Memoirs of the
Queensland Museum.

VOL. XI. PART IIL

ISSUED 17th MARCH, 1939.

EDITED BY THE DIRECTOR, H. A. LONGMAN.

rai :

ISSUED BY THE AUTHORITY OF THE CHIEF SECRETARY FOR QUEENSLAND,
THE HON. W. FORGAN SMITH, LL.D.

MEMOIRS OF THE QUEENSLAND MUSEUM, Vou. XI, Parr IU.

THE SKULL OF AN AUSTRALIAN ABORIGINAL
FOUND AT STRADBROKE. ISLAND, QUEENS-
LAND.

By JosEPH L. SHELLSHEAR, UNIVERSITY OF SYDNEY.
(Text-figures 1 to 7 and Plates XV—XVIII.)

The skull, No. Q.E. 14/561, has been known as the Wynnum skull because
the records show that it was brought to the Museum from Wynnum, a suburb of
Brisbane. A search through official correspondence has, however, brought to light a
letter which reveals that it was found on the beach at Jumpinpin on the east side
of Stradbroke Island on the 28th May, 1914. Further information is contained in the
letter that 20 years previously there were five graves in the vicinity of where the
skull was found, but the sea broke over this part some years ago and washed the
graves away, and made the channel at Jumpinpin, dividing Stradbroke Island.

A brief description of the skull was given by Longman (1918) in which he
referred to its relatively large proportions for an Australian aboriginal. The skull
is, however, definitely that of an Australian aboriginal and no further description
would be warranted were it not for the fact that it has undergone a certain degree of
mineralisation.

The discoveries of the Talgai skull and, later, of the Cohuna skull—both of
which are clearly Australian—vaguely suggest from their state of mineralisation,
and from what we know of their sites of discovery, that the aboriginal was an
inhabitant of the continent for a.long period of time. The geological age of the
Talgai skull is somewhat uncertain ; but the considered opinion of 8. A. Smith (1918),
derived from the observations of Edgeworth David, is that “the claim to high
geological antiquity—the assignation of the fossil to the Pleistocene—is very strongly
supported, and may be regarded as established.” The geological evidence for the
age of the Cohuna skull has not been published but it would appear to be of later
date than the Talgai skull.

Further evidence of the antiquity of man in Australia is provided by the
excavations of Hale and Tindale (1930) in the Lower Murray Valley. These authors,
by employing accurate methods of excavation, have established definite cultural
layers in which human remains were found ; and have given. to us for the first time
a foundation on which to build up cultural sequences and to determine relative age
groupings.

M

170° MEMOIRS OF THE QUEENSLAND MUSEUM.

It is desirable, in the light of these findings, to place on record the morphological
features of any human material which shows evidence of mineralisation in order that
such information may be used for the purposes of confirmation and comparison with
material which may be found subsequently.

Method.—The description of the skull is dealt with, firstly, by making an
accurate survey of the various normae using the Frankfort plane as the base (figs.
1. to 6); secondly, by constructing a centimetre-contoured map on the Frankfort
plane in order to give expression to the various slopes of the skull (fig. 7); and,
thirdly, by the use of photographs to illustrate the non-metrical features (pl. xv,
figs. 8 to 11, and pl. xvi, figs. 12 to 14). The photographs were taken by Mr. L.
Schaeffer, of the Department of Anatomy, and to him I express my thanks. A brief
description of the endocranial cast is also given.

By the use of the above methods the written description can be considerably
curtailed. Detailed measurements of the skull are not given because those who find
scientific satisfaction in their use can read off any desired measurements from the
surveys. Three measurements are, however, of interest because they are unusual
in the skull of the Australian aboriginal. The maximum length is 202 mms., the
maximum breadth 140 mms., and the minimum frontal width 105mms. The length
is the same as that found by Duckworth (1894) in No. 2101 of his series ; it is 2 mms.
less than the skull recorded by Miklucho-Maclay (1883), and is only 2 mms. more than
many of the skulls recorded by Turner (1884) and Klaatsch (1908). The maximum
breadth and minimum frontal breadth are also above the average for the Australian.
The skull is also unusually thick.

The Cranium. Norma verticalis (figs. 1, 7, and 8).—The skull, seen from
above, has the form of a long ovoid, with the widest part of the skull situated 20
mms. posterior to the level of the external auditory meatus. Anteriorly the general
outline is more sharply curved than it is posteriorly. The parietal eminence is not
well marked on account of the general.flatness of the lateral walls of the cranium.
The slopes of the frontal region are uniform ; and the frontal eminences have become
smoothed out in the general outline. A typical Australian form is thus present in
conjunction with a greater average size, which, however, is due is some measure to
the excessive thickness of the skull.

Anteriorly the arci superciliares form two rounded projections extending
from the midline to a point just lateral to the groove for the supraorbital vessels
and nerve. Medially a well marked fossa supraglabellaris is present, whilst laterally
the arci are smoothed off into the supraorbital trigones (fig. 3d). Remaining traces
of the coronal and sagittal sutures permit the identification of bregma.

THE SKULL OF AN AUSTRALIAN ABORIGINAL. 171

Norma occipitalis (figs. 2 and pl. xv, fig. 9).—Viewed from behind the general
contour presents a typical Australian facies. Laterally the side walls are flat and
slope outwards very slightly as they extend from the mastoid processes to the parietal
eminences, The superior surface, extending from the parietal eminence to the middle
line, is flat to slightly concave. There is a slightly raised and rounded keel in the
middle line. The whole appearance is very similar to that of the Australian depicted
by Martin in fig. 370 (1928).

The most salient feature of the posterior region is the well developed torus
occipitalis transversus. ‘The torus is 20 mms. wide and extends laterally on either side
for a distance of 25 mms. as a prominent and uniform ridge. There is no external
occipital protruberance, although there is a small roughened ridge passing downwards
on the nuchal surface from its midpoint. The torus is not symmetrical. On the left
side it forms a fuller curve than on the right side ; and therefore extends higher into
the supraoccipital bone. At a distance of 20 mms. from the middle line the torus
resolves itself into two ridges, an upper and a lower. The upper ridge represents
the linea nuchae suprema and the lower ridge the linea nuchae superior. The condition
is very similar to that described by Burkitt and Hunter (1922) in skull 792 of the
Sydney University series ; but the torus is more pronounced in the Stradbroke skull.
The upper ridge fades away anteriorly in a large flat triangular area situated behind
the superior temporal ridge on the postero-inferior angle of the parietal bone. The
lower ridge retains its prominence and passes uninterruptedly into the mastoidal
crest. The region between the torus occipitalis and lambda is raised into a planum
occipitale, which is a common feature of the Australian skull.

The sutures in the region of asterion are still open; but midway between
asterion and lambda the upper part of the lambdoidal suture becomes obliterated.
The condition of the closure of the sutures indicates that the age of the individual
was between 60 and 70, basing the estimation on the work of Wingate Todd and
Lyon (1925).

On the inner surface of the occipital region the groove for the transverse
sinus is placed at a lower level than the torus occipitalis (fig. 5)—a condition commonly
found in the skull of the Australian—Klaatsch (1908), Burