kind indicated. The grooving
agent has merely selected the divisional planes as convenient lines
for the commencement of operations. Possibly, without such
planes the action would not take place, at least not in such a recti-
linear fashion. The grooved surfaces are mostly fairly smooth-,
but they are not polished.

The grooved rocks extend from three to four feet upwards
from the lake floor. Above this height grooves are absent,
although the rocks form part of the same rock mass as the grooved
ones and have the same divisional planes.

1. — This and other leases referred to below are marked on the geological
maps accompanying the writer’s geological report on the district (2).

Grooved , Pitted and. Pedestal Pocks.

Apart from the grooving, the rocks of the cliff face are break-
ing down under the influence of the weather in the usual way.

The accompanying photograph (PI, XII., Fig. 1) illustrates the
features described.

Grooves in greenstones have also been noted at the following
localities, among others : —

(i. ) At the large “natural quarry ” already referred to, there
is a talus of fallen blocks of the fine-grained greenstone, which
are grooved for a few feet in height at the base of the talus ; whilst
the higher blocks are free from grooves.

(ii.) On the western shore of the lake, about four miles north
of Goongarrie, at the north-western side of a small “ inlet/’ close
to the railway line, grooving occurs in a rock cliff of greenstone,
and is there limited to a height of about two feet from the base,
where there is also an abundance of fine sand.

( iil. ) Grooves occur in fragments of greenstone lying on
the surface of the ground at the foot of the cliffs of the same rock
to the west of the lake and about a quarter of a mile north of the
old Beelzebub Lease, which is situated about one and a half miles
to the east of the Goongarrie railway station. The ground here
is well above the level of the lake floor. Fine quartz sand occurs
in the grooves.

(iv.) A few chains north of the Lady of the Lake Lease,
Goongarrie, there are two small greenstone knobs, not exceeding
eight feet in height, the surfaces of which largely consist of frag-
ments of the rock, the result of breaking down by the weather.
These fragments for a height of not more than two feet from
the floor of the knobs are grooved and also pitted. Above this
height grooving and pitting do not occur. I here is an abundance
of fine-grained quartz sand at the bases of the knobs, and also
associated with the grooved and pitted rock fragments.

(v*) On the western shore of the lake at Comet Vale, near a
small “ inlet” just to the north of the old Planet Lease, there is
a greenstone knob a few feet in diameter and about eight feet
high. The rocks of the. knob: are grooved to a height of about
four feet from the floor of the knob, being higher than usually
noticed elsewhere. On the western side of the knob there is a
sand dune higher than the knob, which bad wind-blown sand scat-
tered around its base and over its flanks to a height of about
five feet. At the time of observation the fine sand at the foot of
the knob was being rapidly removed by the wind, and blown
against the hard rocks of the knob.

~ (vi.) In the Black Diamond Lease at the southern end of the
Goongarrie mining field, a short watercourse, commencing at a
high quartz “blow,” ends in a small alluvial fan, on the sui face
of which are many greenstone fragments, mostly a few inches
only in length and breadth. On the alluvial fan there is scat cel}
any vegetation, and the ground is consequently exposed to the full
force of the wind. The rock fragments on the fan are grooved

J. T. Jutson :

and also pitted. The country rises from the fan, and concurrently
the vegetation increases, but as it does so, the rock fragments
cease to be grooved and pitted. In the actual channel of the
watercourse the rock fragments have neither grooves nor pits.

(vii.) On the western shore of the lake, probably at Comet
Vale, but the exact locality cannot now he indicated, grooving at
the base of the greenstone rocks occurs, and some of the grooves
arc n parallel lines, being evidently along planes of schistosity in.
the rocks.

In none of the above instances have the rocks been polished.

In addition to the general grooving of rock masses and irregu-
larly shaped fragments noticed above, an interesting occurrence
of grooved rocks may now he described. The phenomena have
been observed in one locality only, which is on the western side
of the lake at Comet Vale near the grooved rocks first described
in this paper.

At this point, on the floor of the lake close to the greenstone
cliffs, a number of pebbles of the fine-grained greenstone, possess-
ing distinct horizontal grooves, were observed. The pebbles are
numerous, but are confined to a small area. They rest on the lake
floor by a fairly fiat bottom, but otherwise are generally rounded
and usually from one to two inches in diameter. The grooving,
practically in all pebbles, commences about a quarter of an inch
above the bottom of the pebble, and extends, as a rule, completely
around each pebble. The grooves are about half an inch in
width, and penetrate the rock to a depth of usually less than a
quarter of an inch. The surface of the grooved portions of the
rock is even, but only slightly smoother than the surrounding
parts of the pebble, and there is no indication of polishing. The
pebbles have apparently been quite undisturbed for a considerable
time,

(b) Quartz and Jasperoid Rocks . — At the northern end of
Lake Goongarrie there is a prominent hill bordering the lake
known as “ Poverty Hill" (see text-fig. 1). On account of the
toughness of the rocks (quartz in the form of reefs and jasperoid
banded rocks, the origin of which has not been investigated) com-
posing this hill, the latter projects somewhat into the lake as a dis-
tinct point, and hence this feature is known as “ Poverty Point; ”
At the foot of the steep cliff forming Poverty Point there is a flat
cone of detritus derived from the rocks of Poverty Hill. This
cone rises not more than two feet above the lake floor and in
length is about one chain from north to south and two chains
fmm east to west.

Boulders and pebbles of the quartz and jasperoid rocks (of
many sizes from an inch to 12 inches approximately) are strewn
upon the surface of the cone, and many of them are thin flat-lying
fragments. All are more or less rounded and smoothed (but not
polished) on their sides and upper surfaces, especially in the case

Grooved, Pitted and Pedestal Rocks.

of the larger fragments. These boulders and pebbles are remark-
able for the grooving they have sustained. Some of them have
pronounced horizontal grooves completely around them ; in others
the grooves do not extend so far. Generally speaking, the groov-
ing is much stronger in or entirely limited to those portions which
have apparently been particularly exposed to the eroding agents. In
the jasperoids, when the bands are not parallel to the surface of the
ground, the grooves often follow the bands, and therefore may be
at any angle. 2 Two horizontal grooves, one above the other, may
occur.

The width and depth of the grooves rarely exceed half an inch
and are frequently less. The grooves commence usually about a
quarter of an inch from the base of the pebbles, but in some cases
they were as high as an inch above the base.

In addition to the rounding* and grooving, the rocks were also
markedly pitted. The pits may be round, elliptical or oval at their
mouths and may be drilled at almost any angle from the vertical
to the horizontal. In size the pits vary from about two inches to a
quarter of an inch or less in diameter. The pits in places unite to
form a groove, and, in some instances, the pit has extended into a
hole bored right through the pebble.

The grooves and pits and the rounded faces occur, not only
in the loose rocks on the small alluvial cone, but also in the rocks
forming the base of the cliff. Above a height of about three
feet from the floor of the cone, the rocks are almost invariably
quite angular, not grooved and not pitted, although some groov-
ing and pitting can be traced to a height of about 25 feet. But
where such occurs the cliff is exposed to the action of strong
south-westerly, southerly and south-easterly winds.

Similar quartz and jasperoid rocks occur at the Causeway
Hill (see text-fig. 1), and there the same phenomena of rounding,
grooving and pitting occur as at Poverty Point, to a height, as a
rule, of about three feet from the floor of the lake.

At "The Snout n (see text-fig. 1) similar jasperoid rocks
are grooved and pitted at the base of the cliff, and also on
a shoulder perhaps 20 feet or more above the lake floor, but no
details are available.

(B) Pitted Rocks.

(a) Greenstones . — In various places on the flats, which in
part border the western shore of the lake at Goongarrie, there
are a number of small, roughly circular knobs of resistant, fairly
coarse-grained greenstones, in which felspar and fibrous horn-
blende are quite prominent. These knobs are usually from
about 6 to 15 feet in height, and from about 5 to 12 feet in
diameter. The flats on which they rest are usually open spaces,

2. — In the same rocks the evident difference in texture in the component
bands brings about differential grooving.

J. T. Jutson :

with only a few scattered, low, shrubby plants. The ground
may be covered with much white quartz’ rubble or by sandy or
clayey soils.

The rocks of some of the knobs are pitted to a height of about
three feet above the surface of the ground. The pits at their
openings mostly are circular, with a diameter of usually less than
half an inch, and with varying depths, which probably as a rule
do not exceed half an inch. The pits penetrate the rocks at dif-
ferent angles, and in places are numerous. They pass alike
through the hornblende and felspar, but the hornblende is slightly
more resistant to this mode of erosion than the felspar, since in
places the hornblende projects as unreduced fragments into the
pit, a fact which was not observed in the felspar.

Above the limit in height mentioned, pits are absent, although
the upper portions of the rocks are as much exposed to the action
of the w r eather as the lower . 3

Where pitting occurs, there is generally some fine quartz sand
about the ba^e of the knob and in the pits.

It may be noted that in a hornblende felspar porphvrv (the
precise locality of which cannot now lie given) the hornblende
was in small spherical nests of about equal size. In the lower
portion of the outcrop, the hornblende weathers out more rapidly
than the felspar phenocrysts, and hence a number of small pits of
uniform size have developed.

The pitting of the greenstones in manner described above may
be observed at the following localities: — Towards the northern
end of the Bushman Lease, south-east from Goongarrie town-
ship; in or near the Lady of the Lake Lease, at the southern end
of the Goongarrie field ; in the Lord Nelson Lease, just to the west
of the Lady of the Lake Lease; and north of the Overlander
Lease, which is to the north of the township of Goongarrie.

In addition to the small knobs just referred to, small fragments
of greenstone lying on the surface of the ground in certain areas,
are also extensively pitted.

The surface of portions of the low-lying, gently-sloping
ground, bordering the western shore of the lake at Goongarrie
has an abundance of rock fragments of various sizes. The vege-
tation is sparse, and the fragments, which consist of white quartz
and fine and coarse-grained greenstones, are consequently much
exposed to wind, sun and rain. It is in the coarse-grained green-
stones that pitting, in association with grooving, occurs. Many
of the fragments are thin in proportion to their length and breadth,
and lie flat. This fact, combined with the gentle slope of the
ground, makes the gravitational travelling of the rock fragments
very slow; hence many may remain in the same positions for lung
periods, thus giving erosion an opportunity to make its mark in
any particular manner.

3. — A few small pits may be observed on the tops of some greenstone
ridges and knobs, but these are clearly due to the action of rain, and
they have no relation to the pits of which this paper specially treats.

Grooved, Pitted and Pedestal Rocks.

The pits usually are circular, and may occur on both the top
and sides of the rock. On the top, the pits vary in diameter from
about one-eighth of an inch to half an inch or more, with similar
depths, whilst in some thin fragments the holes have been drilled
through from top to bottom. On the sides, the holes do not ex-
ceed half an inch in diameter and depth, and are usually much
less. Frequently the pits coalesce, and so a more or less con-
tinuous horizontal groove may be formed. Fine sand occur > in
some of the pits, the surfaces of which are usually fairly smooth.

Localities where the phenomena described may be seen are on
the western side of the Golden Sun Lease, Goongarrie; at the
small alluvial fan in the Black Diamond Lease, at the southern
end of the Goongarrie mining field, referred to under the grotw-
ing of greenstones; and generally in various places on the low-
lying exposed ground immediately to the west of the lake at both
Goongarrie and (more rarely) Comet Vale.

The rocks are nowhere polished.

(b) Quartz and Jasperoid Rocks . — The pitting of these rocks
at Poverty Point and Causeway Hill has been described above
when giving details of their grooving.

(^) Other Rocks . — At the western end of one of the southern
arms of the lake, which the writer has named the “ Tombstone
Arm,” a cliff a few feet high occurs, surmounted by a number of
projecting but discontinuous rocks of approximately even size.
They resemble a number of tombstones, somewhat tilted from the
vertical; hence the name, “The Tombstones,” given to the
locality. These slab-like rocks have had their shape determined
by their vertical planes, and by the removal of intervening slabs,
but it is not apparent why such a peculiarly selective mode of
erosion has taken place.

The component rocks are believed to be fine-grained quartz -
porphyries, but the writer’s records on this point are incomplete.

The surface of a “ tombstone ” is coated with a film of iron
oxide, and is free from pits and grooves.

The rocks which form the low cliff referred to are dark grey
shales and slates from which' masses several feet in length have
been detached. The upper surfaces and sides of these detached
blocks often have pits varying in diameter from a quarter of an
inch to about one foot, and in depth from a quarter of an inch to
three inches. They are usually roughly circular in surface outline,
except where two or more pits have coalesced into one elongated
one. The pits may be large shallow saucer Tike hollows or rela-
tively deep narrow ones, or they may have about the same sur-
face diameter and depth.

At a tiny gully close to “The Tombstones,” there is a short,
sharp drop to a lower level, down which rain water occasionally
falls. The rocks at this point are much pitted, and of especial
interest is a concretionary structure which has facilitated the hol-
lowing out of the rocks on vertical faces in a remarkable manner.

J. T. Jutson :

The writer’s records unfortunately are insufficient to state what
these concretionary rocks are.

The shales and slates at “ The Tombstones ” are mostly free
from the iron oxide film mentioned above.

On the western shore of the lake at Comet Vale, and quite close
to the rocky clifTs, several examples of miniature pedestal rocks
occur. The lake floor at the edge of the lake is a " billiard-
table rock floor. The pedestal rocks are part of the same rock
mass ms the floor, and consist of fine-grained greenstones of the
type already referred to. They are so tough that examples of the
pedestal rocks could only be obtained by the writer by wedging
them out from the floor along the close-set and irregular joint
planes of the rocks.

Kirk Bryan (4. p. 123) describes a pedestal rock as an isolated
rock consisting of a larger mass above, supported on a more
slender pedestal.

These miniature pedestal rocks attain a height of two to three
inches above the rock floor. In horizontal section, the portions
above the pedestals are roughly circular or oval, with diameters up
to about two inches; and they project about half an inch or more
beyond the pedestals, hut the extent of the projection varies in
the same pedestal rock and in different rocks. The pedestals them-
selves form short columns about one inch, and less, in height. The
result is the well-known mushroom appearance. The surface of
the pedestals and of the rock floor overhung by the upper masses
is smoothed by abrasion, hut is not polished.

At Poverty Point, some of the quartz or jasperoid rock frag-
ment on the surface of the alluvial cone mentioned above, have
been undermined so as to form miniature pedestal rocks with a
pedestal about an inch in height, and a top that may he six or
seven inches in diameter, and which may project one to two inches
beyond the pedestal. The rock fragments are smoothed and
rounded.

The pedestal rocks described are in miniature only, hut never-
theless they are of interest, inasmuch as they indicate, to some
extent, the nature of the erosion processes operating in the dis-
trict.

V. The Origin of the described Phenomena.

For the sake of clearness and convenience of reference, the
various types of erosion dealt with in this paper have been sepa-
rately described. In considering, however, the possible origin of

4. — The word “pedestal" has heen suggested by, Kirk Bryan (3, 4, and 5)
as a descriptive term for the kind of rocks now described, and is-
preferable, in the writer's opinion, to the old term. “ mushroom.”

Grooved, Pitted and Pedestal Rocks.

the described features, it is convenient to treat the subject as a
whole, touching on the various aspects as they arise.

In the following discussion the semi-arid nature of the country,
with the resulting scarcity or absence of vegetation, and the abun-
dance of blown sand available must be borne in mind. The mode
of occurrence of the sand is described in publications (1) and
( 6 ).

As regards the agents of erosion which have brought about the
effects noted, the abrasive action of streams, lakes and seas, and
of the wind; the action of rain in its mechanical and chemical
aspects ; the effect of the crystallization of salts at the surface of
the rocks ; and differential atmospheric weathering generally, must
be considered. There may, of course, be a combination of forces.

No satisfactory evidence has yet been adduced to show that
the sea has had any influence in moulding the rock cliffs and rock
floors of the lake. The former occurrence of large freshwater
lakes in the interior of Western Australia has been postulated,
but definite evidence is as yet wanting. Even if the sea had
recently occupied large areas of the country, or if lakes of the
type just mentioned had previously existed, all the effects noted
could not be ascribed to such agencies, which may therefore be
disregarded.

Again, the erosive power of the very shallow waters that occa-
sionally cover the lake surface is too weak to produce the various
kinds of grooving and pitting, or the pedestal rocks described
above. Moreover, some of the features observed occur beyond,
although close to the lake, so that there must be some agent more
general in its action than sea or lake waters.

Rain, chiefly by its chemical action, may form pits in rocks con-
taining much soluble material, such as arenaceous limestones; but
such pitting, so far as the writer is aware, is rare in igneous and
most sedimentary rocks. Rain, no doubt, in its combined chemical
and mechanical action, can groove and pit rocks, but such action
would not be limited to a definite height above the surface of the
ground. In the examples described in this paper (omitting the
rocks at “The Tombstones”), there is such a limitation on the
cliff faces, except in occasional special cases, which can be ac-
counted for. Similarly, pitting and grooving are only found
among surface rock fragments where the ground is open and
largely destitute of vegetation, and therefore exposed to the
action of the wind.

Rain, therefore, does not appear to be the primary cause of the
pits and grooves, although, once erosion had commenced, it would
doubtless he hastened by the rain; but this would hardly or only
slightly apply to horizontal pits and grooves.

The effects of the crystallization of salts must be considered.
The water beneath the floor of the lake is heavily charged with
common salt, and much of the underground water at some dis-
tance from the western shore of the lake also contains the same
substance in abundance. Other salts also occur.

J. T. Jutson :

If crystallization takes place when the water rises by capillary
attraction to the surface and there evaporates, the rocks may be
disintegrated to some extent, as the writer in an earlier paper (7)
has indicated.

Where pronounced divisional planes occur in the face of a cliff
(as in the fine-grained greenstone cliffs containing at their base
the rectilinear grooves described above) these planes may possibly
facilitate the ascension of the salt-charged water through the
immediately adjacent areas of the rock; and at the surface, as •
result of the crystallization of the salt, slight disintegration or
internal strain may occur. If wind-driven sand be the chief cause
of the grooves, it would be aided by such disintegration or strain.

The crystallization of salt at Lake Goongarrie cliffs, however,
seems to cause an irregular undermining by a flaking of the rod:
rather than disruption or strain along the divisional planes. Where
grooves are several inches deep, it is improbable that they are
caused wholly or largely by salt crystallization, as it is difficult to
imagine the process working in this regular wav. Moreover, the
grooved rocks forming the actual cliff face have mostly been
broken away from the main mass — apparently before the grooves
were formed — and consequently evaporation of the water and pre-
cipitation of the included salts would doubtless take place at the
surface of the rocks in situ.

So far as the writer's observations and recollections go, no dis-
ruption or weakening along the division planes occurs. The
surface of the grooved rocks is firm, and free from signs of dis-
integration by flaking or crumbling.

With regard to the pits in the small coarse-grained greenstone
knobs, the crystallization of salt may perhaps loosen or detach
•a mineral fragment, and so be the means of starting a pit; but it
is inconceivable that the process should so continue as to form the
spherical fairly smooth pits already described. Rather there
would he a disintegration over practically the whole of the sur-
face of the area affected. Furthermore, these pitted rocks con-
sist in part of blocks detached from the main mass, and it is a fair
assumption that the pits have developed since the detachment—
at least in some of the rocks. If so, the crystallization of salts is
not likely to occur on the surface of the fragmentary rocks.

With reference to the horizontal grooving of the fragments of
quartz, greenstone and other rocks lying loose upon the surface of
and, consequently, not in continuous contact with the ground, the
same difficulty as to the passage of the capillary water into the
fragments again occurs. If, however, this difficulty were overcome,
the crystallization of salts at an even height above the surface of the
ground seems improbable. Crystallization is more likely to occur
over the whole exposed surface. Apart, however, from theoreti-
cal considerations, the rock fragments show no evidence of decay
through crystallization of salts.

Grooved , Pitted., and Pedestal Rocks . 75 -

Other chemical action will tend to weaken the coherence of the
the rocks, and so make them more easily eroded by any eroding
agent.

There is no indication that temperature variation is the cause
of the pits and grooves or that it has aided in their formation.

The abrasive action of the wind being the remaining possible
factor, is therefore apparently the prime cause of the pits and
grooves. Its action is discussed below.

At ‘‘The Tombstones ” the majority of the pits clearly appear
to be due chiefly to the solvent action of rain. This conclusion is
suggested by the fact that there is no definite limitation of the
height at which they occur, and by the large, shallow, saucer-
shaped character of many of the pits. The process appears to
start with the formation of small irregular hollows (due to differ-
ential atmospheric erosion) on the surfaces of the rocks. Rain
water collects in these hollows and acts as a slight solvent, thus
further disintegrating the rocks. Further rain will wash out the
separated material, and, the processes being repeated, the cavities
become enlarged. Some of the smaller pits, and especially the
more or less horizontal ones, may be due to the action of the
wind. The hollowing out of the rocks at the small waterfall is
due mainly to the fall and splash of the water.

Kirk Bryan in various publications (3, 4 and 5) has given in-
stances of pedestal rocks formed otherwise than by wind action.
He considers (3, p. 11) their formation to be due to the work of
rain, of mechanical disruption, of stream action and of chemical
weathering; and he has shown (3 and 5) in lucid and convincing
fashion that in arid areas, in some instances, such rocks are
moulded into their present shapes by the action of a “ drip cur-
tain ” during rain, and by the spreading of a film of water on the
tinder surface of the overhanging rock. Pedestal rock formation,
takes place in this way, especially when a less resistant rock, e.g.,
a shale, underlies a more resistant one. e.g., a conglomerate. In
the case, however, of the miniature pedestal rocks on the floor of
Lake Goongarrie, the homogeneous character of the rocks, the
smoothness of the face of the pedestal and of the floor forming
its base, the absence of the grooves caused by the drip curtain,
and the weakness as an eroding agent of such a tiny drip curtain,,
if formed, suggest strongly that rain action must be eliminated.

The mode of occurrence of the miniature pedestal rocks, wheie
those rocks are in situ, should be favourable to erosion by salt
crystallization, as it is in undermined areas that one would expect
such crystallization to take place. No such effects, however, are
visible megascopically. The surface of the pedestal is smooth and
free from any indication of crumbling or flaking; and apparently
the pedestal is as tough as the rock above and below. If crystalli-
zation is taking place, then its action appears to be very slight, or
even negligible.

J. T. Jut son :

Microscopical examination of the miniature pedestal rocks
—as well as of the pitted and grooved rocks — might throw some
light on the question whether salt crystallization has directly or
indirectly aided in the formation of the pits, grooves and pedestal
rocks.

1 he remarks made above in connection with the pits and
grooves as to further chemical action and the effect of tempera-
ture variation apply to the miniature pedestal rocks.

The wind in its abrasive capacity therefore appears to be the
principal agent in the formation of the pedestal rocks.

By a process of elimination of other possible factors the writer
has arrived at the conclusion that the wind in its direct abrasive
capacity is the chief agent in the production of most of the unique
features described in this paper; and the general conditions pre-
vailing favour this view. These conditions are * limitation in
height of erosion, which is especially characteristic of wind action ;
the dry climate; the sparse vegetation; and the abundance of
quartz sand. In the case of the pits in the coarse-grained green-
stones, Harger\s suggestion (7, p. xxxv) with regard to the
honeycombing of “ augen ” gneiss, that the holes were probably
started by the weathering out of a particular mineral, would prob-
ably apply.

The actual mode in which wind-driven sand brings about the
results stated may now be considered, although the subject is a
difficult one on account of want of direct observation of the pro-
cess.

The restriction of the grooves in the cliffs and of the pits in the
small coarse-grained greenstone knobs to a height of about three
feet above the surface of the ground is probably due — at least in
part — to the wind being unable, as a rule, to lift above this
height particles of sand of a size or in numbers sufficient to erode
a rock surface. This limitation is apparently of wide application.
Kirk Bryan (5, p. 12) states that ail authorities are agreed that
two to three feet above the ground surface is the limit of effective
wind scour ; and he refers to the paper by W. H. Hobbs, who
shows (9, p. 33), among other examples, that in the Great
Oasis of the Libyan Desert the cast-iron telegraph poles lining
the railway were well burnished by the flying sand to a height
above the ground of only about a yard (see also p. 35). Hobbs's
observations are strikingly confirmed by the records given in this
paper, if wind-driven sand has caused the grooves and pits.

The actual process is difficult to visualize, but the following
suggestions are made. The more or less vertical grooves will be
first discussed. The sand must be lifted and driven against and
perhaps up or down the face of the rocks to the height mentioned.
Erosion may take place by this means, but so comparatively evenly
— except in specially favourable places — that there is no definite
record of the work of the sand blast. The “ frosting ” evenly
over the surface of glass by the sand blast in Nature is an Ulus-

Grooved , Pitted and Pedestal Rocks .

/ /

tration of this widespread erosion. It is easily recognized on the
glass on account of the smooth surface of % the glass at the com-
mencement of the bombardment of the sand grains, and the re-
sulting roughening of such surface. In rocks, however, such as
greenstones, the surface would show little recognizable change,
unless the action were very strong, and except, as already noted,
in specially favourable places, such as joint or division planes.
These provide lines of weakness along which the wind-driven
grains may erode faster than the adjacent portions of the rock. In
this way a slight groove may be made, which then supplies a defi-
nite passage along which the rasping sand grains may be pushed
up or down by the wind, which must be assumed, when it ap-
proaches the rock face, to he deflected in various ways. So the
grooves may deepen, widen and lengthen, and are probably most
pronounced close to the ground.

The typical pits on the faces of masses of rock rising well above
the surface of the ground are in the coarse-grained greenstone
knobs. The formation of the pits in these rocks is no clmibt fav-
oured by the comparatively large crystals of felspar and horn-
blende, of which the rocks are chiefly composed. Bombardment
by wind-driven quartz grains, to a height of about three feet above
the surface of the ground, takes place, and if, owing to ordinary
atmospheric weathering or crystallization of salt, a piece of felspar
or hornblende has been detached, a small hollow or incipient pit-
in the rock face will result. Sand grains are thrown against the
rock face to the height mentioned, and some must enter the pit.
Centrifugal action as suggested by Harger (8, p. xxxv), may be
set up, whereby the sand grains are whirled round the walls of
the cavity, thereby increasing its size.

Regarding the horizontal grooving of loose fragments of
rocks and to the undermining that takes place in the formation of
the miniature pedestal rocks, the grooves usually commence
about one -quarter to one-half of an inch above the base of the
fragment, although it has been shown in this paper that
where hands of varying degrees of toughness occur, and even in
apparently quite homogeneous quartz, there may be two horizon-
tal grooves, one above the other. The undermining of the pedes-
tal rock is also just above the surface of the ground. The diffi-
culty is to understand why the groove is formed so uniformly at
the height mentioned, and not only in one kind of rock, but also
in several classes.

Long and patient work would be necessary to determine tins
-question by actual observation; but if it be accepted, owing to the
elimination of all other possible agents as prime factors, that
wind-driven sand is the cause, then it must be assumed that such
sand, owing to the quantity available, or to its coarseness o\ giain,
or to the strength of the wind itself at the height mentioned, oi
all or some of these combined, acts most powerfully at that
height. The sand must act above this height, but appai cutl} so
* evenly that it shows no striking effects.

Harger (8, p. xxxiv) states that in late German South-West
Africa the cutting or eroding action of the sand-laden blast is the
most severe just above the ground level, the heavier grains of
sand acting like a rasp and in time cutting upstanding pillars of
rock right through, an example in granite being given. Another
result is the formation of “mushroom-topped" tors. Harger’s
observations are thus in accord with those recorded in this paper.

The writer’s observations do not show that pits and grooves
occur more frequently on one side of an outcrop than on another,
except in one or two instances where they are more numerous on
the eastern than on the western side.

The dominant winds appear to be westerly, but these may not be
the prevailing winds. The wind, however, probably forms eddies
in the vicinity of rock masses. See Hobbs (9, pp. 35 et seq.) and
Harger (8, p. xxxv), who states that the best and deepest honey-
combing is seen on the lee side of the rock masses.

In support of the wind theory, reference may be made to the
outcrops of a vertically banded siliceous ironstone about 10 to 12
inches thick, occurring at Goongarrie to the west of the lake.
These outcrops form a band at the junction of two other rocks,
and they may be traced in a north-north-westerly direction inter-
mittently for some miles. This band projects, on the average, for
about 12 inches above the surface of the surrounding ground.
Its surface is grooved, pitted, smoothed and rounded, and presents
a striking contrast with the sharp contours of similar rocks else-
where in the district, but situated under different conditions. The
bare surrounding ground and the abundance of fine quartz sand
in the vicinity leave little doubt that the wind is responsible for
the features described.

if the conclusions set out in this paper as to wind erosion, and
particularly with regard to the miniature pedestal rocks, be well
founded, they are important in that they support the idea that the
rock floor of the lake on its western shore is clue to wind planation.

The writer has in another publication (10) pointed out that the
mode of rounding described in that paper of fine-grained green-
stone pebbles does not take place until the iron oxide crust, which
is very widespread, has been broken. In the same group of rocks,
pitting and grooving do not occur, usually, unless this iron crust
is absent. The facts that the fine-grained greenstones from which
the crust has always been absent or from which it has been re-
moved after its formation, and that the pitting and grooving of
these greenstones occur generally at the base of cliffs, are coin-
cidences merely, since the coarse-grained greenstones and the
quartz and jasperoid rocks may be without a distinctive iron crust
at any height, but the pitting and grooving are restricted as shown
in this paper.

At “ The Tombstones,” the sedimentary rocks are free from
the iron crust, and pits abound in them, but as shown above, they
are, in the main, essentially solution hollows. In the adjacent

Grooved , Pitted and Pedestal Rocks.

rocks (probably porphyries), which project well above the sur-
face of the ground as “ tombstones/* an iron crust is well devel-
oped, and pitting is absent. Thus in those' two groups of rocks we
have striking examples of how erosion may be retarded or
hastened according to the occurrence or non- occurrence of the pro-
tective iron crust.

The relations between the crust-bearing and the crustless rocks,
especially those of the same kind, would probably repay close in-
vestigation.

VI. Records of Grooving and Pitting elsewhere by

Wind Action.

That pitting and grooving of rock surfaces have in some
instances has been caused by the action of the wind has been stated
by various writers. The following remarks summarize practically
all records that have come under the writer’s notice. He is in-
debted to Mr. Kirk Bryan, of the United States Geological Sur-
vey, for several of the references.

T. O. Bosworth (11) describes small corrugations and pits in
granite, due to erosion by wind blown sand, on the coast of
Mull, Scotland. The quartz and felspar in the rock have been
highly polished by this action.

W. P. Blake (12) describes the cutting by the sand blast of a
granite surface into “ long and perfectly parallel grooves and little
furrows ” on a mountain pass in California. The rocks were
also smoothed and polished by the action of the wind.

R. F. Rand (13) states that at Angra Pequena, on the south-
west African coast, biotite schist and granite have suffered great
pitting and honeycombing by the action of the wind, which is very
powerful and blows from the coast.

A. Wade (14) points out that the softer limestones of the
Eastern Desert of Egypt are sometimes regularly grooved by the
wind in such a manner as often to simulate bedding planes. Ande-
sites and porphyries are also grooved all over in a peculiar man-
ner by the action of wind-blown sand.

R. D. Oldham (15) describes grooves of varying size in quartz-
ites and sandstones caused by wind erosion.

R. W. G. Hingston (16) records the erosive action of the wind
in a gorge in Tibet. Granite boulders on their windward side
were polished, and were cut into by deep pits and grooves, some of
the latter an inch in depth.

E. de Martonne (17, pp. 663, 664) briefly refers to the disin-
tegration of heterogeneous rocks, such as sandstones, conglom-
erates and granites. Grains become detached and are swept
away by the wind. This results in the formation of a honey-
combed surface. The pits may be enlarged by wind corrasion
until potholes result, such potholes being common in granites (17,.
p. 668).

J . T. J at sov :

Johannes Walther ( 18, p. 16<8 and fig. 132, p. 169) describes
the well-known “ stone lattice ” of the desert.

H. S. Harger (8. p. xxxv and fig. 7) describes the occurrence
of “ augen ” gneiss in the late German south-west Africa, where
the rock has been extensively honeycombed by the corrasive action
of wind and weather. He points out that the best and deepest
honeycombing is seen on the lee side of the rock masses, and
that the holes, which were probably started by the weathering
away or falling out of a particular mineral, have been rounded
and enlarged by loose grit being whirled by the wind around the
walls of the cavities.

VII. Summary.

At Lake Goongarrie, a playa in sub-arid Western Australia, the
conditions are described under which certain rocks are being
grooved and pitted, and others undermined so as to form minia-
ture pedestal rocks.

The rocks concerned are “ greenstones,” quartz, certain jas-
peroid rocks, and shales. They occur as rocky cliffs of the lake ;
as small isolated rocky knobs; and as fragments scattered over
portions of the surface of the lake floor on the western shore, and
over the adjacent wind-exposed, low-lying ground, which has but
scanty vegetation.

These rocks are pitted and grooved. The grooves may run at
all angles in a horizontal plane, although in their inclination to
that plane tending generally towards the vertical, and, where this
occurs, the irregularities are due to the numerous small joints and
other division planes by which the rocks are traversed. Other
grooves are horizontal. The pits are mostly small, but they are
in some cases of moderate size.

Where the rocks occur as cliffs or knobs, the grooving and pit-
ting (except the pits at the locality known as “The tomb-
stones”) are restricted to a height of about three feet from the
base of the cliff or of the knob, as the case may be ; and where
they occur as surface fragments, the grooving and pitting are
restricted to those fragments which lie on nearly fiat wind-exposed
areas with scanty vegetation.

The miniature pedestal rocks are of the fine-grained greenstone
class, and occur on the western shore of the lake close to the
cliffs of the same rocks, which there hound the lake.

From a consideration of the mode of occurrence of the grooves
and pits and of the tiny pedestal rocks, water, both in its mechani-
cal and solvent action, is eliminated as the chief agent in the pro-
duction of the phenomena described. 1 he disruptive power of
salts brought by capillary attraction to the surface, and there crys-
tallizing, is also practically eliminated ; and the wind, acting in its
abrasive capacity, is regarded as the predominant factor. This,
however, does not apply to the pits at “ The Tombstones,” where

Grooved , Pitted and Pedestal Rocks . 81

water, in its solvent action, is considered to be the chief cause of
those pits.

A brief account of extra- Australian records of pitting and
grooving of rocks is given.

The writer is indebted to Professor Skeats for criticism of this
paper.

VIII. References.

1. J. T. Jutson. The Sand Ridges, Rock Floors and other

Associated Features at Goongarrie Playa. Proc.

Roy. Soc. Vic., n.s., xxxi. (1), pp. 113-128, 1918.

2. J. T. Jutson. The Mining Geology of Comet Vale and

Goongarrie, North Coolgardie Goldfield. W. A lust . Gcol.
Surv. Bull . 79, 1921.

3. Kirk Bryan. Pedestal Rocks in the Arid South-West.

US. Gcol. Surv. Bull. 7 60-A, Washington, 1923.

4. Kirk Bryan. Pedestal Rocks in Stream Channels. Ibid.,

Bull. 760-D, Washington, 1925.

5. Kirk Bryan. Pedestal Rocks formed by Differential Ero-

sion. Ibid., Bull. 790- A, Washington, 1926.

6. J. T. Jutson. The Sand Ridges, Sand Plains and Sand

Glaciers ” at Comet Vale in Sub-arid Western Australia.
Proc. Roy . Soc. Vic., n.s. xxxi. (2), pp. 412-420, 1918.

7. J. T. Jutson. The Influence of Salts in Rock Weathering

in Sub-arid Western Australia. Ibid., n.s., xxx., pp. 165-
172, 1918.

8. H. S. Harger. Some Features associated with the Denuda-

tion of the South African Continent. Proc. Gcol. Soc. S.
Africa, 1913, xvi, pp. xxii-xxxix, 1914.

9. W. H. Hobbs. The Erosional and Degradational Processes

of Deserts, with Especial Reference to the Origin of
Desert Depressions. Annals Assoc. Amer. Geographers,
vii., pp. 25-60, pis. xiv. xxvi., 1918.

10. J. T. Jutson. Note on an Unusual Method of Rounding of

Pebbles in Sub-arid Western Australia. Amer. Journ.
Sci., xlviii., pp. 429-434, December, 1919.

11. T. O. Bos wort i \ . Wind Erosion on the Coast of Mull.

Gcol . Mag., n.s. |5], vii. (8), pp. 353-355, pis. xxviii.,
xxix., Aug. 1910.

12. W. P. Blake, On the Grooving and Polishing of hard

Rocks and Minerals by dry Sand. Amer. Journ. Sci.,
1 2], xx., pp. 178-181, 1855.

13. R. F. Rand. Angra Pequena (Liideritzbucht) and Subaerial

Denudation. Gcol. Mag., 1920, pp. 32-35.

14. A. Wade. Some Observations on the Eastern Desert of

Egypt, with Considerations bearing upon the Origin of
the British Trias. Quart. Journ. Gcol. Soc., lxvii., pp.
238-262, pis. xiii.-xvi., 1911.

J. T. Jutson :

15. R. D. Oldham. Note on Blown Sand Rock Sculpture.

Rec. Gcol. Surv. India, xxi., pp. 159-160, pi., 1885.

16. R. W. G. Hingston. Animal Life at High Altitudes.

Gcog. Journ., pp. 185 et seq., March, 1925.

17. E. de Martonne. Traite de Geographie Physique. 2 Ed.,

Paris, 1913.

IS. Johannes Waltiier. Das Gesetz der Wiistenbildung in

Gegenwart und Vorzeit. 4 Ed., Leipzig, 1924.

EXPLANATION OF PLATES.

Plate XII.

Fig. 1. — Rectilinear grooving in the fine-grained greenstone.

Cliff face south of the large “ natural quarry ” on the
western shore of Lake Goongarrie at Comet Vale.
Note the absence of grooving towards the top of the
figure.

Fig. 2. — Miniature pedestal rock of fine-grained greenstone on
the floor of the western shore of Lake Goongarrie at
Comet Vale, near the large “ natural quarry.” Slightly
less than natural size.

Fig. 3. — Pebbles of the fine-grained greenstone, showing the
horizontal groove above the base. From the rock
floor of the western shore of Lake Goongarrie at
Comet Vale, near the large “ natural quarry.” Natural
size.

Plate XIII.

Fig. 1. — A fragment of the coarse-grained greenstone, showing
the almost continuous horizontal groove slightly above
the base. Some of the pits on the top can be observed.
From west of the Golden Sun Lease, Goongarrie.
Natural size.

Fig. 2. — A fragment of quartz with a horizontal groove slightly
above the base, particularly shown in profile at each
end. From flat cone at the foot of Poverty Point, north-
ern end of Lake Goongarrie, Comet Vale. Natural
size.

Fig. 3. — A fragment of jasperoid rock, showing its rounded sur-
face and the deep notch at one end. From near the
Lady of the Lake Lease. Goongarrie. Natural size.

ADDENDUM.

A series of papers by the writer treating of various phases of

physiography in sub-arid Western Australia appeared in earlier

volumes of this Journal. The proofs of most of those papers

Proc. R.S. Victoria, 41 (2), 1929. Plate XII.

Fig. 2

Fig. 3

Proc. P.S. Victoria, 41 (2), 1929. Plate XIII.

Fig. 1

Fig. 2

Fig. 3

Grooved, Pitted and Pedestal Rocks .

were not corrected by the writer, with the result that various
typographical errors have crept in, which it is now desired to
correct, without including all minor palpable errors.

Corrigenda.

Proc. Roy . Soc. Vic.

N.S., XXX. (2), 1918.

Page 163. 2nd line: For 44 water ” read 44 matter.”

Page 171, 25th line: For 41 corrosive ” read 44 corrasive.”

N.S., XXXI. (1), 1918.

Page 119, 19th line: For 44 gravitional ” read 44 gravitational.’ 7

26th line : Delete 44 upon.”

Page 121, 3rd line: For “ corrosion ” read 44 corrasion.”

Page 124, 4th last line of main text: For “corrosion” read
44 corrasion.”

Page 125, 10th line: For 44 erosinal ” read 4< erosional.”

14th line: For 44 solutions ” read 44 solution.”

22nd line : For 44 corrosive ” read 44 corrasive.”

Page 126, 10th last line: For 44 one ” read 44 once.”

Page 128: Add at the end the following: —

Expan ation of Figures.

Fig. 1. — Locality map of the southern portion of West-
ern Australia.

Fig. 2. — Diagrammatic section showing a piedmont plain
truncated at the lake floor. See pp. 117 and 123.

Fig. 3. — Diagrammatic section across an arm (with a
sand ridge on either side) of the lake, showing
the bare rock floor of the lake and the detritus
beneath the sands of the ridges. See pp. 119
and 125-128.

Fig. 4. — Diagrammatic section (after Hobbs) from high
land to a playa surface.

Fig. 5. — Diagrammatic section across the lake, showing
the lake floor abutting the 44 high ” lands. See
page 123.

Fig. 6. — Diagrammatic section across the lake, showing a
narrow piedmont plain intervening between the
rock floor and the 44 high ” lands. See p. 123.

Pages 113-128: Throughout for 44 lowlands,” read 4 ‘ 4 low ’
lands.”

X.S., XXXII. (2), 1920.

Page 314, 5th last line : For “ 38 ” read 44 83.”

Page 315, 19th line: For “ willy-willy,” read 44 willy-willys.”

Page 317 : The first two sentences under 44 General Remarks ”
apply to the first eight records.

Page 319, 3rd last line : For 44 could ” read 4< would ” ; and last
line, for 44 hemispheres ” read 44 hemisphere.”

84 Jutson : Grooved , Pitted, and Pedestal Hocks.

Page 320, 20th and 21st lines: The words, “but the rate of
motion could not be determined should be in paren-
theses.

27th line : Delete the comma after “ dull.’'

Page 321, 12th line: For “ columns ” read “ column.”

14th line: For “ over ” read “ ones.”

24th line: For “ whirls ” read “ whirl.”

Page 322, 2nd last line : For “ 325 ” read “ 52.”

[Pkoc. Roy. Soc. Victoria, 41 (N.S.), Pt. IT.. 1928.]

Art. VII . — Notes on Australian termites ( Isoptera ),
Descriptions of new species.

By GERALD F. HILL.

(With Plate XIV.)

[Read 13th September. 1928; issued separately 30th January, 1929.]

In this paper the following termites are described as new,
namely, two species of Hamiternies from Western Australia, and
one -pecies of J\I iro ter me s from each N.W. Australia, N. Queens-
land and Victoria. The hitherto undescribed alate form of
Eutcrmcs marcebensis from N. Queensland is described.

The genus Hamiternies is widely distributed in Australia, where
it is represented by 16 described and numerous undescribed
species. Included in the latter are several very striking examples
of the sub-genus Drepanotcnnes , which it is considered are best
held over until the alate forms have been discovered.

The genus Miroteruics is represented bv 18 previously described
species, which are listed in a recent paper (Hill. 1927). It is of
interest to note that one of the species described in the following
pages is the first of the genus to be recorded from Victoria.

'Eutcrmcs , the third genus referred to in this paper, is repre-
sented in Australia by 32 described and probably as many unde-
scribed species, many of which are so closely similar in the soldier
caste that the group can be satisfactorily dealt with only in a re-
view of the whole of the available material. The inclusion here
of a description of an undescribed caste of a hitherto incompletely
known form, however, appears to be justified in view of the fact
that the remainder of the writer’s collection contains only com-
pletely described or wholly undescribed species.

Hamitermes westraliensis, n. sp.

(Plate XIV., Figs. 1.2.)

Imago.

Colour. — Head very dark brown, slightly darker than prono-
fcum ; postclypeus and antennae distinctly lighter ; anteclypeus
hyaline ; legs, labrum and palpi yellow-ochre to ochraceous-tawm ;
pleurites and sternites mostly yellow-ochre, the former suffused
with brown, the latter brown around spiracles; nieso- and meta-
thorax and tergites" somewhat paler than pronotum ; wings dark
brown, veins very distinct.

Gerald F. Hill:

Head. — Hemispherical behind the eyes, depressed angularly in
front of fontanelle, a little wider than pronotum, very setaceous,
the setae long and short, as on pronotum. Eyes small, prominent
(0-192 diam.) surrounded by a pale ring. Ocelli oval (0 096 X
0-144) widely separated (0-128) from eyes. Antennae 16-
jointed; 1st joint more than twice as long as 2nd and markedly
wider; 2nd a little longer than wide; 3rd very short and closely
fused with 4th, which is scarcely longer ; 5th about as long as 3rd
and 4th together, longer than 6th. Postclypeus markedly seta-
ceous like labrum, with brown median suture, truncate in front,
markedly convex behind, strongly arched above, 0*48 long x0*65
wide. Fontanelle large, about as large as ocelli, oval, with linear
extension anteriorly, anterior margin of ovate portion on line with
posterior margin of eyes. Anteclvpeus hyaline, nearly straight on
sides, strongly produced in front.

Thorax. — Pronotum large, nearly straight in front, antero-
lateral corners somewhat rounded, sides sloping rather sharply to
the slightly sinuate hind margin. Meso- and metanotum narrowed
sharply to the deeply notched posterior margin.

Wings (PI. XIV., Fig. 1 ). — Large, dark coloured, with dis-
tinctly darker veins, the latter distinct to their extremities, the
proximal half of the two anteriormost veins markedly setaceous;
mierotrichia moderately numerous.

Legs. — Moderately long and setaceous.

Abdomen. — Markedly setaceous; cerci with large basal seg-
ment, as long as apical segment.

Measurements.

Length, with wings -
Length, without wings

Head, from base to apex of labniin. long -
Head, from base to el ypeof rental suture, long
Head, at and including eyes, wide -
Antennae, long -----
Pronotum. long. 0*68; wide
Wings, forewings, long, 12-50; wide
Wings, hindwings, long, 12-00; w‘de
Tibia iii. long -

Soldier.

Colour. — Head vellow-ochre; thorax, mouth parts (excepting
mandibles) and legs light buff; mandibles yellow-ochre at base,
shading to dark chestnut towards apex ; labrum and clypeus
yellow-ochre, margined anteriorly with hyaline, a dark chestnut
spot at articulation of mandibles.

Head (PI. XIV., Fig 2). — Long and very little widened on
sides, wide and only slightly rounded behind, strongly arched
«bove ; clothed very scantily with reddish setae, these most nume-
rous on the f rons. Mandibles relatively short, little more than

mm.

14-25 — 14'50
7*50— 8-00

1-60 — 1-67
0-74 — - 0-S0

1- 30

2- 04
1-24
3*28

3- 40
1-36

A usti xtl i an Ten > i i tes.

half as long as head capsule, stout, strongly curved in at the tips,
each with a short tooth a little nearer base than apex. Labrum
moderately large, conical, with several stout reddish setae, a little
more than half as long as mandibles. Clvpeus shorter than labrum,
anterior margin strongly bilobed and broadly margined with hya-
line. Antennae 15-jointed; 3rd joint very short, much shorter
than 1st and 2nd; 4th longer than 5th, about equal to 2nd; gula
about one-fourth as wide as head.

Thorax.— Pronotum short and wide, with few long reddish
setae, mostly near margins, anterior half narrowed and bent up,
with slight emargination. anterolateral angles markedly produced,
sides narrowed sharply to obscurely sinuate posterior margin;
mesonotum narrower than pronotum, with wide but not deep
emargination, fringed with reddish setae; melanotum wider than
mesonotum. similarly fringed, posterior margin not so strongly
emarginate.

Legs . — Moderately long and slender, with very scanty setae.

Abdomen. — With scanty reddish setae.

Measurements.

Total

length

mm.

5* 50 —

6*00

Head.

to apex of mandibles, long

2-35 •—

2-41

Head.

to apex of labrum. long

2 04

Head.

to labral suture

1-62

Head,

wide -

1-36 —

1-42

Head,

greatest depth, including gula -

1-17

Gula.

long ... -

0-32

Pronotum. long 0*50; wide-

0*93

Tibia

iii, long -

1-24

Worker.

Colour. — Warm buff, frons whitish, antennae a little darker
than head, mandibles chestnut.

Head. — Posteriorly from the insertion of the antennae hemi-
spherical, with very few setae; postclypeus about one-third wider
than long, strongly arched above, anterior margin truncate, pos-
terior margin markedly convex, with scattered reddish setae, a
dark ferruginous spot at each end; antennae 16- jointed.

Thorax. — Pronotum as in soldier.

Measurements.

Total length -

Head, to apex of labrum. long
Head, to cl ypeof rental suture, long -
Head, wide - -

Pronotum. long 0 43; wide -

mm.

4-50 — 4-90

1*55
1*05
1-30
O' 86

Locality. — Western Australia; Darlot (Charles Biddle,
6.12.27).

Types ( imago, soldier, worker) in the author’s collection.

Gerald F. Hill:

Hamitermes (Drepanotermes) tamminensis, n. sp.

(Plate XIV., Figs. 3-6.)

Imago.

Colour. — Head, pronotum and principal veins of wings dark
brown; antennae and legs light yellowish-brown; pleural sclerites
and tergites brown, lighter than head; sternites yellowish, suffused
with dark brown at spiracles ; postclypeus yellowish-brown suf-
fused with dark brown on sides.

Head (PI. XIV., Fig. 3). — Clothed with many small setae,
widest in front, narrowed slightly posteriorly. Eyes very small
and prominent. Ocelli widely separated from eyes. Fontanelle
very large, broadly oval Antennae with 16-17 segments (gene-
rally 17) ; 1st long and stout, cylindrical; 2nd less than half as
long and much narrower than 1st, cylindrical ; 3rd and 4th very
small, closely fused; 5th globose.

Thorax.— Pronotum clothed similarly to head, slightly narrower
than head, anterior margin nearly straight, anterolateral angles
broadly rounded, the posterior margin broadly rounded with
obscure indentation in middle ; posterior margin of meso- and
metanotum widely notched.

Wings (PL XIV., Figs. 4, 5). — Large, all veins very distinct;
the two anteriormost veins very setaceous.

Legs. — Long and slender, with numerous small setae.

Abdomen. — Moderately setaceous; setae small.

Measurements.

Total length -
Length, without wings
Head, to apex of lahrnm. long
Head, to clypeof rental suture, long -
Head, wide - -
Antennae, long -
Eyes, diam. -
Eyes, from ocelli
Eyes, from lower margin of head
Ocelli, Longest diam. -
Pronotum. long 0*93; wide
Wings, forewings, long 16:00: wide
Wings, liindwings. long 15*00; wide
Tibia iii, long -

Soldier .

Colour. — Head, antennae and pronotum dark orange-yellow;
legs, tergites of abdomen light clay colour; anterior margin of
anteclypeus whitish; labrum lemon-vellow; mandibles mahogany-
red.

Head (PI. XIV., Fig. 6)— Elongate oval, with very few
setae; frons rugose; postclypeus strongly bilobed, divided medi-
ally by a deep groove; mandibles long and slender, with broad
tooth on each about the middle, tooth on left jaw larger than that

mm.

10*00 — 20*00
9*00
1*98
1*05
1*60
2*54

0*240X0*288
0*160
0*144
0 160

1*48 — 1*55
4*27
4 52
1*86

Australian Termi tes.

on right; gula long and narrow, rather more than one-fourth
wider in middle than head. Antennae with 16 (very rarely 17)
segments; 1st segment long, moderately wide, twice as long as
2nd, slightly widened towards the apex; 2nd much narrower than
1st, nearly cylindrical; 3rd a little shorter than 2nd, narrow at
base but as wide as 2nd at apex; 4th shortest of all; 5th as long
as 3rd; 6th-llth lengthening progressively.

Thorax. — Pronotum small, with very few setae, these stout and
confined to near the margin except on anterior one-third which
bears scattered setae; the anterior one-third narrowed and sharply
bent up and rounded on anterior margin ; posterior margin
rounded, very slightly sinuate in middle. Mesonotum about as
wide as pronotum, with sinuate posterior margin and setae as on
pronotum ; metanotum markedly wider, but not longer, than
mesonotum ; posterior margin and setae as in the latter.

Legs. — Long and slender, with scanty setae ; claws and spines
small and slender.

Abdomen. — Short and wide, narrowed abruptly to bluntly
pointed apex, with rather scanty, long setae ; circi large, the basal
segment large, nearly as long as the apical segment, which is
slender.

Measurements.

Total length -

Head, to apex of mandibles, long -
Head, to external articulation of mandibles,
long -

Head, wide -----
Gula, at narrowest part, 0-031; long
Antennae, long -----
Pronotum, long O' 62; wide
Tibia iii, long

Worker.

Colour. — Head, upper surface as in soldier, sides and frons
shading to light straw ; postclvpeus suffused with orange-yellow
laterally.

Head. — With very few pale setae, widest in front, narrowed
posterior margin ; postclypeus about half as long as wide, nearly
truncate in front, broadly rounded behind. Antennae with 17 seg-
ments ; 3rd segment shortest.

Measurements.

Total length

mm.
5-80 —

6*20

Head, to apex of labruin, long

1*86

Head, to clypeofrontal suture, long -

1-17 —

1-24

Head, to external articulation
long

of mandibles ,

1*42

Head, wide

1-61 —

1-67

Antennae, long

2-91

Pronotum, long 0 62 - — 0*68;

wide

0-93 —

1-00

Tibia iii, long

1-79

mm.

5-90

2-66 — 2*72

1-67 — 1-79
1*36 — 1*42
0*062
3 r 28

0-93 — 0-99
1S6 — 1-92

Gerald F. Hill:

Localities. — Western Australia: Tammin (type locality) all
castes, Eradu, soldiers and workers, Merredin, soldiers ( J. Clark).
Geraldton, queen, soldiers and workers (Edwin Ashby, in Janu-
ary) .

Allied Species. — This species is most closely allied to Hand -
tenues (Drcpauotcnnes) silvestrii Hill from Townsville, N.Q.,
(Bull. Rut. Res., xii. (4), p. 364, 1922), from which the imago
is distinguished, inter alia, by its much smaller size, darker (less
reddish) wings, and fewer antennal segments, and the soldier by
its oval, smaller and lighter coloured head and fewer antennal seg-
ments.

Types (imago, soldier and worker) in the author’s collection.

Eutermes vf areebe n sis Hill.

Proc. Linn. Soc. N.S.W., xlvii. (2), 1922.

(Plate XIV., Figs. 7, 8.)

Imago.

Colour. — Head and thorax mummy-brown; tergites of abdo-
men very little paler; clvpeus buckthorn-brown, mandibles (ex-
cepting teeth), antennae and trophi a little paler; coxae, trochan-

Fig. 1 . — j Eutermes mareebensis, Hill, imago, a, head, b f thorax.
Mirotermes argutus, n.sp., c, soldier, head.

Mirotermes insitivv.s f n.sp., d, imago, antennae.

Australian Termites.

ters, femora and tibiae somewhat lighter than tergites; tarsi
whitish, sternites paler than legs, the first four mostly pale yellow-
ish-brown suffused with darker colour laterally; wings smoky,
with principal veins same colour as tergites.

Head (Text-fig. \a). —Moderately setaceous, widest in
front, almost hemispherical behind the eyes when viewed
from above, frontal and transverse sutures distinct. Fon-
tanelle large, linear, nearly as long as eyes are wide, the an-
terior end slightly widened. Eyes small (0* 170X0- 170) promin-
ent, about as far from lower margin of head as they are from
ocelli. Ocelli large (0-085 long), oval, oblique. Postclvpeus
large, distinctly lighter than head with a fairly distinct brownish
median suture, twice as wide as long, the anterior margin trun-
cate, with a scanty fringe of setae, these mostly longer than the
few on the remainder of the postclypeus, posterior margin hemi-
spherical; anteclypeus large, whitish, produced in the middle.
Antennae 14- jointed; 1st joint large, cylindrical; 2nd a little
more than half as long as 1st and a little narrower; 3rd markedly
shorter and narrower than 2nd smallest of all; 4th- 13th progres-
sively longer and wider, the 13th nearly as long as 1st; 14th as
long as 1st. narrower than 13th, widest at proximal third.

Thorax (Text-fig. 1 b ). — Pronotum moderately setaceous, a
little narrower than head, slightly arcuate in front, anterolateral
corners, narrow sides sloping sharply to the narrow and slightly
sinuate posterior margin; posterior margin of meso- and meta-
notum more sinuate than in pronotum.

Wings (PI. XIV., Figs. 7, 8). — Slender, the radial sector and
first six or seven branches of the cubitus very distinct, the former
and the margin very setaceous. Membrane with rather numerous
microtrichia, and densely covered with star-like micrasters. The
media passing through the wing a little above the middle, and
joining the margin a little above the apex, generally with one
branch about the distal fifth to the hind margin; the cubitus with
about eleven branches, the first two or three very short, the others
not so well-defined, but easily discernible to their extremity,
mostly unbranched.

Legs. — Dark coloured, short and comparatively stout ; the
femora with scanty setae ; tibiae with markedly stronger and more
numerous setae ; tibiae, spurs and claws long and slender.

Abdomen. —

Measurements.

Length, with wings ^ 7*50 — 8*00; £

Length, without wings J 1 4'10 — 4 • 44 ; $
Head, to apex of labrum. long
Head, to dypeofrontal suture, long -
Head, at and including eyes, wide -
Antennae, long
Pronotum. long 0*34; wide
Forewing, long 6 '00 wide
Tibia iii, long -

mm.

8*00 — 8-50
4 '67 — 4-90

0- 91
0'46
0 ' 69
1*10
0*60

1- 52

0'62 — 0'68

Gerald F. Hill:

Locality. — North Queensland; Meringa (F. H. Taylor, Nov.,
1924).

The identity of the above has been established by comparison of
associated soldiers and workers with the types of these castes
(from Mareeba, Cairns hinterland, N.Q.).

Types (imago, soldier and worker) in National Museum, Mel-
bourne.

Mtrotermes argutus, n. sp.

Soldier.

Colour. — Head, antennae and palpi light orange-yellow, thorax
and legs stramineus.

Head (Text-lig. lr). — Long and narrow, parallel on sides, with
scanty long setae; frontal process large, stout at base, rather
bluntly pointed, the extreme tip bent upwards. Antennae 14-
jointed, slender; 1st joint large, more than twice as long as 2nd
and one-third wider, slightly swollen at apex; 2nd longer than
wide, parallel on the sides; 3rd and 4th smallest, 4th a little
smaller than 3rd; 5th-9th increasing in length progressively; 10th-
14th subequal, long and narrow (about as long as 8th). Labrum
narrow, parallel on the sides, truncate in front with the anterola-
teral corners produced into points. Mandibles very long and
slender. Gula long and narrow, about one-fourth as vide as

Thorax, — Pronotum small, much narrower than head, saddle-
shaped, anterior margin convex in the middle, the anterolateral
corners narrowed, sides and posterior margin together neatly
hemispherical, with scanty long setae.

Measurements.

Total length

Head, to apex of frontal process, long

Head, deep - -

Head, wide -

Gula. at narrowest part, wide

Antennae, long

Mandibles, long - *

Pronotum. long 0*27 — O’ 28; wide
Tibia iii. long -

mm.

4-25

1-42

0'62 — 0-74
0*86
0*228
1*70
1*14

0*56 — 0*57
0*62

Worker .

Colour.— Head light orange-yellow, frons whitish, antennae,
thorax and legs pale stramineus.

Head. — Glabrous, almost spherical as seen from above, widest
at antennae, with scanty setae. Clvpeus large, strongly convex,
with obscure median suture, a pale ferrugineous mark at articula-
tion of mandibles, with a few setae ; anteclypeus large, nearly as
long as post clvpeus. Labrum small, markedly convex, with few
setae. Antennae 14-jointed ; 3rd and 4th joints smallest, closely

A ustral tan Term i tes.

fused, 5th shorter than 2nd, globular; 6th- 14th increasing in
length progressively; 14th noticeably longer than 13th, narrowed
ffrom the proximal fourth to the pointed apex.

Thorax. — Pronotum, as in soldier.

Measurements.

mm.

Total length - - - - - 3-50

Head, to apex of labrum, long - - 0'85

Head, to clypeofrontal suture, long - - 0'60

Head, wide - 0 ’74 — 0*80

Pronotum, long 0*22; wide - - - 0-45

Tibia iii, long - - - - - O’ 60

Locality. — Victoria, Ivewell.

Described from a soldier and three workers; found under a log
(February).

Allied Species.— The soldier differs from the typical form of Miro-
tennes kraepclini Silv. in having a shorter and narrower head,
more angular frontal process, and much more slender mandibles.
From variety “A” of the last named species (Hill, Mem. Nat.
Mns Melb., No. 7. 1927, p. 95) it differs in having a more
angular and straighter frontal process, longer, shallower and
deeper coloured head and different labrum; from variety “ C ” it
differs in its smaller and narrower head, more slender mandibles,
narrower and otherwise different labrum ; from variety “ E ” it
differs in having a shorter and narrower head, more slender man-
dibles, narrower labrum and more slender frontal process.

Types (soldier and worker) and others in National Museum,
Melbourne; collected and donated by Mr. Jas. A. Hill, of Mur-
toa.

Mirotermes jarmuranus, n. sp.

Imago.

Colour. — Head, thorax and tergites of abdomen argus-brown;
legs, antennae and sternites of abdomen buckthorn-brown ;
clypeus ochraceous-tawny. The whole insect densely setaceous,
many of the setae markedly long.

Head. — Almost hemispherical when viewed from above, the
summit depressed, fontanelle obscured by setae. Eyes moderately
small (0*306X0*306) and prominent. Ocelli large (0*170 long)
broadly oval, a little less than their short diameter from eyes.
Antennae 15-jointed, very long; 3rd joint very large, not much
smaller than 1st; 2nd very small, bead-like; 4th and 5th long and
narrow, shorter and narrower than 3rd; 6th longer than 4th and
5th ; remaining joints very long and narrow. Mandibles each with
apical tooth much larger than, and widely separated from, the
next, Postclvpeus small, strongly convex above, with numerous
long and short setae; anteclypeus whitish, short, truncate in front.
Labrum small, longer than wide, not covering apex of mandibles.

Gerald F. Hill ;

Thorax. — Pronotum very large, markedly longer than wide,,
strongly arched, conspicuously concave in front, sides sloping to
the broadly rounded posterior margin. Posterior margin of meso-
ar i metanotum narrowed and deeply notched, the metanotum
more so than mesonotum.

Wings. — Wing stumps small, those of mesonotum not much
larger than those of metanotum.

L e cr S . — Moderately short and stout; very setaceous.

o“ •

Measurements.

Length, without wings

Head, to apex of labnun. long

Head, to clypeofrontal suture, long -

Head, at and including eyes, wide -

Head, deep -

Antennae, long

Pronotum, long 0*85: wide

Tibia Hi, long

mm.
6 15
0*96
0*74
1*19
0*51
2*28
1*19
1*19

Locality. — North-West Australia: 130 miles south-east of
Broome (July or Aug., 1924).

Described from a dealated female, collected and presented to
the National Museum, Melbourne, by A. S. Cudmore.

It is possible that the specimen described above is the macrop-
terous form of M. broomensis Mjob., only the soldiers and work-
ers of which have been described.

Mirotermes tnsitivus, n. sp.

(Plate XIV., Figs. 9-11.)

Imago.

Colour. — Head very dark brown, postclypeus very little lighter
than head, anteclypeus whitish; labrum suffused with yellow;
antennae, mouth parts and legs, light brown ; pronotum nearly as
dark as head ; wings dark smoky.

Head (Plate XIV,, Fig. 9). — Small, rounded when viewed
from above, rather densely clothed with short and long setae.
Eyes rather large (0*289X0*289), prominent, close (0 04) to
lower margin of head, closer than to ocelli. Ocelli large (0*136),
broadly oval, separated from the eyes by a space equal to their
short diameter, Postclypeus moderately large, about twice as wide
as long, strongly convex, hemispherical behind, truncate in front,
with rather distinct median suture and clothed moderately densely
with small setae ; anteclypeus less than half as long as postclypeus,
nearly truncate in front. Labrum small, a little widened in the
middle, moderately convex, densely setaceous, broadly rounded
in front. Fontanelle small, oval, laying within a small depressed
area and in line with the middle of the eyes (in all cleared pre-
parations there is to be seen a dark-coloured, broadly oval pos-
terior extension of the fontanelle).

A ustralian Termites .

Antennae (Text-fig. Id). — 15- jointed; 1st joint short and wide,
two-thirds as wide as long, as long as 14th; 2nd short and wide,
as wide as long ; 3rd very short, but nearly as wide as 2nd ; 4th and
6th about equal to 2nd ; 5th wide, but a little shorter than 4th
and 6th ; 8th to 13th about equal, a little longer than 7th, the latter
wider at base than the following joints; 14th, a little longer than
13th, but hardly as long as 15th, which is elongate-oval, and
widest in the middle. Mandibles with apical tooth on each side
much larger than the next; dentition as shown in (PI. XIV.. Fig.

9).

Thorax. — Pronotum very large, as wide as head, densely seta-
ceous, rather strongly arched, the anterior margin broadly con-
cave, anterolateral angles rounded, sides sloping sharply to the
narrow posterior border, the latter sometimes almost truncate,
but generally markedly emarginate. Meso- and metanotum with
the sides markedly narrowed posteriorly, the hind border of the
former generally emarginate as in pronotum, that of the latter
much more deeply and acutely notched, both sclerites markedly
setaceous, though less setaceous than pronotum.

Wings (PI. XIV.. Figs. 10, 11). — Generally with hindwings
a little longer and wider than forewings, rather wide relatively
to length, of same colour as those of M. kraepelini, very seta-
ceous along border and radial sector ; the two anteriormost veins,
especially the radial sector, very dark; all the veins distinct to
their termination; the media passing through the upper third of
the wing, with 4 or 5 branches, the first generally a little beyond
the middle and sometimes branched, the main stem joining the
margin near the apex of the wing; the cubitus with 8-14 branches,
seven or eight nearest the base very dark and distinct. Mem-
brane with few microtrichia, but densely covered with micrasters.
Wing-stumps small, very setaceous, suture straight.

Legs. — Densely setaceous, femora a little less so than tibiae;
claws and spurs long and slender.

Abdomen. — Very setaceous, the setae shorter and finer than
those on head and thorax ; circi short and very wide at base.

Measurements.

mm.

Length, with wings -
Length, without wings
Head, to apex of labrum, long
Head, to clypeofrontal suture, long -
Head, at and including eyes, wide -
Antennae, long

Pronotum, long 1*60 — 1’70; wide
Forewings, long 9‘00 — 9*50; wide
Tibia iii, long -

11-00 — 11-50
6-00 — 6-50
1*14

0- 57

1- 14

1*60 — 1-70
1-14 — 1-30
2*70
1-14

Locality. — N. Queensland: Townsville (G.F.PI., 22.12.19,

6.1.20, 15.2.21).

Gerald F. Hill : Australian Termites.

<F>

Biology. — The association of this species with Eutermes ver-
noni Hill is referred to in an earlier paper (Hill, P.L.S. NSW.,
xlvii. (2), 1922, p. 148, 2nd line).

Type in the author’s collection.

A CORRECTION.

In my paper entitled ‘‘Termites (Isoptera) from South Sea
and Torres Strait Islands,” Proc. Roy. Soc. Victoria , xxxix. (1),
11th Nov., 1926, the reference to Calotcrmes ( Calotermes ) rcpan-
d us Hill is given as “ Memoirs of the National Museum, Mel-
bourne , No. 7, in Press.” The correct references are “ The Ento-
mologist , yj lix., Nov., 1926, p. 297, and “ Insects of Samoa,” part
vii., 28th May, 1927, p. 6.

It should be noted also that owing to long delay in the Press,
the description of the imago of Calotcrmes (Glyp'totermes) xan-
tholabrum Hill (Mem. Nat. Mus., Melbourne, No. 7, June, 1927)
is antedated by the subsequently written preliminary and full de-
scriptions of the soldier caste in “ The Entomologist,” and “ In-
sects of Samoa,” of the above-mentioned dates, respectively.

ACKNOWLEDGMENTS.

Cordial thanks are extended to Mr. J. A. Kershaw, National
Museum, Melbourne, to the collectors who have courteously made
available the material dealt with in this paper, and to Messrs. G.
McLennan, B.V.Sc., and D. Murnane, B.V.Sc., for the photomic-
rographic illustrations.

EXPLANATION OF PLATE XIV.

Figs. 1, 2. — Hamit ermes westraliensis, n. sp.

„ „ 1. Imago: wings.

,, ,, 2. Soldier: head.

Figs. 3, 6. — Hamitermes tamminensis, n. sp,

„ „ 3. Imago.

„ „ 4. Imago : wing.

,, „ 5. Imago : micrasters.

„ „ 6. Soldier.

Figs. 7, 8. — Eutermes mareebensis Hill.

„ „ 7. Imago: wing.

„ ,, 8. Imago : micrasters.

Figs. 9, 11. — Mir ot ermes insitivus, n. sp.

„ „ 9. Imago : head and

thorax.

„ ,, 10. Imago: wings.

„ „ 11. Imago: micrasters.

Proc. R.S. Victoria, 41 (2), 1929. Plate XIV.

[Proc, Roy. Soc. Victoria, 41 (N.S.), Pt. II., 1929.]

Art. VIII. — The Devonian and Older Palaeozoic Rocks of
the Tabherabbera District , North Gippsland, Victoria .

PROFESSOR ERNEST W. SKEATS, D.Sc., A.R.C.S., F.G.S.

(University of Melbourne).

(With Plate XV.)

[Read 11th October, 1928; issued separately 8th January, 1929.]

I introduction.

The geology of the Tabberabbera district has long been recog-
nised as presenting interesting problems concerning the Middle
Devonian sediments, because their lithological characters, as well
.as the subsequent earth movements which have affected them,
present a contrast with the rocks of the better-known areas of
Buchan and of Bindi in Eastern Victoria. As no geological
work had been done in the region under discussion for over a
quarter of a century, I welcomed the opportunities which arose in
January, 1924, and January, 1925, of paying visits to the area.
Through the kindness of Mr. VV. Baragwanath, Director of the
Geological Survey of Victoria, camping facilities were madie
available. Mr. Baragwanath joined me for the first four or fivb
days of field work, Mr. J. Easton, Geological Surveyor, and his
assistant (Mr. Norman Winter) were with me throughout thp
three weeks spent in the field, and Mr. Keble, of the Geological
Survey, was with us for the last five days of the first trip, and
Mr. Easton was with me during three weeks of the second visit.
While I am responsible for the form and substance of this com-
munication, I owe much to help rendered in the field by the gentle-
men above mentioned.

Previous Literature.

The late Dr. A. W. Howitt, whose pioneering geological work in
Gippsland was so remarkable, was the first to investigate the area
about 50 years ago (1). His report is not only a valuable contri-
bution to the geology of a wide area, but includes an exceedingly
interesting account of a trip in bark canoes downstream from
Tabberabbera through the gorge of the Mitchell River, accom-
panied by two aboriginals, Turnmile, meaning “ one who swag-
gers,” and Bungil Bottle, distinguished for his capacity for the
absorption of strong waters.

The peculiar nature and compressed character of the mid-
’ Devonian Tabberabbera shales was recognised and described and

E. W. Skeats :

their unconformable relations to the flat-lying beds of the Upper
Devonian Iguana Creek series was recognised and figured.

Mr. E. J. Dunn (2) in 1890 published the account of a rapid
survey of the area. His account refers to the presence of Silurian
[Ordovician] rocks in Sandy’s Creek, and of small areas of lime-
stones in the Tabberabbera series. A sketch map accompanying
Mr. Dunn’s report shows the approximate distribution of the
Silurian [Ordovician], Middle Devonian and Upper Devonian
sediments.

Mr. O. A. L. Whitelaw (3) in 1899 published some sections
illustrating the relations of rocks from the district, in a general
account of Devonian rocks in Gippsland.

Mr. H. Herman (4), in June, 1899, published a short account
of the Tabberabbera district with sketch geological map and sec-
tion.

Mr. R. Etheridge (5), in 1899, gave identifications and descrip-
tions of Silurian corals from Sandy’s Creek.

Mr. T. S. Hall (6 and 7), described Upper Ordovician grapto-
lites from Sandy’s Creek.

Location of area and means of access.

The district described in this paper constitutes a roughly rectan-
gular area of approximately 30 square miles. It lies within the
Counties of Dargo and Wonnangatta, and includes parts of the
parishes of Tyirra, Nungatta, Cobbannah and Morekana. The
parish plans, on a scale of 2 inches to the mile, contain little topo-
graphical detail, and parts of them, in this sparsely settled region,
are blank. The name of Tabberabbera does not occur on them,
but is understood to refer to the scattered settlements close to the
junction of the Mitchell and Wentworth Rivers. This lies about
40 miles W'NW. of Bairnsdale, which is about 170 miles east of
Melbourne. Tabberabbera, I understand, is an aboriginal name
meaning Thunder, and in the months of January and February,
during the occasional storms, the noise of thunderclaps reverberat-
ing among the hills of the district makes the naming appropriate.

The conditions of access to the district have much changed in the
last forty years. Then it was difficult to reach the area, as there was
no good graded road from Bairnsdale, but within the area access to
various parts was readily made by good mining tracks, which were
kept in repair, as alluvial and reef gold mining were then fairly
active. Good grazing existed, as rabbits had not then invaded the
district, so that, apart from the miners, there was a fair number
of settlers running cattle, which assisted to keep the tracks open.
Now there is a good graded and metalled road from Bairnsdale to
Bullumwaal, and a good graded and formed road from Bullum-
waal to Tabberabbera. But within the district mining has long
ceased, mining tracks are overgrown, rabbits have come in, and
therefore the country carries only a fraction of the cattle formerly

Palaeozoic Rocks.

grazed. Settlement has, in consequence, declined, blackberries
have over-run many of the gullies, and hop vine and other secon-
dary scrub all contribute to make the district one in which it is not
easy to do geological mapping. The district is rough and hilly,
in places with steep, precipitous gorges, and in the months of
January and February, apt to be uncomfortably hot. In the cir-
cumstances, much of the energy one would like to put into geologi-
cal mapping is necessarily expended in the physical exertion of
•climbing, or of forcing one’s way through scrub.

Nature of work done.

The total period of six weeks spent in the field allowed of some
attention being paid to the stratigraphical and tectonic problems
of the area, but, having regard to the fairly rugged topography,
was insufficient for detailed mapping. The geological map which
accompanies this paper, while it represents a considerable advance
of knowledge over the earlier pioneer work, must be regarded as a
sketch map. In particular, the boundaries shown for the Upper
Devonian rocks are only roughly approximate, as this series was
not the main object of study. The boundaries shown between
Middle Devonian and Silurian rocks and between Silurian and
Upper Ordovician rocks are based on more careful work, checked
by palaeontological determinations kindly made for me by Mr.
Chapman, Palaeontologist to the National Museum. Even these
junctions, away from the sections exposed in the rivers, are only
sketched in. It is clear, therefore, that until a detailed survey is
made, some of the problems of stratigraphy and of tectonics can-
not be completely solved, and such conclusions as are drawn in this
paper are necessarily qualified by this consideration.

The sketch geological sections accompanying the map, and
drawn nearly to the same vertical as horizontal scale, represent an
attempt to illustrate the structure of the area.

Physical Features of the Area.

On approaching the area from Bullumwaal by road over the
Upper Ordovician rocks, the following aneroid heights were
noted: — Bullumwaal 735', Burnett-Merri j ig divide 1475', Merri-
jig River crossing 1005', Merri jig-Sandy’s Creek divide 1475',
road crossing over upper part of Sandy’s Creek 695', Sandy V
Wentworth divide (at the Gooseneck) 1105', Camp No. 1 460',
Wentworth R. level near Camp No. 1 450'. The rough timbered
country of the Upper Ordovician series in the eastern part of the
area rises, therefore, to a maximum of 1000 feet above the level
of the Wentworth River.

The Silurian rocks occupying the northern central part of the
area yield fairly open undulating country in the valleys, especially
near the junctions of the Wentworth and Mitchell Rivers, and the

100

E. W. Skeats :

Palaeozoic Mocks.

101

mouth of Swamp Creek. In following up Swamp Creek to the
north a prominent sandstone hill to the NNE. is seen rising to
about 1450', while to the west a very steep ridge of heavy con-
glomerates, rising to a height of about 1400', about 900' above
Swamp Creek, forms a narrow divide with a steep slope to the
West down to the Mitchell River. About 2^ miles up Swamp
Creek its level by aneroid is 500', and beyond this the country
is un fenced and becomes very rough.

The Middle Devonian rocks form a belt, trending about NW.
across the area except in the south central part, where they are
overlain by the Upper Devonian rocks. They are prominent near
Ostler’s and Horseshoe Bend, and to the north-west, and have
weathered into undulating park-like country, grassed and with
few trees.

The Upper Devonian rocks developed in the central part of the
area form a rough dissected tableland rising in the central part to
nearly 1500', and owing to a gentle south-west dip of about 5°
descending by eroded terrace formations towards the Mitchell
River. The Mitchell and Wentworth Rivers and Sandy’s Creek
all have well developed meanders and horseshoe bends, suggesting
that the country had been formerly maturely dissected. Owing
to late Tertiary uplifts all the streams were rejuvenated, and have
trenched deeply into the underlying rocks. The whole course of
Sandy's Creek below the Merrijig junction constitutes now a
gorge-like valley, with steep cliffs of Upper Devonian on either
side, overlying Ordovician, Silurian and Middle Devonian rocks,
all of which slope steeply down to stream level, and in places form
precipitous river cliffs. The Mitchell River, about 14 miles below
Ostler’s, flows in a picturesque gorge, about 500 feet deep, cut
through the Upper Devonian rocks for about a mile or so above
the junction with Sandy’s Creek. Just south of Camp No. 2, an
abandoned course of the Wentworth River is shown by a broken
line on the map. trending westerly, and then bending south to join
Swamp Creek, just above its junction with the Wentworth River.

Geology.

Upper Ordovician.

Prior to my visits to the district definite Ordovician fossils
had been obtained only from one locality (No. 9 on Map) on
Sandy’s Creek, about 25 chains above the junction with Merrijig
Creek. These were obtained by Mr. Herman in 1897, and include
Glossograptus hennani, Dicranograptiis ram os us and Didymo -
graptus ovatus (7). We revisited the locality and obtained simi-
lar specimens of these graptolites. Our examination of the road
cuttings east of Camp No. 1, on the road going east towards Bul-
lumwaal, yielded a number of graptolites, including Diplograptus
sp. and Glossograptus hermani , from black cherty slates, inter-
bedded with black cherts about half a mile south-east of Camp

102

E. W. Slceats :

No. 1 (No. 5 on Map). Traces of graptolites were found in
similar black cherty slates for about a mile east of this locality.
1 here can be little doubt that all these rocks in the eastern part
of the area belong to the Upper Ordovician series, and as such
they are shown on the map. The road section east of the Camp
shows in addition to black slates and cherts, grey and brown mica-
ceous sandstones and a considerable development of olive coloured
micaceous mudstones, some of which are finely laminated with
thin beds of lighter and darker colour suggesting seasonal band-
ing. Several lamprophyre dykes seen in road section 1 J miles east
of the camp are described later. One other small inlier of rocks
older than the Silurian, and presumably Upper Ordovician in
age. although no fossils were found in them, was found in Sandy's
Creek. It occurs about 1| miles below the junction with the
Merrijig Creek, as an elliptical area elongated in a north-westerly
direction, and only about 200 yards broad. The rocks are black
slates, similar to those containing Upper Ordovician graptolites,
and quite unlike the Silurian rocks with which they are in contact.
The boundaries of this inlier with the Silurian are probably deter-
mined by faults, while the main junction of the Ordovician and
Silurian rocks may be determined either by faults or by an uncon-
formity.

Silurian (Yeringian).

It is under this heading that the greatest changes are shown in
the map accompanying this paper as compared with previous
maps. Hitherto no Silurian rocks have been shown on any map
of this area except in Sandy’s Creek. As the result of my strati-
graphic examination in the field and Mr. Chapman’s valuable help
in the determination of fossils from a number of localities, it is
now known that a broad belt of country in the north-central part
of the area and a limited belt in Sandy’s Creek below the outcrop
of Upper Ordovician rocks consist of Upper Silurian rocks,
probably of Yeringian age. The large genera! geological map of
Victoria, 1902, on the scale of 8 miles to the inch, shows as Silu-
rian all the rocks of Sandy’s Creek from the junction with the
Mitchell River upstream for about 3 miles to just below the Mer-
rijig Creek junction, where the Upper Ordovician rocks come in.
Actually the occurrence of Silurian rocks in Sandy’s Creek is
limited in extent to a strip of country exposed on cither side of
Sandy’s Creek, extending from about 300 yards below the Mer-
rijig-Sandy’s junction downstream for about a mile and a half.
Even within this licit its continuity is interrupted by the small
inlier of Upper Ordovician rocks previously referred to.

Some fossil corals from Sandy’s Creek, probably collected by
Mr. Herman in 1897, were described by Mr. Etheridge (5) in
1899 as Upper Silurian. The locality was probably from an outcrop
of impure limestone about 200 yards upstream from the Upper
Ordovician inlier, since we found at that spot similar genera to
those described by Mr. Etheridge. The forms he described are

Palaeozoic Rocks .

103

.Diphyphyllum ported var. mitchcllensis , var. nov., U. Silurian,
Rhizophyllum intcrpunctatum De Konin^k, U. Silurian, Monti -
. culipora ( Heterotrypa ) australis, sp. nov., U. Silurian.

The 1902 map, eight miles to the inch, shows a great area
•coloured as Middle Devonian limestone, which starts about two
miles south of Tabberabbera, and continues northwards for about
18 miles. It is shown extending for two or three miles east of
the Wentworth River, and has a maximum breadth of about
eight miles, gradually becoming narrower in its northerly exten-
sion. The later general geological map of Victoria, 1909, on a
scale of 16 miles to the inch, gives this area the colour appropriate
to the Upper Devonian sandstones and shales. This is lithologi-
cally more correct, but from the point of view of age appears to
be further from a correct determination than that shown in the
earlier map. My own observations in the field have only extended
to a point on Swamp Creek, about three miles north of Tabberab-
bera, but the rocks up to that point show similar lithological char-
acters to those nearer Tabberabbera, such as localities 1.2.3 and
•6, on the map, from which abundant fossils, determined by Mr.
Chapman as Upper Silurian (Yeringian), have been obtained. It
seems certain, therefore, that the southern part of this area up to
three miles north of Tabberabbera, consists of Silurian and not of
Middle or Upper Devonian rocks, and it is probable that the whole
of this area, extending to about 18 miles north of Tabberabbera,
consists of rocks of Silurian age.

The localities 1, 2, 3, and 6 from which we obtained abundant
fossils occur north and south of the junction of the Mitchell
and Wentworth Rivers, which Howitt named as Tabberabbera.

Loc. 1. Allot. 13, S. Websdale, Tyirra.

Fossils - — Bythotrephis cf. gracilis J. Hall.

Spirifcr aft. crispus (Hisinger).

Yeringian.

Loc. 2. Allot. 14. Tyirra.

Fossils — Spirifer aft. crispus (Hisinger).

Ctenodonta cf. portlocki Chapman.

Yeringian.

Loc. 3. Allot. 9A, E. Desailly, Tyirra.

Fossils — Spirifer aff. crispus (Hisinger).

Pcntamerus aff. lens (Sower by ) .

? Gloss it es or ? Palaeoneilo.

Actinoptcria sp.

Yeringian.

Loc. 6. Allot. 6, Nungatta. E. bank of Mitchell River, about
half mile south of Birch’s.

Fossils — Plant remains, ind.

? Try plasma.

Chollotrypa sp.

cf. Favosites gothlandica Lam.

Roemingeria sp.
cf. Leptaena sp.

Atrypa aspera (Scliloth.). Gerontic forms.
ConcKidium sp.

Spirifer sp., probably new.

Yeringian.

104

E. W. Skeats :

This evidence shows that the type locality for the Tabberabbera.
shales at the junction of the Mitchell and Wentworth Rivers
actually consists of Upper Silurian rocks. Dr. Howitt (1, page-
206) states: “I found a small limestone patch at Tabberabbera,
situated at the junction of the Mitchell and Wentworth Rivers.
No fossils have been procured from the limestone, but associated
with them are black shales, yielding plentifully the Spirifera laevi-
costato [later redescribed as Spirifer yets sen sis] and a Grammy -
sia. They are regarded by Professor McCoy as being of the same
age as the Buchan limestones, and therefore Middle Devonian/*

To reconcile these statements with the evidence I have ob-
tained, it is necessary to interpret very loosely Howitt’s word
“ associated ” in connection with the black shales yielding the
above forms. It is almost certain that Howitt did not obtain them
from the junction of the Mitchell and Wentworth Rivers, and I
was unable to locate such black shales at this junction. It is prob-
able that he obtained them from a locality on the Mitchell River,
about three miles below Tabberabbera.

He figures (1, p. 207) a sketch section No. 16 Tabber-
abbera, showing the fossiliferous black shales on the east side
of the Mitchell River, and just below the junction with the
Upper Devonian (Iguana Creek) beds. His statement is that the
section is “below Tabberabbera / 1 9 Further on (1, p. 215), when
describing his canoe journey down the Mitchell, he states that
they started about two miles below Tabberabbera, and after con-
tinuing some time he landed and examined a limestone which he
states was “ very much upon the line of section given in sketch
No. 16/’ It is not possible to locate this place exactly from
Howitt’s description, but I take it to be about one mile down
stream from Ostler’s. If that is so, the difficulty disappears, since
I have obtained Middle Devonian fossils at Horseshoe Bend and
east of Ostler's and rocks of a similar character continue down
stream for some distance.

Loc. 4, at the road cutting on the north side of the Wentworth.
River, about half a mile west of Camp No. 1, provides another fos-
siliferous locality in the Silurian series. The rocks are finely lam-
inated black and brown cherts crowded with Radiolaria. Mr. Chap-
man has reported on microscopic sections of these rocks as fol-
lows : —

“ The rock is crowded with radiolarian remains, but only very
few are determinable. In some cases the ferruginous staining
and replacement of the siliceous test is an aid to deciphering the
form and structure/*

Genera or species noted —

Distriactis sp,

Acanthosphaera cf, etheridgei Hinde.

Stylos p ha era sp.

Spongoloncha cf. lens Hinde.

Palaeozoic Pocks .

105*

The assemblage closely resembles that from the Tamworth dis-
trict, described by Hinde in Quart. Journ. Geol. Soc., vol. lv., 1899,.
pp. 38-64.”

Lithological types in the Silurian rocks of this district are
numerous and varied. The sections exposed in the road cut-
tings between localities 4 and 2 on the map, are in a zone of
great crushing and contortion. Olive mudstones and dark cal-
careous shales, some very fossiliferous, are common. In places
where the calcium carbonate has been leached out the rocks are
rusty brown in colour, and the fossils are only preserved as casts.
Thin lenticular blue limestones were also noted. Sandstones and
grits are fairly prominent. Calcareous mudstones and thin blue
limestones occur also south and north of Loc. 6, South of Birch’s.
An impure blue limestone in Sandy’s Creek about 200 yards above
the Upper Ordovician inlier yields an abundant supply of Silurian
corals.

Two parallel massive and thick conglomerates with intercalated
grits are shown on the map, extending from east of E. Websdale’s
down to the Mitchell River, and a smaller conglomerate and grit
higher in the Silurian series are shown east of Swamp Creek.

The pebbles in the most westerly conglomerate near E. Webs-
dale’s are up to a foot in length, and are much dimpled and
sheared. Fine-grained laminated pink and white sandstones or
quartzites were noted about two miles up Swamp Creek adjoining
the big dyke shown on the map. and were seen again on the
Mitchell River about one mile west of Sinnott’s, and very similar
types occur in two places in the Sandy’s Creek section, one adjoin-
ing the western boundary of the Silurian and the other near the
eastern boundary of the same series.

Numerous dykes intersect the Silurian rocks. Many appear to
have a similar strike to that of the adjoining sediments, others
cut across the strike of the sediments at various angles. They in-
clude hornblende porphy rites, some fresh and others calcareous
with decomposition, quartz felspar porphyries, black dykes show-
ing quartz, and a tinguaite. Brief petrological descriptions of
some of these are included later in the paper.

Middle Devonian.

The rocks to which a Middle Devonian age can be assigned are
restricted, as shown on the map, to a belt of country occupying the
western part of the area.

The junctions with the Silurian rocks to the east have been defi-
nitely located only in two places. One is near Loc. 7, on the
Mitchell River, east of the saddle of Horseshoe Bend. The other
is in Sandy’s Creek, just above Whitbourne’s Hut, and just east
of a prominent sill or interbedded flow near a crush zone, and just
west of prominent laminated pink and white quartzites.

Below this point on Sandy’s Creek, down to the junction with
the Mitchell River, the whole sequence for about 1J miles across
the strike is in the Middle Devonian sediments, and characteristic.

E. W. She at s :

106

Middle Devonian fossils have been collected from several bands
of blue limestone or dark calcareous shales exposed in the river
cliffs.

One limestone band a few hundred yards below Whitbourne’s
flat yielded in microscopic sections, according to Mr. Chapman's
determination —

Spirifer y ass en sis.

? Coenitcs or Campophyllum ?

Carapaces of Ostracods, chiefly Primitia.

Syringopora ?

Foraminifera including Pulvinulinci ?

Nubecularia ?

Crinoid ossicles.

From Loc. 7, just east of the saddle of Horseshoe Bend, abun-
dant fossils were obtained, chiefly as casts, which Mr. Chapman
has determined as under :

From a shale band —

Spirifer yasscnsis de Kon.

Grammysia sp.
and from a limestone band —

Abundant specimens of Spirifer yasscnsis de Kon.

Similar fossils were also obtained at Loc. 8, south of Horseshoe
Bend and east of Ostler's. The lithological types in the Middle
Devonian include blue limestones, black, brown and yellow shales
or slates, some silicified or flinty shales and siliceous sandstones.
The rocks at Horseshoe Bend are intensely crumpled, faulted and
in places vertical.

Numerous dykes penetrate these rocks, and are especially
noticed in the ridge-like saddle of Horseshoe Bend and south
from that locality; some striking E. and YV., others conforming
more or less to the strike of the sediments either W. or E. of
north. They include hornblende and other types of porphyrites;
at Loc. 8 a tinguaite strikes E. and W. ; at Whitbourne’s flat on
Sandy's Creek an interbedded igneous rock occurring as a sill or
lava flow is a fine-grained porphyrite; and S. of Whitbourne's
flat a big black basaltic dyke cuts across the sediments. South of
Ostler's, where the Mitchell bends to the west, the slopes below
the Upper Devonian series consist of a large area of diorite por-
phyrite, probably intrusive into the Middle Devonian sediments
(see Section C-D).

Upper Devonian.

The rocks of this series, provisionally described as of Upper
Devonian age, occur within the central and south-western parts
of the area shown on the map. They form a series about 900 ft.
in thickness. They rest in turn on the heavily eroded edges of
folded rocks of the Upper Ordovician, Upper Silurian and Middle
Devonian series, and as they themselves are almost horizontal,
their dip being not more than 5° to the SW., their relations with
the older rocks, even with the Middle Devonian, constitute a very
important and striking angular unconformity.

Palaeozoic Rocks.

107

1 hev ha\e not in this district as yet yielded any recognisable
fossils, and the validity of the reference of them to the & Upper
Devonian depends on questions of geological continuity and of
lithological correlations with other areas. The following rough
section, supplied to me by Mr. J. Easton, and trending in an
approximately easterly direction from the Mitchell River above
Horseshoe Bend, and just north of Loc. 7, will serve to illustrate
the sequence of the rocks of this series. At the base, about 150'
above the level of the Mitcheil River, and resting directly and
unconformably on the eroded surface of the folded Middle Devo-
nian mudstones and shales, are about 100 feet of purple grits
and mudstones, then a few feet of purple breccia followed by about
60 feet of purple mudstone. About four feet of nodular or
spherulitic rhyolite comes next, followed bv 90 feet of red and.
grey mudstone, then 20 feet of breccia and conglomerate, 50 feet
of mudstones, 3 feet of rhyolite, 70 feet of conglomerate and
breccia, and continuing to the top of the series developed in this
district about 500 feet of siliceous and pebbly grey sandstone beds.

The precise sequence of these rocks varies" somewhat in different
parts of the area shown on the map. In some places the inter-
bedded rhyolites are much thicker, and just east of the diorite por-
phvrite about a quarter of a mile south of Loc. 8 the sequence
appears to start with spherulitic rhyolites.

The reference of these to the Upper Devonian is not quite cer-
tain, since, as shown on the map, they have a dip of 30°, and it
is just possible that they may unconformably underlie the base
of the Upper Devonian, and may be a small area of Lower
Devonian igneous rocks. If this were so the diorite porphyrite
on which they appear to rest would be older than is shown in the
sketch geological section.

On the whole, however, it is thought to be more probable that
these spherulitic rhyolites are of Upper Devonian age, and that
their relatively high and abnormal dip is due to restricted local
movement. This view is strengthened since A. W. Howitt (18)
m describing the sequence of Upper Devonian rocks in the Snowy
Bluff section, refers to interbedded compact felsites (felstones),
having in places a spherulitic structure, the spherules being from
one to two inches in diameter.

This description corresponds closely with the nature of the
spherulites just east of the diorite porphyrite, and is in accord-
ance with their recurrence higher in the series as noted by Mr.
Easton in the section described above.

In most parts of the area, however, purple mudstones form the
base .of the series. No interbedded basalts (melaphyres of
Howitt) have been found “ in situ ” in these rocks, but they prob-
ably occur, since abundant pebbles of this type of rock have been
found at the Mitchell River at and near Loc. 7. I have not seen
any dykes penetrating these rocks, but Mr. Easton informs me of
the interesting fact that he has obtained a lamprophyre and
several felspar porphyrite dykes intruded into the Upper Devo-
nian sediments.

■108

E. W. Skeats :

The lithological characters of these rocks and their prevalent
purple and red colours suggest that the series is a lacustrine one,
rapidly accumulated in an arid climate subjected to occasional
rain storms.

Tectonic Movements and Structures.

The district has clearly suffered from successive movements of
‘compression in post-Upper Ordovician, post-Upper Silurian, and
post-Middle Devonian times, and in the Silurian and Upper Ordo-
vician rocks it is almost impossible to distinguish the effects of
the earlier from those of the later movements.

The present relations expressed in dips, strikes, trends of fold
axes and trend of boundaries between different formations have
developed as the result of the combined effects of all the earlier
and later structural movements.

The structural features in the Ordovician rocks are fairly
clearly shown in the road section, starting from about half mile
from Camp No. 1, and continuing for about miles in a general
easterly direction. The distribution of dips and strikes shown on
the map indicates that away from the junction with the Silurian the
average strike is about N.20°E. and the average dip about 65°-70°.
Near the Silurian junction the strikes are much more disturbed,
and trend west of north at varying angles from NNW. to \V. It
would seem that near the Silurian junction along this road section
is a zone of special disturbance, and since it will be seen that the
Silurian rocks near this junction also tend to have abnormal strikes
it may be that either this junction was determined bv post-Silurian
fault movements or, if the junction be an unconformity, that the
post-Silurian movements were only able to impress themselves on
the indurated and compressed Ordovician rocks in the neighbour-
hood of the junction with the Silurian.

The section in Sandvs Creek above the Merrijig junction shows
an abnormal strike of N.70°W. This locality is about half mile
from the junction with the Silurian. About a mile below the
junction the small, probably faulted, inlier of Ordovician rocks
has a strike of N.30°W., and an unusually low dip of 20 c . This
latter may well be an effect of over folding.

The general strike of the Ordovician east of north, away from
the Silurian contact, is in harmony with the evidence given by
Teale (10) from Nowa Nowa, and farther east in Croajingolong.
In this part of Victoria the trend of the Palaeozoic rocks and of
their junctions, as seen on a general geological map, is east of
north, and continues in this direction into New South Wales. But
northwards from a line through Mt. Wellington, Waterford to Mt.
Baldhead, the trend of the junctions of Upper Palaeozoic and
Lower Palaeozoic rocks is about N.40°W., and strikes of this
nature are common in the Ordovician rocks north of the line men-
tioned. Both sets of trend lines must be of post-Palaeozoic
-development, or at any rate continued till late Palaeozoic times,

Palaeozoic Rocks .

109

since the trend of the junction of the Upper Devonian and Upper
‘Ordovician rocks in the northern area conforms to the direction
•of N.40°W.

The structural features in the Silurian (Yeringian) rocks are
comparatively simple in the sections seen in Sandy’s Creek. The
strikes are all west of north, varying from 20° to about 40° west
• of north. The lines of junction with Upper Ordovician and with
Middle Devonian rocks appear to trend about N,40°W.

Much greater diversity of strikes and complexity of folding
occur in the central and northern part of the area in sections seen
on the Mitchell and Wentworth Rivers and in Swamp Creek on
the bare exposed saddles within this part of the region.

West of a north and south line through Camp Xo. 2, the strikes
are all west of north at angles varying from X.20°\V., which is a
common strike in the western outcrops, to X.65°W., in several
places near Loc. 6. In the road sections east of Camp No. 2, and
in one or two localities further north, the strikes are all east of
north from 20° -50°, except in one case south-east of Camp No.
1, adjoining the junction with the Upper Ordovician. A dyke
and grits strike north and south about one mile north of Camp
No. 2. On either side of this there is a tendency to a convergence
of strike of the beds to the south, suggesting a syncline pitching
north. A prominent grit bed on the Wentworth River just north
of Birch’s shows an axis of a syncline. However, the prominent
conglomerate beds shown on the western part of the map con-
tinue with a strike of N.20°W. in a southerly direction at least to
the Mitchell River. There may be a strike fault east of these con-
glomerates, and the big dyke seen along Swamp Creek may have
been intruded along such a fault. The average dip of the
Silurian rocks is very high. The only one as low as 45' occurs
near the Middle Devonian junction north-east of Loc. 7. In
many cases the beds are vertical, and perhaps the average dip oil
either side of the fold axes is 70° to 75° f A puckered anticline
and syncline with steep northerly pitch occur at Loc. 7 at the junc-
tion with the Middle Devonian rocks. While the high dips and
the fold axes are the expression of compressional earth move-
ments, the numerous dykes intersecting the Silurian rocks indicate
that tensional cracks either accompanied or succeeded the com-
pressive movements within the same geological period or at later
times.

The majority of the dykes in the Silurian, especially the por-
phyrite dykes, appear to strike nearly or quite parallel to the ad-
joining sediments, but near Horseshoe Bend two dykes, one of
them a tinguaite, cut right across the strike of the sediments in an
east-west direction.

The structural features in the Middle Devonian rocks have been
noted near Horseshoe Bend, near Localities 7 and 8, and in con-
tinuous sections along Sandy’s Creek for about 1^ miles across
the strike upstream from its junction with the Mitchell River.

110

E. W. Skeats :

In the first locality considerable changes in strike direction are
noticeable from N.10°E. to N.20' W„ while on the saddle in the
neck of Horseshoe Bend a porphyrite dyke strikes east and west
and farther south at Loc. 8 a tinguaite dyke strikes in the same
direction. The rocks in this locality are very steeply folded and
crinkled, with dips of 75° to 80°. At Ostler’s and south of the
Mitchell River south of Ostler’s, there is a large area of intrusive
rock, dark green in colour, consisting of diorite porphyrite. Howitt
also noted the occurrence of a similar rock further down the
Mitchell River in a locality which 1 have not been able to visit.
The boundaries and field relations of this rock were not deter-
mined, but it is probable that it represents a hypabyssal intrusion
of post -Middle Devonian age.

On Sandy’s Creek the strikes of the rocks and the fold axes are
uniformly west of north, usually about N.20°W. Several anti-
clinal and synclinal folds are seen in section in the river cliffs, and
usually the dips on either side of the axes are at 45°-50°. Dips
up to 80° are, however, recorded, and at one of the anticlinal folds
severe local puckering and faulting complicate the relations. At
the junction with the Silurian rocks just above Whitbournefs Ilut,
while the dip is westerly, there is a zone of puckering and over-
folding with an intercalated fine-grained sill or lava seen in the
cliff section. It is clear from these facts that in this region there
is evidence of local severe compressive earth movements later
than the Middle Devonian. Tcale (10) has noted that at Hickey’s
Creek on the Macallister River, there occurs a local severe tec-
tonic zone of faulting and synclinal folding, which is of post-
Upper Devonian age since rocks of this age are involved.

In the Grampians in Western Victoria the author (11) has
given evidence of post Lower Carboniferous plutonic intrusions..
The evidence cited from these localities shows that the long main-
tained view that notable compressive earth movements with ac-
companying plutonic intrusions ceased in the Lower Devonian
period cannot now be entirely accepted.

In Central Victoria the similarity of composition of dacites and
granodiorites suggests that although the granodiorites are instru-
sive into the dacites, they probably belong to the same period of
igneous activity, which has been regarded as probably Lower Devo^
nian, since iti various places the dacite series is overlain uncon-
formably by Upper Devonian sediments. At Bindi, the Middle
Devonian limestones and shales rest possibly unconformably on
the Snowy River porphvrites of Lower Devonian age, and are only
gently folded. At Buchan, pyroclastic igneous rocks associated
with the Snowy River porphvrites, are intercalated with the lower
part of the limestones and shales. In the Buchan district the struc-
tural relations of the limestone series are seen from numerous
recent road sections to have been affected in general by only
gentle post-Middle Devonian compressive movements since the
average dips seldom exceed 20° except in one or two places, where
quite local puckers have developed small anticlinal folds with high

Palaeozoic Rocks.

Ill

dips. At Buchan and Bindi, therefore, the gentle folding stands
in marked contrast to the more severe compression which has
affected the Silurian rocks generally in Victoria and the Middle
Devonian rocks of Tabberabbera.

The rocks described as Upper Devonian in this district have
not suffered from any compressive earth movements. They
appear in the sections exposed in the field to be almost horizontal,
but a dip of about 5° to the south-west can be inferred from the
fact that rhyolites and other associated rocks outcropping at river
level along the Mitchell at about 400' elevation are over 1000'
above sea-level about two miles ENE. from that locality.

Significance of the Unconformity between the
Middle and Upper Devonian Rocks.

The most remarkable structural features of the district are
firstly, the severe compression and folding which have affected the
Middle Devonian rocks, whose age is definitely determined by
their fossil content, and secondly, the profound character of the
unconformity which separates these folded rocks from the flat-
lying sediments and lavas which rest on their denuded edges and
also unconformably overlie the Silurian and Upper Ordovician
rocks. These overlying rocks are here described as Upper Devo-
nian, but the question of their age invites some discussion. As
stated above, no fossils have as yet been found in these rocks, but
they appear to be geologically continuous, as stated by R. A. F.
Murray (17). with the series developed further south at Iguana
Creek, and south-west at the Avon River. There is continuity and
similarity of sedimentation in all three areas, but McCoy (11)
described the Avon River beds as Lower Carboniferous, on ac-
count of the presence in them of Lcpidodendron australe, and the
Iguana Creek Beds (12) as Upper Devonian on account of the
presence in them of Cordciites australis and Archaeopteris
Howitti.

A broad belt of similar sediments stretches N.40°W. from the
Avon River through the Mt. Wellington district, described by
Teale (10), to Mansfield. Near Mansfield, on the Broken River,
Cresswell and, later, George Sweet (14) discovered plants and
fossil fish partially described by McCoy (13) as showing forms
of mingled affinities ranging through Lower Devonian, Upper
Devonian to the base of the Carboniferous. McCoy placed the
Mansfield Beds as at the top of the Upper Devonian. It should
be noted that McCoy identified the plant remains as Lcpidoden-
dron Mans field ense, a species quite distinct from the form met
with in the Avon River section. Smith Woodward (15), how-
ever, later described the Mansfield fossil fish as typically Lower
Carboniferous, and the Geological Survey of Victoria, in their
latest general geological map of the State (1909), on the scale of
16 miles to the inch, have distinguished the beds round Mansfield
from the rest of the belt of similar rocks, colouring them as Car-

112

E. W. S heats :

Coniferous, and the rest, including the Avon River beds, as Devo-
nian. In numerous localities in New South W ales, as at Mt.
Lambie and lamworth, New South W ales geologists have shown
that, in that State, Lcpidodendron australe is interbedded with
marine beds containing Devonian marine fossils, at Tamworth
with radiolarian cherts described as of Middle Devonian age, and
at Mt. Lambie and elsewhere interbedded with marine beds con-
taining Spirifer disjunct a, a typical Lpper Devonian brachiopod.
Professor Benson (16) has given a full discussion on the Devo-
nian palaeontology of Australia and discussed the stratigraphical
implications.

Our Victorian problem is to reconcile the geographical con-
tinuity over a wide area of fairly flat-lying beds of similar litho-
logical types, and apparently one series formed under similar con-
ditions. with the palaeontological determinations which would
place the Avon River Beds on plant determinations as Lower Car-
boniferous, the Mansfield beds as Lower Carboniferous on iden-
tification of fossil fish, and the Iguana Creek beds as Upper Devo-
nian on the identification of fossil plants.

The reconciliation of these apparent anomalies will probably
not be achieved until continuous and detailed geological surveys
are made throughout the broad belt of rough mountainous coun-
try between Iguana Creek and the Avon River, and between the
Avon River and Mansfield. Until this work has been accom-
plished the point of view expressed on the Geological Map of Vic-
toria in 1909 — the separation of the Mansfield area from the re-
mainder — appears to have some justification.

The important evidence from various localities in New South
W r ales that Lcpidodendron australe is there an Upper Devonian
form may justify us in Victoria in regarding the Avon River beds
as well as the Iguana Creek beds as of Upper Devonian age.

In this connection it is perhaps pertinent that the broad belt
shown on the Geological Survey Map of Victoria, 1909, as Devo-
nian. is an area throughout which there are intercalated with the
conglomerates, sandstone and shales, important flows of rhyolite
and thinner sheets of basic lavas (melaphyres of Howitt) and
similar intercalated igneous rocks are recorded from several of
the New South Wales areas in which Upper Devonian rocks are
recorded. But in the Mansfield area these intercalated igneous
rocks have not been found. Despite then the similarity of the
•sediments in the Mansfield district to the sediments farther to the
south-east, the absence of contemporaneous lavas in the Mansfield
area, may be regarded as negative evidence supporting the positive
evidence of the fossil fish described by Smith Woodward as fixing
a Lower Carboniferous age for the Mansfield beds.

The foregoing discussion then may justify us in accepting, at
any rate provisionally, the flat-lying sediments with intercalated
lavas of the Tabberabbera district as of Upper Devonian age.

If this view is correct the significance of the gigantic uncon-
formity between these beds and the highly crumpled Middle

Palaeozoic Rocks .

113

Devonian rocks beneath is remarkable and difficult to explain, for
we have to picture that in this part of Victoria in the geologically
short interval between Middle and Upper Devonian the sea re-
ceded, the Middle Devonian rocks were crumpled and elevated,
and denuded to a low-lying area, before the lacustrine conditions
of the Upper Devonian were established.

Petrographic Characters of the Igneous Rocks of
the District.

In this paper, mainly concerned with structural and stratigra-
pliical relations, only brief descriptions of the igneous rocks will
be given. The reference numbers are those of the main collection
of rock sections in the Geological Department of the University
of Melbourne.

Dykes of varying size and petrologic character occur in the
Upper Ordovician. Silurian (Yeringian), Middle Devonian and
Upper Devonian (Easton’s communication) rocks.

A big hvpabyssal or small plutonic intrusion of diorite porphy-
rite occurs in the Middle Devonian rocks as well as an intercalated
sill or lava flow.

Prominent nodular or spherulitic rhyolites, as well as banded
flow rhyolites, occur in the Upper Devonian, and the evidence of
boulders in the river-beds suggests that basic flows (melaphyres)
may also be represented in the Upper Devonian, although they
have not yet been found “ in situ.”

In the Upper Ordovician sediments, apart from spherulitic kera-
tophyres, and a hornblende porphyrite, high up Sandy’s Creek,
near the Bullumwaal road, the only dykes found up to the
present are somewhat decomposed mica lamprophyres. A
boulder of a somewhat similar rock found about two miles
up Swamp Creek suggests that mica lamprophyres may also
penetrate the Silurian rocks, while a boulder of a fresh green
tinguaite, No. 1726 . found about three miles up Sandy's Creek,
indicates that a dyke of this type probably intruded the Ordo-
vician sediments. Within the Silurian (Yeringian) rocks, dykes
of hornblende porphyrite, felspar porphyrite, quartz felspar por-
phyrite, oligoclase trachyte and of tinguaite. No. 1737 . have
been found, and a boulder of fresh tinguaite. No. 1/18. was
found in the Mitchell River, about half a mile below
E. Websdale’s house. This represents material from a
dyke which may intersect either Silurian or Upper Ordo-
vician rocks, since the Mitchell River above this point drains
areas of both these series. Dykes penetrating the Middle Devo-
nian sediments include numerous porphvrites, a basalt and a tin-
guaite, No. 1727 . It will be noted that four felspathoid-bearing
rocks are now known from this district. The author (8) has
given petrographic descriptions of them recently, so that it is not
necessary to refer to them further, except to point out that within

114

E. W. Skeats :

recent years it has become known, by the author's contributions
to recent volumes, as one of the Secretaries to the Alkaline Rocks-
Research Committee of the Australasian Association for the Ad-
vancement of Science, that Eastern Victoria must be regarded as
an alkali -rich province. Phonolites and tinguaites have been re-
corded by him from the Tolmie Ranges, near Mansfield* from
Pretty Boy pinch, west of Tabberabbera, from near Mt. St. Ber-
nard, north-north-west of Tabberabbera, and from near Omeo,
north of Tabberabbera. Many of these are so fresh and unaltered
that they may quite likely be of Middle to Late Kainozoic age, like
the alkali rocks of Mt. Macedon and the Western District of Vic-
toria. If this is so, they may have been intruded along tension
cracks associated with the successive plateau elevating movements,
differential in character, which have uplifted Eastern Australia. It
cannot, however, be said that their association with fault move-
ments has vet been proved or definitely established.

Descriptions of Rock Sections of Igneous Rocks.

1608. Dyke 10' thick, road cutting in Upper Ordovician, H
miles east of Camp No. 1.

A dense dark fine-grained rock in hand specimen, weathering
to a rusty brown colour. Pale to pink “ Schlieren ” occur through
it and occasional large plates of biotite are present. Under the
microscope brown biotite is abundant, green and brown sub-por-
phyritic hornblendes, and small prismatic, green to brown crystals
of the same mineral are abundant. Large clear zoned crystals of
plagioclase are invaded by the ferromagnesian minerals and the
ground mass is partly cloudy through the alteration of the smaller
felspars, while “ Schlieren ” are represented by clear colourless
areas, partly consisting of felspar, and isotropic areas which
occur suggest that a felspathoid such as sodalite may be present.
The rock may be described as a mica hornblende lamprophyre.

1723. Dyke through Upper Ordovician in the upper part of
Sandy's Creek, near the Bullumwaal road.

In hand specimen the rock is rather decomposed, cream-
coloured, and apparently largely felspathic. Under the micro-
scope it is seen to be practically wanting in ferromagnesian
minerals, and to be composed almost entirely of felspar. A
few areas of almost colourless chlorite indicate the former
presence of a small amount of a ferromagnesian mineral. A
number of small quadrate to lath-shaped clear felspars with
fine twin lamellae and almost straight extinction consist of
oligoclase, while the bulk of the rock consists of spherulitic
aggregates of felspar laths. The rock may be described as a
spherulitic keratophyre. *

1738. Dyke penetrating Upper Ordovician high up Sandy’s
Creek near Budumwaal road.

In hand specimen the rock appears to be fairly fresh, dark grey
in colour, fine grained with small porphyritic crystals. Under

Palaeozoic Rocks.

115

the microscope the rock consists mainly of two minerals. Plagio-
clase is abundant as fair sized porphyritic fresh crystals of
quadrate habit and moderate extinction angle indicating ande-
sine. Somewhat later than the felspar is abundant pale horn-
blende, some of which is altered to chlorite. A little magnetite in
crystals and irregular grains is also present. The rock is a rather
basic hornblende porpliyrite.

1730. Dyke cutting Upper Ordovician high up Sandy's Creek,
about one mile below Bullumwaal road.

In hand specimen the rock is cream coloured, with porphyritic
quartz. Under the microscope large corroded crystals of quartz
showing crystal boundaries are common, and large abundant
phenocrvsts of plagioclase ranging from oligoclase to andesine are
set in a fine groundmass of spherulitic aggregates of felspar.
Small microscopic quartz veins penetrate the rock. The rock may
be described as a porphyritic and microspherulitic quartz kerato-
phyre.

1724. Weathered dyke cutting Silurian, 250 yards XE. of Loc.

2 . ’

The hand specimen is a dense fine-grained brownish grey rock,
with porphyritic felspars. In section its altered character is appar-
ent. All the larger felspar phenocrysts are kaolinized. Smaller
quadrate phenocrysts are oligoclase, as is most of the felspar in
the felted groundmass. The ferromagnesian mineral has altered
to chlorite and a small amount of secondary calcite is present. The
rock may be described as an oligoclase trachyte.

1747. From big dyke cutting Silurian shown on map along
Swamp Creek, from quarter mile below top fence.

In hand specimen the rock has a rather coarser texture than
most of the dykes seen, and shows small phenocrysts of felspar
and hornblende. Under the microscope both plagioclase and horn-
blende are abundant, each is in turn porphyritic, and each may he
included in the other, suggesting almost simultaneous crystalliza-
tion. The groundmass consists mainly of small plagioclase fel-
spars. The felspar is mainly oligoclase, the hornblende is pale,
and some amount of minute magnetite is present, and secondary
calcite occurs in small amount. The rock is a hornblende por-
phyrite.

1743. From same dyke as 1747, but 10 chains lower down
Swamp Creek.

In hand specimen its paler colour and more altered appearance
than 1747 is noted. Under the microscope it is distinguished from
1747 by the absence of hornblende, and the abundance of secon-
dary calcite and the presence of a fair amount of quartz in the
groundmass. The rock is an altered felspar quartz porpliyrite.

1606. Dyke penetrating Silurian, road east of Loc. 2, at north-
west corner of Allot. 9, R. J. Oates.

In hand specimen the rock is dense, cream-coloured, with por-
phyritic quartz crystals. Under the microscope large phenocrysts

E. W. Skeats

11(3

of oligocla.se, andesine and of corroded quartz crystals are set in
a micrographic groundmass of quartz and acid plagioclase. The
rock is a micrographic quartz felspar porphyrite.

1719. Dyke cutting Silurian and striking north and south from
hill east of Swamp Creek. Allot. 2, K. Sinnott.

In hand specimen the rock is dense and cream-coloured, with
porphyritic felspars and quartz. Under the microscope pheno-
crysts of untwinned felspar and of corroded quartz are set in
a microcrystalline groundmass of quartz and felspar, consisting of
both plagioclase and orthoclase. Some elongated hiotite more or
less altered to chlorite is also present. The rock is a quartz felspar
porphyry.

1744. Big dyke 110' thick, penetrating Silurian near western
boundary, 250 yards above Whitbourne’s'Tlut, Sandy’s Creek.

In hand specimen the rock is dense and dark greenish in colour,
with small felspar phenocrysts. L nder the microscope the rock is seen
to be considerably altered. The plagioclase phenocrysts are de-
composed, and the hornblende replaced by chlorite and abundant
calclte. The groundmass contains small crystals of magnetite, but
consists mainly of a microcrystalline aggregate of felspar, with
some quartz. The rock is an altered hornblende porphyrite.

1739. Black dyke cutting Silurian conglomerate on Mitchell
River, west of Sinnott’s.

In hand specimen the rock is black and densely crystalline.
Under the microscope it is seen to be fresh with ophitic texture
since lath-shaped labradorite penetrates pale brown augite, a little
of which is altered to chlorite. Irregular crystals of magnetite
are fairly abundant. The rock is a dolerite.

1733. Big dark dyke cutting Silurian quartzites, and striking
N.10 W. near bend in Mitchell River, west of Sinnott’s.

In hand specimen the rock is a black fine-grained but crystalline
rock with felspar phenocrysts. Under the microscope it is seen
that the rock has suffered dynamic alteration. Porphyritic plagio-
clase felspars are set in a finer ophitic intergrowth of felspar laths
and ferromagnesian minerals, but the latter are now fibrous horn-
blende and the plagioclase has been mostly recrystallized to radiat-
ing or needle-shaped secondary minerals. Fine-grained irregular
crystals of magnetite occur in the groundmass. The rock is a fine-
grained dynamically altered dolerite.

1746. Dark dyke striking east and west, eight chains NW. of
saddle of Horseshoe Bend.

In hand specimen the rock is a dark grey fine-grained crystalline
rock. Under the microscope the rock is seen to consist of large
lath-shaped labradorite, with prismatic to irregular pale purplish
to brown augite and a fair quantity of magnetite or ilmenite. Con-
siderable alteration of much of the augite has occurred with the
development of calcite and the introduction of some chlorite. In
the interstices of the rock some of the felspar is somewhat spheru-
litic. The rock is a fine-grained felspathic dolerite.

Palaeozoic Rocks.

117

1729. A rather large mass probably intrusive into the Middle-
Devonian, south of Ostler’s, on the Mitchell River.

In hand specimen the rock is of medium grain size, and dark
grey green in colour. Under the microscope the texture is be-
tween the hypabyssal and the plutonic. There is a tendency for
the irregularly quadrate felspars, oligoclase to andesine, to be por-
phyritic. The ferromagnesian mineral, originally hornblende, is
now largely pale green and fibrous in habit, and is largely chlorite.
A small amount of minute felspars with interstitial quartz, con-
stitutes a second generation of crystals in which are recognised
small magnetite phenocrysts and occasional irregular crystals of
sphene. The rock is a diorite porphyrite.

1731. Big dyke 150' thick, striking N.40°E., penetrating Middle
Devonian, west of Loc. 7.

In hand specimen the rock is a nearly black, fine-grained rock,,
showing minute quartz and specks of pyrites. Under the micro-
scope its fine-grained texture is clear, but the rock is much altered,,
both plagioclase and augite being largely altered. The lath-shaped
plagioclase still shows twinning, but the ferromagnesian mineral
is now changed to chlorite. A little interstitial quartz, and a small
amount of black opaque iron ores occur ; calcite is moderately
abundant. The rock is a fine-grained quartz dolerite.

1732. Dyke 15' thick, cutting Middle Devonian, south end of
Whitbourne’s paddock, Sandy’s Creek.

In hand specimen the rock is black, fine-grained, but crystalline,,
with small porphvritic felspars. Under the microscope the rock is
clearly porphvritic. Clear large lath-shaped labradorite felspar
and pale brown augite phenocrysts are set in a finer textured
ophitic aggregate of the same minerals, with the addition of
granular magnetite; green chlorite and calcite, are noted as secon-
dary products. The rock is a porphvritic dolerite.

1612. Sill or interbedded flow in Middle Devonian five chains
south of south end of Whitbourne’s flat, Sandy’s Creek.

In hand specimen the rock is dark, fine-grained, and somewhat
decomposed. Under the microscope it is seen to be considerably
altered. Phenocrysts of altered plagioclase and hornblende altered
to chlorite are abundant. A fair amount of granular magnetite
is present and the felspathic groundmass contains a little inter-
stitial quartz. The rock is an altered porphyrite.

1745. Acid lava at base of Upper Devonian, ridge south of
Ostler’s, and south of the Mitchell River.

In hand specimen the rock is compact, fine-grained, pink to grey
coloured, showing fluidal banding. Under the microscope the
fluxion structure is well developed. Phenocrysts of corroded
quartz crystals are set in a microcrystalline to cryptocrystalline
groundmass of quartz and felspar in which dark irregular bands
are developed streaming past and round the phenocrysts. The rock
is a banded and fluidal rhyolite.

1740. A nodular or spherulitic lava at the base of the Upper
Devonian, ridge south of Ostler's, and south of the Mitchell
River.

118

E. W. Skeats :

In hand specimen large nodular spherulites up to \\ inches
diameter of dense brown material, with lighter margin, are set in
a dense fine-grained matrix. Under the microscope the section
passes through the margin of one of the nodules, which is brown
in colour and almost completely glassy. The nodules are set in a
rock which shows a remarkable flow structure of cryptocrystalline
to glassy texture, in which occur small phenoerysts of corroded
quartz and of felspar. The rock is a spherulitic or nodular rhyo-
lite.

Summary and Conclusions.

The earlier work of Howitt and others in the Tabberabbera dis-
trict is referred to. The conditions of access to the district have
been improved by the making of good roads to the area, but within
the area the diminution of settlement clue to decay of mining and
introduction of rabbits has led to the overgrowing of tracks, the
growth of secondary scrub and the blackberry pest ; and these com-
bine with the rugged topography to make geological work difficult.
Two periods of three weeks each in January and February of
1924 and 1925 were spent in the area, which lies 40 miles NNW.
of Bairnsdale in Eastern Victoria. The boundaries of the eeolosri-
cal formations are approximately located on the map published
with this paper, and the structure elucidated by three sketch
geological sections. The area of Upper Ordovician rocks, con-
sisting of black shales, cherts and sandstone, previously known in
Sandy’s Creek has been extended.

It has been shown that at Tabberabbera itself, formerly re-
garded as the type area for the Tabberabbera shales of Middle
Devonian age, no Middle Devonian rocks occur, but that a broad
belt of Silurian rocks, consisting of impure limestones, shales, grits,
sandstones and conglomerates, trends NNW. to SSE. across the
area, including Tabberabbera. The Middle Devonian rocks, con-
sisting of blue limestones, shales and sandstones, with character-
istic fossils such as Spirifcr y ass crisis, are restricted to the
western part of the area.

The Upper Devonian rocks, red and purple sandstones, shales
and conglomerates, with interbedded rhyolites, form an unconform-
able plaster in the central part of the area, resting in turn on the
denuded edges of Middle Devonian, Silurian and Upper Ordo-
vician rocks. The Ordovician, Silurian and Middle Devonian
rocks have suffered from very severe compressive earth move-
ments, and are in consequence highly folded, even the Middle
Devonian showing dips ranging from 45° to 80° on either side of
the fold axes. The boundaries between these formations are
determined by unconformities or faults. The Upper Devonian
rocks, with a dip of 5° to the SW., have not been compressed into
folds, but were elevated and tilted with the older rocks by late
Kainozoic differential earth movements. The Mitchell and Went-
worth Rivers have cut steep valleys and gorges, and have given an
immature topography to the district.

Palaeozoic Rocks.

119

The outstanding structural features in the Tabberabbera district
are the local character of the severe compression of the Middle
Devonian sediments, contrasted with their open folding at Buchan
and Bindi, and the gigantic character of the unconformity separat-
ing them from the Upper Devonian rocks. The appreciation of
this has led to the discussion of the age of the rocks called Upper
Devonian, from which no fossils have been obtained in this district.
The comparison with areas of similar rocks at Iguana Creek, the
Avon River, Mt. Wellington and Mansfield in Victoria, and Mt.
Lambie and other areas in New South Wales, has led to the view
that the reference of them to the Upper Devonian can be justified
• on the available evidence.

A brief account of the petrology of the igneous rocks of the
area is appended. Lava flows of spherulitic and of banded rhyo-
lite are interbedded with the Upper Devonian. It is shown that
numerous dykes penetrate the Ordovician, Silurian and Middle
Devonian rocks, and that a few have been noted by Mr. Easton
penetrating the Upper Devonian. The types include mica lampro-
phyres, hornblende porphyrites, felspar porphyrites, keratophyres,
quartz keratophyres, spherulitic quartz porphyrites, dolerites and
oligoclase trachytes. In addition four examples of felspathoid-
' bearing rocks, tinguaites, have been recorded, two as boulders and
two as dykes “ in situ A

It is shown that the Eastern part of Victoria constitutes an
alkali-rich province, since felspathoid-bearing rocks have been
previously recorded by the author from Pretty Boy Pinch, the
Tolmie Ranges, Mt. St. Bernard and from Omeo, localities lying
west, north-west, north and north-east of Tabberabbera.

Bibliography.

1. A. W. Howitt. Notes on the Devonian Rocks of North

Gippsland. Prog. Kept. Gcol. Surv. Vic,, No. 3, pp. 181-
249, 1876.

2. E. J. Duxx. Notes on the geological formation of the coun-

try east and west of the Mitchell River, Gippsland. Rcpt.
and Statistics of Mines Dept ., Vic., Quarter ending
31/3/1890, pp. 22-26.

3. O. A. L. Whitelaw. Notes on the Devonian Rocks of

Gippsland, Mon. Prog. Rcpt. Geol. Surv. Vic., No. 2,
pp. 16-22, 1899.

4. H. Herman, in “ Victoria: Its Mines and Minerals/’ Special

Edition of Australian Mining Standard, June 1st, 1899,
p. 68, with geological sketch-map and section.

.5. R. Eth bridge, jun. Descriptions of new or little known
Victorian Palaeozoic and Mesozoic Fossils, No. 1. Prog.
Rcpt. Geol Surv. Vic., No. 11, pp. 30-36, 1899.

6. T. S. Hall. Graptolites from Sandy’s Creek, Mitchell
River. Ibid., Nos. 10 and 11, p. 104, 1899.

120

E. W. Skeats : Palaeozoic Rocks.

7. T. S. Hall. Reports on Graptolites. Rec. Geol. Surv Vic

i (1), pp. 33-34, 1902.

8. E. V . Skeats. from Report of Secretaries, Alkaline Rocks

of Australia and New Zealand Committee. Rept. Aust.
Assoc. Adv. Sci, xviii. (Perth Meeting 1926) o 41,
1928. 1

9. E. \\ . Skeats, from Report of Secretaries, The Alkaline

Rocks of Australia Committee. Ibid., xvi. (Wellington
Meeting. 1923), p. 107, 1924.

10. E. O. Teale. A Contribution to the Palaeozoic Geology of

Victoria, with special reference to the Districts of Mount
Wellington and Nowa Nowa respectively. Proc Roy
See. Vic., n.s., xxxii. (2), pp. 67-146, 1920.

11. F. McCoy. Prodromus of the Palaeontology of Victoria.

Decade I., pp. 37-39, 1874.

12. F. McCoy. Ibid.. Decade IV., pp. 21-23. 1876.

13. F. McCoy. Report on Palaeontology for the Year 1889.

Ann. Rcpf. Secretary of Mines , Vic., for 1889, pp. 23-24,
1890. '

14. G. Sweet. On the Discovery of Fossil Fish in the Old Red

Sandstone Rocks of the Mansfield District. Proc. Roy.
Soc. Vic., n.s., ii., pp. 2 and 13. 1890.

15. A. S. Woodward. On a Carboniferous Fish Fauna from

the Mansfield District, Victoria. Mem. Nat Mus
Mclb., No. 1, Jan., 1906, pp. 1-32.

16. A) . N. Benson. Materials for the study of the Devonian

Palaeontology of Australia. Rec. ' Gcol. Surv. N S
Wales, x. (2), pp. 83-204. 1922.

1/. R. A. F. Morray. Progress report on the Geology of por-
tion of the country between the Thomson and Wonnan-
gatta Rivers, N. Gippsland. Prog. Rcpt. Gaol. Surv
Vic., No. 4, pp. 53-54, 1877.

18. A. W. Howitt. Notes on the Geological Structure of N.
Gippsland. Ibid., pp. 77-78, 1877.

Proc. K.S. Victoria, 41 (2), 1020.

Plate XV.

GEOLOGICAL SKETCH MAP

OF-

TABBLRAB&CRA DISTRICT.

BY £. V/. SYC ATS. 0.5c.

W m C Hustle /

UPPER DEVONIAN

MIDDLE DEVONIAN

D Webs dak

UPPER 5ILURIAN

UPPER ORDOVICIAN

D10P1TL

P0RPHYR1TE

DYKES L

H 3/nnott

MILE

J Sinru

MOREKANA

5 Websdale

Oates

Scott's

? 5 E- 5 HOr

\atiiO \ :

Ostlers

NUHGAT.TA

1 1

ond)\ Fault ?d

[Proc. Eoy. Soc. Victoria, 41 (N.S.), Ft. II., 1929.]

Art. IX . — The Buildivg Stoves of Victoria, Part If'.
The Igneous Rocks.

By KATHLEEN McINERNY, M.Sc.

(Assistant Lecturer and Demonstrator in Geology,
University of Melbourne).

(With Plate XVI)

[Read lltli November, 1928; issued separately 30tli January, 1929.]

Introduction.

Previous Literature.

The Igneous Building Stones of Victoria:

Granites: Harcourt. Wangaratta, Cape Woolamai, Gabo
Island. Orbost. Trawool, Dromana, Colquhoun,
Tynong.

Daeite: Aura.

Porphyry : Tallangatta.

Basalts: Malmsbury, Footscray. Kyneton.

Tables of Tests and Chemical Analyses.

Summary.

Bibliography.

Introduction.

In the following* paper some Victorian igneous rocks used as
building stones are described. Of the fourteen rocks included
here, twelve have been used for constructional or ornamental pur-
poses as well as in monumental masons' work, and one in monu-
mental work alone, and one, the Tynong granite, has recently been
selected for use in the Victorian Shrine of Remembrance,

H. C. Richards (13) in 1909 published a description of eight
Victorian sandstones used as building stones. The title of his
paper was “ The Building Stones of Victoria — Part I. : l he Sand-
stones.” Therefore, this paper is styled 14 The Building Stones of
Victoria, Part II.: The Igneous Rocks.'

In addition to the building stones described here, there are
other igneous rocks occurring in Victoria which have been used
in the past for building stones, and of course many others which
may be used in the future, but so far as the writer is aware those
deferred to here include most of those being quarried at the
present time for purposes of building.

From Victorian igneous rocks entire buildings have been
erected, or they have served as basecourses, as ornamental pillars
or columns, and for monumental works, and at the present time
thin slabs are frequently cut for use as a veneer on concrete build-
ings.

'122

Kathleen M cl nervy ;

Igneous rocks used as building stones are divided into three
classes by stonemasons. The first is that of the “ granites; ” which
in this connection includes all coarse, even-grained types of
igneous rocks, and these arc usually capable of taking a polish.
The second is the M porphyry ” class, including all rocks with
large porphyritic felspar crystals. As a rule, these are used in
small polished slabs for ornamental purposes. The “ Milestones 99
or basalts form the third class. These are sombre, blue-grey
•coloured fine-grained rocks, whose chief use as building stones is
for basecourses, where their dark colour makes an effective con-
trast with a lighter coloured main structure.

The following scheme has been adopted in the description of
•each building stone. In an introductory paragraph the site of the
quarry, the amount of stone available there, and the systems of
jointing with their corollary, the size of blocks obtainable, are
described. In a paragraph headed 4k Appearance; ” the colour and
structure of the stone are referred to as well as any blemishes it
may possess. Under “ Working Qualities; ” the ease of sawing
and polishing the stone and the. quality of its polish are treated.
Following this, all the tests done on each stone are grouped to-
gether, and in a final paragraph a tabulation of some of the build-
ings erected of the stone is given, with a summary of the principal
•characters which either recommend or forbid its use.

The rocks are described in the order of the classes recognised
by technical workers in the trade. Within each class, where there
is more than one representative, that stone most in use for build-
ing is treated first, and is followed by the rocks in descending
•order of their use up to the present time.

Of the quarries described here all have been visited by the
writer, with the exception of those at Gabo Island and Tallan-
gatta. All crushing strength tests except that of the Gabo Island
granite have been carried out in the Melbourne University Engin-
eering School, on test pieces, most of which have been prepared
by the writer. The remaining tests have been done by the writer
unless the contrary is stated.

The writer desires to acknowledge gratefully the helpful advice
and criticism offered to her throughout this work by Professor
Skeats, and to thank Associate-Professor Summers also for his
continuous assistance.

Messrs. William Train and Co., and the owners and managers
of the various quarries visited, have been always most courteous
in throwing open their works for inspection, and in giving much
practical advice.

Previous Literature.

In 1860 a committee of the Royal Society of Victoria published
a report of building materials (1) occurring in Victoria. The
report which, so far as is known, is the first record in print of
such occurrences, describes rather fully the basalt or “ bluestone 99

The Building Stones of Victoria.

1 23

of the colony, without, however, specifying localities from which
it was then obtained. It records various > quarries for granite,,
indicating that stone for building in Melbourne had been obtained,
at that period from Gellibrand’s Hill, at Broadmeadows, and
from Gabo Island.

In 1864 a treatise on “ Australian Building Stones/’ by J. G.
Knight (2), was published in London. As the author was the
chairman of the Committee of the Royal Society referred to>
above, this treatise contains in a fuller form practically the same
information given in the Committee’s Report.

Later reports are confined to lists of localities of quarries in.
granite, greenstone, basalt, serpentine, etc., until in 1915 R. T.
Baker published the “ Building and Ornamental Stones of Aus-
tralia ” (17), which includes notes on many of the igneous rocks
of Victoria, referring to their use or possible use as building
stones. A similar list, though a shorter one, appears in the Com-
monwealth Year Book for 1909 (18). Until the present time
the only other references have been reports in Geological Survey
Records on single quarries and a few references in reports on the
building stones of other States.

Publications dealing with the use or possibilities of Victorian
igneous rocks as building stones as well as other books and papers
referred to here are listed in the bibliography at the end of the
paper.

The Igneous Building Stones of Victoria.

Granites.

The “ granite ” of the worker in the stone trade has been defined
above as a coarse, even-grained type of igneous rock which can
usually be polished, this group of rocks is subdivided according
to the predominating colour of each type into red or pink, grey,
green and black granites. In a red or pink granite the felspar
present is usually orthoclase or an alkali plagioclase which has
become reddish brown by iron staining. J his felspar being
present in comparatively large proportion, imparts its reddish
brown colour to the whole rock. These “ red or pink granites 9
conform most nearly to the granite of the petrologist. In a grey
granite the felspar present is a white one unco loured by iron,
which occurs with small amounts of black mica, giving a, peppei
and salt ” or grey colour to the stone. The petrologist’s grano-
diorite is included here. The green granites derive their colour
from the minerals hornblende and epidote. I he latter of these
occurs as an alteration production of plagioclase felspar. Such
rocks are diorites. A black “ granite ” may be composed ot
orthoclase or plagioclase felspar, augite and biotite, with an iron
oxide, when it will fall into either the syenite or diorite petrological
class. With the addition of olivine and the subtraction of some of
the felspar, the rock becomes a gabbro. ihe combined effect of

124

Kathleen Mclnerny :

these minerals approximates to a black stone. The first two
classes include a far greater proportion of rocks than do the last
two.

In addition to this subdivision by colour, granites are classified
according to their grain size. A convenient method of classifica-
tion is outlined by T. Nelson Dale (10). By it a granite contain-
ing felspars of more than 1 centimetre (2/5 inch) diameter is
classed as coarse-grained, one with felspar whose diameters lie
between 0-5 cm. (1/5 inch) and 1 cm. (2/5 inch), is medium
grained, and all those with felspars below 0-5 cm. are fine-
grained. Throughout this paper this scale is referred to
when the terms coarse-grained, medium-grained, and fine-
grained are used. The lower limit for the coarse-grained division
-seems rather a high one, but since this is the most distinctly enun-
ciated scale of grain size, and is quoted by J. Allen Howe in the
“ Geology of Building Stones ” (15), it has been adopted here.

In the quarrying of granite the system of joints which occur
in the rock are important. In most granite quarries it is found
that the rock will split most easily in one definite direction, which
is known always as the “ rift.” In a direction at right angles to
the rift a granite will also split with ease, but slightly less well
than along the rift. This second direction of splitting is known as
the “grain.” The terms “rift ” and “grain ” are used through-
out this paper with the same significance. In most quarries the
rift and grain are vertical, and the joint system in the third dimen-
sion is usually horizontal, so that a sheet-like structure in the
granite is suggested. This third joint is never so perfect as either
rift or grain, the break being usually concave or convex, and the
“ sheets ” more or less lens-shaped.

Another term of almost universal application in quarries, which
needs explanation, is the word “ dry.” A “ dry ” is a direction in
a rock mass along which a block of the stone tends to fracture,
but may not do so until after it has been quarried and exposed for
some time. The fracture does not usually take place along a
plane, but along a curved direction and penetrates for but a short
distance into a block. “ Drys ” are spaced quite irregularly, and
their existence in a block of stone is often not suspected until
after cutting and dressing is completed, when if a “dry” shows
up the block must be rejected.

Harco urt G ranite.

The granite used most widely in Victoria as a building stone
outcrops over an area of some 150 square miles in the neighbour-
hood of Harcourt and Ravenswood, SO miles north of Melbourne
and 20 miles south of Bendigo. It is quarried on the side of
Mount Alexander, three miles east of the Harcourt railway
station, on the Melbourne to Bendigo line, where quarrying com-
menced over sixty years ago, and during this period a large
quantity of stone has been removed.

The Building Stones of Victoria .

The joint system in this rock mass is exceedingly favourable
for the extraction of large blocks. The “ rift ” or easiest direc-
tion of splitting the stone runs vertically north and south, and the
“ grain ” is also vertical, and runs east and west. Rift and grain
are so spaced that very large blocks can be obtained; in 1921 a
block 84 feet long, 28 feet wide and 25 feet deep, which weighed
5000 tons was moved by a single charge of powder.

Appearance. — This granite is a light grey one, containing large
crystals of white felspar and glassy quartz, and a smaller quantity
of biotite mica. The felspars average 1/5 inch in diameter, so
that the granite just falls within the medium-grained division of
the Nelson Dale scale. The grain size is very even through the
rock except where “ black spots ” or “ heathen ” occur. These
are patches of dark fine-grained material averaging two square
inches in size, although much larger ones occur. They are rich
in the mineral biotite, and form basic segregations. They show
prominently on sawn and polished blocks and occur at an average
distance apart of two feet. Less frequently small acid veins
about half an inch wide occur, which contain quartz and felspar
alone. These are not very noticeable on account of the prevail-
ing light grey colour of this stone. Its light grey colour is this
granite's most noticeable feature, and is especially marked on
smooth, unpolished blocks. The polished stone has a darker
-colour, which becomes somewhat lighter after exposure, appar-
ently on account of the gradual evaporation of quarry damp. The
granite placed in 1926 in the additions to the State Savings Bank
in Elizabeth Street was distinctly darker at first than that in the
first part of the building erected in 1911, but now the junction
between the two cannot be distinguished. Specimens from all
parts of the quarry are very similar, and it is noticeable that the
rock outcropping at Big Hill, ten miles to the north of this quarry,

■ does not differ in grain size nor mineral composition.

Working Qualities. — Rift and grain in this granite are so well
developed as to make the ease of working this stone at the quarry
a standard of excellence among granite masons. Blocks of all
sizes and shapes required are obtainable, and since the supply is
practically inexhaustible any type of work can be undertaken in
this stone. At the mason's yard this stone takes a good edge or
*' arris,” and it polishes well, although biotite is inclined to flake
off from the surface, leaving it uneven.

Resistance to Crushing. — This stone has been tested in the
Melbourne University Engineering School three times for its
resistance to a crushing stress. Three inch cubes were used for
the tests, which were conducted on the dry stone. The cubes
• crushed at 11,444 lbs. per sq. in. (736 tons per sq. ft.), 11,333 lbs.
per sq. in. (728 tons per sq. ft.), and 8510 lbs. per sq. in. (547
tons per sq. ft.) respectively. The stone has a somewhat lower

■ crushing strength than most of the granites described here.

126

Kathleen Mclnerny :

Absorption. — The percentage of water absorbed was deter-
mined by immersing a small weighed and dried block of the stone
in water. The rectangular shape of the block made the conditions
approximate to those experienced by the stone in a building in wet
weather. After four days’ immersion the block absorbed 0-11%
of its weight of water, so that Harcourt granite may be called
impervious for all practical purposes.

Chemical Analysis. — A chemical analysis of this rock has been
made by Mr. G. Arnpt, and the result published in a paper by Dr.
H. S. Summers (16). It is included here under Chemical
Analyses at the end of the paper.

Specific Gravity and Weight per cubic foot. — The specific
gravity of this granite is 2*678, and hence the weight of a cubic
foot is 167*5 lbs., which is a normal weight for a granitic rock.

Microscopic Examination. — A thin section of this rock shows-
idiomorphic crystals of felspars, interstitial quartz, in some cases
under strain, and highly pleochroic biotite. There are a few
occurrences of the accessory minerals apatite, zircon and mag-
netite. The relative grain size has been calculated for the three
principal constituents by an adaptation of Rosiwal’s method for
measuring the dimensions of minerals ( 7). This rock is coarse-
grained enough for the measurements to be made in millimetres
on the polished surface of the rock itself. The result of twelve
traverses gave the ratio Quartz: Felspar: Biotite, as 4:5:2.

Felspar is in the form of plagioclase and orthoclase in the pro-
portion of 3:2. The plagioclase was determined as A^An^ or
Na 2 0 CaO 2Al 2 0 r! .8Si0 2 by measurement of its angle of extinc-
tion. Orthoclase has altered to kaolin, which has become iron-
stained, and some of the plagioclase has changed to epidote. This
rock belongs to the adamellite class, since more than one-third and
less than two-thirds of the felspar is orthoclase. In the American
Classification the rock falls into Class 1. Persalane; Order 4, Brit-
tanare; Rang 2. Toscanase; Subrang 3, Toscanose.

A portion of a basic segregation was examined under the
microscope. It is distinctly fine-grained. Plagioclase is more
abundant than orthoclase and the former shows marked alteration
to epidote. The section is crowded with small, stumpy biotite
crystals in greater abundance than in the normal rock. Quartz is
also present. This section was difficult to obtain since the basic
segregations are crumbly*

Uses. — This granite is widely used in Melbourne. Some of
the better known buildings in which it appears are the Colonial
Mutual Life Assurance, formerly the Equitable Life Assurance,
where the upper storeys have been constructed of smooth, un-
polished blocks, the Commercial Travellers’ Club, the State Sav-
ings Bank, the Herald Newspaper Office, the Union Bank, the
Flinders Street Railway Station, and many others. The stone is
seen throughout the city in polished ornamental panels, pillars,
steps and basecourses.

The Btiilding Stones of Victoria .

127

Its very light grey colour must be regarded as a defect in this
stone, because it becomes dirty rapidly in ^ city atmosphere. The
gateway of the Fish Market in Flinders Street, and Rocke, Tomp-
sitt’s warehouse in the same street are examples of dirty Flar-
court stone.

The dark basic segregations or “ heathen ” are a disfigurement.
These may be seen in the wall of the head office of the State Sav-
ings Bank.

No other granite, either Victorian or imported, has been used
to the same extent as this stone for building, ornamental and
monumental purposes in Melbourne.

Wangaratta Granite.

Granite from this district has been used locally and in Mel-
bourne. It is quarried in the Warby Ranges, about seven miles
SW. of Wangaratta. The Warby Ranges consist of a granite
inlier rising abruptly from a plain composed of Recent material.

The quarry for building stone has been made in an area where
segregation of pyrites has occurred in the granite, giving it an
appearance distinctly different from that of the pyrites-free
granite found at no great distance. The quarry is on a hillside,
and after the blocks are dislodged and shaped into roughly rectan-
gular blocks they are rolled downhill and levered on to lorries.
The working face slopes nearly parallel with the slope of the hill-
side. This quarry face is very uneven, since there is trace neither
of rift nor of grain in the granite, but “ drys,” whose nature is
defined above, are found irregularly through the stone. On ac-
count of the " drys,” the size of the blocks obtainable is very
uncertain, and a great deal of material has to be rejected. Blocks
up to six feet in length have been got out, but there is no guaran-
tee that blocks of this size can be secured frequently. The size
more usually obtained is 2 ft. 6 in. long by 1 ft. square. This lack
of regular jointing somewhat restricts the use of this granite as a
building stone.

Appearance. — The granite is pink and even-grained, but its
appearance varies with the amount of pyrites present, and the pro-
portion of this mineral which has been oxidised. Three distinct
types can be recognised, and are described here as A, B and C.

Type A is a pale pink, fine-grained granitic rock, containing
abundant creamy felspar, averaging 1/20 inch in diameter, grains
of quartz, and scattei'ed pyrite cubes. Some of these have been
lost, leaving small cavities in the rock.

Type B is a very soft friable cream-coloured rock. It contains
felspar, kaolin, quartz, but no pyrites, and is very porous.

Type C is a dark, fine-grained stone, of a colour ranging from
pale pink to purple. Felspar is very abundant, and there is a
good deal of quartz. Pyrites is absent, though occasionallv cubic
cavities, which contained originally pyrites crystals, are to be seen.

128

Kathleen Mclnemy .-

Limonite resulting from the oxidation of the sulphide mineral has
penetrated the felspar, colouring it dark pink and purple, and
probably causing the greater hardness of this type. Type C has a
warm and attractive appearance on either smooth or rock-faced
surfaces due to the alternating red and cream patches, according
to the varying richness of the stone in ferric oxide.

Working Qualities. — The buildings in which this stone has
been used have been constructed of comparatively small blocks,
averaging 2 feet 6 inches by 1 foot square. Types A, B and C
have been used for slightly different purposes in building construc-
tion. Type A, rich in unaltered pyrites, is used with a rock- faced
finish in the construction of walls. Men who have worked on
both say that this stone may be worked with about the same ease
as Melbourne basalt. When it is being chiselled a strong smell
of sulphur dioxide is noticed. It will not work up to a particu-
larly sharp arris. Type B being a soft stone is very easy to work,
and is used with the axed finish required in window surrounds.
Type C, the hardest stone, is selected for rock-faced work, and is
used chiefly in walls and foundations.

The rock rich in pyrites (type A) was found to take a good
polish; square cross-sections of pyrites, prismatic crystals of
cream felspar and quartz grains showing up well against a pale
pink groundmass. The only undesirable feature is the presence
of some small pits on the surface. Type C, which is coloured
purple-red, is much too porous to look well when polished, though
the solid parts take a high polish. Holes l/8th in. in diameter
and l/16th in. deep are commonly seen.

Resistance to Crushing. — Specimens of types A. B and C were
tested for their resistance to crushing. Type A. which is the
stone containing unaltered pyrites crystals, is much the strongest
of the three, since it broke only under a load of 19,600 lbs. per sq.
in. (1261 tons per sq. ft.). Type B, the soft stone, fractured
under a load of 7,110 lbs. per sq. in. (457 tons per sq. ft.), which
is a low value for the crushing strength of any igneous rock. Type
C proved rather stronger, breaking beneath a load of 9,670 lbs.
per sq. in. (622 tons per sq. ft.). The comparative weakness
under a crushing load of the two latter stones compared with
normal igneous rocks can be attributed to the changes suffered by
the stone in the oxidation of its pyrites. The figures indicate,
however, that even these two stones are quite strong enough for
use for ordinary purposes in a building.

Absorption Percentage. — Rectangular blocks of all the stones
were tested for their absorption percentages. They were im-
mersed in distilled water until they ceased to gain in weight, when
they were judged to be completely saturated. This took a dif-
ferent period for each stone. All their absorption percentages are
above the average of normal granitic types, due to the cavities
left, when pyrites cubes are lost, and the general alteration suf-
fered by the stones.

The Building Stones of Victoria.

129

Type A immersed for 9 days gained 1*45% of its weight.

„ B „ „ 13 „ „ 3-7,5% „

» C „ „ 14 „ „ 4*08% „ „

Specific Gravity and Weight per Cubic Foot. — Type A has a
specific gravity of 2-512, and weighs 157 lbs. per cubic foot. Type
B weighs 145 lbs. per cubic foot and its specific gravity is 2-324.
The specific gravity of type C is 2-446, and its weight per cubic
foot is 152-5 lbs.

These figures are all low, which is probably due to the fact that
the stones are rather porous. The resulting low weight per cubic
foot is a factor in favour of the use of this stone.

Microscopic Examination. — In a thin section of type A, felspar
makes up two-thirds of the rock, quartz bulks largely, and there
are some cubic crystals of pyrites. The felspar is allotriomorphic,
much of it being clouded by formation of kaolin, which is stained
bv ferric oxide. Epidote and sericite have formed also, and some
of the unaltered felspar shows lamellar twinning. Kaolin is
formed typically from alkalic felspar, while sericite and epidote
come from calcic plagioclase, and since a greater proportion of
the felspar present has altered to iron stained kaolin than to epi-
dote and sericite, this rock may be termed an altered granite.

Type B is very similar to type A, except that fresh, unaltered
pyrites cubes are absent from B. Clouded felspar is the most
abundant mineral. Kaolinization and limonitic staining are
marked, and the development of sericite from plagioclase is more
noticeable than in type A.

A thin section of type C is distinguished from types A and B
by the greater abundance of hematite present. After its forma-
tion by oxidation from pyrites, the hematite penetrated along
cleavage cracks of the kaolinized felspar, making a rectangular
network within the mineral (PI. XVI., Fig. 1), which has
strengthened and hardened the stone. Little quartz is present,
and no unaltered pyrites.

Uses. — The stone has been used in two churches in Wanga-

ratta. The first part of the Anglican Cathedral was built about
1908 of stone from this quarry, and in 1922 the quarry was
reopened to obtain stone for additions to this building. Blocks
of the hard red material (type C) are used with a rock- faced
finish in the main structure, while the softer type B is used for
the window surrounds. Rock was extracted from this quarry 60
years ago, when blocks for the Catholic Church in Wangaratta
were obtained. In Melbourne sawn blocks of Wangaratta granite
have been used in Collins House, Collins Street. The stone
used appears most like type C. For the keystone of the arch over
the entrance a block of Sydney sandstone was introduced.

The blocks are light reddish in colour, and show patches of a
darker colour due to the oxidation and leaching of iron of the
pyrites crystals originally contained in the rock. Such differential
staining is more usually associated with sedimentary rocks, and
the rock in this building is often mistaken for such.

•€a

130

Kathleen Mclnerny :

The warm reddish colour of this stone is very attractive, and
should make it a popular one for city use, since it discolours less
readily after exposure to a city atmosphere than do stones of
paler tints.

Cape Woolamai Granite.

The granite outcrop of Cape Woolamai forms the south-eastern
point of Phillip Island in Westernport Bay, and has provided
stone for building in Melbourne. Cape Woolamai is two miles
across Newhaven Strait from San Remo, a township on the main-
land 80 miles by road and rail south-west of Melbourne. By
another route the granite may be taken about 15 miles by water
to Stony Point, which is 46 miles from the city bv rail. The dis-
tance from Cape Woolamai to Melbourne directly by water is
approximately 65 miles. The depth of water at the stone landing
stage at the Cape is 2 fathoms. Three hundred yards out it has
increased to 12 fathoms.

The Cape is formed of a granite cliff, rising out of the sea to a
height of about 300 feet. At its widest, the Cape is one mile
across, and the granite is nowhere covered by more than a few
inches of unconsolidated sands. From the headland a jetty of
granite blocks was built out into Westernport, from which boats
removed the stone. At present the main quarry, which is con-
nected to this jetty by a tramline somewhat out of repair, is under
water at high tide. The perpendicular sides of this disused quarry
show that large well-shaped blocks were obtained by fracture
along regular joint planes, one of which strikes north and south
with the face of the joint plane dipping 60° to the east, while the
second strikes east and west and dips 30° south. Blocks up to 6
ft. in length by 2 ft. square are still lying at the stone landing
stage, while pillars 12 ft. high by 2 ft. square, and blocks 7 ft. long
by 3 ft. 6 in. wide by 2 ft. 6 in. high, were used in the base-courses
and portico of the Equitable Building, now the Colonial Mutual
Life Assurance Building.

The granite mass contains cream-coloured acid veins and vughs
of large pink felspar crystals, which mar the evenness of grain of
the rock. Segregations of basic material do not commonly occur
in this stone, which is remarkably free from any dark mineral.

Appearance.- — This granite has a pleasant colour varying be-
tween a light and a dark pink, according to the amount of altera-
tion suffered by the felspar present. The felspar crystals average
three-tenths of an inch in diameter, so that the grain size of the
rock is medium. It is composed mainly of pink felspar and
quartz. In addition a little green-stained felspar and a subor-
dinate amount of black mica are present.

The granite when polished has a darker colour, and makes a
handsome ornamental stone. Narrow veins about 2 inches wide,
containing large quartz and felspar crystals from 1 inch to 2
inches in length, cut across blocks of the normal coarse-grained
granite. More rarely portions of the stone are marred by dark

The Building Stones of Victoria .

131

•streaks caused by the segregation of ferromagnesian minerals in
narrow veins. Some of these can be seen 'in the base-course of
the Equitable Building. One vein measures 18 inches long and 2
inches wide. These veins are not so dark-coloured as the “ black
spots ” in Iiarcourt granite, because the black minerals are not so
closely packed, and therefore are less of a disfigurement to the
stone.

Working Qualities. — Little is known of other working qualities
than the polish of this stone, because it is over 30 years since it
was worked. On a test piece in the laboratory a surface was
smoothed and an extremely fine polish was obtained with rela-
tively little work.

Resistance to Crushing. — This granite has a remarkably high
crushing strength for a rock of this grain size. A block measur-
ing approximately i\ sq. in. by 2 in. broke under a load of 27,100
lbs. per sq. in. (1743 tons per sq. ft.).

Absorption Percentage. — A smooth block of this granite was
immersed in distilled water for 12 days, during which time it
absorbed only 018% of its weight of water.

Chemical Analysis. — The result of a chemical analysis of this
rock, which has been carried out by Mr. A. G. Hall for Dr. H. S.
Summers (16), appears at the end of this paper. This granite
is the richest in silica of the eight granites and granodiorites,
whose analyses are published in the paper cited.

Specific Gravity and Weight per Cubic Foot. — The weight per
cubic foot of this stone is 165 lbs., calculated from the specific
gravity of 2-643, given with the chemical analysis (16),

Microscopic Examination. — The minerals present are felspar,
quartz, biotite, apatite and zircon. Felspar is in the form of
microperthite altered to kaolin, and of plagioclase near oligoclase.
Some microperthite crystals show a thin film of iron oxide, which
is the cause of the reddish tint seen in most of the felspar in hand-
specimens. The tinge of green seen in others is due to small
crystals of epidote, formed from plagioclase. Large grains of
quartz are abundant, while flakes of biotite in a dark-coloured,
corroded form are rare. Some of these are altered to chlorite.
Apatite and zircon are included in mica. The proportion of ortho-
clase to plagioclase is greater than 2 to 1 ; therefore this rock is
a true granite.

Uses. — Large polished blocks ( 7 ft. long by 3 ft. 6 in. wide by
2 ft. 6 in. high) form the base-course of the Colonial Mutual Life
Building, and pillars 12 ft. high flank the entrance. This was built
for the Equitable Life Assurance Company in 1893, and the Cape
Woolamai quarry was opened to supply stone for this building.
So far as is known, it is used nowhere else in the city. Vertical
cracks have developed across the face of some blocks. It is likely
that these have arisen from “ drys ” in the granite, while the
appearance of some blocks is marred by quartz veins. It is re-
ported that specks of gold can be seen on some of the polished
blocks.

132

Kathleen Mclnerny :

The size of blocks obtainal.)le, the excellent polish and colour of
this granite are in its favour, and though the quarry is rather in-
accessible, its position at the water’s edge makes possible direct
water transport to the city by boats of shallow draught.

Gabo Island Granite.

This small island is composed of granite, which has been quar-
ried and used for building. It is close to the coast near the
boundary between Victoria and New South Wales, and lies near
the sea route between Sydney and Melbourne, 242 miles from the
former and 333 from the latter. It is thus accessible by boat
from either capital. Admiralty charts record the depth at the
jetty as 5 fathoms. Blocks measuring 2 feet high by 3 feet square
are in use.

Appearance. — The presence of abundant red felspar gives an
attractive pink colour to this rock. It is composed of compara-
tively small crystals, all uniform in size. Since the felspars have
an average length of one-tenth of an inch, the rock falls into the
fine-grained group of building stones. In a polished block the
colour is dark pink, though rectangular pale green felspars fre-
quently occur. The dark red colour of rock- faced blocks of this
stone can be seen in the Elizabeth Street Post Office, Mel-
bourne. A vein, half an inch wide* of very fine grained quartz
and pink felspar crosses one of the hand-specimens examined.

Working Qualities, — Blocks of this stone have been left ex-
posed for 60 years in a stonemason’s yard in Melbourne. They
have retained a good “ arris ” and polish until the present time.

The granite polishes fairly well, although small pits are left on
the surface where hornblende has been torn out while the stone
was being ground smooth.

Resistance to Crushing. — The result of crushing strength tests
on this granite is published by Baker and Nangle (12). Three
3-inch cubes were tested and their strengths in lbs. per sq. in. were
15.200, 14.900, and 17,500 respectively (9 79, 950, and 1128 tons
per sq. ft.).

Absorption Percentage. — A small block of the stone dressed to
a rectangular shape was immersed in distilled water for eight-
days, when it was found to have increased in weight by 0-39%.

Chemical Analysis. — The chemical analysis is recorded at the
end of the paper.

Specific Gravity and Weight per Cubic Foot. — The weight per
cubic foot of this granite is slightly under 165 lbs., calculated
from the specific gravity (16), which is 2-635.

Fire Test. — Baker and Nangle (12) have carried out tests to
discover the effect of heating and sudden cooling by streams of
water on Gabo Island granite. These tests imitate the effect of
fire and fire-fighting apparatus on a granite building. A cube was
heated gradually to 783°C., and removed after 35 minutes, when,
it was found to be badly cracked. A second cube was heated

The Building Stories of Victoria . 133

gradually to 544° C. and plunged suddenly into cold water. This
cube was almost unaffected.

Microscopic Examination. — This granite contains altered fel-
spar, abundant quartz and altered hornblende. Most of the fel-
spar is in the form of a microperthitic intergrowth beneath a film
of iron-stained kaolin, while some of it shows lamellar twinning
and a cloud of alteration products (epidote, sericite, etc.), in which
are caught up many small flakes of chlorite, which is the cause
of the green felspar noticed in the hand specimen. Ilmenite and
apatite also occur. 1 he rock is best described as a normal granite.

Uses. — Polished columns of this granite are used in the build-
ing of the Australian Travel Service, 493 Collins Street, rock-
faced blocks of the stone support Tasmanian sandstone columns
in the Elizabeth Street Post Office, and smooth-dressed blocks
form the base-course of the Customs House, Flinders Street.

The colour of this stone readily recommends its use, and its
strength is great enough to fulfil any requirement. Unfortu-
nately, the long distance of Gabo Island from a city will operate
against the frequent use of the granite from there.

Orbost Granite .

About two miles east of Orbost, on the road to Mallacoota
Inlet, and 233 miles east of Melbourne, Young’s Creek has ex-
posed a large face of granite which was quarried for use in the
Commonwealth Bank in Melbourne in 1923. No soil overburden
covers the granite, and there is an exposure about 40 feet high by
60 feet in width. The face of stone in the quarry is remarkably
irregular, and shows more or less conchoidal breaks, which prove
that there is no continuity in the jointing system in the stone.
The most marked joint runs on a sloping plane at right angles to
the face of the quarry, but is not continuous for any distance.

Appearance. — This is a greenish-grey granite, considerably
darker in colour than the Harcourt stone. Its green tint comes
from stained felspar, and is an attractive colour, especially when
seen on a polished block. The felspars average one-tenth of an
inch in diameter, and the granite is therefore fine-grained.

The granite mass is traversed by many veins of quartz up to an
inch in width, with some of which epidote and carbonate minerals
are associated. Large, dark-coloured segregations of basic
minerals, and narrow veins of dark-coloured minerals, also mar
the appearance of the stone, and in addition blocks up to 12 inches
square of fine-grained sedimentary material occur as inclusions.
Some of these are surrounded by a rim of partially absorbed
material. Even small hand specimens cannot be obtained free
from disfiguring u black spots.” One block of stone outcropping
near the quarry is traversed by three narrow dark veins, one of
which on examination under the microscope was found to consist
of a string of chlorite crystals altered from hornblende and biotite.
The minerals from which the string of chlorite has been derived

134

Kathleen Mclnerny :

have resisted weathering to a greater degree than has the re-
mainder of the rock, with the result that the narrow dark veins
are the centres of three ridges standing about one inch above the
general surface of the rock.

Working Qualities. — This stone was cut into two 2-inch cubical
blocks for testing purposes, and it was noticeably easier to saw
than most other granitic types, presumably on account of decom-
position suffered by the minerals present. The stone polishes
rather well, though the polished surface is somewhat pitted. These
pits are due to the loss of biotite during the grinding of the rock.

Resistance to Crushing. — Two rectangular blocks each approxi-
mately a two inch cube of this granite were tested. One crushed
beneath a load of 15,300 l'bs. per sq. in. (984 tons per sq. ft.),
while the other did not crush under the heaviest load of which
the machine is capable, 100,000 lbs., or 25,400 lbs. per sq. in.
(1633 tons per sq. ft.). It was noticed that although the speci-
men had not actually broken, it w T as just on the point of breaking.
It should be pointed out that although the crushing strengths of
these two cubes vary rather widely, they were prepared in a similar
manner by the writer from a single block of the stone, and were
crushed in the same machine by the same operator on the same
day.

Absorption Percentage. — A smoothed block of Orbost granite
absorbed 0T5% of its weight of distilled water after immersion
for eight days.

Specific Gravity and Weight per Cubic Foot. — The specific
gravity of this stone is 2-803, hence the weight per cubic foot is
175 lbs.

Microscopic Examination. — A thin section shows quartz in allo-
triomorphic and interstitial grains and felspar in relatively small
idiomorphic crystals, which are altered considerably, though in a
few crystals the lamellar twinning of the plagioclases can be
detected. The plagioclase has been saussuritized, causing the for-
mation of grains of zoisite, a little epidote and small, brightly
polarizing fibres of mica, probably the soda mica, paragonite, since
it has developed from plagioclase. Biotite is present, showing
very extensive alteration to chlorite, which imparts a green tint
to the rock, and a small proportion of hornblende also. Magne-
tite, apatite and zircon are accessories. The rock may be termed
a granodiorite.

A thin section was cut of one of the foreign included blocks.
This consists essentially of small angular interlocking quartz-
grains set in a felspathic matrix. Flakes of chloritized biotite and
cubes of pyrite occur sparingly. The inclusion is an indurated
sandstone or quartzite. The junction between the inclusion and
the normal granodiorite is marked by a band of quartz which has
recrystallized and forms a polysynthetic mosaic.

Uses. — The sole use of this granite in Melbourne has been for
the base-course of the Commonwealth Bank in Collins Street.

135

The Building Stones of Victoria .

The stone is polished here, but the polish is not good. Most of
the blocks are marred by black spots or inclusions of foreign
rocks.

This rock is found at such a great distance from Melbourne,
and is so variable in appearance, that it will probably never be
widely used. The heart of the stone would almost certainly be
more uniform in appearance. The colour is distinctly attractive,
and in comparison with other granites this one is more easily
worked.

Trawool Granite .

Granite from this locality was quarried for building about 30
years ago. The Monthly Progress Report of the Geological
Survey for 1899 (8) records the value of the granite obtained
here in 1897 to be £2,100. The quarry site is 60 miles north of
Melbourne, and within two miles south-west of the railway siding
at Granite, on the Mansfield line. Quarrying was commenced on
an outcrop beside the Trawool Creek, the water from which has
now filled the quarry hole to within three or four feet of the top
of its walls. The vertical walls of the quarry striking approxi-
mately north and south and east and west show that the joint
system in the stone is good, which is further proved by a polished
block measuring 4 ft. by 3 ft. 6 in. by 1 ft., lying near the quarry.
This block was rejected because large felspar crystals 2 in. long
by ^ in. wide have broken out of it, and cracks have developed in
the stone.

The quarry was opened to supply stone for the Equitable Build-
ing, hut early in the work it was abandoned on account of flaws
such as those found in the rejected block described above.
Machinery for all processes of dressing and polishing the stone
was brought to the quarry. Gabo Island stone was also taken
there to be dressed and polished.

Appearance. — It is a grey granite, containing white well-shaped
felspar crystals in a finer-grained groundmass. The average
grain-size of the felspar measured in a hand specimen is slightly
over one-tenth of an inch, so that this granite is a fine-grained
one. However, scattered through the rock are occasional large
felspar crystals, over an inch in length, making the rock almost a
porphyritic granite. These prominent felspar crystals make the
stone more suitable for small pieces of ornamental work than for
the construction of walls. The colour of the groundmass is
darker than Harcourt stone, and consequently it does not change
colour markedly after some years' exposure in city air.

Working Qualities. — On a polished surface of this stone nume-
rous pittings occur which are due to the softness of the abundant
biotite, which is torn out when the surface is being ground smooth.
The stone therefore cannot be said to take a good polish. The
tendency of large plagioclase phenocrysts to drop out during the
working of the stone has already been noted.

136

Kathleen Mclnerny :

Resistance to Crushing. — This has not been determined.

Absorption Percentage. — After immersion for twelve days a
rectangular block of Trawool granite absorbed 0-28% of its
weight of water.

Chemical Analysis and Specific Gravity. — The chemical analysis
of the Trawool granite has been published (16), and the specific
gravity determined as 2-666.

Weight per Cubic Foot. — The weight per cubic foot of this
rock is 167 lbs.

Microscopic Examination. — Plagioclase near labradorite,
quartz, perthite and biotite are found in the rock. Some of the
biotite is chloritized and apatite and zii-con are abundant acces-
sory minerals. The rock is named adamellite by Dr. Summers
(16). “ Xenoliths ” such as inclusions of micaceous hornfels and
cordierite hornfels found in the Trawool stone show that the
quarry is in an area close to the contact between sedimentary
rock and adamellite (Tattam, 22).

Uses. — Polished blocks of this stone have been used in the
base-courses of Sargood’s and Griffiths’ warehouses in Flinders
Street, Melbourne, in pillars at 459 Little Collins Street, and in
ornamental bands, and the steps of the Australian Mutual Provi-
dent Building, Collins Street.

Dromana Granite.

Granite outcrops in the hills behind Dromana township, 40
miles south of Melbourne, where a quarry has been made, which
is distant about two and a half miles east of the Dromana jetty.
The outcrop at this point is very extensive. Joints in the quarry,
approximately east and west and north and south, are evenly
spaced and are sufficiently far apart for the extraction of blocks
7 ft. long by 3 ft. square. In 1920 a private road was being made
to the quarry, and in 1924 a report appeared in the press that a
tramway was being constructed which would junction with the
Melbourne road, not far from Red Hill station.

Appearance. — This is a green granite whose colour is derived
from abundant felspar crystals altered to a bright apple-green.
These felspars are rectangular, and stand out almost as pheno-
crvsts from a finer-grained groundmass in which the felspar is-
creamy -yellow, Zoning in some of the green felspars can be de-
tected by the unaided eye.

The average size of all the felspars is slightly less than one-
twentieth of an inch, so that the Dromana stone ranks as a fine-
grained building stone. Veins half an inch wide of honey-
coloured quartz crystals cross some polished blocks of the stone.
No basic segregations mar any specimens which have been exam-
ined, nor were they seen in the stone at the quarry on a visit in
1920.

Working Qualities. — The working qualities have been tested in
Train’s yard. South Melbourne, where large-sized slabs have been:

The Building Stones of Victoria .

137

worked up. The granite is reported as comparing favourably
with Harcourt granite for ease of working. It spalls off well, and
works to a line edge and polishes well and easily. It repays the
work put into it better than does the Harcourt stone, since no un-
sightly dark patches appear upon the polished surface. The green
tint is seen to better advantage on the polished stone.

Resistance to Crushing. — Two tests have been carried out on
specimens of this stone obtained in 1918, when quarrying com-
menced first. Dry cubes measuring approximately two cubic
inches were tested. The cubes crushed under loads of 17,870 and
16,300 lbs. per sq. in. respectively (1149 and 1048 tons per sq.
ft.). The stone broke with the columnar fracture usual among
granitic rocks. The values are high, and since the two tests gave
results of approximately the same magnitude, it is safe to forecast
for the stone obtained from Dromana strength sufficient for any
purpose whatsoever.

Absorption Percentage.— The absorption percentage was de-
termined on a small smooth rectangular block of the granite.
After three days’ immersion the block was found to have in-
creased in weight by only 0* 18%. Hence, like the Harcourt stone,
it is nearly impervious to water.

Specific Gravity and Weight per Cubic Foot— The specific
gravity of this stone is 2-605; therefore, the weight per cubic foot
is 163 lbs.

Microscopic Examination. — This examination shows a liolo-
crystalline, even-grained rock which has undergone a good deal
of alteration. The minerals present include felspar, quartz, horn-
blende, biotite and in very small quantity apatite, zircon, magne-
tite, pyrites and copper pyrites. Specks of gold are sometimes
noticed on polished surfaces. The alteration of some of these
minerals has given rise to others. Chlorite has resulted from
changes in biotite and hornblende; kaolin and sencite from fel-
spar, limonite from the iron-containing minerals. Oithoclase
showing perthitic intergrowth w T ith another alkali type, and a
plagioclase, near labradorite, are both present. Minute flakes of
chlorite occur throughout the felspar, which are no doubt respon-
sible for its green colour. The proportion of orthoclase to plagio-
clase felspar being less than 1:2, this rock is classified as a giano-
diorite.

Uses. This stone has been used in the steps of the entrance to

and the facings for the block of shops in the Argus building,
Elizabeth Street, and a Soldiers’ Memorial in Daylesford contains
a polished block of it.

It can be obtained in great quantity, the quarry is reasonably
near Melbourne, the stone works up well, and has a handsome
appearance from its bright green colour, which should give it
special architectural value.

It is a stone that could be used on bigger pieces of work than
has been the case in the past.

138

Kathleen Mclnerny :

Colquhoun Granite.

Near the railway siding of Colquhoun, 28 miles east of Bairns-
dale in Gippsland, and 195 miles east of Melbourne, a red granite
has been quarried and worked up in Melbourne for monumental
work. The quarry is one mile west of the Colquhoun railway
siding and 200 yards south of the railway line. The country in
this district is thickly timbered, and the side of a small gully
has been chosen for the site of the quarry. There is no over-
burden on the patch of granite where quarrying was commenced,
and though the rock mass appears lens-shaped and dips away
front the surface to the north, the covering for some distance is
thin, and is composed of unconsolidated sands whose removal
should not represent a costly item in the quarrying operations.

A description of this quarry has been published by A. H.
Sharpe (20).

The stone has been quarried over an area approximately 40 ft.
long by 20 ft. wide and to a depth of 15 ft. Jointing is good in this
stone. There are two systems of vertical joints which persist
throughout the quarry. The “ rift ” trends north-north-east and
the “ grain ” makes an angle of approximately 100° with the
4 ‘ rift.” The joints are sufficiently far apart for the extraction
of blocks of 4ft. 6 in. long by 2 ft. square. The granite exposed
at the surface is somewhat iron- stained, but on a quarried surface
is free from quartz veins and basic segregations.

Appearance. — This granite has a warm pink colour, and when
polished is almost a brick-red. Its grain-size is fine, since its
felspar crystals average slightly less than one- tenth of an inch in
diameter. It contains beside felspar and quartz only a small
amount of biotite mica. Blocks lying at the quarry have a margin
of 6-9 inches wide of greyish-green granite, surrounding the
normal red rock. Outside this rim again, is a brown, iron-stained
band a quarter of an inch thick. A faint brown stain discolours
the surface of some polished red blocks. It is only visible when
the stone is polished. This rust mark is attributed to the passage
of iron-bearing solutions carried upwards through the rock by
evaporation. As described under the paragraph “ Microscopic
Examination,” the origin of the rim of grey-green granite sur-
rounding the red stone is ascribed to the same cause. All the
stone worked up to the present has been taken from the surface or
but slightly below it, and within a zone likely to be affected by
evaporation. Neither the stain nor the greyish-green rim occur-
ring along joints is likely to be found in blocks taken from greater
depths.

Working Qualities. — This granite is reported to be difficult to
work, but the stone takes and keeps a good arris. It takes an
excellent and uniform polish, since it is composed almost entirely
of quartz and felspar, which are of nearly equal hardness.

Resistance to Crushing. — The crushing strength of this

139

The Building Stones of Victoria.

granite, which was determined by a test on a two-inch cube, is
14,750 lbs. per sq. in. (946 tons per sq. ft).

Absorption Percentage. — This granite absorbed 0-32% of its-
weight of water after immersion for eight days.

Specific Gravity and Weight per Cubic Foot. — The specific
gravity is 2-616 and the weight of a cubic foot of this granite is
163 lbs.

Microscopic _ Examination. — In thin section the following
minerals are found, angular grains of quartz, orthoclase felspar
showing perthitic intergrowth with an alkali plagioclase, subor-
dinate plagioclase of composition between oligoclase and andesine,
and grass-green and greenish-yellow biotite which occurs very
sparingly. Much of the felspar is clouded by alteration to
kaolin and epidote. The former is coloured brown by a thin film
of iron oxide. The large proportion of orthoclase to plagioclase
present in this rock places it among the true granites.

An examination of a thin section of the greenish-grey granite
found as a rim 6-9 inches thick around the margin of some of the
exposed blocks of the normal red stone shows perthitic felspar,
plagioclase of composition near oligoclase, quartz, and a few
flakes of chloritized biotite, some of which are green, while one
or two whose ferrous iron has been oxidized show a brown stain-
ing. In the amount of iron-staining in the perthitic felspar lies
the difference between this grey rock and the red granite which it
surrounds. In the former, iron-staining is not so marked, and it
is suggested that iron has been leached out from the originally
stained felspar of the now grey rock, and carried to the surface
of the block, or to a joint plane channel. A narrow, very much
iron-stained rim, a quarter of an inch wide, was noted around the
extreme edge of the block, while the grey rock, which may be re-
garded as a bleached type, lies immediately inside the narrow rim
to a depth of about 9 inches, and inside it again is found the
normal red rock unaffected by iron-bearing solutions rising to the
surface by evaporation. The original red staining of the felspar
is regarded as the work of magmatic vapours. Plagioclase has
undergone alteration to epidote in both types and in both is subor-
dinate to orthoclase in amount.

Uses. — A few monumental headstones have been worked up
out of the red Colquhoun stone, but otherwise it has not been
used. The rock has an attractive colour, a uniform texture, takes
a good polish, and is easily accessible to a railway line. The
difficulty of working it up and the distance of the quarry from
Melbourne, while adding to the expense, should not prevent its
use in the future.

Tynong Granite.

The Victorian Shrine of Remembrance is to be constructed of
granite obtained from Tynong, Gippsland. Tynong is 43 miles
south-east of Melbourne, to which it is connected by rail. Several

140

K a tldeen Me 1 v ei ■?? y ;

large domes of granite form the crest of a small hill one mile
north of the railway station, and the quarry is situated on the
north side of one of these. There is a downward grade from
quarry to railway station, which facilitates transport. This dome
of granite rises about 25 feet above the ground, and outcrops over
.an area of about 30 feet in diameter, with no overburden what-
ever, and with but a quarter-inch rim of weathered iron-stained
material covering the fresh granite. As is usual with granite
exposed to the sun’s heat, thin sheets tend to exfoliate from this
'Outcrop and split off parallel to the domed outcropping surface.
The uppermost one is about 6 inches thick. Very little quarrying
has yet been done, but work is proceeding to expose the stone over
a larger area in order to obtain material for the Shrine, In the
weathered dome two major joints trending north-west and south-
east are evident about 12-15 feet apart. These are south of the
present small quarry, work in which aims at reaching them. In
the preliminary workings now going on the stone is reported to
split best along the 44 board,” that is to say, horizontally. When
splitting the stone vertically, north and south and east and west
directions are at present selected. Lines of holes are drilled in
these directions by means of a jack hammer driven by compressed
air. Plugs and feathers are then inserted in the holes, and on
hammering them a clean break occurs. The object of this pre-
liminary work is to cut back to the major joint or “ dry ” referred
to above, and down to a “board” or horizontal joint. The size
of blocks to be obtained is said to be limited only by the capacity
of the crane, which can lift 10 tons. The blocks so far removed
average 2 ft. 6 in. square by 3 to 5 ft. in length. In the faces of
stone exposed bv workings in the quarry, perhaps 250 square feet,
only two small 44 black spots ” each about 1 inch in diameter were
observed. One light-coloured vein about 1 inch in width, com-
posed of coarse felspar crystals, was seen passing through the
stone to the surface. Around this vein occurs a good deal of a
pvritic mineral. This mineral is fortunately only observed in any
amount coating joint planes, and does not extend into the body of
the stone, and in a thin section very little pvritic mineral was
found, so that if careful selection of pyrites-free blocks is made,
little trouble from discoloration by oxidation should be experi-
enced.

Appearance. — The rock is a very light grey granite composed
of large white felspar crystals, glassy quartz and a little black
mica. A small quantity of an iron sulphide mineral can be seen.
The axed surface of the stone is nearly white. As described in
the previous paragraph “black spots” are of very infrequent
occurrence. Indeed, dark minerals are only rarely found in the
main body of the rock, though patches about 12 inches in diameter
occur where dark minerals are more plentiful, but are not so con-
centrated as to constitute a 44 black spot.”

Working Qualities. — This granite is reported to correspond to
'that from Harcourt in its working qualities. The hardness of the

The Building Stones of Victoria .

141

two is nearly equal. The Tynong stone will work up to a smooth
axed surface, and into rounded capitals and pediments, and will
take a sharp “ arris/' It polishes well and easily, and looks dis-
tinctly grey when polished. The small infrequent biotite flakes
being so much softer than the quartz and felspar grind away more
quickly than these, leaving small pits in the polished surface.

Resistance to Crushing. — A rectangular block of this granite
of approximately two cubic inches volume in a compression test
crushed under a load of 25,700 lbs. per sq. in. ( 1652 tons per sq.
ft.).

This high value is comparable only with the crushing strengths
of the granites from Cape Woolamai and Orbost and the dacite
from Aura of the stones described here.

Absorption Percentage. — A block of this granite with smoothly
ground surfaces absorbed 0-28% of its weight of water after
eleven days' immersion.

Specific Gravity and Weight per Cubic Foot. — Th3 specific gravity
was determined as 2*- 633, and the weight per cubic foot calculated
to be 165 lbs.

Microscopic Examination. — In thin section the following
minerals are seen — allotriomorphic quartz in large and small
grains, of which some are interstitial, while some small grains
are contained within felspar crystals and collections of other
small grains are suggestive of chalcedonic quartz. The
greater proportion of felspar consists of large phenocrysts of
albite perthitically intergrown with orthoclase, and the remainder
is plagioclase of composition between Ab fi0 An 40 and Ab iM An,. 0 ,
which occurs in zoned phenocrysts as well as in smaller interstitial
crystals. The alkali felspar is somewhat kaolinized, while the
core of some of the plagioclases has altered to a sericitic aggre-
gate. In addition to these constituents there is only a small pro-
portion of dark-coloured minerals of which the principal ferro-
magnesian is brown biotite, in places altered to a green chloritic
product. Zircon, apatite, fluorite, pyrite and pyrrhotite are acces-
sory minerals. The two latter occur very sparingly, and their
oxidation should do no harm to the colour of the rock when
present in such small proportions. Undoubtedly these minerals
occur more freely along joint planes in some parts of the quarry
which should be avoided in the selection of blocks for building.
The texture is holocrystalline and hypidiomorphic. The rock is
a granite.

Uses. — Granite was quarried in this locality some years ago
and used as pitchers in the yards at Spencer Street Station.
However, as it was found to be slippery for such a purpose (a
quality inherent in all granites, and not confined to this particular
stone), the granite pitchers were removed. Blocks of granite
from this quarry have been used as pedestals for statues in the
Queen Victoria Gardens. For facing the exterior of the Shrine
<of Remembrance this stone is being used in smoothly dressed

142

Kathleen Mclnerny :

axed blocks. The stone with this treatment appears almost dead
white, which colour is desired for the Shrine.

The Tynong granite outcrops within 45 miles of the city, is
near a main railway line, and there is in sight a very large
quantity of stone, so that it should prove a useful stone for build-
ing purposes in Melbourne.

Dacite.

Aura, Dcmdenong Ranges.

A great part of the Dandenong Ranges consists of the rock
dacite, which in places has been quarried for building stone and
road metal. One such quarry, a quarter of a mile east of the
Aura railway station, and about 30 miles east of Melbourne,,
which has supplied stone for building, was visited. The place
chosen for quarrying was at the outcrop of a hemi-cylindrical
block about 30 feet in length. The quarry is ; within five yards of
the narrow gauge railway line, on the high or northern side, so
that drainage from the quarry should be excellent. The outcrop
has a semi-circular upper surface, showing traces of exfoliation,
which is stained slightly by rust and lichens. The exposed block
disappears into the hill-side under an overburden of about two
feet of soil containing boulders.

Two vertical joints, one due north and south and the other due
east and west, are excellently developed, allowing the removal of
well-shaped blocks at least 2 ft. long by 2 ft. 6 in. wide, since
blocks of this area have been cut into slabs 3 inches thick and used
for a veneer on a concrete building. A block now lying at the
quarry measures 5 feet by 2 feet by 2 feet. No horizontal joint
plane is apparent. Very little material has been removed from
the quarry up to the present. The cavity worked is 4 ft. high by
10 ft. long by 2 ft. 6 in. wide.

The colour of the exposed stone is uniformly dark, only one
acid vein a quarter of an inch wide being observed traversing the
outcrop. The stone has been quarried by drilling holes 6 to 8
inches deep, and inserting plugs and feathers in these. No
machinery has been installed, and the quarrying done up to the
present has been only in the nature of scratching at the surface.

Appearance. — This rock is fine-grained, and the fractured sur-
face is coloured dark-grey to black. Though petrologically a
dacite, it falls into the trade class of “ black granite; ” A very
slight tendency to parallel arrangement of the mica is apparent in
freshly-broken pieces in the field. The rock when polished is
darker in colour, since there are only a few sparkling crystals of
felspar to break its uniform blackness. A smoothly ground, un-
polished surface of the stone has a blue tint. Quartz veins a
quarter of an inch wide occur so rarely that only one out of every
five finished blocks of stone contains one. The colour is unpopu-
lar in the building trade, where it is condemned as “ cold.” For

The Building Stones of Victoria .

143

special purposes, such as monumental work, however, the distinc-
tive almost black colour of the polished face should be an asset.

Working Qualities. — The cutting of a small cube for a test did
not present greater difficulty than is to be expected with any
quartz-bearing igneous rock, though the stone is said by one man
who has had experience with it, to be difficult to work. This
man spoke also of the existence of “ drys,” whose occurrence on
a dressed block makes it necessary to reject the block. A sharp
knife-like “ arris ” was obtained on the test block. The rock takes
a high polish, and its surface remains smooth and without pits after
grinding. However, as the individual minerals are small and dark-
coloured, there is no relief on a polished surface except for some
milky quartz and a few grains of a metallic mineral. On this ac-
count probably, the polished Aura stone when used for building
is relieved by blocks of unpolished stone finished by patent ham-
mering, which gives it a light grey colour.

Resistance to Crushing. — A rectangular block approximately 2
sq. in. by 16 in. high crushed under a load of 26,400 lbs. per sq.
in. (1672 tons per sq. ft.).

Absorption Percentage. — After immersion in distilled water
for 12 days a rectangular block of this stone absorbed 0T6% of
its weight of water.

Specific Gravity and Weight per Cubic Foot. — The specific
gravity of this stone is 2-765, and the weight per cubic foot is
172-5 lbs.

Microscopic Examination. — This rock is remarkably free from
alteration products. Felspar of composition between labradorite
and andesine in zoned hypidiomorphic crystals is the most abun-
dant mineral. A few allotriomorphic crystals of quartz occur.
Of ferromagnesian minerals biotite is more prominent than pyro-
xene, which is represented by hypersthene. A few irregular
crystals of iron sulphide are present. In a dark- coloured rock
such as this one, discoloration due to the oxidation of an iron
sulphide mineral need not be feared. The groundmass is granular
and micro-crystalline in texture, and consists of small circular,
equidimensional felspar grains and some quartz. The felspar
crystals are very small, and the determination of their species is
difficult, but many of them being clear and untwinned suggest
orthoclase. Accessory minerals are magnetite, apatite and zircon.
The rock is a biotite hypersthene clacite.

Uses. — This stone was used in a branch of the English, Scot-
tish and Australian Bank, at the corner of Swanston and^Little
Bourke Streets, Melbourne. This was demolished in 1927, and
upon rebuilding the dacite was not used. Polished slabs of the
stone were used mainly, but around the doorway relief was given
to the dark Aura stone by specimens of a grey rock with an axed
finish. The locality of this rock is uncertain.

The abundant supply, the proximity to a railway track, and the
uniformity of colour are all points in favour of the use of this

144

Kathleen Mclnerny :

stone, though it should be noted that Aura is on a narrow gauge
railway line, and any quarried material has therefore to be trans-
ferred at Upper Fern Tree Gully to wide gauge trucks.

Porphyry.

Tallangatta .

This district contains many igneous rocks, one of which, a
porphyry, has been used in Melbourne as an ornamental stone.
Details of the position of the quarry with respect to the town of
Tallangatta. which is 212 miles from Melbourne in a north-east
direction, or of the nature of the outcrop of the rock, are not
known to the writer.

Appearance, — The stone used is light pink in colour, and has
large porphyritic crystals of a cream-coloured rhombic-shaped
felspar and corroded crystals of quartz set in a fine-grained pink
groundmass. Its attractive colour shows well when the stone is
polished. Other types from the same district have been examined,
which should also serve well as ornamental stones. One of these
has the greenish-brown colour and fine-grained appearance
familiar in the “ trachyte ” of Bowral, New South Wales. On a
polished surface numerous rectangular porphyritic crystals of
green felspar are seen. This stone polishes excellently.

Working Qualities. — A specimen of the pink rock was ground
smooth and polished in the laboratory. A very even surface was
obtained, and a high polish appeared on it. Other working quali-
ties are unknown to the writer.

Resistance to Crushing. — This has not been determined.

Specific Gravity and Weight per Cubic Foot. — The specific
gravity of this porphyry is 2-565, and its weight per cubic foot is
160- 5 'lbs.

Microscopic Examination. — Examined in thin section the rock
is found to be much altered. Large quartz crystals, some idiomor-
phic and some corroded by the surrounding groundmass, stand
out from among the rest of the minerals, which are clouded with
decomposition products. Small porphyritic crystals of felspar
occur, but all traces of twinning and cleavage are masked by a
thick film of secondary products, such as kaolin, sericite and iron
oxide, so that the species cannot be determined. Fibrous aggre-
gates of a ferromagnesian mineral, which is a chloride product of
the original biotite, are seen. The groundmass of the rock is crys-
talline, and consists mainly of small felspars. The rock is a quartz
felspar porphyry.

Uses. — A large rectangular block of the pink stone has been used
in the Eight Hours Day Monument, erected in 1888 near Parlia-
ment House, and moved in 1923 to its present site, at the corner
of Victoria and Lygon Streets. The steps and pillar of the monu-
ment are constructed of Harcourt stone, while the central block at
the base of the pillar bearing the inscription comes from Tallan-
gatta.

The Building Stones of Victoria .

145

Its warm pink colour, the large well-shaped, cream-coloured
felspar crystals, and its good polish make this stone a very hand-
some one.

Basalts.

Mai ms b ury Basalt.

The basalt of Malmsbury and district was quarried for building
stone and road metal before 1861, as notes on Quarter Sheet
9 NW. published in that year record quarries for this stone.
Many of the early quarries were in the neighbourhood of Green
Hill, a point of eruption three and a half miles north-east of
Malmsbury. Basalt quarries are working now about two miles
east of Malmsbury, and half a mile west of the railway siding of
Edgecumbe, on the Redesdale branch line. Several small quarries
.are in operation on a low rise about twenty feet in height, which
.at one time formed the outer bank of a meander of the Campaspe
River. The river now flows to the east of its old course, and this
•deserted meander is now a marsh. The quarries are working into
the old river bank, which by this means has been cut back about
30 feet. Pillars of stone which are useless on account of the
amount of honeycomb basalt in them are left standing in the. quar-
ries. The vertical jointing system in this stone is not particularly
marked in any given direction. Those working on the stone say
it may be split vertically with equal ease in any direction, and
vertical joints intersecting at various angles were observed in the
quarry. Some of these joints are filled with a weathered layer
of clay or “reef,” up to one foot in thickness, between the solid
basalt. These are called “ clay ” or “ open ” joints, and those
where no clay appears are known as “ tight joints.” The stone
with “ tight joints ” is more difficult to quarry. The vertical joint
planes meet at angles which suggest that columnar jointing, with
each column of a large diameter, is prevalent in this flow as in
other flows of basalt. The horizontal joints or “ bed ” are rather
uneven, and follow more or less the upper surfaces of layers of
honeycomb basalt which are found through the solid stone. A
characteristic section in the quarry is seen in a face of basalt about
16 feet in height. A layer of soil 1 foot in depth covers 4 feet
of solid basalt, which overlies 14 inches of honeycomb basalt.
Below this, 15 inches of solid stone overlies a thin “reed” or
sheet of porous stone half an inch in thickness. Below this
depths of solid stone of about 18 inches width are separated by
layers of honeycomb basalt perhaps 6 inches in thickness or by
narrower “ reeds.” The thicknesses of the layers of solid stone are
not uniform throughout the quarry. The “ honeycomb ” basalt —
i.e., the extremely porous stone — since it is developed in bands 6
to 14 inches thick, may mark the quickly cooled upper surfaces of
individual flows, to which a great deal of gas finds its way from
the body of each flow, and so forms the “ honeycomb ” basalt at
the surface of the flow. Later, this honeycomb layer is covered

146

Kathleen Mclverny :

by a new flow of basalt. The ease with which horizontal jointing
takes place immediately above a layer of “ honeycomb ” basalt is
probably due to this junction between flows. The “ reeds/'
which are thin sheets of porous basalt, are never more than half
an inch thick. These often occur in a horizontal plane, but also
an “ up and down reed ” occasionally passes through solid stone
from one honeycomb layer to another. It is thought that the
“ reeds ” represent the tracks of bubbles of gas which are carried
horizontally for a certain distance and then may find their way to-
the surface by travelling vertically or diagonally up through the
flow, making an “ up and down reed.” Their origin is thus pic-
tured as similar to that of the “ corks ” found commonly in Foot-
scray basalt and described later. In the “ reed,” however, there
has been no subsequent filling of the steam cavities. A type of
“ reed ” difficult to account for is one which follows a horizon-
tal plane for some distance, bends at right angles to a vertical
plane for about 6 inches, then back to a horizontal plane again;
for a foot, following this the “ reed ” bends down again along a
vertical plane, and finishes up after another bend in the same hori-
zontal plane in which it commenced. The bubbles appear to be
surmounting an obstruction of stone already solidified, perhaps a:
block of foreign stone being carried along by the molten basalt.

ooDoooOoo,,

O g

o x

0 0

O <00 O

Fig. 1. — Path of 44 reed ” in basalt, Malmsbury.

In the stone in this part of Malmsbury no “ corks ” are seen,,
but half a mile away basalt outcropping on the main road con-
tains them. They are described and their origin is discussed in
the description of the Footscray basalt. As in many other igneous
rocks, “ drys ” occur in this basalt. They are sometimes marked
by a brown iron-stained thread in the good stone as though some-
oxidation of iron had taken place along them.

Appearance. — The stone is fine-grained, porous and slate-grey
in colour. It contains many colourless, needle-like zeolite crystals..
Its uniform appearance is monotonous and unrelieved by any
sparkling mineral. Two grades of stone are recognised, depend-
ing upon the porosity. First quality stone is porous, but none of
the pores is larger than one-fortieth of an inch in diameter.
Second quality stone is porous and contains these minute pores,
but in addition some larger ones, whose diameters vary between;
one-eighth and one-half inch, are scattered irregularly through it.

Working Qualities. — The Malmsbury basalt is quoted by
quarrymen as the standard of excellence among building stones
when working qualities are considered. By means of a scavelling
pick a block of stone obtained from the quarry is readily dressed:

The Building Stones of Victoria. 147

to a regular shape. 44 Drys ” in the stone, as described above, are
the only flaws against which the workman has to guard when
selecting blocks for dressing. The average size of the finished
blocks obtained is 4 ft. long by 2 ft. by 1 ft. In a report by
Lidgey published in 1894 (6), the Malmsbury basalt is described
as taking “ a fine polish,” with which statement the writer cannot
.agree. A piece of first quality Malmsbury was smoothed and
polished, but as was expected, the pores in the stone are so nume-
rous that the smallest polished surface is broken by gaps where
the pores intervene, and in many of them the rouge powder used
for polishing lodges and is extremely difficult to remove. To
overcome this disadvantage Canada balsam was poured on the
smooth surface of a heated block of the basalt, then baked, and
allowed to cool. Grinding removed the superficial layer of
balsam, and the surface was then polished. An even polish,
though a poor one, resulted, since there were few pores unfilled by
balsam.

Resistance to Crushing. — A two-inch cube of the stone was
crushed under a load of 8,620 lbs. per sq. in. (554 tons per sq.
ft.).

Absorption Percentage. — The stone absorbed 2-16% of its
weight after immersion in water for five days. Absorption was
complete after this period, since three weeks later no material in-
crease in weight was found in the test block.

Specific Gravity and Weight per Cubic Foot. — The specific
gravity is 2-595, and the weight per cubic foot 162 lbs.

Microscopic Examination. — The rock contains plagioclase
felspar laths whose mean composition is that of labradorite, abun-
dant squat prisms of faint green augite, corroded crystals of
olivine, the larger ones completely changed to brown iddingsite,
and the smaller ones colourless with a border alteration only. Mag-
netite is common, many crystals of it occurring in long narrow
flakes. The texture is ophitic, since some of the felspar is en-
closed by later crystallizing augite. On account of the extreme
porosity of the rock, a thin section is rather broken. The rock is
an olivine basalt.

Uses. — Malmsbury basalt was at one time used very extensively
for the basecourses of large Melbourne buildings. The stone
from the Green Hills quarries probably figures in many of these.
Nowadays some of the stone from the quarry visited near Edge-
cumbe is used for basecourses, but it is generally sawn into steps
and cemetery kerbing. Malmsbury basalt forms the basecourses
of the following buildings in Melbourne : — English, Scottish and
Australian Bank (head office). Bank of Australasia, Australian
Mutual Provident, Royal Insurance, London and Lancashire In-
surance, Northern Assurance, Alliance Insurance and Guardian In-
surance Buildings, and many others. It is used in the gateway in
the main entrance to the University. In combination with boot-
scrav basalt it is used in the Melbourne Grammar School. Malms-

148

Kathleen Mclnerny :

bury basalt is not accepted for use on the roads, as on account
of the ease with which it is worked it is regarded as too soft for
this purpose. The ease with which it may be worked recom-
mends this stone for use for building, when its dull appearance-
will serve as an effective contrast to a lighter-coloured stone, but
it is too unattractive to be used alone in construction.

Foot sc ray Basalt.

The quarry visited at Footscray lies north of the Footscray rail-
way station, and is four miles west of Melbourne, Quarrying
commenced here more than 25 years ago, and in that time stone
has been removed over an area 150 yards square for an average
depth of 20 feet. The ground here was practically horizontal
originally, and the result of quarrying has been to leave a hole in
the ground with the dimensions given above. Several basalt flows
have occurred in this area, the later ones being superimposed on
the earlier, and marked junctions occur between them. Drainage
from the quarry is reported as good, since all rain water flows
away through a vesicular, iron-stained basalt which forms the
bottom of the main quarry. The depth below the surface at which
the iron-stained band lies, varies in different parts of the quarry.
At the northern end this band is 21 feet below ground level. Im-
mediately above it lies 15 feet of solid rock, from which building
stone is obtained. In the eastern part of the quarry this same
vesicular band is seen 10 feet below the surface of the ground,
and stone is quarried from beneath it. presumably in an earlier
flow. The top of the quarry is formed by 5 feet of stone showing
irregular columnar jointing. The jointing system is very irregu-
lar. even in the solid stone. In quarrying, a vertical drill-hole per-
haps 26 feet deep is made with a pneumatic drill, until the iron-
stained honeycomb layer is reached. A charge of powder in this
hole blows out the side of the quarry, and may dislodge blocks
large enough for building stone purposes. Such a block with
very irregular surfaces and measuring 8 feet long by 3 feet square
was seen by the writer. It was to be cut up by the use of plugs
and feathers into regular-shaped kerb stones 6 or 7 feet long by
12 inches by 7 inches. Stones which spall well are cut into larger
blocks suitable for basecourses. Otherwise the building stone
market is not now catered for.

Horizontal layers of vesicles traverse the solid rock. They
average half an inch in width, but occasionally reach two inches.
They are known as “ reeds,” and represent the paths of bubbles
of gas which have travelled through the lava along the direction
of the “ reed.” The gas has taken a horizontal track along the
level where the viscosity of the crystallising lava has prevented its
further passage upward. The lava being still in motion, has
drawn out the bubbles of gas into ellipses, whose long axes are
arranged parallel to the direction of movement. In some u reeds ’ r
the vesicles are lined with white carbonate crystals, which pene-

The Building Stones of Victoria.

149

trate into a central cavity. In such a case the quarrymen speak
of a “ silver reed.” Elsewhere the solid ^tone is pierced by verti-
cal cylindrical pipes known as “ corks.” These are channels
whose diameters vary from one to three inches, which pass from
the bottom honeycomb layer of the quarry vertically up through
the stone until within 14 feet of the surface. In the uppermost
14 feet of the quarry is stone showing a great development of
platy and horizontal jointing, which may represent a different flow
of basalt, and through which the “ corks ” do not pass. The chan-
nels or “ corks ” show up in the normal basalt, since their margins
are defined by rows of vesicles, and they are more porous than the
normal stone, carrying as they do an average of 20 pores to the
square inch. The pores are the shape of irregular triangles, or
quadrangles, or they are circular. Some are filled with a white

i 1-

1 Inch

Fig. 2. — Longitudinal section through “ cork.” One-fourth of the larger
pores are filled with carbonates. The shapes of the pores are character-
istic.

carbonate mineral. The rock forming the matrix of the cork
is finer-grained than is the normal rock. Near the bottom of the
flow the u corks ” are narrow, being about one inch in diameter,
while towards the top they expand to about three or four inches
across. At the bottom of the “ cork ” the white carbonate coating
to the pores, which makes a “ silver cork,” is more commonly
found than at the top of the cork,” where the vesicles remain

150

Kathleen Mclnevny :

empty, forming a “ black cork.” The “ corks ” are practically ver-
tical, but owing to the irregularity with which a stone splits after
a charge of gelignite, they appear in circular, elliptical or half-
moon-shaped cross-section and look like sporadic occurrences.
One “ cork ” was traced vertically by the writer for twelve feet

Fig. 3. — Characteristic cross-sections through “ corks.”

through the stone, while experienced quarrymen state that they
have traced a single “ cork ” through the 40 feet of the stone
which is worked. “ Corks ” are the channels by which the gases
imprisoned in a lava escape to the surface. On account of the
passage of gases through such a channel, the lava in its neigh-
bourhood is rendered very porous, and is retained in the liquid
state for a longer period than the surrounding lava, one result of
which is the larger percentage of isotropic material seen in a thin
section of a “ cork,” as described later under “ Microscopic
Examination.” The ” cork ” expands when nearing the surface,
since the overlying pressure is less.

Du Toit describes (11) in the diabasic lavas of Barkly West,
South Africa, “pipe amygdales ” or “bubble trains,” structures
very similar to “ corks,” which he ascribes to the “escape of
steam generated in the flowing of molten rock over moist sur-
faces.” In Victoria, since many of the basalt flows have filled old
stream valleys, an analogous origin is not impossible. As de-
scribed earlier, the floor of this quarry is formed by an iron-
stained layer of stone which may represent the lower surface of
the flow in contact with the bed of the old river.

Appearance. — This basalt is a fine-grained rock coloured dark-
grey, with a slight bluish tint, which earns it the trade name of
“ bluestone.” The whole rock is pierced by fine pores, and in
addition larger circular vesicles averaging one-tenth of an inch
in diameter occur about one inch apart throughout the stone. The
rock is denser than the basalt from Kyneton, in which the pores
and vesicles are more numerous. It contains small sparkling
felspar crystals and occasionally a yellowish-green powdery
mineral, probably halloysite. Blocks of otherwise solid stone con-
tain elliptical or circular patches of varying size, which are cross-
sections through the “ corks ” described above. In paving stones
the “ corks ” stand up above the surface of the rocks after a
certain amount of traffic over the pavements. This is seen especi-
ally on the north side of Bourke Street, between Swanston and
Russell Streets. This greater resistance to wear is considered to
be due to the coarser texture of the “ cork,” as compared with

The Building Stones of Victoria . lol

the finer-grained nature of the matrix. Another type of abnor-
mality found in this rock is the “ flint/’ which is a patch rich in
•calcium carbonate, where all pore spaces are filled with this white
mineral, which has crystallized from solutions perhaps imprisoned
in the rock when crystallization of the main flow prevented their
escape. The “ flints ” are so known because it is reported that
the rock in which they occur is more difficult to work than the
normal basalt. They are also found in the Kyneton stone.

Working Qualities.— In spite of the irregular jointing this
stone spalls remarkably well. An experienced worker easily
breaks it into rectangular blocks with a hammer. However, it
is not so easy to work as the more vesicular Malmsbury and Kyne-
ton basalts. A sample was ground smooth and polished with
rouge powder, which filled all the small pore spaces in the basalt,
and no amount of scrubbing would remove it. The solid portions
between the pores took a moderately good polish, showing a
greyish -brown colour, but the general effect is far from pleasing.
This result illustrates the common saying, 44 Basalt will not
polish.”

Resistance to Crushing. — A three inch cube of Footscray
basalt which was crushed in 1891 fractured under a load of
10,577 lbs. per sq. in. (680 tons per sq. ft.). A two inch cube of
this stone was made from material obtained in 1926 from a quarry at
Footscray, and this stone proved stronger than that crushed
earlier. Its crushing strength was 16,300 lbs. per sq. in. (1048
tons per sq. ft.).

Absorption Percentage. — The absorption percentage of this
stone is L45. The absorption of water by the basalt is gradual
and continuous. In five days’ immersion a block measuring about
one cubic inch increased in weight by 1-09%, but after thirteen
days its weight had increased by 1-44%. At the end of seventeen
days, when the stone had absorbed 1*45% of its weight of water,
saturation was considered complete.

Chemical Analysis. — A chemical analysis of the basalt from
this quarry has been made in the Victorian Mines Department
Laboratory by Mr. A. G. Hall, but has not been previously pub-
lished. By the courtesy of the Geological Survey permission has
been given to publish it in this paper with other analyses.

Specific Gravity and Weight per Cubic Foot. — The specific
gravity of the stone obtained from this quarry determined by
weighing a specimen first in air and then in water is 2*570. From
this value the weight per cubic foot is found to be 161 lbs. It will
be noticed that the specific gravity given with the chemical analysis
is 2-839. This value was obtained by weighing the powdered
basalt, and since basalt is a porous rock the disagreement between
the two determinations is intelligible. The first method is more
useful for building stones, though it is difficult to carry out in the
case of a porous rock, where some of the water is absorbed while
weighing is proceeding. The result obtained is known as the

T52

Kathleen Mclnerny :

“ apparent specific gravity,” and from it the weight per cubic foot
of the stone is calculated.

Microscopic Examination. — A thin section of the normal
stone of the quarry contains laths of plagioclase whose composi-
tion is between labradorite and bytownite. Many of these are set
inside titaniferous augite crystal's in the typical ophitic texture.
Some augite shows strain polarization. Olivine is abundant, and
has a brown alteration product, iddingsite, around its edges and
along its cleavage planes. Occurring interstitially between some
felspar laths is a colourless substance thickly studded with black
globules of iron oxide and some long laths of an iron oxide
mineral, magnetite or ilmenite. Since this mineral occurs in
hexagonal plates, and in brown skeletal crystals, forms more
characteristic of ilmenite than of magnetite, it is more likelv to be
the titanium-bearing iron oxide, ilmenite. Titanium is also present
in the augite of this rock. The colourless matrix in which the
ilmenite occurs was at first taken for volcanic glass, but Professor
Skeats has pointed out that while some of this material is
isotropic, much of it is not, and also the refractive index is too
low for a basaltic glass, nor has it the characteristic greenish-
brown colour of such a glass, and further glass in basalt is found
only in a narrow tachylytic margin of a basalt flow, always less
than one inch in width, while the specimen from which this
section was cut comes from within a uniform mass of basalt, cer-
tainly 30 feet in thickness. In ordinary light the refractive index
and colour of this material are similar to the felspar of the basalt.
Where it is anisotropic its polarization colours are low in the first
order, and occasionally there is a suggestion of zoning in the inter-
stitial material. These considerations point to the interstitial
material being felspar. It has been the last material to crystallize
from the liquid state, and has thus filled up interstices’ a role
commonly taken by the quartz of quartz-bearing rocks. The iron
oxide carried in this liquid has separated in the form of
globules. In some cases this liquid has been supercooled below
the temperature of crystallization of felspar, and has eventually
solidified in the form of a felspathic glass, and hence is isotropic.
In ordinary light this isotropic material is in colour and refractive
index indistinguishable from that which is anisotropic, and the
former also contains the iron oxide globules found in the latter.
Where an augite crystal is set in this matrix the iron oxide has
been drained from the latter to go to the formation of augite,
and the augite crystal is surrounded by a narrow rim of clear,
colourless, felspathic matrix. The rock is a porous olivine basalt.

In a thin section of a “ cork ” or pipe-amygdale, the materials
of the normal rock are found. The interstitial felspathic glass is
more abundant here, and in this case practically all of it is iso-
tropic. In this section in addition to the globular form the iron
oxide also occurs in hexagonal plates, and in brown feathery
skeletal crystals (Plate XVI., Fig. 2) suggestive of incipient

The Building Stones of Victoria. 153 ;

crystallization brought about by the mother liquor being retained
in the liquid state longer than where the iron oxide is found in the
globular form. When it is remembered that this material comes
from the former channel for the passage of gases its less crystal-
line state is explicable. While passing through the “cork” the
gases would tend to keep the neighbouring lava in a state of flux,
with crystals of augite, olivine, felspar and iron oxide suspended
in a liquor which, drained of other constituents by their crystalliza-
tion and largely felspathic in composition, became supercooled
below its freezing point. After the passage of most of the gases,
that is to say of the fluxing agent, sudden solidification or
quenching would cause the formation of felspathic glass as the
matrix binding together the crystalline material Many pores are
filled with a concentrically or radially arranged calcium carbonate
mineral, probably aragonite. In a thin section of a “flint” the
minerals of the normal rock may be recognized. Many of the
pores in the rock are filled with aragonite. Here too the fels-
pathic matrix occurs in both isotropic and anisotropic forms, the
former preponderating. Carried in it is iron oxide in both
globular and skeletal-crystal forms. Hematite is noted very occa-
sionally in the section.

Uses. — The stone from this quarry is used for screenings, for
foundations in concrete roads, for gutter pitchers, paving slabs,
doorsteps and staircases and, when large enough blocks are ob-
tainable, for building stone. It is commonly used in basecourses,
and can be seen in the base of the Melbourne Town Hall and of
the Telephone Exchange. St. Patrick's Cathedral, Melbourne,
has been constructed entirely of Footscray basalt, as have been
numerous Melbourne warehouses.

The large supply, the proximity of the outcrop to Melbourne,
and the comparative ease of dressing it, are to be reckoned in
favour of this stone's use in Melbourne. Its dark colour consid-
erably lessens its suitability as a building stone, and while it should
make it more suitable for monumental work, the fact that it will
not polish curtails its use for this purpose.

Ky net on Basalt.

Basalt in the neighbourhood of Kyneton, as well as near
Malmsburv, has been quarried for building stone for many years.
A quarry visited three miles south-east of Kyneton, and two miles
north-west of Carlsruhe, is on the site of a very old one on a low
ridge running parallel to and about one mile east of the railway
line, and is 55 miles north of Melbourne.

An area 30 ft. by 40 ft. has been quarried to a depth of 25 ft.
By the lease under which the stone is obtained, the quarry has to
be filled as stone is taken out, so that only a small pit is left where
present quarrying operations go on.

The quarry face shows a layer of overburden 2 ft. in depth,
while for a further 5 ft. below the stone is broken into large

a 54

Kathleen Mclnerny :

boulders showing spheroidal or “ onion ” weathering. Below this
the stone is solid though at intervals very vesicular bands about
2 in. in width occur. The stone in the upper portion of the
quarry contains some sporadic vesicles or “ blow-holes,” which are
larger than the pores of the normal vesicular basalt. The stone in
the lower 15 ft. of the quarry is free from them. A series of
horizontal and north-south and east- west vertical joints traverse
the stone, along which weathering agents have found a track,
forming a band of clay one inch in width as a result of the altera-
tion of the basalt. Hence the joints are called “clay joints.”
From the u clay joints ” the flaws called M drys ” pass in to the
good stone. The stone is liable to split along a “ dry ” and a block
must be rejected if a “ dry ” appears on a dressed surface. The
dry may only show after the dressing of the block is complete.
The stone is found to be without “ blow-holes ” or u drvs ” about
25 feet below the surface, where flawless regular blocks 10
feet by 8 feet by 7 feet can be extracted. Vertical holes 6 inches
deep are drilled 8 inches apart by means of a jumper drill along a
vertical ** clay joint.” After the block has cracked vertically, that
is, along the“ cut." holes passing horizontally into the stone for 18
inches are drilled 2 feet apart on the front of the stone just above
a line of honeycomb basalt if any is present. Plugs and feathers
at first and. later, *’ lifters ” are inserted in these holes, and a hori-
zontal crack occurs which is called the “ board.” Along the east
and west vertical joints a good face of stone is obtained. Along
other vertical joints the stone is apt to break irregularly.

Appearance. — This basalt is a drab-grey vesicular one, with
here and there groups of vesicles filled with an opaque white
mineral which effervesces with acid, and when examined in thin
section is found to be calcite. The patches which contain calcite
are known as “ flints,” and are said to blunt tools used on the
basalt. It is difficult to see why the soft mineral, calcite, in these
patches should have earned them the name of “ flints,” though
when it is remembered that elsewhere the vesicles are empty the
assertion that the “ flints ” are harder than the normal stone is
probably explained. Some vesicles are penetrated by needle-like
crystals of probably a zeolite mineral, natrolite. An axed surface
of this basalt is lighter grey in colour than is a rock-faced block,
but both colours are monotonous and rather unattractive.

Working Qualities. — This stone is reported to be easier to work
than the Footscray basalt, though it is not so easy as that from
Malmsbury. In common with these other basalts it is easily
broken by the hammer along plane surfaces at right angles, giving
nearly smooth rectangular blocks. A great deal of quarry damp
is noticed in the stone when a chip is flaked off a block just after
it has been quarried.

This basalt is too vesicular to be susceptible of polishing.

Resistance to Crushing. — This porous basalt has a lower crush-
ing strength than the denser Footscray basalt. The Kyneton

The Building Stones of Victoria. 155,

stone crushed under a load of 9,220 lbs. per sq. in. (593 tons per
sq. ft.).

Absorption Percentage. — Absorption ' percentages of both,
normal and “ flint ” types of the stone were obtained. The nor-
mal stone absorbed water slowly, and saturation was not complete
until the stone had been immersed for 24 days, when it was found
to have absorbed 2-92% of its weight of water. The absorption,
percentage of the “ flint ” type was 1-77, which indicates that in a
“ flint” nearly half the pore space of the normal basalt has been
filled with secondary minerals.

Normal Footscray basalt has an absorption of 1-45%, i.e., only
half that of normal Kyneton stone. The porosity of the latter
makes it the easier stone to work.

Specific Gravity and Weight per Cubic Foot. — The specific
gravity of the stone is 2-615, and its weight per cubic foot is
164 lbs.

Microscopic Examination. — A thin section shows fresh, unal-
tered labradorite felspar and augite, olivine surrounded by a rim
of reddish-brown iddingsite, and some magnetite. The structure
is vesicular, but many of the vesicles are coated with calcite. The
mineral content and texture of this rock are typical of normal
basalt.

Uses. — The main use to which the Kyneton stone is put is the-
construction of basecourses. Its dark colour makes an effective

Tabulated List of Tests.

Name and locality

Specific

Gravity

Weight per
Cubic Foot
in Lbs.

Absorption

Percentage

Crushing
Strength
Lbs. per
Sq. In.

Crushing
Strength
Tons per
Sq. Ft.

Granite, Harcourt

- 2-678

- 167-5

- Oil -

11,444

- 736

11,333

- 728

8,510

- 547

Granite, Wangaratta —

Type A

- 2*512

- 157

- 1-45 -

19,600

- 1261

„ B - -

- 2 324

- 145

- 3-75 -

7,110

- 457

„ c -

- 2A46

152*5

- 4-08 -

9,670

- 622

Granite, Cape Woolamai

- 2-643 (16)

- 165

- 0-18 -

27,100

- 1743

Granite, Gabo Island

- 2-635 (16)

- 165

- 0-39 -

15,200

- 979

14,900

- 950

17,500

- 1128

Granite, Orbost

- 2-803

- 175

- 015 -

25,400*

1633

15,300

984

Granite, Trawool

- 2-666 (16)

- 167

- 0-28 -

not determined

Granite, Dromana

- 2-605

- 163

- 018 -

17,870

- 1149

16,300

- 1048

Granite, Colquhoun -

- 2-616

- 163

- 032 -

14,750

- 946

Granite, Tynong

- 2-633

- 165

- 0-28 -

25,700

- 1652

Dacite, Aura

- 2-765

- 172-5

- 0*16 -

26,000

- 1672

Basalt, Footscray

- 2-570

- 161

- 1-45 -

10,577

- 680

16,300

- 1048

Basalt, Kyneton

- 2-615

- 164

- 292 -

9,220

- 593

Basalt, Malmsbury

- 2-595

- 162

- 2-16 -

8,620

- 554

* Not actually broken, but on the point of breaking-.

156

Kathleen Mclnerny :

contrast with a lighter-coloured superstructure. The basecourses
of the New Arts Block at the Melbourne University are con-
structed of basalt from this quarry in blocks measuring 8 ft. 9 in.
by 5 ft. by 1 ft. 6 in. Waste pieces left after the shaping of larger
blocks are trimmed up for gutter pitchers, etc.

The large quantity available, the ease of extraction and work-
ing, and the possibility of obtaining big blocks, are all points in
favour of this stone.

Chemical Analyses.

Si0 2

70-94

7631

72-49

69-19

50-86

.U 2 0 3

1399

13 09

13-48

13-45

13 84

Fe 2 0 3

035

0-41

1-16

2 71

470

FeO

302

107

•

209

2-78

6-56

MgO

0-80

0*36

0-49

106

8-94

CaO

2-35

0'65

1-31

2-04

8-45

Na 2 0

3 94

3-00

3-38

289

2-59

k. 2 o

3 66

4-76

4-06

3 94

O' 75

h 2 o +

0-21

0-29

0-76

0-77

0-82

H 2 0 —

o-ii

Oil

0-18

0-16

0 57

Ti0 2

0-58

tr.

046

0-51

1-93

p 2 o 5

tr.

—

tr.

0T8

023

C0 2

■

0-66

tr.

007

nil

MnO

—

o-ii

0*13

014

020

Li 2 0

—

* tr.

* tr

tr.

—

)

tr.

NiO

—

001

nil

001

S0 3

—

nil

[ nil

nil

—

CoO

—

nil

tr.

BaO

tr.

Cr 2 0 3

—

005

Less 0

= S

—

•

—

UUo

001

Total

99-95

100-43

99-99

99-89

100*52

Specific Gravity

2*643

2-635

2666

2-839

1. Adamellite, Hareourt Quarry (Analyst, G. Ampt) (16).

2. Granite, Cape Woolamai (Analyst, A. G. Hall) (16).

3. Granite, Gabo Island (Analyst, J. Watson) (16).

4. Adamellite, Trawool Quarry (Analyst, A. G. Hall) (16).

5. Basalt, Eldridge's Quarry, Footscray (Analyst, A. G. Hall).

Summary.

In this paper are described fourteen Victorian igneous rocks
used as building stones. It has been found that most of these
are excellently adapted for such a purpose so far as their dura-
bility is concerned, but the long distance at which some of them
occur from a market must add to the expense of using them, as
for example the rocks from Gabo Island, Colquhoun, Orbost and
Wangaratta.

The Building Stones of Victoria .

157

Some are considered most suitable for ornamental purposes, as
they consist of large crystals included in a fine-grained ground-
mass. Such are the stones from Tallangatta and Trawooh
Another set, including the Aura dacite and the three basalts, being
very dark and sombre in colour, needs to be combined in a build-
ing with a lighter-coloured rock to give relief.

The rock so far obtained from the quarry at Orbost is found to
be too uneven in texture to be usable with success. As is pointed
-out in the report on that stone, probably a more even-textured
stone freer from “ black spots ” and sedimentary inclusions will
be found at depth in this quarry.

So far as working qualities are concerned, the basalts stand out
as those most easily worked ; next in order of ease is probably
the Wangaratta stone, followed by that from Orbost. The stone
from Harcourt is well known to be, for a granite, not difficult to
work, while the Tynong granite is reported to approximate closely
to it in ease of working.

So far as attractiveness of appearance, convenience of situation
to the capital, and amount of stone available are concerned, the
stone from Dromana seems to merit development ; while, though
its situation is somewhat remote, and the quantity perhaps limited,
the working qualities and appearance of the Wangaratta stone
should recommend its further use.

In the case of the granites from Wangaratta, Orbost, Dromana,
Colquhoun, Tynong, the dacite from Aura, the porphyry from
Tallangatta and the basalts from Footscray and Kyneton, the
petrological descriptions of these rocks are here published for the
■first time.

Bibliography.

The following is a list of the publications cited in the text as
well as others which have dealt with the use or possibilities of
Victorian igneous rocks as building stones. The arrangement is
• chronological.

1. Report on the Resources of the Colony of Victoria. Trans .

Phil Inst. Vic., iv. (2), p. 11, 1860.

.2. Knight, J. G. Australian Building Stones. London, 1864.

3. Newbery, J. C. On the Ornamental Stones of the Colony.

Trans. Roy. Soc. Vic., iv. (2), p. 79, 1869.

4. Krause, F. M. Report on Sandstones of the Grampian

Range. Geol . Surv. Vic. Prog. Rept., No. 1, p. 125,

1874.

5. Newbery, J. C. Laboratory Report. Ibid., No. 4, p. 164,

1877.

6. Lidgey, E. Report on the Malmsbury and Lauriston Gold-

field Ibid., No. 8, p. 20, 1894.

7. Rosiwal, A. Verh. Wien Geol. Reichs-Anst ., xxxii., p. 143,

1898.

• 8. Foster, H., jr. Report on Certain Clays and Felspars. Geol.

Surv. Vic. Mon. Rept. Prog., No. 3, p. 8, 1899.

158

Kathleen Mclnerny :

9. Foster, H., jr. Report on Building Stones, Pigments and
Clays. Geol. Surv. Vic. Prog. Reft., No. 11, p. 27, 1899.

10. Dale, T. Nelson. The Granites of Maine. U.S. Geol.

Surv. Bull. 313, 1907.

11. Du Toit, A. L. Pipe-Amygdaloids, Geol. Mag., n.s., [5],

iv., p. 13. 1907.

12. Baker, R. T., and N angle, J. On some Building and

Ornamental Stones of New South Wales. Journ. Roy*.
Soc. N.S.IV. , xliii., p. 190, 1909.

13. Richards, H. C. The Building Stones of Victoria, Part I.

The Sandstones. Proc. Rov. Soc. Vic., n.s., xxii. (2), p.
172, 1909.

14. Richards, Id. C. The Building Stones of St. John’s Cathe-

dral, Brisbane. Proc. Roy. Soc. Qld., xxiii. (2), p.
199, 1912.

15. Howe, J. Allen. A Geology of Building Stones. (Arnold) r

1910.

16. Summers, PI. S. On the Origin and Relationship of some

Victorian Igneous Rocks. Proc. Roy. Soc. Vic., n.s.,
xxvi. (2), p. 256, 1914.

17. Baker, R. T. Building and Ornamental Stones of Aus-

tralia. N.S.IV. Technical Education Series, No. 20,.
1915.

18. The Building Stones of the Commonwealth. Official Year-

Book of Commonwealth of Australia , ix., p. 446, 1916.

19. Richards, H. C. The Building Stones of Queensland.

Proc. Roy. Soc. Qld., xxx. (8), p. 97, 1918.

20. Sharpe, A. H. Granite at Colquhoun, near Bruthen. Rec .

Geol. Surv. Vic., iv. (4), p. 453, 1925.

21. Kenny, J. P. L. St. Elmo Granite Quarries, Castertom

Ibid., p. 453, 1925.

22. Tattam, C. M. Contact Metamorphism in the Bulla Area

and Some Factors in Differentiation of the Grano-
diorite of Bulla, Victoria. Proc. Roy. Soc. Vic., n.s.,
xxxvii. (2), p. 230, 1925.

EXPLANATION OF PLATE XVI.

Fig. 1 — Microphotograph — Altered Granite, Warby Ranges,.
Wangaratta. Ordinary light, x 26.

1. Orthoclase.

2. Hematite.

3. Hematite penetrating along cleavage crack parallel to c

(basal pinacoid).

4. Hematite penetrating along cleavage crack parallel to b

(clinopinacoid).

Proc. R.S. Victoria, 41 (2), 1929. Plate XV J.

Fig. 2.

Building Stones of Victoria.

159

Fig. 2 — Microphotograph — Basalt, Standard Quarries, Footscray,
showing texture of " cork ” or pipe amygdale. Ordinary
light, x 115.

1. Brown skeletal crystal of ilmenite.

2. Hexagonal cross sections of ilmenite.

3. Flakes or flattened crystals of ilmenite.

4. Augite.

5. Plagioclase.

6. Isotropic felspathic matrix clouded by globules of iron

oxide.

7. Anisotropic felspathic matrix clouded by globules of iron

oxide.

[Proc. Rot. Soc. Victoria, 41 (N.S.), Pt. II., 1929.]

Art. X . — Long Haw ga Haiwfall Forecasting J rum Tropical
(Darwiv) A tr Pressures.

By E. T. QUAYLE, B.A.

[Read 11th October, 1928 ; issued separately 3rd April, 1929.]

Darwin has come to occupy a position of singular importance
in world meteorology, especially with regard to its air pressure
records. These have not only proved valuable as aids to fore-
casting Indian weather, but show striking correlations with the
meteorological phenomena of many other areas, chiefly tropical.
It therefore seemed reasonable to hope that since our Southern
inland rains are mainly of tropical origin, they also would show
some relation to Darwin air pressures. This paper gives the
results of an attempt to show whether this is such as to be of use
for forecast purposes.

That tropical conditions have a large and direct control over our
Southern weather I have already shown by Bulletin 1 5, Common-
wealth Bureau of Meteorology. In this case the minimum tempera-
tures, which give some indication of the total blanketing effect
upon the earth’s surface of the humidities of the air at all levels
of the atmosphere, were used. From these it was deduced .that
even in winter vast bodies of moist tropical upper air not infre-
quently invade the continent, and that the rain production of
storm systems generally is dependent upon their being met by
these invasions. It was found, too, that the semi-permanence of
tropical conditions made possible during the winter half of the
year forecasts of rain probabilities as much as three weeks ahead,
and for this Darwin was the station mainly relied upon.

Of the data up to the present available those provided by the
surface air pressures are probably the best for tracing changes in
the general atmospheric circulation due, say, to the varying out-
put of solar heat, the interplay of ocean currents and storm
systems, etc. And any change in the distribution of pressure
over the globe must have its influence upon the development and
paths of storm systems, and so upon the rainfall of any locality.
It is in the tropical belt that such changes might be expected
to reveal themselves first.

This investigation consists mainly of comparisons between the
monthly means of air pressure at Darwin, and of the rainfall at
ten representative stations in Northern Victoria. These are Swan
Hill, Echuca, Yarrawonga, Warracknabeal, Charlton, Bendigo,
Shepparton, Dookie, Horsham and St. Arnaud.

As with the minimum temperatures, so with the air pressures
in tracing rainfall relations, it will be seen that the tropical con-
trol of our Southern inland rains is apparently limited to the

161

Long Range Rainfall.

winter half of the year. This is sufficiently well shown by the
numbers of times during the 45 years, 1884-1928, in which the
individual months show agreements between the departures from
normal of the Southern rainfalls and of the Darwin air pressures,
counting agreement when lower barometer readings go with
higher rainfall, and vice versa. Expressed in percentages of the
possible number (45), these are as follow:— January,' 55 ; Feb-
ruary, 57; March, 53; April, 45; May, 67; June, 67;’ July,
71; August, 72; September, 64; October, 73; November. 72;
December, 50.

With a view to rainfall prediction the Darwin pressure depar-
tures for each pair of months were compared with our Southern
rainfalls for the following pair. Agreements, reckoned as above,
resulted as follow: —

Darwin Pressure
Departures

January-February with

February-March „

March-April „

April-May „

May- June ,,

June-July „

July-August „

August- September „

September-October „

October-November „

November-December „

December- January „

Northern Victorian Percentage of

Rainfall Agreements

Departures p.c.

March-April 47

April-May 50

May-June 67

June-July 76

July- August 77

August- September 82

September-October . . . . 70

October-November . . . . 71

Noveinber-December . . . . 50

December- January 55

January-February 59

February-March 56

which are actually better than the synchronous monthly agree-
ments.

As the foregoing suggests, the best forecast results are got by
using the Darwin June-July air pressures to indicate the Southern
August-September rainfall This is of economic importance, the
August-September rainfall having almost a critical value in cereal
production, as well as determining the state of Spring and Sum-
mer pastures. The graph, Figure 1, in which pressure departures
are reversed, shows the remarkably consistent way in which our
August-September rainfall follows the June-July Darwin pres-
sure departures. These curves give the high correlation co-efficient
°f — *7 9d= -038. The proportionality between the extreme varia-
tions is good enough to suggest possibilities of forecasting
drought or flood conditions.

Correlation of the June-July Darwin air pressures with the
rainfalls of the three following months taken separately, gave co-
efficients of — *62 with August, — -58 with September, and — -29
with October, which confirm the advisability of taking the months
in pairs.

The following table gives the forecast relation between the suc-
cessive two-monthly Darwin air pressure means and the Southern
inland rains for the two months following: —

162

E. T. Quayle :

Fig. 1.

Long Range Rainfall.

163

Pressure Departures Rainfall over Correlation

at Darwin for Northern Victoria for Co-efflcients,

Marcb-Apnl with May-Juue' ■ — • 15± • 098

April-May „ June-July — -39±-085

May- June „ July-August — -65± -057

June-July „ August-September . — ■ 79 ± • 038

July-August „ September-October — - 52 ± 073

August-September „ October-November . — •37±-088

In Figure 2, the August-September rainfalls for Northern
Victoria (ordinate) are plotted against the Darwin air pressure
departures (abscissa) for June-July. Each unit represents for
the former one inch, for the latter ten-thousandths of an inch.

0 $
c

Northern Victoria
Aucfust + September
Rainfall Departures
l» I, I, I I I I 1

U NJ ^

C ^ M ^ h

•

■ i-

• •

‘8-

<=>--

5—

•

•••

o—

•

* • * .*

•

Fig. 2.

The proportionality between them is so well maintained that if
we take the rainfall increase as nearly three-tenths of an inch for
each one-hundredth of an inch fall in the monthly barometric
mean at Darwin, we find for the 45 years under review that fore-
casts of the amount of rain so based upon the air pressures would
have been less, or not more, than one inch in error on 35 occasions,

164 E. T. Quayle : Long Range Rainfall.

and over two inches in error on two occasions only. If it were
necessary to say only whether the rainfall would be above or below
average, the percentage of forecast accuracy would have been
82.

It is to be noted that the rainfall normals used for this paper
are based upon the 30-year period, 1885-1914. By using the
whole 45 years the principal correlation co-efficients are slightly
improved by 01, i.e., — -79 becomes — '80.

[Proc. Hoy. Soc. Victoria, 41 (N.S.), Pt. II., 1929.]

Art. XI. — Solitary Waves at the Common Boundary
of two liquids."

By FRANCES E. ALLAN, M.A.

(Communicated by J. H. Mich ell).

[Read lltli October, 1928 ; issued separately 3rd April, 1929.]

The form and the velocity of solitary, or indefinitely long, waves
in a single liquid have been examined experimentally by Scott
Russell and mathematically by Boussinesq and Rayleigh. The
much wider problem of the possible aperiodic wave forms at the
common boundary of two superposed liquids does not seem to have
received similar treatment. Those who have treated the subject
of waves of finite height at the surface of separation of two liquids
have dealt rather with the case of periodic waves, for which a
different method is suitable. (Priestly, Camb. Phil. Soc. Proc., 1910 ;
Lamb, ibid., 1922 ; Kolchine, Math . Ann., 1927-8.)

The discussion here given follows the method used by J. H.
Michell in unpublished work.

The motion is supposed two-dimensional, and will be treated as
steady by choice of an origin of coordinates moving at the rate of
the wave-form. The axis of % is taken horizontal and the axis of y
directed upwards. The independent variables are changed from
'fjto*, \p where \j/ is the stream function for the motion. This
simplifies the treatment of the conditions over the boundaries, the
coordinate ij/ being constant over each of them. The dependent
variable to be found in terms of x and \f/ is now y, for which, there-
fore, a differential equation must be found. When y is found the
form of a boundary is given in Cartesian coordinates by ascribing
the corresponding constant value to i)/.

In carrying out the process of approximation we take as the
general mathematical characteristic of the long-w r ave motion that
the variation of a quantity specifying it (in particular, the gradient
of the wave form) , in a distance equal to the depth of either liquid,
is a small fraction of the quantity itself. Thus, if we take the unit
of length as of the order of magnitude of the depth of either liquid,
the second derivative d 2 y/dx% is to be a small fraction of dyjdx, and
so for higher derivatives. The assumption is to include the small-
ness of dyjdx itself. The general discussion terminates in the
expression of the gradient dyjdx of the w r ave form in terms of y.
I have considered the conditions under which the gradient takes
the factor form appropriate to either a crested or an inverted
(trough) wave form. The expression of x in terms of y in general
involves elliptic integrals of the third kind. W here the undisturb-
ed depth of the low’er liquid is small we may find an approximate

166

Frances E. Allan:

equation involving an elliptic integral of the first kind only to deter-
mine the form of the symmetric wave. I have dealt, finally, with
ajcase of asymmetric wave (bore) where the gradient-equation for
the form can be integrated without further approximation.

The Differential Equation for y.

In terms of independent variables y, the corresponding com-
ponents of velocity are given by

u — —dif/dy,

V— dlfr/dX,

and the vorticity by

to — d^ipjdx^ -f- d^ip/dy^.

When the independent variables x , \p t are introduced we have

v =m

\9x/y const.

and

Also

when y is a function of x and ip,

9yjx const.

^ywhen ip and x are the independent variables.
9\p

\9x)

9v 9y

= ( 9v \ _fffp 9x

:)y const. \9x/'l' const. 9y ’

9ip

and 9u

/9u\ _ 9\p

\9y)x const. — ’

9\p

Therefore, as a function of x and < p,

9v 9y

9\p 9x

9ip

W ~ 9x "

■dy

9\p

9ip

Whence, substituting for u and v,

f &s y

.. _ \ Vx*

9y 9 2 y
9x 9x9\p

- _L -

'9fy 9y
9x9\p 9 x

( 9 y\ 2 )

dxjj 2 \3x) 1

1 *>y

y 9\p

(9yV

\9xp/ j

(9y) 2

^ V 9xp /

0 J

dij/ 2

Solitary Waves.

167

that is,

— CJ

(9yY = d*y(dyy_ 2 9y 9y 9*y_ f ,,( 9 yy\ 9 *y
\9\pj 9x 2 \9 1 /r/ dx 9 iff 9x9xfr^~\ ^\9x) 1 9$*

Therefore for irrotational motion, where w=0, we have

9y 9y 9 2 y
9x 9\f/ 9 x 9ip

To investigate a type of irrotational waves we must now find an
approximate solution of this equation which will satisfy also the
boundary conditions of the problem.

J. H. Michell has used this process as an alternative method of
determining the well-known results for the infinitesimal and solitary
long waves at the free upper surface of a liquid. The method
applies equally well to problems on superposed liquids, and I have
used it to find the equation to the form of the wave of finite height
and wave length as far as the terms of the sixth order in the wave
height.

The question to be considered here, however, is the form of the
long wave at the boundary between two liquids in relative motion,
the whole being confined between parallel planes at a distance h
apart.

Let y— 0, y=h be the fixed horizontal planes between which the
liquids lie. Let 1^=0 aty =0, \l/=a at y=h and i p=c at the interface
of the liquids. Finally, let p , p be the densities of the lower and
upper liquids and U, V their respective “undisturbed” velocities.

At the first step in the approximate solution of the differential
equation (1) for y, we neglect the first two terms as of the second
order and the equation then reduces to

*y=o.

9ifr ' 2

On integration, this gives, for the lower liquid.

( 2 )

y = #> (3)

where rj is a function of x. (There is no term independent of ^
since y=0 when ^=0.)

Substituting the value of y given by (3) in the second order
terms of (1), and integrating again, we find

-%!)■)=* <•>

and putting y=rj\j/ in the second order terms of (4), we obtain

Using the result

168

Frances E. Allan :

and its consequence

vx!/== ^y + U y ^y 2y & y }

9xj/ ~ \ 9% 2 J 9x J

we can write (5) in the form

J ^ 9x 2 y 9v}) VP

(6)

For the upper liquid, when we integrate the equation 9 2 yj9 ip 2 =0‘
we get

y-h — nty-a), (7)

since y=h when x// =a.

Following the same steps as in the case of the lower liquid we
get the equation

- *) (g y } = (t - “)% <»>

Since the pressure must be continuous across the interface, we
deduce from Bernoulli's pressure equation the result

^_^2 = (A _ 2gy)(p _ A (9)

for points on the interface, where q and q' are the velocities in the
lower and upper liquids respectively at the point considered, and
A is some constant.

But

and ip — c, at the interface, so from (6) we find that at the interface

;{>+*>

gv d _*y

dx 2

and in a similar way we find

*(£)*}■

a f2 _ {C^JlY

H (y - h) 2

Hence (9) becomes

pc 2 _ p'(c - a) 2 , 2 fpc 2 _ p'(c -a) 2 \ 9 2 y
y 2 ( y-h ) 2 s \ y ( y - h) ] 9% 2

( 10 )

(11)

Solitary Waves.

169 '

_ i jpc 2 P '{c-a) 2 \ (9y\ 2

(y-h) 2 )\9x) -( A ~ 2 gy)(p-P')> ( 12 )
and this is the differential equation for the, form of the interface.
We may write it

pc 2 _ p(c - a) 2 i fpc 2 _p , (c-a) 2 ) d (dy\*
y 2 ( y-h ) 2 3 t y y-h )dy\dx /

i (pc 2 p(c - a) 2 \ (dv\ 2

-ny-XT^WfKdx) = ( a -2^Kp-a m

that is,

(14)

Integrating this we obtain

(15>

where D is a constant of integration.

Thus

(dy\ 2 _ -<! D + By - 3g(p— p)y 2 y y(y—h)+3 P c 2 (y — h) -3 p '(c—a) 2 y
'd%/ po 2 (y - h)— P {c - a) 2 y

(16)

where 3A (p — p')=B.

This is the expression found by J. H. Michell for the gradient.
We now assume that this expression will factorize in such a manner
as to give the desired wave form, and then consider the further
conditions which will make such a form possible. That is, we
suppose

(dy\ 2 _ -Sg{p-p)(y-k)*(y-k 1 )(y-k i )

\dx) pc 2 (y -h) — P '(c - a) 2 y ' '

This makes dyfdx — 0 and d 2 y/dx% = 0 when y = k; and dy/dx= 0
when y = k v and when y=k 2 .

Thus with this form the condition that the surface may be
horizontal wheny=&, is satisfied.

Now for (16) to be equivalent to (17) we must have, by equating
coefficients of y,

< 18 >

3 g{p-p’)i %kk 1 k 2 +k 2 (k 1 -{-k t ) = -m+z\ P c 2 -p(c-a) 2 y, ..(i9>

-Zg{p-p')\ k 2 + '2k{k 1 +k i )+k l k 2 [-=D-BA (20)

s g(p-p)i 2k+k t +k 2 [ =B + 3 gh(p-p'), (21)

and from these equations (18)-(21) we deduce : —

k +k p(c-a) 2

1 8 "^(rp) “ (h-k) 2 g( P -p') + h ’

( 22 ),

170

Frances E. Allan :

and k 1 k 2 =

pc 2 h

k 2 g(p- P y

so that k v k 2 are the roots of the equation
pc 2 p'(c-a ) 2

( 23 )

a 2 -j

k 2 g( P - p) [h - k) *g(p - p) > ^k 2 g ( P - p)

pc 2 h

7\=0* -(24)

We therefore have

2 k 1= k + - , f 2 -- P'( c - a ) 2 _

k 2 g(p-p ) ( h—kfg{p-p )

- v/lu P c2 _ p'(c-a) 2 | 2 _ 4pc%

k 2 g( p -p) (h—k) 2 g(p - p’)J k 2 g(p-p)’

2k,=h + - . Pfc-a) 2 . _

k 2 g{p~p) (h-k) 2 g(p- p ’)

j _ , / a._ F ^ 2 _ p'(c-a ) 2 \ 2 _ 4 p C 2 h

Vi ( h-k) 2 g(p- P y k 2 g(p- P y

We may write

^=U 2 and
Rr

(c ci) 2 _ y 2

(h — k) 2

since U is the velocity at infinity of the undisturbed lower liquid
ofjdepth k, and V is the velocity at infinity of the undisturbed
upper liquid of depth (h-k).

If we also write p = Xp and V 2 =pXJ 2 we have : —

2 k i — h -f-

2k 2 =h +

U 2

S(l-A)

U 2

4AU 2
g( 1-A)

(I -A)

Necessary Conditions for such a Wave.

We have put the equation for the gradient into the form

( dy\ 2 %( 1 - *)(y - k) 2 (y - k ± )(y ~ k 2 )

\lx/ ~ \J\k 2 h- \ k 2 ~\p{h-k) 2 yy] ‘

Now the denominator may be written k 2 (h - y ) -f A^u, (h - k) 2 y and
y is less than h at all points on the interface. Therefore the denomi-
nator is always positive. Hence, assuming Xel (i.e., pep), we
must have y-k ± and y - k 2 of the same sign, to make dyjdx real.

But y lies between k and either kj or k 2 , since k, k lt and k 2 are
the turning values of y. Therefore, either
(i) keyekxeki*
or (n )k>y>ki>k v
These alternatives represent
(i) a crested wave,
or (ii) an inverted wave.

Solitary Waves.

171

There is no wave for a value of k between k x and k 2 .

Thus, for values of k between 0 and k * there is a crested wave,
and for values of k between k 2 and h there is an inverted wave.

Lamb has treated the infinitesimal wave at the interface between
two liquids (see Lamb's Hydrodynamics, Arts. 231-234), and if
in Lamb's result we make the wave length tend to infinity, we
find, as we should expect, that the two heights at which infinitesimal
long waves are possible are k x and k 2 .

Now, since k Y will be the height of the crest when a crested wave
exists and k 2 will be the depth of the lower liquid at the trough in
the case of an inverted wave, it will be necessary for k x and k %
to be real if there is to be a wave form at all. Therefore, referring
to the equation (24), we deduce the condition

( TJ 2 (I-V.)) 2 4hU 2

g{ i-A) 1 <^TTa)-

Approximation-Method for High Waves.
If we take k very small we find approximately

_ r y(y — %k)(y — k 1 )(y - k 2 )

= C(y-2k)(y-kJ(y-k&

for values of y near the crest, where C is a known constant,
makes

{y -Vk){y - k t )(y -k 2 )

This

Hence we can find an approximate form for the wave in terms
of an elliptic integral when the wave is near its greatest^height.

The Asymmetric Long Wave.

There is, however, a type of long wave whose form can be deter-
mined from the differential equation without further approximation.
This is the wave which we get on putting k x =k 2 . Its differential
equation is

- U \k 2 h- k 2 -^(h-k ) 2

and therefore wheny=&, or y — k^ dy/dx= 0, and d*y/dx 2 = 0.

This means that the wave has no crests but rises gradually,
through an infinite horizontal distance from y=k to y=k 1 . The
motion is here of the nature of a “bore."

Since k x and are the roots of equation (24), the condition that
k x should be equal to k% is

S p c 2 p'(c-fl) 2 , 7 „! 2 V 2 *

l h*g(p - p) - [h - k)*g(p - p') +n $ ~g(p - p')h 21

( 25 )

172 Frances E. Allan :

that is,

( U 2 V 2 A 4U%

~g(l-V‘ (26)

This gives

2 (U 2 — XV 2 ) 2

=° < 27 >

and the roots of this are always real, since (U 2 +AV 2 ) 2 >(U 2 — AV 2 ) 2 .
This means that for any given pair of values of the velocities U, V,
of the currents, there are two possible values of h, the distance
apart of the horizontal boundaries ; they are given by

; >=^rZT)( U ± /AV) 2 (28)

If we regard equation (26) as an equation for V in terms of h and U
we find

| v I = U±y ^ (I - X ) (29)

v A

When condition (26) is satisfied, we have, from (22)

r U 2 V 2 A ■>

k x -k. 2 -j\h +^, _ — g(i — X)J > ( 30 )

and on substituting for V from equation (29) we deduce

: ' <31)

The positive sign with the root in (29) would give

1 V gh{ I -A)

and we consider, therefore, only the negative sign. That is, we take
AV 2 - U a 1 1 ~ j 2

= u *(rr 1 ) 8

Now, returning to the equation for ( dyjdx) 2 , these results give
(dvidx ) 2 = 1 ( y-ky^-y) 2

i yi**) u* jpn ^ k 2 - (h—k) 2 (i -h/kj) 2 Yy]

_ 3g(l-A)V (y-mh-y) 2

U 2 h \k 2 k 2 - \ k x 2 -(h-k) 2 (h-k i) 2 'j-y/h]

= 3 (y-knk x -y) 2 .

[k 2 k x 2 ~ -J k 2 k 2 -{h-k) 2 (h~k i) 2 }- y/h]

Solitary Waves .

173

Three cases now arise, depending on whether

(i) kk x = (h—k)(h— k x ),

(ii) kk x 7> \h — k) (h - kj,

(iii) kk^yi — fytfi — k i).

We shall now consider these separately.

(i) Here we have kk x = (h —k)(h — k x )
and, therefore, k-\-k t -h.

This means that the highest and lowest levels of the wave are
equidistant from the mean height hj 2 of the liquids.

In this case

( W 3(y-k 2 )(k 1 -y 2 )

\dx) ~ k 2 ki 2

and therefore

dy_ _ Vs

dx kk-y

(y-k^h-y)

where y lies between & and

On integrating, if we choose the origin so that the constant of
integration is zero, we find

kk x x

k+ky

k x — k

artanh

k-\-k x

y—r~

— h

Now changing over to a horizontal axis along the mean level, so
we find the equation to the wave form is
2 /

that y'=y 2

w '2 3 *- artanh v^

that is, y’ =a tanh mx,

where a

_ k x k

_^3/l 1

\k v

(ii) Consider now the case — ^(h — k^.

Here

3(^-^) 2 (^i-y) a

w =[^v- ^ k^~(h-knh-k ,) 2 }•>»/*]

a^y-kYik.-yY

P-y

174

Frances E. A Ran :

where

Now, since kk^ty—k^h—ki),

therefore h<.k+k v

and therefore h — k^k-^ and h — k^k.

k*k 2 h

But p -tfkf-th-kYih-ktf
h

{h-mh-k.Y
1 _ k*k*

= = — 7x > where O<0<1.

1 — u

Therefore

>k.

Now — — 1 ^P 2 —y therefore

dy * (y-fy^-y)

(Al _£ )a * = /V^ ^ ^f^Zldy

which is the equation to the wave form in this second case,
(iii) If kk±*<(h — k)(h—k^),

where

(h~-k)*(h-kd*-k*kf

and

Therefore

dx__ f Vfi 2 + y

dy~* (y-k)ik x -yY

Solitary Waves.

175

Therefore

(k 1 -k)ax = 2yk 1 +p 2 ai-tarrh + ^ -2 JY+FP arcoth ^+^ a .

The present paper embodies a small portion of the work done
under the terms of a Research Grant for Mathematics in the
University of Melbourne. The publication of the rest of the work
has been postponed owing to the author’s departure for England.

[Proc. Rot. Soc. Victoria, 41 (N.S.), Pt. II., 1929.]

Art. XII. — The Geology and Ralaeontography of the Cathedral
Range and the Blue Hills , in North- Western Gippsland.

By EDWIN S. HILLS, B.Sc.

(Howitt Research Scholar in Geology iD the University of Melbourne.)

(With Plates XVII., XVIII.)

[Read 13th December, 1928; issued separately 3rd April, 1929.]

Summary.

1. The Cathedral Beds, formerly believed to be of Upper
Palaeozoic age, are demonstrated to be of pre- Upper Devo-
nian, and probably of Upper Silurian age.

2. The western edge of the large area of igneous rocks which
contains Mt. Torbreck and the Cerberean Ranges, is shown to
consist of Upper Devonian rocks, mainly rhyolites, and these
are shown to occur to an unknown extent to the east.

3. A newly discovered fish fauna from the Upper Devonian is
described.

4. Silurian fossils from a new locality are listed, and a new
stelleroid noticed.

Introduction.

Location.

The area studied is a roughly rectangular block of country
stretching, along its western boundary, from Buxton to Taggerty,
a distance of seven miles. It extends eastwards for five and a
half miles from Taggerty, and three and a half miles from
Buxton, and consists mainly of the southern two-thirds of the
Parish of Taggerty, in the County of Anglesea.

The location was suggested in December, 1927, by Professor
Skeats, as likely to reveal the relations between the Cathedral
Beds (supposed Upper Devonian or Lower Carboniferous), and
the dacites (supposed Lower Devonian).

The latter are, however, not present in the area, and the Cathe-
dral Beds appear to be Silurian in age, while the supposed dacites
are Upper Devonian rhyolites.

Previous Workers.

The amount of previous work done in this area is extremely
small. In 1899 W. H. Ferguson reported as follows after a very
rapid examination : “The rocks are layers of coarse and moderately

177

Cathedral Range and the Blue Hills .

coarse-grained sandstones, with some shales . . . the sandstones
, . . contain a few well waterworn pebbles. . . . Ripple markings
and filled-in desiccation cracks were noted. In some of the
coarse-grained sandstones spots of very fine sedimentary rock
were seen. ... No fossils were found in the sandstones of the
Cathedral Mountain, which is three or four miles in length. The
rocks near this mountain are Upper Silurian porphyries, slates and
sandstones. No clear contact could he found showing the rela-
tion of the Silurian rocks to the Cathedral sandstones, but the
general lithological inference is that the Cathedral rocks are
younger than the Silurian, and probably of Upper Palaeozoic age,
Upper Devonian or Lower Carboniferous. On the east the sand-
stones join the fragmental porphyry formation at the Little River.
These rocks, in other parts of the colony, have been referred to
Upper Palaeozoic age.” (Ferguson, 1899).

On this evidence, the Cathedral sandstones were coloured as
‘Upper Palaeozoic on the Geological Map of Victoria, an inch
to 8 miles, published later by the Survey.

Professor J. W. Gregory probably visited the area, and he pub-
lished his idea of the structure in his handbook on the Geography
of Victoria (Gregory, 1912, p. 72), showing the Cathedral sand-
stones as older than the volcanic rocks, which he believed to be
dacites, and thought to be of Tertiary age. A rock section [400]
at the Geology School is labelled “ Dacite: Between Chapel Hill
and the Cerbereans,” and was probably collected by him. It is
not, however, a dacite, but a section of the main rhyolite flow.

Professor Skeats (1910) writes, following Gregory: — ■“ The
map indicates a great mass of granitic rocks extending from near
Narbethong. through Marysville, north-east to Mount Torbrech,
whereas over most of this area the rock is certainly dacite.” 1 he
granitic rock at Buxton is coloured as such by the Survey. r I he
only other relevant record of geological work the author could
find is a report by Dunn (1907) on gold and tin workings just
outside the area here considered, and dealing with Silurian sedi-
ments only.

Physical Features.

The area is one of strong, even precipitous, relief. The Cathe-
dral Range runs in a general SSE. direction from the Cathedral
to the Sugarloaf (see map), and is a composite hogback grading
to a razorback, with the dip slopes (often bare sandstone) to the
east, and the escarpment on the west. There the slopes are
covered with thick talus, and in profile this presents a curve of
great beauty.

The crest of the range is determined by the outcrops of two beds
of hard sandstone, extending the whole seven miles, and separated
by soft shales. At the northern end the highest ridge, with the
Cathedral Mount, is formed by the lower standstone, while in the
south the highest ridge, with the Sugarloaf, is formed by the

178

Edwin (S'. Hills :

upper sandstone dipping easterly at 65°, and constituting a razor-
back. The trough-shaped subsequent valley between the sandstones
is called the Tableland. It is well grassed, and is used as a sheep
run.

In the valley west of the Cathedral Range the Acheron River
flows northwards to the Goulburn. while the valley east of that
Range is occupied by the Little River, flowing north-north-west
and later west to join the Acheron at Taggerty. The western
boundary of the area studied is marked by the Crestline of the
Blue Hills, a range higher than the Cathedral Range, running
generally NNE.-SSW. The slopes of this range are covered
with rounded rhyolite boulders in the upper parts, but become
clearer on the Silurian lower down. Chapel Hill lies in the
traingle between the Little River and the Blue Hills. Immediately
to the east of Buxton a resistant granodiorite-porphyrite has de-
termined the formation of a steep and, especially at its southern
end, high ridge. The river flats are swampy.

Nature of the Present Survey.

The combination of steep bouldery slopes, sometimes precipi-
tous, dense growths of bracken, and in the gullies, tangled fern
flora, made the work of surveying without a companion both
arduous and difficult. The boundary of the Upper Devonian north-
east of the Little River, the Buxton granodiorite, the Little River
alluvium (where shown on the map), and a small but important
area east of the Sugarloaf were traversed by compass, distances
being paced. The Cathedral Beds were mapped approximately
by radiations from fixed points on the main road, in the absence
of a reliable method of measuring distances up steep slopes, and
the talus from the Cathedral Range was marked in from sketches
of the range. The outcrops on Chapel Hill, the talus from the
Blue Hills, and the boundary of the Silurian to the east of
Buxton, were sketched in from a knowledge of the country from
memory, with the help of notes and sketches. 1 he Lands Depart-
ment’s parish plans on a scale of 2 inches to the mile were used as
the basis for all work.

Geology.

Yeringian (Upper Silurian).

1. Typical Silurian.

Shallow marine Silurian mudstones, sandstones and shales occur
both in the north and the south. Fossils were found near the foot
of the Blue Hills, in red sandstones showing miniature current
bedding due to ripple-mark. The fossils are for the most part
fragmental, and occur in small groups, suggesting re-sorting along
a beach or on a sandbar. The following forms have been identi-
fied (the last four by Mr. Chapman) : —

Cathedral Range and the Blue Hills .

] 79

Anthozoa. Cladopora sp.

Asterozoa. Taeniaster ( ?) , sp. jiov. aff. spinosus Billings.

Brachiopoda. Orthis ( Dalmanella ) testudinaria Dalman.
Chonetes sp.
cf. Coelospira sp.

Pteropoda. Coleolus cf. aciculmn J. Hall.

Cephalopoda. Orthoceras sp.

The assemblage, especially Cladopora and Dalmanella , which
.are characteristic, indicates that the beds are Yeringian (Upper
Silurian) in age. Indeterminate fragments of plants were found
in several places along the Little River east of Taggerty. No
fossils were found in the south near Buxton, but the Silurian
rocks are apparently continuous on the west side of the Acheron
from Taggerty to Buxton, as is shown on the Geological Survey’s
map, and in part confirmed by Dunn (1907). The beds are litho-
logically quite similar in the north and the south.

In the north the fossiliferous beds are overlain unconform-
ably by the Upper Devonian, and are faulted against the Cathedral
Beds along an east-west fault line. This fault is pre-Upper
Devonian, beds of that age not being displaced by it. In the south
the relations with the Cathedral Beds are different, the two
apparently forming a conformable series. The Buxton granodi-
orite-porphyrite ( ? Lower Devonian) is intruded, perhaps as a
sill, into the typical Silurian sediments, which are only slightly
metamorphosed. Quartz veins, in part auriferous, are associated
with the granodiorite.

On the north side of the Little River, the dip is 58° to the south-
south-west, and the same dip and strike were found in a small
creek just to the north of the fault, on the northern slopes of the
Cathedral Range. No other outcrops were found where a definite
dip could be obtained (see, however, p. 180).

2. The Cathedral Beds.

The rocks — hard sandstones, soft sandstones, and shales —
which constitute the Cathedral Range and Chapel Hill, form a
triangular outcrop between the Blue Hills and the Acheron
Valley, ending abruptly in the north against a fault line. On the
evidence before cited (see p. 177), they were previously regarded
as of Upper Palaeozoic age.

Field Relations.

Ferguson could find . . no clear contact . . . showing the
relation of the Silurian rocks to the Cathedral sandstones.” Fur-
thermore he says, “ On the east the sandstones join the frag-
mental porphyry formation at the Little River.”

(a) At their northern termination the Cathedral Beds are
brought against the typical Silurian rocks by an east-west fault.
Huge monoliths of crushed and shattered sandstone occur along

180

Edwin S. Hills:

the fault line, and the strikes are locally contorted. The evidence
of faulting may be traced a short distance to the east before
being covered by talus, and the further extension of the fault to
the east is inferred from the abrupt termination of the Cathedral
Beds along the east-west line. Evidence as to the hade is indefi-
nite. but the general dip of the Cathedral Beds (especially on
Chapel Hill away from the fault line may indicate that the down-
throw is to the south. No indication of displacement of the
Upper Devonian beds by this fault was seen.

( b ) East of Buxton, the relations are different. Traversing
eastwards from Buxton, we notice that

1 . The typical Silurian beds are intruded by the granodiorite-
porphyrite.

2. The Silurian can be traced without apparent break to the
foot of the Blue Hills, and there the dip and strike are the same
as those of the Cathedral sandstones which occur close by, and
have the same relations to the Upper Devonian as does the typical
Silurian. The following observations are significant: —

(a) 1 ^ miles W.20°N. from Buxton, the Silurian beds strike
N*32°\V. dip 40° westerly (Dunn, 1907). Compare

(a) About a mile northwards along the Cathedral Range from
the Sugarloaf, Cathedral Beds strike N.25°W. dip easterly at
60°.

( b ) 3 miles E.10°S. of Buxton, the Silurian beds strike
N.30°W. dip easterly at 70° (Dunn, 1907). Compare

(b r ) Southern end of the Cathedral Range, Cathedral Beds strike
N.35°W, dip 65° easterly. Furthermore, the Cathedral Range
shows a marked parallelism with the trend lines in the Silurian
in the southern part of the area, where no faulting occurs.

3. Just east of the Sugarloaf, in a small creek, the relations
between the Cathedral beds and the Upper Devonian rhyolites can
be clearly seen. The latter, striking N.25°E., and dipping at 30°
easterly, overlie the Cathedral Beds, striking N.35°W., dipping
65° easterly, with a strong unconformity. The basal conglome-
rate. though developed further north, is absent here; but that we
are dealing with the base of the rhyolite is shown by the well-
developed prismatic jointing, which disappears higher up in the
flow, the numerous linearly arranged sandstone inclusions which
lessen in number and lose their linear arrangement higher up, and
the slightly vesicular nature of the flow, which elsewhere is
devoid of vesicles. Further, there is no contortion of strikes or
brecciation in the Cathedral sandstones, such as are found where
these beds are faulted in the north. Similar relations hold every-
where at the edge of the Upper Devonian series, being especially
well shown south-east of Chapel Hill, where the characteristic
oscillation ripple-marked sandstones are overlain by the Upper
Devonian basal conglomerate.

The pre-Upper Devonian age of the Cathedral Beds is thus
established, and their conformability with the Silurian strongly

Cathedral Range and the Blue Hills.

181

indicated. In the absence of palaeontological evidence as to their
more precise age, the author has placed them high in the Upper
Silurian series.

Lithology.

Ferguson's original description is in the main correct, though,,
even after extensive search, the present author could find none of
the “ well-waterworn pebbles ” mentioned by him, the largest par-
ticles found being only a few millimetres in diameter. The sand-
stones are very thick-bedded, and where not cross-bedded, are
even-grained. The jointing is somewhat irregular, curved cracks
often forming in the thick, even beds. When not strongly
weathered the sandstones are red, but leaching of the outer few
millimetres of weathered blocks produces a white skin. Ripple-
markings. mud-cracks, current-bedding, and hardened clay pellets
were found.

Along the Cathedral Range, subaqueous current ripple-marks
occur, rarely. On Chapel Hill, however, ripple-marks of sym-
metrical shape, and formed by wave action (oscillation ripple-
marks), occur in profusion in fine sandstones of lighter colour
than the Cathedral sandstones. At some horizons the successive
ripple-marked layers are separated by only fractions of an inch,
and the lengths of the ripples from crest to crest (wave-length)
remain fairly constant, averaging a little over an inch. The
direction of propagation of the water waves which produced the
ripples remained fairly constant, also. Examples were found of
superimposition of the ripples in parallel and slightly inclined
directions. No current-ripples (asymmetric) were found on
Chapel Hill.

Mud-cracks are rare, but were found at the Sugarloaf, the
north end of the Cathedral Range, and on Chapel Hill. The former
two examples are in very fine, indurated sediment, a thin parting
between heavy sandstones. The polygons are small, about six
or eight inches across. The latter example (Chapel Hill) is in a
thick sandy mud bed, which is cracked to a depth of about eigh-
teen inches, the polygons ranging up to fifteen inches across. The
edges are here turned down conspicuously.

Sandstones showing current-bedding were found at the Sugar-
loaf, but nowhere else. The current-bedded layers are usually
bounded by plane surfaces, though a very few show curved
boundaries. The slopes are fairly steep, especially in the thicker
beds, whose thickness is up to three feet between the bounding
planes.

Inclusions of very fine sediments similar to the material of the
mud-cracked layers on the Cathedral Range are common there,
but are absent in the oscillation ripple-marked sandstones of
Chapel Hill. They probably represent clay pellets rolled into the
sands by current action.

The interbedded soft sandstones and shales are grass-covered,
and the only outcrops are blocks in the roots of fallen trees. The

182

Edivin S. Hills :

bed separating the two sandstones of the Cathedral Range is a
soft chocolate sandstone, and differs from the more yellowish
shales of Chapel Hill. No fossils were found in any of the
Cathedral Beds.

The Cathedral Beds differ from the Upper Palaeozoic rocks
with which they were formerly included, in the absence of the
conglomerates which are so characteristic of the latter, as de-
scribed by Howitt (1876), Murray (1877), Kitson (1899), and
Teale (1920), the absence of interbedded lavas, and the absence
of fossils. They also dip at a much greater angle than the Upper
Palaeozoic rocks, except where these are faulted. The author
considers that much more detailed work than was possible in the
time available will be necessary to elucidate the very interesting
question of the conditions under which the Cathedral Beds were
deposited. He would point out, however, that —

1. The Chapel Hill beds resemble somewhat the Berea sand-
stones of North America in the profuse development of the
oscillation type of ripple-mark, with a generally constant direc-
tion of propagation, the absence of current ripples, and the lack
of fossils (Hyde, 1919).

2. The Chapel Hill and the Cathedral beds were not deposited
under exactly the same conditions, as is shown by their different
physical characteristics (see above).

3. In the Grampians sandstones, a thick series in which no
fossils were found for many years, the few remains now known
indicate a marine origin for the beds in which they occur (Chap-
man, 1917).

4. The absence of cut-and-fill structures and coarse materials,
and the scarcity of muds and sun-cracked layers, as well as the
extent and thickness of the beds, are against either a flood-plain
or sub-aerial deltaic origin.

(?) Lower Devonian.

Intruded into the Silurian sediments at Buxton and to the east
of Taggerty is a granodiorite-porphyrite, which is placed in the
Lower Devonian by analogy with the other Victorian granodio-
rites, which are supposed to be of this age. Such granodiorites
occur near Marysville, to the south of the occurrence at Buxton.
The latter is much more extensive than the outcrop at Taggerty,
being a long and narrow outcrop traced for two miles in a NNW.
direction, and continuing still further to the southwards. The
obvious relation of its outcrop to the trend of the Silurian near
by, and its long, narrow outcrop, have led the author to believe
that it may be a sill. The amount of metamorphism of the adja-
cent sediments is but slight, being merely an induration, with
some si licifi cation and development of muscovite. Quartz veins
are associated with this intrusion, and these have proved
auriferous to the south of the area considered.

The occurrences at Taggerty are very small, perhaps offshoots
from a larger concealed mass. Of note is the occurrence in these

Cathedral Range and the Blue Hills. 183

•rocks of numerous pink garnets, sometimes showing dodecahe-
dral outlines but more often rounded by reaction with the magma.
Cognate (basic) and foreign xenoMths also occur.

Upper Devonian.

Beds of this age were discovered, forming the top of the Blue
Hills and the ranges east of the Sugarloaf, and formerly believed
to be Lower Devonian dacites. The main development is a series
of acid lavas (rhyolites) of undetermined thickness, which out-
crop along the whole of the eastern boundary of the area studied.
Basalts and sediments also occur. There are numerous flows, the
lower ones being interbedded with fossiliferous lacustrine and
fluviatile deposits of limited areal extent; but the main rhyolite
is recognisable continuously for over seven miles along its western
edge, and the further extent to the south was not determined.
Through the courtesy of the Director of the Victorian Geological
Survey, Mr. Baragwanath, and his officers, the author had the
privilege of examining some specimens collected by Mr. O. A.
L. Whitelaw from the Mount Torbreck region, some nine miles
to the east of the Blue Hills. The specimens proved to be of
extreme interest, as they revealed the presence of exactly similar
rhyolites, both in hand specimen and under the microscope, in that
area, associated with typical dacites (as developed near Iieales-
ville, in the Dandenong Ranges, and at Mount Macedon) of sup-
posed Lower Devonian age. A porphyritic granitic rock, very
similar to that found at Buxton, is also in close proximity. The
problems of the extent of the rhyolites and their relations to the
granitic rocks are thus opened up.

A generalized sequence of the Upper Devonian is as follows

6. Rhyolite a (main flow, very thick).

5. Rhyolite ft (small flow, in the north).

4. Basalt (in the north and south only).

3. Sediments (sandstones and shales in the north, bedded
tuffs and volcanic breccias in the south).

2. Basal rhyolite.

L Basal conglomerate (in the north).

The absence of the basal beds in places is due to their development
only in the valleys of the Upper Devonian land surface.

They are overlapped by the main rhyolite flow, and where this
rests on the Silurian directly, the upper Devonian land was high,
so that the Sugarloaf was probably an upstanding peak then, as
now (see Section E.-F., Fig. 1).

1. The Basal Conglomerate .

This extends uninterruptedly for nearly four miles along the
Blue Hills, east of Chapel Hill. It is rather variable in thickness,
reaching a maximum of about ten feet in the central part, and
Tthinning out towards the north and south, where it is absent. The

184

Edwin S. Hills

Fig. 1 . —Diagrammatic section along the crest of the Blue Hills, parallel to the edge of the Upper Devonian outcrop.

Cathedral Range and the Blue Hills . 185*

coarseness varies in a like manner, the boulders in the centre being
larger than elsewhere, reaching a foot or more in diameter. Some
pebbles are well rounded, others flat and' with the edges rounded
off. They consist entirely of coarse sandstones, probably derived
from the Cathedral Beds, and resemble fluviatile (valley-plain),
deposits.

2. The Basal Rhyolite.

Locally, rhyolites of relatively small extent and thickness over-
lie the basal conglomerate, or rest directly on the Silurian where
this is absent. These rhyolites generally show prismatic and
platey jointing, and are lighter coloured and more weathered than
the main flow. They are also more calcic, containing less free
quartz and more plagioclase felspar (see p.192).

3. The Sediments.

The basal flows of rhyolite apparently blocked the Upper
Devonian streams, so that lakes were formed in which yellow and
red sandstones, and red, blue and green shales were deposited.
These sediments have yielded a fish fauna and remains of plants.
The following forms occur: —

Ostracodermi. Bothriolepis gippslandiensis ; sp. nov.

Dipnoi. Eoctenodus micro soma, gen. et sp. nov.

Holonema cf. rugosum Newberry, 1889..
This assemblage is a typically Upper Devonian one (see p. 198).

Numerous round impressions in the fossiliferous blue shales*
probably represent bubble impressions, the gas coming from the
decomposition of the inclosed organisms (Twenhofel, 1921,
1926, p. 289). Round the organic remains, especially the plants,,
and along joint planes, the blue shales are buff in colour. The
jointing, especially in the sandstones, is much less regular than:
in the Silurian beds which underlie the Upper Devonian.

East of Buxton, bedded tuffs and volcanic breccias overlie
the basal rhyolite, and underlie the basalt and main rhyolite. T he
tuffs are of two kinds, one a blue-black, well- jointed, compact
rock, grading with increased size of its component particles into
the volcanic breccias, and the other a light, chocolate, poorly-
jointed, fine-grained rock. Both are products of the basic igneous
activity.

The volcanic breccias compare almost exactly in hand specimen
and microscopically with Teale’s “ basal beds ” from the Mt.
Wellington area, labelled by him “ basal breccia.” In that area
the relations of these beds were not clear, but at Buxton they are
evidently part of the basal beds of the Upper Devonian series, as
thought by Teale to be the case. Above them comes the basalt
to which they are undoubtedly related.

4. Basalt (pp. 191).

Very dense, fine-grained, hard blue-grey basalts occur fairly
constantly above the sediments and below the main rhyolite-.

186

Edwin S. II ills .*

They are occasionally absent, as at the Sugarloaf and in the
•extreme north. At their farthest northerly occurrence they rest
directly on the basal conglomerate, without any intervening sedi-
ments. There they are very amygdaloidal at the base, the amvg-
dales being filled with chlorite and chalcedony, and there are
small amygdales in most of the basalt sectioned from higher up
in the flow. In other parts of Victoria the basalts are interbedded
with the rhyolites, and always occur “ consistently higher in the
series than the Wellington rhyolites ” (Teale, 1920). Although
the name “ melaphyre ” has usually been applied to similar rocks
by workers in other areas, the author feels that in view of their
relatively fresh state, the term basalt is most suitable in the
present case.

5. Rhyolite ft (pp. 190).

In the north, a small flow of a black aphanitic rock with small,
clear quartz crystals occurs above the basalt and beneath the main
rhyolite. It is, again, less acid than the main flow.

6. Rhyolite a (pp. 188).

The most constant of the Upper Devonian rocks is the main
rhyolite. Its thickness must be very great, though this was not
determined, only the lower edge of the flow being examined. Its
coarseness of grain led it to be identified as granite by the early
surveyors.

The problem of the separation and distribution of the Upper
Devonian and Lower Carboniferous series in South-Eastern
Australia has long been a vexed one, and the discovery of the fish
remains at Taggerty serves to give some definition to the data.
Not only as regards structure and field relations, but also both
macroscopically and microscopically, the rhyolites, “ melaphyres/ 4
tuffs, conglomerates, shales and sandstones in the Cathedral dis-
trict compare with similar rocks in the Upper Palaeozoic belt in
Eastern Victoria, and the growing ideas as to their Upper Devo-
nian age are strengthened. It should be noticed that the Taggerty
fishes occur in beds stratigraphically beneath the main rhyolite
flow, while many of the Lepidodendron sandstones in the great
Upper Palaeozoic belt occur well above this flow, as at Mt. Well-
ington.

Pleistocene ( ?) to Recent.

Deposits of greater age than the alluvium of the river valley-
plains are represented in the alluvial fan from the Blue Hills, east
of Taggerty, and the numerous alluvial cones and piedmont
alluvial plain on the western slopes of the Cathedral Range. The
creek which enters the Little River after flowing west past
Andrews’ house, aided by smaller streams, both tributaries of it-
self and of the Little River, has built an extensive alluvial fan.
Owing to the removal of the lower edge of this by the Little

Cathedral Range and the Blue Hills.

187

River, these streams are all now incised into the fan. It is com-
posed of boulders, grading downwards in size to the individual
quartz grains weathered from the rhyolite's. The water-table is
everywhere near the surface, as is shown by the numerous seep-
ages which occur, both on its surface and more especially along
its lower edge, where it is terraced by the Little River. The soil
is rich, and the abundance of water and sheltered position make it
valuable as farm land.

The talus produced by the rapid erosion of the Cathedral Range
has formed extensive and very thick deposits between that range
and the Acheron River. The numerous small wet weather
streams which flow westwards down the escarpment face have
each built a steep alluvial cone, and the coalescence of these has
produced a very thick, continuous sheet of sediments. The
streams which produced the sheet are now incised into it, because
of the terracing of the lower edge by the Acheron. These deposits
began to accumulate when the Cathedral Beds were exposed, in
the process of dissection of the peneplain which was developed
over Victoria in early Tertiary time. The date of the deforma-
tion of this peneplain is uncertain, but it is very probable that
some of the above described deposits are at least as old as the
Pleistocene.

The Acheron and Little Rivers and many of the smaller
streams have alluvial flats, composed mainly of coarse boulders
with interstitial fine silts. These flats are often swampy, especi-
ally in the upper reaches of the Little River. There, also, perfect
small examples of alluvial cones are developing at the mouths of
small streams from Chapel Hill. These cones have developed on
top of the river alluvium since the last big flood, as they are
unmodified by the river, the streams which formed them wander-
ing indiscriminately over their surfaces.

Summary.

After the deposition of the Upper Silurian beds in shallow epi-
continental seas (and perhaps other environments in part), they
were compressed in Siluro-Devonian or Lower Devonian times,
into a series of large folds. Probably in the Lower Devonian, asso-
ciated with the final earth movements, a large sill or dyke of grano-
diorite was intruded in the south, and some smaller apophyses in
the north. Before the Upper Devonian, normal faulting occurred.
Owing to the acceleration of erosional processes consequent on
the high relief imposed by the fold movements, Lower and
Middle Devonian times are represented by an unconformity. Con-
tinental deposits accumulated in the Upper Devonian in lakes
formed by the dislocation of the drainage system by small flows
of rhyolite. In late Upper Devonian time basalts and huge masses
of acid lavas were extruded. Subsequently, the region was sub-
jected to pressure from ESE. and WNW., producing the present
strikes and dips of the Silurian and Devonian rocks.

a 88

Edwin S. Hills

From the Upper Devonian to the Pleistocene, the resultant of
earth processes has been erosion, and no deposits are found of an
.age intermediate between these two periods. In the early Ter-
tiary a peneplain was produced, which later suffered uplift, so
that it is now in process of dissection. Locally, continental
deposits of Pleistocene and Recent age were produced, and these
.are even now being removed by the streams.

Petrography.

(The numbers in square brackets refer to slides in the collection
at the Geology School, University of Melbourne.)

Rhyolite «. Cordicritc Nevadite.

(PI. XVII., Fig. 1.) [2244; 2245.]

Macroscopically and microscopically, this flow is extremely con-
stant over the seven miles studied. It is a compact rock with
dark cryptocrystalline groundmass and very numerous pheno-
crysts, a few millimetres in diameter, of clear quartz and cloudy
felspar, usually white, but sometimes pink, and an occasional
biotite. Fluxion structure is developed only along the lower edge,
and even then is infrequent, though some beautiful specimens were
found, coloured by weathering. Throughout the whole flow to
some extent, but becoming more numerous towards the base, are
xenoliths of sandstone and shale. Where the flow rests on the
Cathedral sandstones the sandstone xenoliths are more numerous,
and where it rests on the more shaly typical Silurian beds, the
shale xenoliths, often altered to a spotted shale and always indu-
rated, are more frequent. A few xenoliths of the basalt were
found.

Chemical Composition.

Si0 2 . .

. . 7472

74-39

78-64

ai 2 o 3 . .

. . 13 05

14-28

9-85

Fe 2 0 3 . .

. . 0-52

0-52

0-54

FeO . .

1-42

1-09

2-00

MgO . .

. . 0-41

0-27

0-10

CaO . .

. . 0-66

0-24

0-80

Na 2 0 . .

. . 3-62

2-78

2-03

K.,0 . . .

. . 4-31

5-33

5-16

h 2 o+ .

. . 0-61

0-22

0-40

h 2 o— .

. . 013

0-56

0-14

co 2 . . .

. . 0 08

—

TiO a . .

. . 016

0-29

0-67

P 2 o 5 . .

. . 0-38

tr.

MnO . .

. .

n. det.

—

F . . . .

. . n. det.

n. det.

Cl ... .

. . tr.

n. det.

Total

100- 07

99-97

100-33

Cathedral Range and the Blue Hills.

189

1. Rhyolite, Blue Hills, Taggerty. Analyst, E. S. Hills.

.2. Rhyolite, Archer’s Lookout, Narbethong, Analyst, N. R.
funner.

3. Rhyolite, Mount Wellington. Analyst, E. O. Thiele.

Norms and Classification.

Quartz . .

. . . . 35-58

36-30

44-52

Orthoclase .

. . . . 25-58

31-69

30-58

Albite . . .

. . . . 30-39

23-63

17-29

Anorthite .

. . . . 0-56

2-22

Corundum .

. . . . 2-24

3-47

—

Hypersthene

. . . . 2-85

1-76

2 -68

Magnetite .

. . . . 0-70

0-70

Ilmenite . .

. . . . 0-30

0-61

1-22

Apatite . .

. . . . 0-93

—

Class —

Persalane

Order —

Quart el ic

Quarfelic

Rang —

Peralkalic

Domalkalic

Sub-Rang —

Sodipotassic

Do-sodipotassic

Magmatic Name-

— Alaskose

Alaskose

Mihal-Tehaniose

Under the microscope the quartz phenocrysts are seen to be
rounded and embayed, often very deeply. The felspar is micro-
perthitic orthoclase, the included felspar being quite abundant.
It has a greater refractive index and double refraction than the
orthoclase, and shows polysynthetic twinning. The high soda
(and low lime) content of the rock points to the perthitic inter-
growth as being albite or oligoclase-albite. The orthoclase pheno-
crysts are nearly as numerous as those of quartz, and about the
same size; they show some rounding of the corners and edges,
and are very fresh. Small plagioclase phenocrysts are of in-
frequent occurrence. Though the maximum extinction on the albite
twinning lamellae ranges up to 26° in these, the refractive index
is less than that of quartz. They may be oligoclase-andesine, A
small amount of biotite is present in all sections, while in one it is
more abundant as small flakes in a rather coarser groundmass
than usual. Colourless cordierite, fresh in part, but often altered
to muscovite (pinite) or almost isotropic chlorite, is also present.
One section exhibits a trilling, but often all that remains of the
original mineral is a brown micaceous or chloritic mass. Small
blue tourmalines occur as single crystals and as radiating aggre-
gates, and are often associated with the material filling the em-
bayments in the quartz crystals. Small black specks in the
groundmass are probably ilmenite.

The groundmass is micro- to cryptocrystalline, and always
shows well-developed flow structure, the contorted lines curving
round the phenocrysts of quartz and felspar, which are not
arranged linearly. Biotite crystals curve with the flow lines, and

190

Edwin S. Hills:

wrap round the quartz and felspar. On solidification, the ground-
mass was evidently a glass, varying in composition from point to
point. Some bands are coarser than others, and consist of colour-
less mica and material with undulose extinction and fairly high
double refraction, which is probably a soda felspar. Other bands
are cryptocrystalline, much of the material having a higher refrac-
tive index than quartz. Again, in some cases biotite flakes,
quartz, felspar and colourless mica can be recognised, and occa-
sional microspherulitic aggregates occur. Apatite needles are
seen in the quartz and orthoclase, but the high norm of this
mineral indicates that some is present in the groundmass.

Junner (1914) has described rhyolites from near Narbethong,
which resemble very closely the above rock. He mentions, among
other things, the occurrence of blue tourmaline, corroded quartz,
and perthitic orthoclase. The analyses show an evident similarity
also.

Rhyolite [3 [2246] .

This is a black aphanite with small, clear quartz crystals and
turbid felspar. Under the microscope it is seen to be crowded
with angular fragments of quartz and felspar, which appear to
represent broken crystals. This might be due to continuation of
the process of embayment till the crystals are eaten through, or
(Rosenbusch; Osann, 1923) to rapid cooling of the rock giving
rise to shattering of the crystals. The felspar is both plagioclase
and orthoclase, the former giving a maximum paired extinction
of 22° on the albite lamellae, having a lower double refraction
than that of quartz, and a high refractive index. It is thus ande-
sine. Carlsbad twinning is sometimes shown as well as the albite,
and zoning is well developed. The orthoclase is in general un-
twinned, and is about equal in amount to the plagioclase. A few
fragments of pink garnet, and biotite flakes are present. The
groundmass is cryptocrystalline, and micro-fluxion structure, due
to an originally heterogeneous magma liquid, is sometimes seen.
Apatite, inclosed in quartz crystals, is a common accessory, and
veinlets of epidote and zoisite traverse the rock. A somewhat
clastic appearance is shown under the microscope, and chertv-
looking aphanites from near Narbethong have been described as
silicified tuffs by Junner (1914). However, in the present case,
the evidence points rather to solidification with sudden chilling of
a lava flow, perhaps under water. Thus,

(a) Some of the shattered phenocrysts have not been sepa-
rated, and the fragments as seen may be imagined as fitting
together.

( h ) The micro-fluxion structure shows that the groundmass is
not clastic but igneous, representing a devitrified glass.

( c ) The rock does not contain lapilli, and is quite distinct from
the pyroclastics which occur at Buxton.

(d) The epidote and zoisite veinlets may be due to the action
of caught up and heated water on the groundmass of the rock.

191

Cathedral Range and the Bine Hills.

Small (microscopic) xenoliths of sandstone and basalt occur,
but are not large enough to be seen in hand specimen.

An analysis of the rock gave the following result : —

SiO„

. . . 70-81

Al 2 (3 3 ....

. . . . 15-73

Fe 9 0. 5 . . . .

. . . 0-76

FeO ....

. . .. 1-97

MgO

. . . 1-30

CaO

. . . 1-68

Na 2 0 . . . .

. . . 1-75

K.,0

. . . 5-44

H,0+ . . .

. . . 0-54

H,0- . . . .

. . . 0-10

co 2

—

TiO

. . . 0-09

p„o 5 ....

. . . 0-22

Total

. . . 100-09

Norms and Classification.

Quartz 33-48

Orthoclase 32-25

Albite 14-67

Anorthite ..... 6-95

Corundum 4-39

Hypersthene .... 6-47

Magnetite 0-70

Ilmenite 0-15

Apatite 0-47

Class — Persalane

Order — Quarfelic: Columbare
Rang Domalkalic : Alsbachase
Sub-rang — Dopotassic :

Mihalose
Analyst — E. S. Hills.

Basalt [2250] .

This is a hard, dense, fine-grained, grey rock. Under the
microscope small phenocrysts of pale green augite are common,
and these are sometimes glomeroporphyritic. Labradorite also
occurs as phenocrysts and felted laths in the groundmass. Small
rectangular crystals of black iron oxide are numerous, and car-
bonates are present in the groundmass. Irregularly shaped
vesicles are scattered throughout, and contain mainly chalcedony,
though in some concentric bands of chlorite are seen.

The rock is moderately fresh, but some chloritization of the
pyroxenes has gone on, and an occasional large felspar shows
complete decomposition to colourless mica, though others are
quite unaltered. Local variations are mainly textural, one section
showing macroscopic crystals of iron oxide, but no plagioclase or
augite phenocrysts. In the north the flow is amygdaloidal, the
amygdales being about a quarter of inch in diameter, and very
numerous. They are filled with chalcedony and chlorite.

Volcanic breccia [2248].

This rock is dense, and contains numerous lapilli of grey-black
altered basalt, set in a lighter-coloured matrix. The lapilli do not
exceed an inch in length, and they grade downwards to small
grains in the tuffs. Under the microscope the lapilli are seen to
be of porous altered basalt, the vesicles being filled with chlorite
and chalcedony, also epidote and carbonates. Numerous rhomb-
shaped sections of an altered mineral occur, composed of serpen-
tine and showing irregular cracks along which iron oxide is segre-

.192

Edwin S Hills ;

gated. These are almost certainly altered olivine, and they are
present in both the lapilli and the groundmass, in the latter case as
fragments bounded by the curved fracture lines. The absence of
olivine in the basalts above the tuffs is interesting, and indicates
that the augite was produced by the reaction of the olivine with
the magma. Small iron oxide crystals are numerous. In the
groundmass of the rock detrital quartz occurs, with fragments of
altered olivine and small fragments of the basalts.

Basalt tuffs.

The finer grained pyroclastics have weathered very strongly,
giving chocolate and purple soft, light and friable material. The
breccias and tuffs are both well bedded and jointed.

Basal rhyolite [2247].

The flows at the base of the series are less acid than the main
rhyolite. They are light or dark grey, with a few quartz crystals
and abundant tabular felspars set linearly in a fine-grained ground-
mass. Under the microscope, altered felspar and biotite and em-
bayed quartz phenocrysts are seen. Acid oligoclase apparently
predominates over orthoclase, though the alteration makes deter-
mination uncertain. Numerous blebs of limonite are apparently
pseudomorphous after magnetite, the original parallel growth of
the cubic crystals being readily made out. The groundmass is
microcrystalline to crvptocrystalline.

Granodiorite-porphyrite [2249] .

In hand specimen this rock shows moderate sized phenocrysts
of quartz, white felspar and biotite, set in a finer groundmass.
Much fissured pink garnets occur sporadically. They show dode-
cahedral outlines and are about half an inch in diameter. A few
biotite-rich “ basic segregations ” were seen, and some xenoliths of
country rock, altered to horn f els.

Under the microscope, the quartz is seen to be embayed. The
rock weathers readily, all the specimens examined having cloudy
felspars. The plagioclase gives a maximum extinction of 20°
on the albite twin lamellae, and is andesine. Orthoclase is sub-
ordinate in amount to the plagioclase. The biotite is bleached and
chloritized, and has numerous inclusions of zircon as elongated
crystals arranged along the cleavage planes. Some zircon and
apatite occur in the groundmass, which is composed of quartz,
biotite and felspar. The latter is in part twinned plagioclase of
lower refractive index than that of quartz, probably oligoclase,
and in part untwinned felspar of low double refraction and lower
refractive index than that of quartz, probably orthoclase. The
cracks in the garnets are filled with chlorite and inclusions of
biotite and apatite are present.

Cathedral Range and the Blue Hills .

193

The Buxton and Taggerty granodiorites are exactly similar
-both macroscopically and microscopically, except that in a slide
oi the latter a fairly large piece of blue tourmaline showing
radiate structure occurs.

Palaeontography.

Fish and Plant Remains from Taggerty.

Plantae .

The fragmental plant remains found in the blue shales which
yielded the fishes described below are incapable of exact definition.
Some are narrow, alternately branching stems, with indications of
a relatively large central woody (?) cylinder. Miss I. Cookson,
who examined these remains in the hope that they might be
Psilophytales, says that they yield no indication of structure on
treatment with hydrofluoric acid, and are indeterminate.

Others are unbranched and show indications of longitudinal
ribbing, both coarse and fine. These resemble the “ Cordaites ”
australis McCoy, from the Avon River beds.

Class PISCES.

Sub-class DIPNOI.

Order CTENODIPTERINI.

Family CTENODONTIDAE.

Eoctenodus, gen. nov.

Eoctenodus microsoma, gen. et sp. nov.

(Plate XVIII., Figs. 2-7; Text-fig. 2, Nos. 1, 2, 3, 5, 6.)

Type Material. — Dentaries of mature and immature individuals.
Parasphenoid and median occipital bones of mature individual.
Scales. Bones of the shoulder girdle.

The Palate. — Specimen JA. Impression of the lower surface
of the left dentary (pterygopalatine with attached dental plate),
preserved in fine blue shale, and cleaned by weathering. Dental
plate elliptical, 0-55 cm. wide and 1-50 cm. long, bearing 10 den-
ticulate sub-parallel ridges. The ridges increase in size from
1 to 10 (see Text-fig. 2, No. 1), and the denticles increase in
size from inside to outside, being directed apically outwards.

Pterygopalatine. Maximum length 1*85 cm., maximum width
0-90 cm. (Text-fig. 2, No. 1, and PI. XVIII., Fig. 7). JA shows
the usual transverse crack across the alate extremity, but even so
a distinct downward turn of this part is noticeable. No scar is
present on the inner edge such as has been described for Sac/e-
nodas and Ctenodus (Watson and Gill, 1922-24), due to overlap
of the parasphenoid. In Eoctenodus the latter bone apparently
abutted against a ridge on the pterygopalatine. As is usual, the

194

Edwin S. Hills :

symphysis of the left and right pterygopalatines was weak, the
three specimens found being separate. Both extremities of the
alate portion are rounded and the lateral edges smoothly curved.

Specimens JE 3, JB 1. Mould of the left dentary, preserved
in the same blue shale as JA, but unweathered. General descrip-
tion as for JA, but 9 ridges only present on the dental plate, and
the whole somewhat smaller in size, indicating a younger individual.
Differs also in that the ratio of breadth to length is greater than in
JA. Pterygopalatine unbroken, and showing marked downward
curving of the extremity. Thickness of dental plate at outer edge
1-25 mm.

Specimens C. XVII., a,b. Mould of left dentary of a very
small form, probably the young of Eoctenodus. The dental plate
measures only 0-4 cm. by 0-2 cm., and bears eight ridges, strongly
denticulate, and all slightly concave forwards. Pterygopalatine
relatively large compared with the dental plate, and of a peculiar
shape (see PI. XVIII., Fig. 2). Four denticles present on the
larger ridges, directed apically outwards.

Specimens JFA, JF. Mould, in the same matrix, of a parasphe-
noid, the posterior shaft being incomplete. Length 3-5 cm., width
1-4 cm. JFA shows the impression of the cranial (dorsal) sur-
face, and a plasticine squeeze reveals a ridge round the anterior
part of the lozenge, which may have abutted against a similar
ridge ( vide ante ) on the buccal surface of the pterygoids. Pos-
teriorly a central ridge extends from the centre of the main
lozenge into the attenuated basisphenoid. A peculiar median pit
is situated on the anterior part of the lozenge. (PL XVIII., Fig.
5; Text-fig. 2, No. 5).

Cranial Roof Bones. — Specimens C.MO 1, 2. Mould, in the
same matrix, of a median occipital, incomplete. Anterior pro-
cess relatively large and sharply defined. Estimated length (of
median occipital) twice the breadth. One surface smooth (outer

surface), and bearing only slightly developed radiating ridges,
the other, of which only the anterior part is preserved, bearing
finely sculptured lines, radiating from the centre of the bone.
Maximum thickness, L5 mm. Length of anterior process 8 mm.,
breadth of bone about 2-0 cm.

Specimen JD. Impression, in the same matrix, of a circumor-
bital (?) bone. 7-8 mm. by 4 mm., 1 mm. thick. It shows
ridges radiating from a central rosette, and is bevelled along the
two long edges, the surfaces of the bevel being roughened.

The Shoulder Girdle. — The Clavicle. Specimen C.C. Mould, in
the same matrix, of the upper surface of the right clavicle. Shaft
long and narrow, 2-4 cm. long, 0-6 cm. wide, tapering at the arti-
culating end, which bears a well developed ridge and hollow.
Head imperfectly preserved, but apparently not very long. The
end of the shaft is cracked across and slightly displaced, and
this may indicate that it has been flattened from a twisted shape.

195

Cathedral Range and the Blue Hills.

The Cleithrum. Specimens C XVIII. a,b. Mould, in the same
matrix, of the left cleithium. Outer surface with a very strong
central ridge, continued into a strong articulating process. Outer
edge thickened. The bone is concave inwardly (the orientation
being as given by Watson and Gill, 1922-24), but has been crushed
m preservation due to collapse of the thick, spongy, bony struc-
ture, so that it is probably incomplete at the expanded end. I PI
XVIII., Fig. 4; Text-fig. 2, No. 6.)

Bones, indet.- C XV. 1, Group of small cylindrical bones,
associated with a longer linear group, dhese may he fin-bones.

2. Fine straight bones up to 1 inch long. These may be ribs

or fin rays.

C.C. Strong, curved bone with expanded end, oval in section.
This may be a neural spine.

Squamation.— Scales thin, often subquadrate with rounded
corners and slightly concave edges. Internal structure of fine
radiating ridges and furrows. Exterior smooth, marked by con-
centric growth lines, the centre of growth being excentric. The
largest found measured 14-25 mm. by 12-75 mm. A few sub-
rhomboidal scales were found, and these are smaller. Some show
•what is thought to be remnants of the lateral line (PI XVIII
Fig. 3.)

Relationships. The dipnoan above described shows relation-
ships with Ctenodus in some of the characters of the dental plate,
with its sub-parallel ridges and elliptical shape, the well defined
anterior process of the median occipital, the relatively large, thin,
sub-quadrate scales and the bones of the shoulder girdle. It is
related most closely to the small or moderate sized C. interruptus
of the Lower Carboniferous, but is distinctive in the fewer
ridges on the dental plate, and is separated from all forms of
Ctenodus ( by its distinctive parasphenoid, which indeed serves to
separate it from all other dipnoans. In size it probably approxi-
mated to that of Phaneropleuron and S caumcnacia.

Subclass OSTRACODERMI.

Order ANTIARCHA.

Family ASTEROLEPIDAE.

Bothriolepis Eichwald, 1840.

Bothriolepis gippslandiensis, sp. nov.

(Plate XVIII., Fig. 8; Text-fig. 2, No. 4.)

. Type Material. — Impression of some plates of the head;
impiession of the external marginal of the right appendage.

Plates of the Head. — Specimen C.B. Impression, in fine, irre-
gularly fiacturing sandstone of the dorsal surface of some of the
head plates of a Bothriolepis. Surface ornamentation of
tubei cles, fused more or less at their bases, producing distinctly

196

Edwin S. Hills:

Fig. 2.

Cathedral Range and the Blue Hills.

197

nodose ridges. The tubercles are often arranged linearly, e.g.,
parallel to the sensory grooves, which are united by a V-shaped
commissure behind the orbits, and appear to bend round at the
sides of the latter, instead of further anteriorly as in B. canadensis
and B. hydro phila. (PI. XVIII., Fig. 8.) Fracturing and dis-
placement of the plates during preservation has, however, ren-
dered the paths of these grooves outside the median occipital
uncertain. The latter is wider than long, width 1*8 cm., length
0*9 cm. The post-median is missing, as are the orbitals.

Lateral Appendages. — Specimen C.L. Impression, in the same
fine sandstone, of the external marginal plate of the right appen-
dage. Marginal denticulations very strong. Ornament of fine
parallel ridges, parallel to the line of junction with the interior
marginal plate, and hence making an acute angle with the denti-
culate margin. Maximum width, 0*75 cm., exclusive of the den-
ticles, length along the junction with the interior marginal 1-7
cm. The sudden constriction distallv indicates that the distal end
of the appendage is narrower than the proximal (Fig. 2, No. 6).

Relationships. — The above Bothriolepis resembles B. canaden-
sis in the strongly denticulate margins of the appendages, but
differs in the relatively short, compressed median occipital, in the
disposition of the sensory grooves, and in size.

Sub-Class DIPNOI.

Order ARTHRODIRA.

Family PHYLLOLEPIDAE.

Holonema Newberry, 1889.

Holonema cf. rugosum Newberry, 1889.

Specimen C.H. Mould, in fine sandstone, of a plate of a placo-
derm fish, incomplete. Specimen measures 5 cm. by 2-5 cm.
Ornament of radiating rugae, which are rounded, and in general
equal in width to the separating grooves. One edge shows a
slightly obtuse angulation. Maximum thickness shown 2 mm.,
which is near an edge, so that probably the plate was thicker in
other parts. Compares almost exactly in pattern and size with
the figure given by Newberry (1889, pi. xviii., fig. 4). “Holo-
nema rugosum. Portion of lateral plate of carapace(?), natural
size.”

Fig. 2.

1. Eocte nodus micro soma, sp. nov. Dentary, drawn from a wax squeeze.

R., Denticulate ridges (denticulations not shown), numbered from 1 to
10. P.P., pterygopalatine, x 4 -

2. Eoctenodiis microsoma. Median occipital, in part restored, showing
anterior process, x 2 -

3. Eoctenodos microsoma. Clavicle, drawn from a wax squeeze. S', shaft.

Xi-

4. Bothriolepis i rippslandiensis , sp. nov. External marginal of the right
appendage. I, inserted area. X 4 / 3 -

6. Eoctevodus microsoma . Parasphenoid restored. Cranial surface, from
a squeeze. The shaded area is a pit. x 43 / 5 -
6. Eoctenodus tndcro&oma. The cleithrum, from a wax squeeze, A,F„ ar-
ticulating process. X 3 / 2 *

198

Ed win 8 . Hills :

Discussion.

The assemblage of fishes above described is a typically Upper
Devonian one, and resembles in a remarkable way the suite of
Chemung (Upper Devonian) age at Scaumenac Bay, Canada. In
both localities we find a primitive small dipnoan, a coccostean form,
and Bothriolepis, associated with plant remains (and in Canada,
more fishes). This is interesting in view of the close resemblance
Benson has shown to exist between the marine Upper Devonian
rocks of New South Wales and the Chemung marine beds of the
Eastern United States (Benson, 1922).

It is interesting to note that Dr. Smith Woodward writes
(1904) : “ It may be said that . . . Bothriolepis and Astcrolepis
characterise the Upper Old Red Sandstone or Upper Devonian
wherever it occurs — in Britain, Belgium, Germany, Russia, Spitz-
bergen, Greenland, Canada and the Catskills of New York.” As
the only other Upper Devonian fossils which had been found in
Victoria up to the present were plant remains, some of which
are of doubtful value, the present discovery serves to give some
definition to our ideas of the Upper Palaeozoic succession in Victoria.

In conclusion, I wish to acknowledge my indebtedness to those
who have freely given their advice and help during the progress
of the research. To Professor Skeats especially, under whose
direction the work was carried out, I must convey my thanks for
his ready counsel on all occasions, his companionship for a few
days in the field, and his criticism and encouragement. To Mr.
Chapman, who has been my guide on all palaeontological ques-
tions, who determined some of the Silurian fossils, and who
allowed me to use specimens in his care at the National Museum
for comparison; to Miss L Cookson, for help with the plant re-
mains; to Mr. D. McCance, who smoothed out my paths in
chemical analysis; and to Mr. J. S. Mann, who photographed the
specimens figured in this paper, 1 must also express my thanks.

Bibliography.

Chapman, F., 1917. — On the Occurrence of Fish Remains and a
Lingula in the Grampians, Western Victoria. Rec. Geol.
i Surv. Vic ., iv. (1), pp. 83-86.

Dunn, E. J., 1907. Gold and Tin Workings at Tin Creek, near
Buxton. Ibid., ii. (2), pp. 105-108.

Eichwald, — , 1840. Die Thier- und Pflanzen-reste des alten
rothen Sandsteins und Bergkalks im Novgorodschen
Gouvernement. Bull . Sci. St. Petersbourg, iii., p. 79.
Ferguson, W. H., 1899. Notes on a Rapid Examination of the
Country near Taggerty. Gcol. Surv. Vic. Monthly Kept.
Prog., n.s., No. 2, pp. 4-5.

Gregory, J. W., 1912. The Geography of Victoria. Melbourne:
Whitcombe and Tombs Ltd.

Howitt, A. W., 1876. Notes on the Devonian Rocks of North
Gippsland. Geol. Surv. Vic. Kept. Prog., No. 3, pp.

181-249.

Cathedral Range and the Blue Hills .

199

Hyde, J. E., 1919. The Ripples of the Bedford and Berea forma-
tions of Central and Southern Ohio, with notes on the
Palaeo-geography of that Epoch. Journ. Geol. , xix.
(1), pp. 257-269.

Junner, N. R., 1914. The Petrology of the Igneous Rocks near
’ Healesville and Narbethong. Proc. Roy. Soc. Vic., n.s.,
xxvii. (2), pp. 261-285.

Kitson, A. E., 1899. Notes on the Geology of the Main Range
from Tolmie to Mount Howitt. Geol Surv . Vic.
Monthly Rept. Prog., No. 2, pp. 5-10.

Murray. R. A. F., 1877. Progress Report on the Geology of
Portion of the Country between the Thomson and Won-
nangatta Rivers, North Gippsland. Geol. Surv. Vic.
Rept. Prog., No. 4, pp. 52-57.

Newberry, J. S./1889. Palaeozoic Fishes of North America.

U.S. Geol. Surv. Mem. No. 16, pp. 1-343.

Rosenbusch, H., Osann, A., 1923. Elemente der Gesteinslehre,
p. 348. Stuttgart: E. Schweizerbart.

Skeats, E. W., 1910. The Volcanic Rocks of Victoria. Pres.

Address, Sect. C., Rept. Anst. Assoc. Adv. Sci xii.
(Brisbane Meeting, 1909), pp. 173-235.

Teale, E. O., 1920. A Contribution to the Palaeozoic Geology of
Victoria, with special reference to the Districts of Mount
Wellington and Nowa Nowa, respectively. Proc. Roy.
Soc. Vic., n.s., xxxii. (2) , pp. 113-127.

Twenhofel, W; H., 1921. Impressions made by Bubbles, Rain-
drops, and Other Agencies. Bull. Geol. Soc. Amer .,
xxxii., pp. 359-372.

, 1926. A Treatise on Sedimentation. London: Baill-

iere, Tindall and Cox.

Watson, D. M. S., and Gill, E. L., 1922-24. The Structure of
Certain Palaeozoic Dipnoi. Journ. Linn. Soc., Zool,
xxxv., pp. 163-216.

Woodward, A. S., 1904. Proc. Geol. Assoc., xviii., p. 433.

EXPLANATION OF PLATES.

Plate XVII.

Geological Map of the Cathedral District.

Notes on the Map.

The absence of contours renders the interpretation of this map
a little difficult, because of the strong relief. The bending round
of the outcrops of the Cathedral Beds at the north and south ends
of the range is due to the form assumed by the beds on erosion.
Between Chapel Hill and the Cathedral Range there is probably
a fault, whose line is hidden under alluvium.

The wave-like edge of the talus along the Cathedral Range is
meant to represent the numerous regularly developed alluvial
cones descending from the escarpment.

200

Edwin S. Hills :

W£.

* \ h t

SCALE Of MILES

UPPE R
SI L U*l

UPPE R

D E V O V I A N,-''

hhtiu r*
AWTOLPTE |S
8<SHLT

SWALE 1 INMRr
. , I F«*iluie£

L iAHO^ronc J

AwroLI re
OASAL COivtLSM tnare

Cii^ '•«
&.-r-

r fC

t Wf«

-Lo K

Fig. 3. — A: Sketch section along the line A B from SW, to NE. through
the Sugarloaf, Buxton. 1

B: Diagrammatic section from west to east across the summit
of the Blue Hills, showing the relations of the Upper Devo-
nian and Silurian rocks in the north.

C : Sketch section along the line C D from NW. to SE., across the
north end of the Cathedral Range, Taggertj.

NIAGAROON

Proc. R.S. Victoria, 41 (2) 1929. Plate X\II.

GEOLOGICAL MAP OF THE CATHEDRAL DISTRICT

Proc. R.S. Victoria, 41 (2), 1929. Plate XVTII.

Devonian Fishes, Taggerty.

Cathedral Range and the Blue Hills. 201

Plate XVIII.

Fig. 1. — Section of Rhyolite a (cordierite nevadite), from the
Blue Ranges, Taggerty. Ordinary light, showing em-
bayment of quartz, and flow-structure in the ground-
mass. X 25.

Fig. 2. — Eoctenodus microsoma , sp. nov. Impression of buccal
surface of left dentary of young. X2. [770].

Fig. 3. — Eoctenodus microsoma. Scale, showing lateral line.

X2. [771].

Fig. 4. — Eoctenodus microsoma. Impression of the outer sur-
face of the left cleithrum. X5/3. [772].

Fig. 5. — Eoctenodus microsoma. Impression of the cranial sur-
face of the parasphenoid, incomplete posteriorly. X2.

[773].

Fig. 6. — Eoctenodus microsoma. Scale, showing internal struc-
ture. X2. [774].

Fig. 7. — Eoctenodus microsoma. Impression of buccal surface
of the left dentary, and another small bone, indet. Xl>
[ 775 ]. . . t

Fig. 8. — Bothriolepis gippslandi ensis, sp. nov. Impression ot
the exterior surface of part of the head, showing
median occipital bone with V-shaped sensory canals.
XI. [776].

Note. — Owing to an optical illusion, it is possible that the im-
pressions of the bones of Eoctenodus (Figs. 2, 4, 5, and 7) may
appear as the actual or positive, rather than the impression.

Numbers in brackets refer to registered specimens in the col-
lection of the Department of Geology, University of Melbourne.

[Proc. Roy. Soc. Victoria, 41 (N.S.), Pt. II., 1929.]

Art. XIII . — On the Flanged Coxary 1 Palliocypraea gastroplax.
By FREDK. CHAPMAN, A.L.S., F.G.S., etc.

(With Plates XIX., XX.)

[Read 13th December, 1928 ; issued separately 3rd April, 1929.]

Occurrence of Specimen.

The remarkable cowry described by the late Professor Sir
Frederick McCoy, under the name of Cypraea gastroplax (McCoy,
1875, p. 20, pi. xvi., fig. 1 ; pi. xvii.; pi. xviii., fig. 2), remained a
unique specimen, so far as the National Museum collection is con-
cerned. until the subject of this note was found by Mr. Walter
Greed, of Hamilton. Mr. Greed discovered his specimen, a nearly
perfect example, in the lower beds at Clifton Bank, Muddy Creek,
Hamilton. He presented it to the National Museum on the 3rd
March, 1924 (Reg. No. 13273). The fragility of the Muddy
Creek specimen makes it surprising that the shell was obtained in
so perfect a condition. As it is, however, a portion of the thin
shelly flange has developed cracks more or less parallel with the
periphery, and portions that came away had to be supported with
paper. Other fractures seen in the shell run in zig-zag fashion
across these peripheral cracks right through the flange into the
dome of the shell. The prevalence of these fractures in the shell
and flange seems to suggest that there was an abnormal amount
of organic basis in this type of shell, which, on the extraction of
the specimen from the stratum gave rise, by rapid drying, to con-
traction and compensatory rifting.

Detailed Description of Fossil.

The form of the body of the shell is broadly pyriform, roundly
contoured anteriorly, and tapering rapidly posteriorly. The
profile shows a strong humping of the body of the anterior, with
the spire nearly flush with the general shape, comparable with the
roundly based Cypraea sphaerodoma Tate. The lower surface
of the shell is nearly flat, with a gently furrowed or depressed
margin on the upper surface, indicating the junction of the body
of the shell with the explanate flange. The aperture runs the
whole length of the shell, is gently sinuous in the middle, strongly
arched towards the anterior, and slightly undulose posteriorly.
The usually crenate margin of the aperture is well marked, the
teeth becoming obsolete at about one-fifth from either end. Both
apertural openings are slightly expanded and tubular. During the

Flanged Cowry , Palliocypraea gastroplax. 203

examination of this fragile specimen a portion of the flange, with
the surface of the body of the shell, became loose, thus revealing
a stouter shell layer beneath, over which the thin enamel lay like
a glaze.

The length of the shell, from the anterior edge of the flange
to the posterior, measured along the apertural region, is 97 mm.
The greatest width from side to side is 90 mm. The length of
the body whorl, from the centre of the spire at the apex to the
base of the body within the siphonal extension, is 56 mm. The
greatest width of the body whorl is 52 mm. The greatest height,
of the shell, measured from the base, is 33 mm.

Cossmann and Vredenburg on Palliocypraea.

Cossmann, in his original descriptions of his subgenus Palliocy-
praea , makes reference to Dr. G. B. Pritchard’s specimen from
Mornington, which 'lie figures (Cossmann, 1906, pi. ix., figs. I0 r
11), as the genotype of ” Rhynchocypraea ( Palliocypraea ) gas -
troplax McCoy.” Pritchard's specimen is really a plesiotype, for
the genotype (which can be a name only) is Cypraea gastroplax
McCoy. This latter is represented by the holotype of the species,
in the National Museum (Reg. No. 12140). Cossmann states the
age of the Mornington specimens as u Eocene/' but both the
Mornington and Muddy Creek (Lower) beds are now usually re-
garded as Oligocene, and certainly not Eocene.

In an exhaustive summary of the Family Cypraeidae, E.
Vredenburg (Vredenburg, 1920, pp. 126. 128), refers McCoy's
Cypraea gastroplax to a section of Gisortia under Cossmann's
name of Palliocypraea. He cites two species — C. gastroplax McCoy
and C. midden Tate. From a consideration of the form of the
latter shell, however, it appears quite incompatible to associate the
two species, for C. midderi has a broadly ovoid contour to the
body whorl, and the only approach to a flange is in the depressed
and round edged base of the shell.

Morphological Considerations.

In his original description of this cowry McCoy said : “ The
enormously extended circular thin flange into which the base is
extended, renders this cowry totally unlike any previously known
living or fossil species.” This statement still seems to hold good,
and the genus Palliocypraea is therefore monotypic. One calls to
mind the families of the Aporrhaidae and the Strombidae in
which the outer lip of the body whorl is extended, flattened and
fingered, and which extension sometimes involves the entire
length of the shell. This extension does not, however, surround
the shell, and possibly only in Cypraea could this be effected,
since in that genus or its allies there is a greatly expanded mantle
which covers, or nearly so, the whole body and base of the shell.

204

Frederick Chapman :

In Trivia the edges of the mantle do not quite junction, as may
be seen in the median dorsal furrow dividing the costate orna-
ment.

The flange in this species has no morphological connection with
an expanded lip, as in the genera mentioned, nor with the thin
everted lip of an embryo cowry, which in after life becomes in-
troverted and crenulate. The shelly flange is, therefore, an
exogenous growth in continuity with the periphery of the shell
and was probably the result of using up a redundancy of shell
material as a secretion of the basal part of the mantle, which
otherwise would have been utilised in adding to the body whorl
of this extraordinarily thin cowry shell. The shell flange, more-
over, would obviously be advantageous to the cowry in creeping
over an even-surfaced oozy sea -bed.

Although we consider it better to regard Pailiocypraea as
worthy of separate generic rank, there is no doubt that the body
form shows it to be related to the group often referred to as
Gisortia. Thus we may cite C. leptorhyncJia, C. ampullacca, C.
cximia and C. sphacrodoma as representative of this type of shell,
which, indeed, was already established in Lower and Middle Ter-
tiary times, whilst in C. umbilicata we have a survival to the pre-
sent in Austral seas.

Conditions of Deposition in the Muddy Creek and
Balcombe Bay Beds.

The bed at Clifton Bank, where Mr. Greed’s discovery of Pal -
liocypraea gastroplax was made, is a yellowish, friable shelly sand
or marl containing numerous polyzoa and the pteropod, Vagindla ,
together with some foraminifera. The finer muddy portion of
the matrix seen in the Balcombe Bay specimen is wanting in the
Muddy Creek deposit, but in both localities fairly deep water con-
ditions are indicated by the occurrence of pteropod shells.

The foraminifera of the Balcombe Bay impure limestone add
somewhat to the depth estimation of water in that locality in Bal-
combian times, and this is further substantiated by the prevalence
of glauconite grains.

At a rough estimate Pailiocypraea lived in the Balcombian sea
in the Port Phillip area at between 200 and 400 fathoms, whilst
in the Muddy Creek (Clifton Bank area) its probable depth would
be 100 fathoms or less.

Comparison of the Present Specimen with the
Balcombe Bay Holotype.

McCoy’s type specimen is stated to come from 4< the Qligocene
Tertiary limestone of the tract between Mount Eliza and Mount
Martha on the shores of Hobson’s Bay.” In explanation we may
remark that the precise locality, seeing that the shell occurred in a

Proc. R.S. Victoria, 41 (2), 1929. Plate XIX.

Fio. 1.

K. C. photo.

Fia. 2.

Palliocypraea gastroplax.

Proc. P.S. Victoria, 41 (2), 1929.

Plate XX.

F. 0. photo.

Fit*. 3*

Palliocydraea gastroplax.

Flanged Cowry , Palliocypraea gastroplax.

205

septarian block, is the old Cement Works at Balcombe Bay. The
locality cited by McCoy as “ Hobson’s Bay ” is probably a mis-
print for Balcombe Bay or Port Phillip Bay.

The type specimen (holotype) in the National Museum (Reg.
No. 12140) is undoubtedly conspecific with the beautifully pre-
served Muddy Creek specimen. The latter is only slightly smaller
in proportion, the difference in the length of the body being about
8 mm.

The type specimen is preserved in a hardened concretionary and
impure limestone, the matrix of which is foramini feral, and
besides this, contains innumerable shells of the little pelagic
pteropod, Vaginella elignwstoma Tate. The interior of the shell
has been naturally filled with this V aghiella-hz&rA ng mud, and
where the shell of the cowry has peeled away, the glass-polished
surface of the mud-cast is revealed. In this condition, of a par-
tially-fractured shell reposing on its cast, the tenuity of this species
is remarkably well brought out, showing an inner, prismatic layer
and the outer, enamelled shell. The inner layer is seen on a
weathered surface to show that the prismatic structure radiates
across the flange, whilst the enamel layer of the flange itself has
a fibrous structure concentric with the periphery.

Bibliography.

Cossmann, M., 1906. Essais de Paleoconchologie comparee,
vii., July, 1906, Cypraeidae, pp. 238-240, pi. ix., figs. 10,

McCoy, F.. 1875. Prodromus Palacont. Victoria, dec. ii., pp.
1-37, pis. xi.-xx.

Vredenburg, E., 1920. Classification of the Recent and Fossil
Cypraeidae. Rec. Geol. Surv. India , li. (2), Nov., 1920,

pp. 65-152.

EXPLANATION OF PLATES.

Plate XIX.

Fig. 1. — Palliocypraea gastroplax (McCoy). Aspect showing
profile of body whorl. Circ. nat. size.

Fig. 2. — P. gastroplax (McCoy). Another view of dorsum.
Circ. nat. size.

Plate XX.

Fig. 3. — Palliocypraea gastroplax (McCoy). Ventral aspect,
Slightly enlarged.

[Proc. Roy. Soc. Victoria, 41 (N.S.), Pt. II., 1929.]

Art XIV . — On some Trilobites and Brachiopods from the
Mount Isa District , N.W. Queensland .

By FREDK. CHAPMAN, A.L.S, F.G.S.

(Commonwealth Palaeontologist, National Museum, Melbourne.)

(With Plates XXL, XXII.)

[Read 13tli December, 1928 ; issued separately 3rd April, 1929.]

The following are descriptions of a small but very interesting
series of Cambrian Trilobites and Brachiopods, collected by
Messrs. Campbell Miles and E. C. Saint-Smith, from the head of
the Templeton River, twelve miles west of Mount Isa, and from
Thornton River, NW. Queensland. These fossils were submitted
to me for examination through the courtesy of Mr. B. Dunstan,
F.G.S. , Government Geologist of Queensland. Included in these
descriptions is a specimen, viz., Marjumia conspicabilis, submitted
later by Mr. Campbell Miles, and now incorporated in the Queens-
land Collection.

The specimens were received with other fossils (Ordovician),
on 14/4/25, and a preliminary report was furnished, 7/ 5/26,
which, however, was not published. Since writing this report I
have been able to devote more time to the study of these fossils,
and in some cases the former tentative determinations have been
somewhat modified. Their generic affinities are such as to con-
firm the horizon in which they are found, as being of Middle to
Upper Cambrian age. The genera and species herein described
are as follows : —

Brachiopoda : —

Lingulella niarcia Walcott, var. templetonensis, nov.

Acrothele bulb o ides , sp. nov.

Trjxobita : —

Agnostus chinensis Dames.

Bathyuriscus saint -smithii , sp. nov.

Bathyuriscus nitidus, sp. nov.

Bathyuriscus olenelloides , sp. nov.

Marjumia milesi, sp. nov.

Marjumia conspicabilis , sp. nov.

Marjumia el e gam , sp. nov.

Dikelocephalus dunstani, sp. nov.

Milesia templetonensis, gen. et sp. nov.

207

Trilobites and Brack iopods.

Phylum MOLLUSCOIDEA.

Class BRACHIOPODA.

Order ATREMATA.

Family OBOLIDAE.

Genus Lingulella Salter.

Lingulella marcia Walcott, var. templetonensis, nov.

(Plate XXL, Figs. 1, 2.)

Lingulella marcia Walcott, 1911, pp. 74-75, pi. xiv., figs. 3, 3a
Id., 1913, p. 69, pi. ii., figs. 6, 6a,

Observations. — This new varietal form is distinguished by its
large size, compared with the type species. There are two
examples in the present series. One is an internal cast of a pedicle
or ventral valve, showing umbonal and lateral scars. The charac-
teristic divergent striae of this specific form are even impressed
on this cast, as was also shown in the internal shell surface figured
by Walcott ( loc . at.) from a Chinese example. The outline of
this Queensland specimen represents the internal surface of a
shell of the narrower variety of Lingulella marcia . This agrees
with the figure 6d of Walcott’s series from the Middle Cambrian
of China. The internal striae and pedicle channel are very dis-
tinctly seen, and the shell is of porcellanous whiteness on a yellow
matrix.

Dimensions. — Length, 14-5 mm.; width, 11-5 mm.

The length of the typically broad form of Lingulella marcia
from the Cambrian of China, figured by Walcott on his pi. ii., fig.
6e, is 3 mm. On account of the large dimensions of our speci-
mens we are justified in regarding the Queensland form to be at
least varietal to the Chinese species, and have therefore named it
as a local variety.

Occurrence. — Twelve miles west of Mt. Isa, head of Temple-
ton River; collected by Messrs. Campbell Miles and E. C. Saint-
Smith, July, 1924.

Order NEOTREMATA.

Family ACROTHELIDAE.

Genus Ac rot he I e Linarsson.

Acrothele bulboides, sp. nov.

(Plate XXI., Figs. 3-5.)

Description. — Shell slightly wider than long, rounded to almost
subquadrate. Ventral valve with a fairly long hinge-line; anterior
border rounded and meeting the hinge-line at a decidedly sharp
angle. Posterior sloping away to the anterior border. The tumid
area occupied by a deep subquadrate fossette indicating the
pedicle opening, whilst towards the hinge-line there is the usual
u

208

Frederick Chapman :

V-shaped depression common to the genus. The shell-structure
is corneous and glossy and generally white, whilst the surface is
concentrically marked with very fine growth lines. Dorsal valve
slightly convex to flat, broadly rounded, with fine concentrically
marked surface, the striae on which are seen to slightly undulate
or even anastomose.

Dimensions. — Length of holotype (ventral valve), 3-5 mm.;
width, 4 mm.; height, circ. 1 mm. Length of paratype (dorsal
valve), 5 mm.; width, 5-5 mm.

Observations. — From the relative tumidity of the ventral valve
it might appear that the relationship of the above species was
with Acrotrcta rather than Acrothcle. The broadly expanding
anterior area, however, shows it to belong to the type of Acrothcle
represented by Walcott’s A. matthewi , var. oryx, from the Middle
Cambrian of China (Walcott, 1913. p. 73, pi. iii., figs. 6, 6a-h).
Acrothcle bulboides differs in the greater tumidity of the pedicle
area and in the practical absence of radial surface striae. From
Micromitra the above form is distinct in the position of the pedicle
opening, although at first sight the shells appear to belong to that
genus, as indicated on the Field List supplied with this collection.

Occurrence. — Twelve miles west of Mount Isa, head of Tem-
pleton River ; collected by Messrs. Campbell Miles and E. C.
Saint-Smith, July, 1924.

Phylum ARTHROPODA.

Class TRILOBITA.

Order HYPOPARIA.

Family AGNOSTIDAE.

Genus Agnostus Brongniart.

Agnostus chinensis Dames.

(Plate XXL, Fig. 6.; PI. XXII., Fig. 20.)

Agnostus chinensis Dames, 1883, p. 27, pi. ii., figs. 18, 19.
Walcott, 1913, p. 99, pi. vii., figs. 4-6. 6a.

Observations. — Three pygidia occur on two of the chips under
field No. 26 of the present collection. They agree specifically in
having a semi-circular border with a sub-acuminate axial lobe;
there is also a sub-central tubercle adjacent to a transverse and
slightly curved ridge. On each side of the margin of the pygidial
border, towards the posterior third, there is an obscure and blunt
spine.

A species of the genus Agnostus , viz., A. clkcdracnsis Eth. fil.,
has already been recorded from the same area, by Dr. Whitehouse
in his “ Note on a Collection of Cambrian Trilobites from the
South Templeton River, Queensland” (Whitehouse. 1927). A .
clkcdracnsis ,, according to Etheridge jun., who described it from
the Barkley Tableland, has no lateral pygidial spines, but differs

Trilobites and Brack iopods.

209

from the present species in having a transversely divided pvgidial
lobe with a tubercle on the anterior portion, whilst the lateral
spines are more posteriorly situated. Etheridge’s comparison of
A. elkedracnsis with A. ac adieus of Hartt, shows it to be distinct
from the Thornton River form here referred to A. cliincnsis.

Dimensions, — Plesiotype, length of pygidium, 2-75 mm.; great-
est width, 2-5 mm.

Occurrence. — Found in whitish porcellanised rock with iron-
stains, Thornton River, NW. Queensland ; collected by Mr.
Campbell Miles.

Order OPISTHOPARIA.

Family BATH YURI DAE.

Genus Bathyuriscus Meek.

Bathyuriscus saint-smith it, sp. nov.

Description. — (Based on type, collected by Mr. D. Smith.)
Form of carapace roundly ovate. Head broadly rounded, with
solid and lengthened genal spines extending to the line of the
third thoracic ring. Glabella prominent, well-rounded in front,
straight at sides. Fixed cheeks and palpebral lobes semicircular
and strongly curved. Frontal limb finely lineate, increasing in
strength towards the genal angles. Transverse furrows of the
glabella well-marked. Neck-ring distinct, apparently without
spine.

Thoracic segments twelve, narrow, with sillon. The segments
of the lobe broad and well defined, with indications of a short
basal spine. At junction with the axial furrows the surface of
the pleura rise to tubercles which may have supported the short
spines. Extremities of pleura terminate in short backwardly
curved spines. Pygidium comparatively small, semicircular, con-
sisting of four segments, with a terminal ovate axial lobe.

Dimensions. — Length of carapace, 48 mm. Width of cephalon
measured at the genal angles, 35 mm. Length of cephalon, 17
mm. Length of thorax, 23 mm. Length of pygidium, 8 mm.
Greatest width of axial lobe, 9*25 mm.

Observations. — The nearest related species to the above is per-
haps Bathyurisciis anax Walcott, which occurs in the Middle
Cambrian of Salt Lake Country, Utah. Although agreeing in
general form and character, the present species differs from B.
anax in having twelve instead of eight thoracic segments, and
in having longer genal spines. The carapaces are often so abun-
dant that one lies upon the other and they appear to have drifted
into a closely packed pool. The rock in which they are found
varies from a whitish tuff-like and silicified sediment to a similar
hard rock much stained with iron, varying from yellowish to
perhaps brown.

210

Frederick Chapman :

Occurrence. — This handsome species is by far the commonest
trilobite in the Mount Isa Cambrian series. Twelve miles west of
Mount Isa. at the head of the Templeton River ; collected by Messrs.
Campbell Miles and E. C. Saint-Smith. Also the holotype by D.
Smith, per E. H. Muir (presented to the Commonwealth Collec-
tion). Named in honour of Mr. E. C. Saint-Smith.

Bathyuriscus nitidus, sp. nov.

(Plate XXI., Fig. 9.)

Description.* — Carapace elongate-ovate. Cephalon rounded,,
paraboloid, with slender, dependent genal spines. Glabella rather
narrow, long, and with four well-marked transverse furrows.
Frontal limb sulcate within the margin, becoming finely grooved
towards the genal spines. The fixed cheeks and the palpebral
lobes wide and expanded, margined by a lunate crest. Neck-ring
well developed. Thoracic segments probably eight, very narrow,
the pleura terminating in sharp spines. The axial lobe is about
half the width of the thoracic lobe. Pygidium unknown.

Dimensions. — Length of cephalon, 6 mm. Width at genal
angles. 11 mm. Width of thorax, circ. 9-5 mm.

Observations. — This neat little species appears to find some
relationship with Bathyuriscus rotimdatus (Rom.), which is found
in the middle and base of the Upper Cambrian in the Mount
Stephen district, British Columbia (see Walcott, 1916, p. 346, pi.
xlvii., fig. 2, 2a, b). Thus the present form has similarly sharp
genal spines and thoracic margins, whilst the shape of the glabella
is also identical. On the other hand our species has more widely
expanded fixed cheeks.

Occurrence. — A single specimen found 12 miles west of Mount
Isa, at the head of the Templeton River ; collected by Messrs.
Campbell Miles and E. C. Saint-Smith.

Bathyuriscus olenelloides, sp. nov.

(Plate XXI., Fig. 10.)

Description. — Form of carapace elongate-ovate. Cephalon
broadly semicircular. Glabella roundly expanded in front, con-
cave laterally; border of fixed cheeks strongly convex, with the
palpebral lobes small and lunate. Genal spines long and diver-
gent. Thorax of 10 segments, narrow, with sharply terminated
pleura. Pygidium obscurely preserved and apparently small,
rounded at extremity.

Dimensions. — Height of cephalon, 9 mm. Approximate width
at genal angle, 20 mm. Length of thorax, 14 mm. Height of
pygidium, circ. 3*5 mm.

Observations. — This is a much larger form than B. nitidus,
which occurs on the same slab, and further differs from it in its

Trilobites and Brachiopods.

211

broader carapace and strongly divergent genal spines. This latter
character suggested the trivial name „ olcnelloides. This type of
cephalon is also seen in Bathyuriscus primus (Walcott, 1916, p.
352, pi. xlvi.j fig. 6 d), from the Lower Cambrian, Alberta,
Canada, which otherwise differs in having a shorter carapace
with fewer thoracic segments.

Occurrence. — A single individual on a slab of white porcel-
lanous shale with B. nitidus, sp. nov., 12 miles west of Mount Isa,
at the head of the Templeton River; collected by Messrs. Camp-
bell Miles and E. C. Saint- Smith.

Family OLENIDAE.

Genus Marjumia Walcott.

Marjumia milesi, sp. nov.

(Plate XXL, Fig. 11.)

Description. — Cephalon (glabella and fixed cheeks only).
Glabella elongate-ovate, broad anteriorly, and at the base with
moderately well-marked transverse furrows. Border of fixed
cheeks subcircular with a small palpebral lobe at the posterior
lateral angle. Pygidium transversely ovate, the lateral margin
curving outwardly and downwards, forming conspicuous falcate
or sickle-shaped spines. Posterior lateral margins continued to
basal extremity, entire but for a small blunt posterior spine on
each side of the flattened concave border of the pygidial extremity.
The pygidium has four segments. Pygidial axis moderately con-
vex, flattened towards the posterior excepting at the extreme end,
which is swollen. The flattened pygidial border below the ex-
tremity of the pygidial axis has the surface finely and concentric-
ally furrowed as in both Bathyuriscus and Dikcloccphalus.

Dimensions. — Height of cephalon (paratype), 13 mm. Gla-
bella at widest part, 6*5 mm. Pygidium (holotype), length, 12-75
mm. Greatest width above principal spines, 27 mm. Length of
pygidial axis, 8-5 mm. Width of axis at junction of thorax, 8-5
mm. Width of posterior extremity, 5 mm. Width of pygidial
border at posterior extremity, 4*25 mm.

Observations. — Several of the species of the genus Marjumia
which Dr. Walcott has described from the Middle Cambrian of
Millard County, Utah (Walcott, 1916, p. 402, pi. lxv., fig. 3b),
have points of agreement with the above species, but differ in
some essentials. Thus the pygidia figured by that author as Mar-
jumia callas have the lateral spines falcate, but the pygidial ex-
tremity is rounded and not obtusely concave. Another species,
which was not determined by Walcott, but placed under Bathynns -
cus and compared with Marjumia callas (Walcott, 1916, p. 348,
pi. lxv., fig. 5), and in the text with Bathyuriscus adacus , is
even closer in pygidial characters.

Frederick Chapman :

91 ?

Occurrence. — 12 miles west of Mount Isa, at the head of the
Templeton River; collected by Messrs. Campbell Miles and E. C.
Saint- Smith.

Marjumia conspicabilis, sp. nov.

(Plate XXII., Fig. 13.)

Description. — Holotype, consisting of large part of thorax and
pygidium, shows the carapace to be of large size, and of a long-
ovate form. The thorax has the lateral margins broadly rounded,
and the 14 segments of the genus are represented. The longi-
tudinal axis is wide, and tapers only slightly towards the pygidium.
It is on this depressed area that the ends of the pleura separate into
salient backwardly directed spines. The last of the series belongs
to the anterior segments of the pygidium, where it represents the
falcate spine typical of the genus Marjumia. The thoracic seg-
ments are fairly narrow, and the pleura are each marked by a con-
spicuous diagonal sillon. The pygidial border in the specimen is
not sufficiently well preserved to indicate the number of spines it
carries, but there is an indication of at least one pair of spines
below the anterior, falcate ones. The pygidial margin is finely,
concentrically striate, as in M. milcsi.

Dimensions. — Length of thorax, circ. 34 mm. Greatest width
of carapace. 47 mm. Length of pygidium, circ. 11 mm. Greatest
width, 26 mm.

Observations. — Apparently the only species comparable with
Marjumia conspicabilis is M. typa Walcott. The carapace of the
Shepherd Creek species, however, is more broadly ovate, and the
marginal depression of the pleura and pygidium is more pro-
nounced. The longitudinal axis in M. typa is much narrower,
and the axial furrows are nearly straight. The spinose ends of
the pleura in M. typa are sharper and more salient than in M. con-
spicabilis, where they curve rather suddenly towards the posterior.
The axis of the pygidium in M. typa is proportionately longer than
in M. conspicabilis and its marginal border has apparently a large
number of spines.

Occurrence. — Shepherd Creek, near Miles Creek, north branch
of the Templeton River, NW. Queensland; presented to the
Geological Survey of Queensland by Mr. Campbell Miles.

Marjumia elegans, sp. nov.

(Plate XXII., Figs. 14-16.)

Description. — Carapace rather small, broadly ovate anteriorly,
tapering posteriorly ; cephalon broadly semi-circular, with a pyri-
form and anteriorly expanded glabella. Outer limb grooved inter-
nally, and finely longitudinally, sulcate towards the genal angles.
Genal spines moderately long, dependent. Thorax consisting
of 14 segments. Pleura comparatively narrow, grooved diagon-

Trilobites and Brachiopods. 213

ally, spinose at the curved extremities. Pygidium comparatively
small, consisting of four segments, and bearing four pairs of
lateral, somewhat hook-shaped spines. Axial furrows somewhat
deeply impressed and straight.

Dimensions. — Length of carapace, 27 mm. ; width of cephalon,
21 mm. Greatest width of thorax, 18 mm. Length of pygidium’
6 mm.

Observations.— This species is perhaps the commonest of the
genus here described, and the holotype has been selected from a
complete carapace (No. 13a, Queensland Collection). The
diagnostic characters have been based on additional specimens,
among which are some well preserved cephala and pygidia. In
the pygidial characters the species also resembles M. typa Wal-
cott, already referred to, but the carapace, as a whole, is a rather
different form.

Occurrence. — 12 miles west of Mount Isa, at the head of the
Templeton River. Also occurring in a collection from Mr. Camp-
bell Miles, from Shepherd Creek, near Miles Creek, north branch
of Templeton River.

Family DIKELOCEPHALIDAE.

Genus Dikelocephalus Owen.

Dikelocephalus dunstani, sp. nov.

(Plate XXII., Figs. 17, 18.)

Description. — Based on remains of cephalon and pygidium.
Cephalon, large, broad, the cranidium showing an expanded
glabella, extending to the frontal margin. On either side, the
frontal limb gradually expands towards the genal angle. The
glabella with four transverse furrows. Palpebral lobes of the
fixed cheeks wide and strongly curved. Pygidium broad, almost
flabellate, with a short pygidial axis, almost triangular in outline,
with four segments. Marginal border of the pygidium broad in
outline and conspicuously incised with fine linear grooves. At
the lateral posterior angles there are two large falcate spines
which more closely approximate to one another than in D. min -
nesotensis. On the posterior segment of the pygidial axis is an in-
dication of an incipient spine, sometimes also seen in the genus
Saukia.

Dimensions. — Height of cranidium, 24 mm. ; width of crani-
dium, including palpebral lobes, 30 mm. Length of pygidium,
measured from the tips of the spines, 22 mm. Width of pygidium,
circ. 44 mm. Average width of marginal flange, 8 mm.

Observations. — This species resembles D. minnesotensis Owen,
in general characters. The glabella, however, is more globosely
expanded in the present species, whilst the pygidial spines are
situated near the axis. This species is named in honour of the

214

Frederick Chapman :

Government Geologist of Queensland, Mr. B. Dunstan, F.G.S.,
through whose courtesy I received the fossils for reporting upon.

Occurrence. — Thornton River, NW. Queensland; collected by
Mr. Campbell Miles.

Genus Milesia, gen. nov.

(For generic characters, see Observations infra.)

Milesia templetonensis, sp. nov.

(Plate XXII., Fig. 19.)

Description. — Holotype, of a nearly complete specimen.
General form elongate ovate. Cephalon transversely ovate, show-
ing cranidium with large palpebral lobe. The frontal limb is
wanting, but the free cheeks are well represented in outline by a
sunken impression of a broad head-shield with short and stoutly
falcate genal spines. The axial lobe with indication of the two
basal transverse furrows, the posterior furrow strong, extending
across the glabella. Thorax with 12 well-marked segments. Axial
lobe moderately convex. Pleural lobes flattened, each segment
having a distinct transverse furrow to the spinous margin.
Pygidium semicircular, well rounded has ally, with a deep, de-
pressed flange radially furrowed and transversely wrinkled, and
with traces of the strong pygidial spine near thoracic suture.
Axial lobe of pygidium tapering di stall y to a point and pinched
or rigid at apex. Lateral lobes of pygidium flattened, numbering
about five.

Dimensions. — Approximate height of cephalic shield, 20
mm.; width of cephalic shield, including free cheeks, 42 mm.
Height of thoracic series, 34 mm. Greatest width of thorax, 40
mm. Height of pygidium, circ. 18 mm. Greatest width of axial
lobe, 13 mm.

Observations. — This handsome form shows some relationship
to Dik otocephalus, but it is better to refer it to a new genus. This
is named in honour of one of the discoverers of this interesting
collection, Mr. Campbell Miles.

The furrowed glabella and the large number of thoracic seg-
ments — 12 — separate Milesia from Bathynriscus, which it other-
wise resembles. The expanded base of the pygidium also agrees
with Dikelocephalus. In Marjumia the glabella is narrower and
not so conspicuously furrowed, though the number of thoracic
segments agrees in that particular. The genal spines are not so
slender and prolonged as in Dikelocephalus , but more nearly
resemble those of Bathyuriscus.

Occurrence. — Preserved in sub-cherty shale of a whitish tint,
iron-stained on joints and showing as a brown chert on a frac-
tured surface. 12 miles west of Mount Isa, at the head of Tem-
pleton River, NW. Queensland (Queensland Geol. Survey Coll.).

Proc. R.S. Victoria, 41 (2), 1929. Plate XXI.

F.C. photo.

Cambrian Fossils. Mt. Isa, Queensland.

‘1 * 1

Proc. R.S. Victoria, 41 (2), 1929. Plate XXII.

F.C. photo, et del.

Cambrian Fossils. Mt. Isa, Queensland.

Trilobites and Brachiopods .

215

Bibliography.

Dames, W., 1883. In Richthofen’s “ China.” Ergebnisse eigener
Reisen und darauf gegrundeter Studien, iv. Palaeon-
tologische Theil., Abth. i., Abhandl. i., Cambrische Trilo-
biten von Liau-Tung, pp. 3-33, pis. i, ii.

Walcott, C. D., 1911. Cambrian Geology and Palaeontology,
Smithsonian Misccll. Coll., lvii. (4), Cambrian Faunas
of China, pp. 69-108, pis. xiv.-xvii.

, 1913. Research in China, iii. The Cambrian Faunas

of China, pp. 1-228, pis. i.-xxiv.

, 1916. Cambrian Geology and Palaeontology. Smith-
sonian Mis cell. Coll., lxiv. (5), Cambrian Trilobites, pp.
303-456, pis. xlv.-lxvii.

Whitetiouse, F. W., 1927. Abstract of Proceedings, Roy . Soc.
Qld ., May 2nd.

EXPLANATION OF PLATES.

Plate XXI.

Fi g. 1. — Lingulella marcia Walcott, var. templetonensis, nov.
Pedicle valve. X circ. lj.

Fig. 2. — L. marcia Walcott, var. templetonensis , nov. Dorsal
valve. X circ. 2.

Fig. 3. — Acrothele bulboides, sp. nov. Holotype. Pedicle valve
(on right). X 2.

Fig. 4. — A. bulboides, sp. nov. Dorsal valve. X circ. 2.

Fig. 5. — Slab showing group, mainly pedicle valves, of A. bul -
boides. X 2.

Fig. 6. — Agnostus chinensis Dames. Two pygidia. X circ. 2.

Fig. 7. — Bathyuriscus saint-smithii, sp. nov. Holotype. Natu-
ral size.

Fig. 8. — B. saint-smithii, sp. nov. Slab with two carapaces.
Natural size.

Fig. 9. — Bathyuriscus nitidus, sp. nov. Holotype. Circ. natural
size.

Fig. 10. — Bathyuriscus olenelloides, sp. nov. Holotype. Give.
natural size.

Fig. 11. — Marjumia milesi, sp. nov. Holotype, -f natural size.

Fig. 12. — M. conspicabilis, sp. nov. Pygidium, Natural size.

Plate XXII.

Fig. 13. — Marjumia conspicabilis, sp. nov. Holotype. Natural
size.

Fig. 14.— Marjumia elegans, sp. nov. Paratype. Natural size.

Fig. 15. — M. elegans, sp. nov.; a, cephalon; b, pygidium. Natural
size.

Fig. 16. — M. elegans, sp. nov. Holotype. X circ. 2.

216 Frederick Chapman :

Fig. 17 . — Dikelocephalus dunstam, sp. nov. Holotype. Natural
size.

Fig. 18. — D. dunstani, sp. nov. Paratype, cephalon. Natural
size.

Fig. 19 —Milesia templetonensis, gen. et sp. nov. Holotype.
Natural size.

Fig. 20 . — Agnostus chinensis Dames. Pygidium. Enlarged
drawing. X 3.

Fig. 21 . — Bathyuriscus nitidus, sp. nov. Holotype. Enlarged
drawing. X 6.

Note. — The holotypes and paratypes are in the Queensland
Geological Survey Collection, with the exception of the holotype
of Bathy arisen s saint-smithii y which is in the Commonwealth
Collection, and the paratype of Marjumia elegans Chapm. (pi.
xxii., fig. 14) which is in the author's collection.

[Proc. Roy. Soc. Victoria, 41 (N.S.), Pt. II., 1929.]

Art. XV . — On a New Species of Capulus found attached
to a Pterygotus Carapace.

By FREDK. CHAPMAN, A.L.S., F.G.S., etc.

(Commonwealth Palaeontologist, National Museum, Melbourne.)

(With Plate XXIII.)

[Read 13th December, 1928 ; issued separately 8tli April, 1929.]

Introduction,

Some years ago Sir Frederick McCoy (1899) described a new
species of Pterygotus ( P . australis) from the hard blue shale of
South Yarra, near Melbourne. The rock in which it was found
belongs to the lower or Melbournian stage of the Silurian.

This specimen of Pterygotus is represented by both the holotype
of a somite figured by McCoy and the counterpart on which the
shells of Capulus are present. These specimens are in the National
Museum collection.

The Pterygotus somite measures, post-anteriorly, 61 mm., and
it has an estimated breadth of 165 mm. (6^ inches).

Soon after my arrival in Melbourne, in 1902, whilst examining
the holotype 1 noticed that the surface of the carapace bore,
besides the conspicuous scale-like markings, several depressed
ovate areas which were not noted in McCoy’s description. A wax
or plasticine impression from these showed a valve in relief which
then seemed to be referable to either Pholidops or Orbiculoidea ,
the latter genus as well as Crania being already known as a
brachiopod commensal on cephalopod shells from the Palaeozoic
of the United States of America.

Lately, whilst in conversation with Sir Edgeworth David on
the subject of his recent discoveries of Eurypterids in the Pro-
terozoic rocks of South Australia, he mentioned the occurrence of
generally similar commensal limpet-like fossils. My interest in
the Silurian specimens here described was thus revived.

The chief obstacles in referring the present fossils either to
Crania , Pholidops or Orbiculoidea are : —

1. The absence of a cemented valve, or trace thereof.

2. The irregular growth stages of the shell surface, which
in Pholidops are evenly developed.

3. The absence of a pedicle opening either of Crania or
Orbiculoidea.

218

Frederick Chapman:

The only form to which we can reasonably assign the present
specimens seems to be that of a member of the gasteropod genus
Capulus , which ranges from Cambrian to Recent. I have been
greatly helped in this comparison and diagnosis by finding a
recent specimen of Chlamys bifrons from South Australia in the
Dennant Collection, the valves of which are peppered over with small
attached shells of an allied genus, Hippo nix. A common charac-
ter of both this recent — H. conicus (Schumacher) (May, 1921,
1923) — and the Silurian examples is the cleared area around the
shell, a feature also to be noticed with living limpets.

It is very remarkable to find so close a resemblance in form,
external structure and nearly equal dimensions, in the two Silurian
and living genera and species.

In selecting his holotype of Pterygotus from the two specimens
(counterparts) discovered by F. P. Spry in the Silurian mud-
stone, McCoy seems to have been influenced by the character of
the ornament of the carapace. In the holotype this is a salient
squamation, and agrees in the main with that seen in well-pre-
served specimens from England, North America and Bohemia.
Apart from this general character of salient squamation, there
appear parts of the carapace in other specimens which have an
impressed ornament. So that my selection of the counterpart of
McCoy’s type of the South Yarra specimen of Pterygotus as the
actual positive surface, on account of the appearance of the shells
of Capulus in relief, is not without reason.

Description.

Class GASTEROPODA.

Family CAPULIDAE.

Genus Capulus Montfort.

Capulus melbournensis, sp. nov.

(Plate XXIII., Fig. 1.)

Shell ovate, expanding in ephebic stage. Apex moderately high
to depressed and incurved. Concentric growth lines distinct, with
strong growth stages at intervals. In some examples an apical
ridge extends down the middle of the shell. The shell is seated in
a depressed area on the carapace of Pterygotus australis , similar
to that seen in living examples of the allied genus Hipponix
attached to Chlamys and other bivalves.

Length of holotype, 3 4 25 mm. Greatest width, 2-25 mm.
Height, circ. 1 mm.

Occurrence.

Eleven specimens attached to the carapace of Pterygotus aus-
tralis McCoy. Silurian (Melbournian). Domain Road, South

Pi’ 015 . R.S. Victoria 41 (2), 1920. Plate XXIII.

F.C. photo. Fig. 2.

Sessile Gasteropods : Capulus (Silurian),
Hipponix (Recent).

New Species of Capulus.

219

Yarra (main sewer tunnel). Found by F. P. Spry, and donated
to the National Museum. Registered No. of present counterpart
No. 1085. Registered No. of McCoy's Holotype, No. 577.

Bibliography.

McCoy, F., 1899. Note on a New Australian Pterygotus. GeoL
Mag., [4], vi., pp. 193, 194, text-fig.

Hedley, C., 1904. Studies on Australian Mollusca, Part VIII.

Proc. Linn. Soc. N.S.W., xxix., p. 190, pi. viii., figs.
15, 16.

May, W. L., 1921. A Check-List of the Mollusca of Tasmana, p.
57.

, 1923. Illustrated Index of Tasmanian Shells, pL

xxvi., fig. 15.

EXPLANATION OF PLATE XXIII.

Fig. 1. — Portion of somite of Pterygotus australis McCoy, with
attached shells of Capulus mclbournensis, sp. nov.
Silurian (Melbournian). S. Yarra. X circ. 2.

Fig. 2. — Chlamys bifrons (Lam.). Living, S. Australia.

Valves with attached shells of Hipponix conicus
(Schum.). 2/3rds natural size.

[Proc. Roy. Soc. Victoria, 41 (N.S.), Pt. II., 1929.]

Art. XVI . — Notes on and Additions to Australian
Fossil Polyplacophora (Chitons).

By EDWIN ASHBY, F.L.S., etc.

(Communicated by F. Chapman, A.L.S.)

(With Plate XXIV.)

[Read 13th December, 1928; issued separately 8th April, 1929.]

Introduction.

Mr. Francis A. Cudmore has placed in my hands for descrip-
tion a large number of valves of Fossil Chitons, both from the
Table Cape beds in Tasmania and the Balcombian beds in Vic-
toria. The Rev. George Cox, of Mornington, and Dr. H. J.
Finlay, of Dunedin, N.Z.. have also permitted me to study impor-
tant material from Balcomhe Bay. Two species are added
to the fossil fauna: one, a unique example of Ischnochiton
( H eterozona ) cariosus Pilsbry, is the first fossil representative of
its genus to be found in Australia; the other discovery, for which
the new genus 06 chiton is instituted, is still more remarkable, the
nearest apparent relatives being two rare deep water forms from
Cape Horn and Antarctica, one of which is figured for compari-
son. A discussion of the systematic position of the new dis-
coveries is given and a classified list of the Australian Fossil Poly-
placophora is furnished.

Systematic Description.

Lorica compressa Ashby and Torr, 1901.

From the Crassatella Beds, Table Cape, Tasmania, Mr. Cud-
more has taken complete, or portions of, 24 median valves and
three portions of anterior valves, one almost complete, all refer-
able to the above species.

Lorica compressa var. affinis Ashby and Torr, 1901.

In the collection is one median valve and two fragments of
median valves of this variety in which the longitudinal ribbing is
much more widely spaced than is the case in L. compressa f s.str.

Australian Fossil Polyplacophora. 221

Lorica cudmorei Ashby, 1925.

From the same bed as the foregoing' two imperfect median
valves of this species were taken.

Mr. Cudmore has found 32 valves or portions thereof belong-
ing to the genus Lorica from one bed. It not only evidences that
the genus Lorica was numerically very strong in the ancient sea
in which this Crassatella bed was laid down, but also that this
genus of Chitons was almost the only one represented in associa-
tion with the Crassatella. In the seas of to-day the genus Lorica
is but poorly represented as compared with other groups of Poly-
placophora, and it is only recorded from Australasian waters.

Loricella gigantea Ashby and Torr, 1901.

(Plate XXIV., Fig. 9.)

One beautiful example of the head valve of this species is in
the collection ; it was taken from the Lower Bed, Table Cape,
Tertiary (Janjukian).

In the original description the locality was given as Morning-
ton, although thought to have been a mistake. This, the second
example of this valve, is a small replica of the holotype, measures
24 X 12-5 mm., and settles the question as to the true locality of
the original find. The median valve described by Hull as Loricella
maguifica, which, as I have already indicated, is referable to this
species, was also from Table Cape: I therefore indicate the Lower
Bed, Table Cape, as the type locality and horizon.

Oochiton, n.gen.

This new genus is proposed for the reception of a new and
unique form which is herein described under the name Oochiton
halli , n. sp., which species I designate as type of this genus.

The median valve has in common with the genus Notochiton a
very strongly carinated shell with very steep side-slope, the
sutural laminae joined across the middle line, insertion plate in
median valve broad, edge smooth, slits 1/1, broad and deep. It
differs from Notochiton in the absence of regular longitudinal
ribbing, and possesses peculiar ovate pustules which stand erect in
irregular rows or widely scattered over the whole of the tegmen-
tum; these pustules apparently are associated with the "nerve
fibres, for most of them have a minute aperture at the summit,
and differs also in the greater width of the insertion plate. The
name is suggested by the peculiar sculpture which suggests strings
of minute eggs.

Since the above definition was written, the tail valve has been
discovered. This valve differs widely from the genus Notochiton ,
and, to the best of my belief, is quite unique in its characters. The

222

Edwin Ashby:

upturned and greatly thickened extremity, with the deep sinus
immediately behind the mucro and the extended fold of the teg-
mentum into this sinus, in a limited degree, resemble the genus
Lorica ; the entire absence of the insertion plate immediately
behind the mucro together with the greatly thickened extension of
the insertion plate laterally with its single slit on either side,
faintly reminds one of some members of the Mopaliidae.

The contour of the anterior valve is remarkably like that of
Notochiton Mirandas Thiele, the insertion plate is also similar in
being grooved and bevelled, but the slits in Ode hi ton are propor-
tionally broader. Whereas Notochiton mirandus possesses ray-
ribs corresponding with the slits, the species under discussion has
no ray-ribs and no correspondence between the sculpture and the
slits, Jt will be seen that both the anterior and the median valves
show some affinity with the genus Notochiton , but the tail valve
is strikingly dissimilar and unique. I consider the genus Oochiton
more primitive than the genus Notochiton , but it might well be
placed immediately preceding that genus.

Oochiton" halli, n.sp.

(Plate XXIV., Figs, la,b; 2; 3 a-c; 8a, b.)

Mr. F. A. Cudmore has placed in my hands two median valves
of an entirely new species of Chiton; the one I am making the
holotype was found by him at Balcombe Bay, near Mornington,
Victoria, Tertiary (Balcombian) ; the other is also in Mr. Cud-
more's collection, and was collected by the late Dr. T. S. Hall at
Belmont, Geelong, Victoria, Tertiary (Barwonian), and is separ-
ately described herein.

Since writing the following description I have received from
the Rev. George Cox, of Mornington, through Mr. R. A. Keble,
the Palaeontologist of the National Museum, Melbourne, a tail
valve and some additional median valves of the same species. Mr.
Cox writes as 'follows : “The tail valve and several median valves
were found [in the Balcombian Beds] at Mornington, by a lad
aged 12 years, named Evan Chitts; two median valves and the
tail valve were washed out of one cubic inch of clay, and may
have belonged to the same animal.”

Still more recently Dr. H. J. Finlay of Dunedin, N.Z., has sent
me an example of the anterior valve of Oochiton halli, which had
been collected by himself in the Balcombian beds at Mornington.
He is generously allowing me to keep this specimen, which I am
describing hereunder as the type of the head valve of this species:
up to the present this example of the head valve is unique.

I am naming this interesting species at the suggestion of Mr.
Cudmore after the late Dr. T. S. Hall, the discoverer of the first
median valve found.

Australian Fossil Polyplacophora.

223

Median valve.

Holotype, Balcombe Bay, Victoria. — PL (XXIV., Fig.

1 a,b). Strongly carinated, very elevated, side-slope straight and
steep, angle of divergence 70°, surface smooth and polished,
areas indistinguishable, one or two shallow growth lines parallel
with the margin towards the girdle. The ornamentation is unique,
and consists of six longitudinal broken strings of minute bead-
like pustules; the pustules are ovate, and together resemble
strings of minute white eggs, which feature has suggested the
name of the genus. The first row nearest the jugum has 12 of
these pustules ; the second has nine only, traverses only half way
across the pleural area and is bowed upwards; the third has 14
pustules ; the fourth has 9; the fifth has 10; and the sixth has only

2 pustules. It must be noted that all these rows have gaps, the
string not being continuous, but this is in some places undoubtedly
due to the breaking off of some pustules. The dorsal ridge is
slightly raised, anteriorly a little broader than at the beak, and in
a faint degree is subgranulose. The foregoing is as seen under a
simple lens, X 20.

Under a Zeiss binocular microscope, X65, some very interest-
ing features are made clear. The whole of the surface of the
shell is highly polished, and everywhere thickly perforated with
megalopores. It is also transversely, concentrically crossed by
numerous growth grooves or lines, these running across the
jugum from side to side. To these grooves is due the apparent
subgranulose appearance of the dorsal ridge. The bases of
detached pustules are visible, the pustules themselves are defi-
nitely ovate, attached by the smaller end and almost vertical ; each
pustule has a small perforation at the summit, looking like a
black dot, which is a little larger than the megalopores of the
normal surface of the shell ; except for this aperture the pustules
are solid, not hollow, as in Protochiton. The channel connecting
nerve fibres with the black dot can be seen in places where the
pustules have been broken. Corresponding with the rows of pus-
tules is an irregular series of deep pits with a black, probable eye-
dot at the bottom ; these rows of pits are on the lower and outer-
side of the pustules, and are overhung and almost hidden by these.
The perforations at the base are much larger than the megalo-
pores, and therefore must have functioned much like what are
known as “ eyes ” in recent species. This description is taken
from the right side of the shell; the other had met with some
injury during life, and the process of mending has caused the
outer part of the lateral area to bend upwards, and the strings of
egg-shape pustules have somewhat merged into one another.
The inside of the valve is white, and the tegmentum is folded
over at the beak, the margin of the fold being coarsely pustulose.

Dimensions. — The holotype, median valve, is 4 ’5 mm. in width
and 3-75 mm. in length; angle of divergence, 70°.

224

Edwin Ashby:

Paratype, Belmont. — (PI. XXIV., Fig. 2). Median valve,
beaked, carinated, side-slope very steep, dorsal area arched except
near the beak, where it is narrowed to a mere ridge, smooth
except for several narrow ridges unsurmounted by pustules,
separating this area from the pleural. The character of the
strings of egg-like pustules is similar to that of the type, but the
rows are shorter and in places a narrow ridge connects widely-
spaced pustules ; near the insertion plate, grains are scattered.
The sutural laminae are broken, but are joined across the centre
line; the lateral area is separated from the pleural by a shallow
diagonal fold. The colour of the tegmentum is silvery grey, the
pustules opaque white. The pitting, although present, does not
appear to be associated with the pustules, as is the case in the
holotype. Interior creamy white, insertion plate undamaged on
one side, teeth sharp, slits 1/1, well-defined and broad, callus
imperceptible, tegmentum extensively folded over at the beak
forming a “pocket.” This median valve, Nat. Mus. No. 13497, is
longitudinally narrow, measuring 4x3 mm. ; dorsal area without
pustules.

Paratypes, Balcombe Bay. — No. 1 measures 4*25 x 3*75; No.
3 certainly has the articulamentum joined across the middle line
between the sutural laminae; No. 4 is imperfect, has a V-shaped
notch in the articulamentum between the sutural laminae; No. 5
is fragmentary, dorsal area ornamented with egg-shaped pustules
but without raised dorsal ridge; No. 6 is a fragment only.

Tail valve .

Paratype, Balcombe Bay. (Type of tail valve.) — (PI. XXIV.,
Fig. 3a-c.). Small, measuring longitudinally 2*75 mm., laterally
2 *25 mm., very strongly carinated ; mucro at the posterior margin
or more correctly subposterior, because the tegmentum is bent
over at the mucro and turned down vertically ; the portion immedi-
ately behind the . mucro is concave and in this cavity or sinus, are
two of the egg-shaped pustules common to the sculpture of the rest
of the tegmentum ; from the mucro is a raised diagonal rib or fold,
the strings of egg-shaped pustules of the pleural area are con-
tinued across this fold to the posterior edge of valve. This valve
is upturned at the mucro and the extremity very much thickened,
the insertion plate here is subobsolete, and reduced to a mere
callus or ridge behind the mucro, but on either side the insertion
plate is developed into a highly thickened extension of the articu-
lamentum with one diagonal slit on either side, and in addition on
one side a supplementary groove, but not a true slit. The sutural
laminae are well developed, the sinus between being very narrow,
and are joined to the thickened posterior insertion plate by a
broad extension of the articulamentum which is suggestive of the
Acanthochitonidae.

Australian Fossil Polyplacophora .

225

• Anterior valve.

r Paratype, Balcombe Bay. (Type of anterior valve).— (PL
Figs. 8a, h.). Valve highly elevated, apex slightly re-
curved, anterior slope very steep and concave (due to recurved
apex). The ornamentation consists of strings of egg-like pus-
tules similar to those in the other valves ; the arrangement is
generally speaking longitudinal, the strings commencing at the
posterior margin and continuing to the insertion plate with con-
siderable irregularity, several strings bifurcate, and in some places
there are short intermediate rows; the strings or rows of pustules
do not seem to have any relationship with the slits in the insertion
plate.

A rt i cul am entum , or inner layer of shell creamy white, highly
polished, smooth, without any grooves; the tegmentum infolded
at the apex, this infolded portion is thickly studded with the egg-
like pustules ; the insertion plate is well produced, perfect, except
for a few minute chips; slits 7, broad and short, spacing irregular;
the upper side of the insertion plate is numerously grooved, the
plate is broad and proportionally thick, but the upper edge is
bevelled off, so that the actual edge is sharp, the grooves not
continuing to the inner edge. Valve measures 4-5 X 2-25 mm.

Notochiton mirandus Thiele.

(Plate XXIV., Figs. 4, 5, 6a, b.)

(N. mirandus Thiele, Subantarktischen Chitonen, pp. 12, 13.)

In the preparation of this paper comparisons have had to be
made with this species, of which I have in my collection a cotype
given to me by Major Dupuis. Pilsbry does not refer to it,
Thiele (in Re vis. des Syst. der Chitonen, p. 107) neither figures
nor describes it, making a bare reference and stating that Edgar
Smith considered it a Chetopleura, but Thiele considers it allied to
the genus Nutt alio chit on, and is probably correct. For purposes
of comparison with Odchiton halli, n. sp., figures are given.

Ischnoci-iiton (Heterozona) cariosus Pilsbry, 1892.

(Plate XXIV., Fig. 7.)

The Rev. George Cox has sent me a single median valve of
the above Ischnochiton , collected by Master Evan Chitts in the
Balcombian Beds at Mornington. This is the first true record 1 of
the discovery of a fossil Ischnochiton in Australia.

This example appears to have a well-defined diagonal rib, and
for that reason I at first thought it would likely prove to be a new
species, but on careful examination I find that this apparent

1.— F. Chapman was in error when he referred P rotoehiton gmnvlomn to this genus (Proc. Roy.
Soc. Vic., n.s., xx. (2), pp. 218-220, 1008).

226

Edwin Ashby :

feature is due to a slight wearing of the anterior of the raised
lateral area. As compared with a half -grown example from
Marino in South Australia, in which the valves are of a corre-
sponding size, I find the sculpture similar, though a little more
deeply cut, in this respect corresponding with the form from
Western Australia, but it differs slightly in that the infolding
of the tegmentum under the jugum is about double the width as
compared with the Marino example, but in the lateral extension
of this infolding it is similar. There are no differences to be dis-
tinguished in this valve to justify separation, but such may be
revealed when fossil end valves are discovered. The fossil valve
measures 5-5x2-25 mm.

Discussion on Classification.

I have retained the genera Lorica and Loricclla under Pilsbry’s
subfamily Liolophurinae, while recognising that this is not their
true setting. Thiele found that the radula showed relationship
with the Ischnochitonidae, and treated these genera as advanced
forms of that group, but I feel that more work needs to be done
on characters other than that of the radula, before their true
niche in the Natural Taxis can be determined. I therefore retain
them in the setting in which Pilsbry placed them, until the study
of this problem, from the points of view suggested above, supplies
added data upon which \ye may form a considered opinion.

All students of the Polypiacophora are greatly indebted to Dr.
Thiele for the specialized work he has done in the radula of that
order. He has laid a good foundation, and it is unfortunate that
since the production of his valuable work, “ Revision- des Systems
der Chitonen,” no material work has been done on the radula of
this group. One should hesitate to accept too hastily conclusions
based chiefly on one feature alone, until such time as other sup-
porting features have been studied and made known.

Thiele has pointed out that in the family Lepidopleuridae there
is some variation in the characters of the radula. Iredale and
Hull have assumed that this discovery of Thiele’s means that the
absence of insertion plates and other accepted primitive charac-
ters, are the result of degeneracy, and have founded their classifi-
cation on this assumption. Thiele himself drew no such conclusion
from his discovery, and proposed a suborder, Lepidopleurina, for
this group, numbering it (I.). I feel sure a right conception of
taxonomic values will endorse Thiele in his treatment, and I have
suggested that the Chitons living in the seas of to-day have not
arisen from primitive stock in one phylum alone, but through
more than one. This I have demonstrated in the case of the Acan-
thoid group, and have expressed the opinion that the existence
of divergences in the radula of members of the Lepidopleuridae is

Australian Fossil Polyplacophora.

227

important evidence that gathered together in this group are the
progenitors of more than one phylum that have developed along
parallel lines.

The discovery, in addition to the median ones, of the end
valves of the 'fossil Protochiton granulosus (Ashby and Torr), has
made it quite clear that this species could not have been derived
through any members of the Lepidopleuridae, and its evident rela-
tion to the Acanthoid Group of Chitons makes necessary a partial
revision of our previous conception of the Classification of Poly-
placophora.

This revision was foreshadowed in my Phylogenetic Diagram,
page 75 (l.c.) ; I have endeavoured in the following Classification
List to give expression to this revised conception, made necessary
by the recent discoveries named above.

The proposal of Iredale and Hull to substitute the word “ Lori-
cates ” for the universally used “ Chitons,” dating as this latter
does from the days of Linne, surely can commend itself to no
one. The proposal to substitute the word “ Loricata ” for “ Poly-
placophora/’ and 14 Crvptoconchidae ” for “ Acanthochitonidae,”
is not compulsory, and surely can serve no good purpose. The
law of priority does not apply to ordinal and family names ; also
the use of the term “Type Genus” is understood by most workers
to mean “ typical genus,” which the specialized form Cryptocon-
chus certainly is not.

Since the issue of my Monograph on Australian Fossil Poly-
placophora, Iredale and Hull have described the cast of a Chiton
from the Permo-Carboniferous beds of Bundanoon, New South
Wales, and have called it Permochiton australianus. This speci-
men is a very interesting one, in face of the fact that Etheridge’s
Chelodes calceoloidcs has already been disallowed, for although
its true character is still in doubt, there seems a consensus of
opinion that it is not a representative of the order Polyplacophora ;
Permochiton australianus comes from the oldest series of beds
in which Chitons have yet been found in Australia. These gentle-
men suggest some resemblances between P. australianus and
the genera Ischnochiton and Lepidopleurus, but judging from
their figures I can see no resemblance, though certainly there is a
general resemblance to the Palaeozoic genus H elininthochit on, and
it is quite natural to suppose that members of that genus would
persist from the Carboniferous into the Permo-Carboniferous.

228

Edwin Ashby :

Classification of Australian Fossil Polyplacophora.

Class AMPHINEURA.

Order POLYPLACOPHORA (Blainville em.) Gray, 182L

[Primitive.]

Suborder EOPLACOPHORA Pilsbry, 1900.

[Fossil only.]

Family GRYPHOCHITONIDAE Pilsbry, 1900.

Genus Permochiton Iredale and Hull, 1926 (without definition)
Permo chit on australianus Ire. and Hull, 1926.

Suborder PROTOCHITONINA Ashby, 1928.

Family PROTOCHITONIDAE Ashby, 1925.

Genus Protochiton Ashby, 1925.

Protochiton granulosus (Ashby and Torr, 1901).

Family ACANTHOCHITONIDAE Hedley, 1916.
Subfamily AFOSSOCHITONINAE Ashby, 1925.

Genus Afossochiton Ashby, 1925.

Afossochiton cudmorci Ashby, 1925.

A. rostratus (Ashby and Torr. 1901).

[Advanced.]

Subfamily ACANTHOCHITONINAE Ashby, 1925.

Genus Acantiiochiton Gray em., 1821.
Acanthochiton chapmani Ashby, 1925.

Subfamily CRYPTOPLACINAE Thiele, 1910.

Genus Cryptoplax Blainville, 1818.

Cryptoplax pritchardi Hall, 1905.

C. gatliffi Hall, 1905. j

[Primitive.] •

Suborder LEPIDOPLEURINA Thiele, 1910.

Family LEPIDOPLEURIDAE Pilsbry, 1892.

Genus Lepidoplerus Risso, 1826.

Lepidoplciirus magnogranifer Ashby, 1925.

Australian Fossil Polyplacophora.
[Advanced.]

Suborder CHITONINA Thiele, 1910.

Family CALLOCHITONIDAE Thiele, 1910.
Subfamily TRACHYDERMONINAE Thiele, 1910.

Genus Notochiton Thiele.

Notochiton mirandus Thiele. Recent.

N. hyadesi Rochebrune, 1889. Recent.

Genus Oochiton Ashby, n. gen.

Oochiton halli Ashby, n. sp.

Family MOPALIIDAE Pilsbry, 1892.

Genus Plaxiphora Gray, 1847.

Plaxiphora concentrica Ashby and Torr, 1901.

Family ISCHNOCHITONIDAE Pilsbry, 1892.
Subfamily ISCHNOCHITONINAE Pilsbry, 1892.
Genus Ischnochiton Gray, 1847.

Subgenus Heterozona (Cpr. MS.) Dali, 1878.
Ischnochiton ( Heterozona ) cariosus Pilsbry, 1892.

Subfamily CALLISTOPLACINAE Pilsbry, 1892.

Genus Callistochiton Carpenter, 1882.
Callisto chiton meridionalis Ashby, 1919.

Family CHITONIDAE Pilsbry, 1892.
Subfamily CHITONINAE Pilsbry, 1892.

Genus Chiton Linne, 1758.

Subgenus Rhyssoplax, Thiele, 1893.

Chiton ( Rhyssoplax ) fossicius Ashby and Torr, 1901.

Subfamily LIOLOPHURINAE Pilsbry, 1893.

Genus Lorica H. and A. Adams, 1852.

Lorica comprcssa Ashby and Torr, 1901.

L. comprcssa var. af finis Ashby and Torr, 1901.

L. endmorei , Ashby, 1925.

Genus Protolorica Ashby, 1925.
Protolorica atkinsoni Ashby, 1925.

Genus Loricella Pilsbry, 1893.

230 Edwin Ashby : Australian Fossil Polyphacophora.

Loricella gigantea Ashby and Torr, 1901.

L. paucipustulosa Ashby and Torr, 1901.

Subgenus Pseudoloricella Ashby, 1925.

Loricella ( Pseudoloricella ) sculpta Ashby, 1921.

Literature Cited.

Pilsbry, H. A. Man. Conch., vols. 14 and 15.

Thiele, J. Revision des Systems der Chitonen, vols. 1 and 2.

Iredale, T., and Hull, A. F. B. Monograph on Australian Lori-
cates. Roy. Zool. Soc. NSW.

Ashby, E., and Torr, W. G. Fossil Polyplacophora. Trans .
Roy. Soc. S. Anstr., xxv. (2).

Ashby, E. Monograph on Australian Fossil Polyplacophora.
Proc. Roy. Soc. Vic., n.s., xxxvii. (2).

EXPLANATION OF PLATE XXIV.

Fig. 1. — Oochiton Haiti, n. sp. Balcombe Bay, Vic. ; Balcombian.

Holotype, median valve, (a) side view, ornamenta-
tion and broken insertion plate, X 7; (Z?) upper side,
X 7. Nat. Mus. No. 13496.

Fig. 2. — O. halli, n. sp. Belmont, Vic. ; Balcombian. Para-
type, median valve, showing ornamentation and com-
plete insertion plate, X 6*5. Nat. Mus., No. 13497.

Fig. 3. — 0. halli, n. sp. Balcombe Bay, Vic. ; Balcombian. Para-
type, taken as type of tail valve, (a) posterior of valve
tilted upwards, to show truncated posterior, also orna-
mentation, X 6-5; ( b ) side view, showing complete in-
sertion plate, slit and truncated posterior, X 7. Nat.
Mus. No. 13494.

Fig. 4. — Notochiton mirandiis Thiele. Antarctica, dredged : Re-
cent. Cotype, anterior valve, side view showing inser-
tion plate and sculpture for comparison with Oochiton
halli, X 5. Ashby Coll.

Fig. 5. — N. mirandiis Thiele. Median valve, side view. X 5.
Ashby Coll.

Fig. 6. — N. mirandiis Thiele. Tail valve, (a) side view show-
ing sculpture, mucro and insertion plate, X 5; (Z?)
same valve, inside, showing teeth, X 7. Ashby Coll.

Fig. 7. — Ischnochiton (Heterozona) cariosus Pilsbry. Balcombe
Bay; Balcombian. Median valve, X 7.

Fig. 8. — Oochiton halli. Ashby. Balcombe Bay; Balcombian.

Paratype, here taken as type of anterior valve, (a)
showing sculpture and insertion plate, X 6; ( b ) side
view, showing anterior slope, sculpture and insertion
plate, X 12. Ashby Coll.

Fig. 9. — Loricella gigantea. Ashby and Torr. Table Cape, Tas.,
Lower Bed; Janjukian. Anterior valve showing shape
and sculpture, X 3*5. Nat. Mus. No. 13499.

Proc. R.S. Victoria, 41 (2), 1929. Plate XXIV.

JS. Ashby, photo,

Australian Fossil Polyplacophora.

§togal ^crietg of ffrlorm.

1928 .

patron :

HIS EXCELLENCY THE RIGHT HON. BARON SOMERS, K.C.M.G., D.S.O., M.C.

Jfrmbent :

Prof. VV. E. AGAR, M.A., D.Sc., F.R.S.

t8tcc-$rcsibents :

F. CHAPMAN, A.L.S
D. K PICKEN, M.A.

3jmt. ‘Cv ea surer :

N. A. ESSERMAN, B.Sc., A. Inst. P.

3Bmt. librarian :

F. A. CUDMORE.

(Secretary :

R. T. PATTON, B.Sc., D.I.C., M.F.

3W*w. (Secretary, Jttatltcmatical anb $h;ysical (Section:

J. S. ROGERS, B.A., M.Sc.

Council :

Prof. E. W. SKEATS, D.Sc., A R.C.Sc.,
F.G.S.

Prof. A. J. EWART. D.Sc., Ph.D., F.R.S.,
F.L.S.

Prof. T. H. LABY, M.A., Ph.D., Sc.D .,
F.Inst.P.

Assoc. Prof. H. S. SUMMERS, D.Sc.
Capt. J. K. DAVIS.

Assoc. Prof. YOUNG.

Prof. W, A. OSBORNE, M.B., B.Ch.,
D.Sc.

E. J. DUNN, F.G.S.

J. SHEPHARD.

W. RUSSELL GRIM WADE, B.Sc.

W. HEBER GREEN, D.Sc.

.J. M. BALDWIN, D.Sc.

Committees of the Council

publication Committee:

THE PRESIDENT.

THE HON. TREASURER
THE HON. SECRETARY.

Stjonornvj) JUtbitors :

C. A. LAMBERT.

J. SHEPHARD.

Igottorarj) Architect :

\Y. A. M. BLACKETT.

‘Crtistees :

PROF. SIR W. BALDWIN SPENCER, K.C.M.G., F.R.S
J. A. KERSHAW, F.E.S.

1929.

LIST OF MEMBERS

WITH THEIR YEAR OF JOINING.

[Members and Associates are requested to send immediate notice of any
change of address to the Hon. Secretary.]

Patron.

His Excellency, The Right Hon. Baron Somers, K.C.M.G.,
D.S.O., M.C.

Life Members.

Ewart, Prof. A. J., D.Sc., Ph.D., F.R.S., F.L.S., Univer- 1906
sity, Carlton, N.3.

Gregory, Prof. J. W., D.Sc., F.R.S., F.G.S., University, 1900
Glasgow.

Oliver, C. E., M.C.E., c/o J. E. Minifie, 12 Martin-street, 1878
Elwood, S.3.

Selby, G. W., Glenbrook-avenue, Malvern, S.E.5 1889

Ordinary Members.

Agar, Prof. W. E., M.A., D.Sc., F.R.S., University, Carl- 1920
ton, N.3.

Anderson, George, M.A., LL.M., M.Com., 222 Beacons- 1924
field-parade. Middle Park, S.C.6.

Aston, R. L., B.E., M.Sc., Trinity College, Parkville, N.3 1927

Austin, E. G., Boeri Yallock, Skipton 1922

Baldwin, J. M., M.A., D.Sc., Observatory, South Yarra, 1915
S/E. 1

Bale, W. M., F.R.M.S., 83 Walpole-street, Kew, E.4 .... 1887

Balfour, Lewis J., B.A., M.B., B.S., Burwood-road, Haw- 1892
thorn, E.3.

Baragwanath, W., Geological Survey Department, Trea- 1922
sury Gardens, Melbourne, C.2.

Barrett, A. O., 25 Orrong-road, Armadale, S.E.3 1908

Barrett, Sir J. W., K.B.E., C.M.G., M.D., M.S., 105 Col- 1910
lins-street, Melbourne, C.l.

Chapman, F., A.L.S., F.R.M.S., F.G.S., National 1902

Museum, Melbourne, C.l.

Cudmore, F. A., 12 Valley View-road, East Malvern, S.E.6 1920

Davis, Captain John King, “ St. Carols,” Caroline-street, 1920
South Yarra, S.E.l.

234

List of Members.

Dunn, E. J., F.G.S., “ Roseneath,” Pakington-street, Kew, 1893
E.4.

Dyason, E. C., R.Sc., B.M.E., 92 Queen-street, Melbourne, 1913
Cl.

Elliott, R. D., 395 Collins-street, Melbourne, C.l 1927

Esserman, N. A., B.Sc., A.lnst.P., Research Laboratories, 1923
Maribyrnong, W.3.

Gault, E. L., M.A., M.B., B.S., 4 Collins-street, Mel-
bourne, C.l.

Gepp, H. W., Kurrajong House, 175 Collins-street, Mel-
bourne. C.l.

Gilruth, J. A, D.V.Sc., M.R.C.V.S., F.R.S.E., 7 Clowes-
street, South Yarra, S.E.l.

Green, W. Heber, D.Sc., University, Carlton, N.3

Greenwood, Prof. J. N., D.Sc., University, Carlton, N.3. . .
Grimwade, W. Russell, B.Sc., 420 Flinders-lane, Mel-
bourne, C.l.

Hurst, W. W., B.Sc., Ph.D., Urquhart-street, Hawthorn, 1927

E. 2.

Jut son, J. T., B.Sc., LL.B., “ Darlington,” 9 Ivanhoe-par- 1902
ade, Ivanhoe, N.21.

Kelly, Bowes, Glenferrie-road, Malvern, S.E.4 1919

Kenyon, A. S., C.E., Lower Plenty-road, Heidelberg, N.22 1901

Kernot, Assoc. Prof. W. N., BCE, M.Mech.E., M Inst.C.E., 1906
University, Carlton, N.3.

Kershaw, J. A., F.E.S., National Museum, Melbourne. 1900

C. l.

Laby, Prof. T. H., M.A., Ph.D., Sc.D., F.Inst.P., Univer-
sity, Carlton, N.3.

Lewis, J. M., D.D.Sc., 44 Whitethorns,” Boundary-road,
Burwood, E.13.

Littlejohn, W. S., M.A., Scotch College, Hawthorn, E.2 . .

Lyle, Prof. Sir Thos. R., M.A., D.Sc., F.R.S., Irving-road,
Toorak, S.E.2.

MacCallum, Prof. Peter, M.C., M.A., M.Sc., M.B., Ch.B., 1925

D. P.H., University, Carlton, N.3.

Mahony, D. J., M.Sc., Geological Survey, Public Offices, 1904
Treasury Gardens, C.2.

Mann, S. F., Caramut, Victoria 1922

Masson, Prof. Sir David Orme, K.B.E., M.A., D.Sc., 1887

F. R.S.E., F.R.S., 14 William-street, Sth. Yarra,

S.E.l.

Merfield, C. J., F.R.A.S., Observatory, South Yarra, S.E.l 1913
Merfield, Z. A., F.R.A.S., University, Carlton, N.3 . . . . 1923

1915

1921

1920

1889

1899

1926

1909

1896

1928

1912

List of Members.

235

Michell, J. H. M.A., F.R.S., 52 Prospect Hill-road, 1900
Camberwell, E.6.

Millen, Senator J. D., 90 William-street, Melbourne, C.l 1920
Miller, Leo F., M Moonga/* Power-avenue, Malvern, S.E.4 1920
Miller, E. Studley, 396 Flinders-lane, Melbourne, C.l . . 1921

Monash, Lieut.-General Sir John, G.C.M.G., K.C.B., Doc. 1913
Eng., LL.D., State Electricity Commission, 22
William-street, Melbourne, C.l.

Mullett, H. A., B.Ag.Sc., Dept, of Agriculture, Melbourne, 1923
C.2.

Osborne, Prof. W. A., M.B., B.Ch., D.Sc., University, 1910
Carlton, N.3.

Patton, R. T., B.Sc., M.F., Hartley-ave., Caulfield, S.E.8 1922

Payne, Prof. H., MJnst.C.E., M.LMechE., University, 1910
Carlton, N.3.

Penfold, Dr. W. J., M.B., Alfred Hospital, Commercial- 1923
road, Prahran, S.l.

Petrie, A. H. K., B.Sc., University, Carlton, N.3 1925

Picken, D. K., M.A., Ormond College, Parkville, N.3 .... 1916

Piesse, E. L., 43 Sackville-street, Kew, E.4 1921

Pratt, Ambrose, M.A., 376 Flinders-lane, Melbourne, C.l 1918

Quayle, E. T., B.A., 27 Collins-street, Essendon, W.5 . . 1920

Rae, F. J., B.Sc., B.Ag.Sc., Botanic Gardens, South Yarra, 1927
S.E.l.

Reid, J. S., 498 Punt-road, South Yarra, S.E.l 1924

Rivett, Dr. A. C. D., M.A., D.Sc., Council for Scientific and 1911
Industrial Research, 314 Albert-street, East Mel-
bourne, C.2.

Rogers, J. Stanley, B.A., M.Sc., University, Carlton, N.3 . 1924

Ryan, Rev. Wilfrid, S.J., M.A., F.G.S., Newman College, 1926
Carlton, N.3.

Schlapp, H. H., 31 Queen-street, Melbourne, C.l 1906

Shephard, Tohn, “ Norwood,” South-road, Brighton Beach, 1894
S.5,

Shiliinglaw, Godfrey V., 64 Dandenong-road, Caulfield, 1925
S.E.7.

Singleton, F. A., M.Sc., University, Carlton, N.3 1917

Skeats, Prof. E. W., D.Sc., A.R.C.Sc.,F.G.S., University, 1905
Carlton, N.3.

Smith, B. A., M.C.E., Mutual Building, 395 Collins- 1924
street, Melbourne, C.l.

Spencer, Prof. Sir W. Baldwin, K.C.M.G., M.A., D.Sc., 1887
F.R.S., National Museum, Melbourne, C.l.

Stillwell, F. L., D.Sc., 44 Elphin-grove, Hawthorn, E.2 . . 1910

Summers, Associate Prof. H. S., D.Sc., University, Carl- 1902
ton, N.3.

23 6

List of Members.

Sweet, Georgina, D.Sc., Cliveden Mansions, Wellington- 1906
parade, East Melbourne, C.2.

Thirkell, Geo. Lancelot, B.Sc., 4 Grace-street, Malvern, 1922
S.E.4.

Thomas, Dr. D. J., M.D., 12 Collins-street, Melbourne, 1924

C.l.

Tiegs, O. W., D.Sc., University, Carlton, N.3 1925

Trinder, E. E., M.I.H.V.E., “ Ruzilma,” Orrong-grove, 1922
Caulfield, S.E.7.

Wadham, Prof. S. M., M.A., Agr.Dip., University, Carl- 1927
ton, N.3.

Walcott, R. H., Technological Museum, Melbourne, C.l . . 1897

Weber, E. K., 49 Armadale-street, Armadale, S.E.3 . . . . 1927

Wickens, C. H., F.LA., F.S.S., Commonwealth Statistician, 1923
Canberra, F.C.T.

"Woodruff, Prof. H. A., M.R.C.S., L.R.C.P., M.R.C.V.S., 1913
Veterinary School, Parkville, N.2.

Young, Assoc. Prof. W. J., D.Sc., University, Carlton, N.3 1923

Country Members.

Caddy, Dr. Arnold, 44 Chandpara,” Tylden, Vic 1924

Crawford, W., Gisborne, Vic 1920

Drevermann, A. C., Dookie Agricultural College, Dookie, 1914
Vic.

Easton, J. G., “ Kiewa,” Murphy-street, Bairnsdale, Vic. 1913

Harris, W. J., B.A., High School, Echuca, Vic 1914

Hart, T. S., M.A., B.C.E., School of Mines, Bairnsdale, 1894
Vic.

Hope, G. B., B.M.E., 44 Carrical,” Hermitage-road, New- 1918
town, Geelong, Vic.

James, A., B.A., M.Sc., High School, Colac, Vic 1917

Kitson, Sir Albert E., C.M.G., C.B.E., F.G.S., 75 Cornwall 1894
Gardens, Kensington, London, S.W.7, England.

Langford, W. G., M.Sc., B.M.E., 44 Vailala ,” Elizabeth- 1918
street, Gordon, Sydney, N.S.W.

Lea, A. M., F.E.S., 241 Young-street, N. Unley, S. Aus. 1909

Mackenzie, H. P., Engr. Commr. R.N.(Ret), Trawalla, 1924
Vic.

Parker, L. C., B.Sc., High School, Ballarat 1927

Sutton, J. W., 127 Doncaster-avenue, Kensington, Sydney, 1924
N.S.W.

237

List of Members .

Trehilcock, Captain R. E., M.C., Wellington-street, Kerang, 1921
Vic.

White, R. A., B.Sc., School of Mines, Bendigo, Vic 1918

Corresponding Member.

Lucas, A. H. S., M.A., B.Sc., Sydney Grammar School, 1895
Sydney, N.S.W.

Associates.

Albiston, H. E., M.V.Sc., Veterinary School, Parkville, 1925
N.2.

Allen, J. M., M.A., 41 Nirvana-avenue, East Malvern, 1924
S.E.5.

Allen. Miss N. C. B., B.Sc., University, Carlton, N.3 . . . . 1918

Archer, Howard R., B.Sc., c/o J. M. Moffatt, Faulkner- 1921
street, Armidale, N.S.W.

Ashton, H., ‘'The Sun,” Castlereagh-street, Sydney, 1911
N.S.W.

Bage, Mrs. Edward, “ Cranford,” 7 Gellibrand-street, 1906

Brisbane, Old.

Baker, F. H., 167 Hoddle-street, Richmond, E.l 1911

Barkley, H., Meteorological Bureau, Melbourne, N.3. . . 1910

Barnard, R. J. A., M.A., University, Carlton, N.3 1926

Bordeaux, E. F. J., G.M.V.C.,B. fes L., Mangalore-street, 1913
Flemington, W.2.

Breidahl, H., M.Sc., M.B., B.S., 23 Chatsworth-avenue, 1911
North Brighton, S.5.

Brodribb, N. Iv. S., Ordnance Factories, Maribyrnong, 1911
W.3.

Buchanan, Gwynneth, D.Sc., University, Carlton, N.3 . . . . 1921

Carter, A. A. C., “ Fairholm,” Threadneedle-street, Bal- 1927
wyn, E.8.

Chapman, W. D., M.C.E., “Hellas,” Heidelberg-road, 1927
Clifton Hill, N.8.

Chappie, Rev. E. H., The Manse, Warrigal-road, Oakleigh, 1919
S.E.12.

Cherry, R. O., M.Sc., Natural Philosophy Dept., Umver- 1927
sity, Carlton, N.3.

Clinton, H. F., Produce Office, 605 Flinders-street, Mel- 1920
bourne, C.l.

Collins, A. C., Public Works Department, Treasury Gar- 1928
dens, Melbourne, C.2.

Cook, G. A., M.Sc., B.M.E., IS Elphin-grove, Hawthorn, 1919
E.2.

238

List of Members .

Cookson, Miss I. C., B.Sc., 154 Power-street, Hawthorn, 1916

E.2.

Coulson, A. L., M.Sc., D.I.C., F.G.S., “ Finchley,” King- 1919
street, Elsternwick, S.4.

Cox, E. H., Literary Staff, M The Argus/’ Elizabeth-street, 1924
Melbourne, C.l.

Crespin, Miss I., B.A., 67 Studley Park-road, Kew, E.4 . . 1919

Dare, J. H., B.Sc., State School, Brunswick, N.10 . . . . 1917
Deane, Cedric, “ Cloyne,” State-street, Malvern, S.E.4 . . 1923

Feely, J. A., Observatory, South Yarra, S.E.l 1924

Fenner, C., D.Sc., Education Department, Flinders-street, 1913
Adelaide, S.A.

Ferguson, W. H., 37 Brinsley-road, E. Camberwell, E.6 . . 1894

Finney, J. M., 36 Toorak-road, Malvern, S.E.4 1925

Flecker, Dr. H., 71 Collins-street, Melbourne, C.l , . 1922

Gabriel, C. J., 293 Victoria-street, Abbotsford, N.9 . . . . 1908

Hardy, A. D., F.L.S., Forests Department, Melbourne, 1903
C.2.

Hartung, Prof. E. J., D.Sc., University, Carlton, N.3. 1923

Hauser, H. B., M.Sc., Geology School, University, Carl- 1919
ton, N.3.

Hercus, E. O., D.Sc., A.Inst.P., University, Carlton, N.3 1923

Heslop, G. G., D.V.Sc., 7 Hudson-street, Caulfield, S.E.7. 1923

Hill, Gerald F., Council for Scientific and Industrial Re- 1924
search, 314 Albert-street, East Melbourne, C.2.

Hills, E. S., B.Sc., Geology School, University, Carlton, 1928
N.3.

Holmes, W. M., M.A., B.Sc., Observatory, South Yarra, 1913
S.E.l.

Howitt, A. M., Department of Mines, Melbourne, C.2 ... 1910

Jack, A. K., M.Sc., 49 Aroona-road, Caulfield, S.E.7 .... 1913

Jessep, A. W., B.Sc., M.Ag.Sc., Dip. Ed., Horticultural 1927
Gardens, Burnley, E.l.

Jona, J. Leon, M.D., B.S., D.Sc., “ Hazelmere,” 104 1914
Wattle Tree-road, Malvern, S.E.4.

Kannuluik, W. G., M.Sc., Natural Philosophy Dept., Uni- 1927
versity, Carlton, N.3.

Keartland, Miss B., M.Sc., Cramer-street, Preston, N.18 1919

Keble, R. A., National Museum, Melbourne, C.l 1911

Lambert, C. A., Bank of N.S.W., Melbourne, C.l 1919

Lambert, J. R., 135 Drummond-street, Carlton, N.3 .... 1928

Leslie, J. R., 99 Toorak-road, South Yarra, S.E.l .. .. 1923

Lewis, Miss R. M., 52 Campbell-road, East Kew, E.4 .... 1925

List of Members. 239

Llewelyn, Miss Sybil, M.A., M.Sc., St. Quentin, Broadway, 1928
El wood, S.3.

Luly, W. H., Department of Lands, Public Offices, Mel- 1896
bourne, C.2.

Macdonald, B. E., Dairy Export Branch, Rialto, Collins- 1920

street, Melbourne, C.l.

Mackenzie, G., 1 High-street, Prahran, S.l 1907

Maclean, C. W., 56 Cole-street, Elsternwick, S.4 1879

McLennan, Ethel, D.Sc., University, Carlton, N.3. . . 1915

Melhuish, T, D’A., M.Sc., Adelaide Chemical and Fertilizer 1919
Co., Currie-street, Adelaide, S.A.

Mollison, Miss E., M.Sc., Royal-crescent, Camberwell, 1915
E.6.

Moore, F. E., O.B.E., Chief Electrical Engineer’s Branch, 1920
P.M.G.’s Department, Treasury Gardens, Mel-
bourne, C.2.

Morris, P. F., National Herbarium, South Yarra, S.E.l. . . 1922

Nelson, Miss E. A., M.A., M.Sc., University, Carlton, N.3 1924

Newman, B. W., Meteorological Offices, Melbourne, C.l 1927
Oke, C., 56 Chaucer-street, St. Kilda, S.2 1922

Orr, D., B.Sc., 860 Mount Alexander-road, Essendon, W.5 1927

Parr, W. J., 17 Bokhara-road, Caulfield, S.E.8 1927

Pern, Dr. Sydney, M.R.C.S., L.R.C.P., 16 Collins-street, 1920
Melbourne, C.l.

Petersen, Miss K., B.Sc., 56 Berkeley-street, Hawthorn, 1919
E.2.

Pretty, R. B., M.Sc., Technical School, Warrnambool, Vic. 1922

Raff, Miss J. W., M.Sc., F.E.S., University, Carlton, N.3 1910

Richardson, Sidney C., 2 Geelong-road, Footscray, W.ll 1923

Rosenthal, Newman, H., B.A., B.Sc., 11 Wimbledon-ave., 1921
Elwood, S.3.

Ross, Miss D. J., M.Sc., Merton Hall, Anderson-street, 1924
South Yarra, S.E.L

Sayce, E. L., B.Sc., A. Inst. P., Research Laboratories, Mari- 1924
byrnong, W.3.

Sharman, P. J., M.Sc., “ Glenalvie,” 9 Daphne-street, Can- 1916
terbury, E.7.

Shearer, J., B.Sc., Queen’s College, Carlton, N.3 1924

Sherrard, Mrs. H. M., M.Sc., “ Balnagowan,” Bromley- 1918
ave., Cremorne, N.S.W.

Shiels, D. O., M.Sc., Ph.D., Chemistry School, Univer- 1927
sity, Carlton, N.3.

Simpson, H. W. C., 64 Dandenong-road, Caulfield North, 1927
S.E.7.

240

List of Members.

Smith, J. A., 25 Collins-place, Melbourne, C.l

Stickland, John, 433 Brunswick-street, Fitzroy, N.6

Sutton, C. S., M.B., B.S., Education Department, Mel-
bourne, C.2.

Thomas, R. G., B.Ag.Sc., c/o Dr. Thomas, Northam, W.A.

Thompson, Mrs. G. R., 26 Fawkner-street, St. Kilda, S.2

Thorn, Win, 37 Chrystobel-crescent, Hawthorn, E.2 ....

1 raill, J. C., B.A., B.C.E., 630 St. Kilda-road, Melbourne,
S.C.3.

Treloar, IT. M., Meteorological Offices, Melbourne, C.l . .

Triidinger, W., 27 Gerald-street, Murrumbeena, S.E.9 ..

Turner, A. H., M.Sc., Natural Philosophy Dept., Univer-
sity, Carlton, N.3.

Turner, A. W., M.V.Sc., Veterinary School, Parkville,
N.2.

Wilcock, E. L., B.Sc., University High School, Carlton, N.3.

Williamson, H. B., F.L.S., “The Grange,” 231 Waverley-
road, East Malvern, S.E.5.

Wilson, F. E„ F.E.S., 22 Ferncrofl-avenue, E. Malvern,
S.E.5.

Wilson, Major H. W„ O.B.E., M.C., C.de G., B.Sc., 630
Inkerman-road, Caulfield, S.E.7.

Withers, R. B., 10 Nicholson-street, Coburg, N.13

Woodward, J. H., Queen’s Buildings, 1 Rathdown street,
Carlton, N.3.

1905

1922

1908

1922

1907

1903

1922

1918

1927

1925

1925
1919

1921

1923

1926
1903

INDEX.

The names of the new genera and species are printed in italics.

Acrothele bulboides, 207. ^

Age and Area Hypothesis, 60.

Aglaophenia acanthostoma, 25 ; laxa,
25.

Agnostus chinensis, 208.

Air Pressures, Forecasting from
Tropical, 160.

Allan, Frances E., 165.

Anchylodiscus gadopsis, 45.

Ancyrocephalus bassensis, 51.

Annual Fluctuations of Temperature,
32.

Ashby, Edwin, 220.

Aura Dacite, 142.

Australia, Flora of, 59.

Australian Fossil Polyplacophora, 220 ;
Termites, 85.

Barrett, A. O., 32.

Basalts, 145 ; Footscray. 148 ; Kyneton,
153 ; Malmsbury, 145.

Bathyuriscus nitidus , 210 ; olenelloidcs,
210 ; saint-smithii, 209.

Bingham, R., 55.

Blue Hills, Geology and Palaeonto-
graphy of, 176.

Bothriolepis gippdandiensis , 195.

Boundary of two liquids, Solitary
Waves at, 165.

Brachiopods from the Mount Isa Dis-
trict, 206.

Building Stones of Victoria, The
Igneous Rocks, 121 ; Basalts, 145 ;
Chemical Analyses, 156 ; Dacite, 142 ;
Granites, 123 ; Porphyry, 144 ; Sum-
mary, 156; Tests, 155.

Campanulina humilis, 8.

Cape Woolamai Granite, 130,

Capulus melbournensis , 218.

Cathedral Beds, 179.

Cathedral Range, Geology and Palaeon-
tography of, 176,

Chapman, Fredk., 202, 206, 218.

Chitons, Fossil, 220.

Colquhoun Granite, 138.

Common Boundary of two liquids,
Solitary Waves at, 165.

Cowry, Flanged, 202.

Dacite, Aura, 142.

Darwin Air Pressures, Forecasting
from, 160.

Devonian Rocks of Tabberabbera Dis-
trict, 97.

Dikelocephalus dunstani, 213.

Diurnal Fluctuations of Temperature,
32.

Drepanotermes, 88.

Dromana Granite, 136.

Dynamena quadridentata, 23.

Eoctenodus ■ microsonia , 193.

Endocrypta huntsman i, 1.

Eutermes mareebensis, 90.

Ewart, Alfred .1-, 69.

13a

Filellum serratum, 4 .

Fishes, Trematode Parasites on, 45.

Flanged Cowry, 202.

Flora of Australia, 59.

Fluctuations of Temperature in In-
terior of a Large Tree, 32.

Footscray Basalt, 148.

Forecasting from Tropical Air Pres-
sures, 160.

Fossil Polyplacophora, 220 ; Classifica-
tion, 228 ; Discussion on Classifica-
tion, 226 ; Systematic Description,
220 .

Gabo Island Granite, 132.

Geology of the Cethedral Range and the
Blue Hills, 176; Cathedral Beds, 179;
(?) Lower Devonian, 182; Petro-
graphy, 188 ; Physical Features, 178 ;
Pleistocene (?), 186; Recent, 186;
Summary, 176, 187 ; Upper Devonian,
183; Yeringian (Upper Silurian),
178.

Gippsland, North-Western, 176.

Glass, Reflection of X-rays from, 55.

Goongarrie, Lake, 63.

Granites, 123 ; Cape Woolamai, 130 ;
Colquhoun, 138 ; Dromana, 136 ; Gabo
Island, 132 ; Harcourt, 124 ; Orbost,
133; Trawool. 135; Tynong, 139;
Wangaratta, 127.

Grooved Rocks, 63> 66.

Halecium corrugatissinium, 7 ; expan-
suvis 7 ; flexile, 6 ; ienticulare, 6.

Hamitermes tamminensis, 88 ; westra-
liensis, 85.

Harcourt Granite, 124.

Hebella calcarata, 4; corrugata, 4.

Hill, Gerald F., 85.

Hills, Edwin S.. 176.

Holonema cf. rugosum, 197.

Hughes, Winifred Kent, 45.

Iiydroida, New Zealand, 1.

Igneous Rocks, Building Stones of Vic-
toria, 121.

Ischnochiton (Hcterozona) cariosus,
225.

Isoptera, 85.

Jutson, J. T., 63.

Kyneton Basalt, 153.

Laby, T. H., 55.

Lake Goongarrie, W.A., 63.

Lineolaria flexuosa, 8.

Lingulella marcia, var. templetonensiSt
207.

List of Members, 233.

Long Range Rainfall Forecasting, 160.

Lorica compressa, 220 ; compressa
var. affinis, 220; cudmorei, 221.

Loricella gigantea, 221.

242

Index .

Macrophylla, 48 ; antarctica, 48.

Malms bury Basalt, 145.

Marjumia conspicabilis, 212 ; elegans,
212 ; mile Hi, 211.

Melnerny, Kathleen, 121.

Members, List of, 233.

Milcsia templetonensis, 214.

Mirotcrmes argutus, 92 ; insitivus, 94 ;
jarmuranus , 93.

Mount Isa District, N.W. Queensland,
206.

Naturalized Aliens of Victoria, 59.

New Species of Capulus, 218.

New Zealand Hydroida, 1.

Noith Gippsland, Victoria, 97.

Notochiton mirandus, 225,

Obelia australis, 2 ; geniculata, 2.

Octobothrium thy rites, 49.

Older Palaeozoic Rocks of Tabberab-
bera District, 97.

Obckiton, 221 ; halli, 222.

Orbost Granite, 133.

Orthopyxis crenata, 3 ; delicata, 3 ;
jonnosa, 2.

Palaoontography of the Cathedral
Range and the Blue Hills, 176, 193 :
Silurian, 179, 193 ; Upper Devonian,
185, 193.

Palliocrypraca gastroplax, 202.

Parasites, Trcmatode, 45.

Pedestal Rocks, Miniature, 63, 72.

Pennaria australis, 1.

Perisiphonia quadriscriala, 4.

Pitted Rocks, 63, 69.

Plumularia hyalina, 24 ; pulchclla, 24 ;
setacea, 24 ; sotacooides, 24.

Polyplacophora, Fossil, 220,

Porphyry, Tallangatta, 144.

Pterygotus Carapace, 218.

Quartz, Reflection of X-rays from, 53.

Quayle, E. T., 160.

Queensland, Mount Isa District, 206.

Rainfall Forecasting, Long Range,
160.

Reflection of X-rays, 55.

Rocks, Grooved, Pitted, ami Pedestal,
63.

Saaba ( 7) scandens, 2.

Selaginopsis monilifera, 9.

Sertularella, 9 ; crassiuseula, 13 ; fusca,
13 ; irregularis , 13 ; macrogona, 11 ;
vrocera, 11 ; pygmaea, 10 ; rentoni,
10 ; robusta, 16 ; robusta var. flucti-
culata, 18 ; robusta var. quasiplana,
18 ; simplex, 15 ; simplex-robusta
group, 15.

Sertularia bispinosa, 22 : divergens, 23 ;
episcopus, 22 ; fasciculata, 22 ; mini-
ma, 23 ; trispinosa, 22 ; trispinosa
var. inarmata, 22 ; unguiculata, 23.

Shearer, J.. 55.

Silicularia bilabiata, 3 ; campanularia,
4.

Skeats, Ernest W., 97.

Solitary Waves at the Common Boun-
dary of two liquids, 165.

Squalonehocotyle antarctica, 46.

Stereotheca elongata, 23.

Synthecium patulum, 9.

Tabberabbera District, Devonian and
Older Palaeozoic Rocks of, 97 ;
Igneous Rocks, 113; Location, 98;
Middle Devonian, 105 ; Physical
Features, 99; Silurian ( Yeringian) ,
102; Summary, 118; Tectonic Move-
ments and Structures, 101 ; Uncon-
formity between M. and U. Dev-
o7i ian, 111; Upper Devonian, 106;
Upper Ordovician, 101.

Taggerty, Fish and Plant Remains
from, 193.

Tallangatta Porphyry, 144.

Temperature Fluctuations in Interior
of a Large Tree, 32.

Termites, Australian, 85.

Thecocarpus formosus var. inarmatus,
26.

Thecocaulus heterogona, 25 ; minutus,

25 .

Thuiaria buski, 19 ; buski var. tenuis-
sima, 20 ; farquhari, 19 ; spiralis, 21.

Thyroseyphus simplex, 8.

Trawool Granite, 135.

Trebileoek, R. E., 1.

Tree, Fluctuations of Temperature in
Interior of, 32.

Trematode Parasites, 45.

Trilobites from the Mount Isa District,
206.

Tropical Air Pressures, Rainfall Fore-
casting from, 160.

Tynong Granite, 139.

Victoria, Building Stones of, 121 ;
Naturalized Aliens of, 59 ; Rocks of
Tabberabbera District, 97.

Victorian Fishes, Trematode Parasites

on, 45.

Wangaratta Granite, 127.

Waves, Solitary, 165.

Western Australia, Lake Goongarrie,
63.

X-rays, Reflection of, 55.

END OF VOLUME XLI.

[Part II. Published 11th April, 1929.]

S/3AI

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Proceedings of the Royal Society of Victoria. New series"

See other formats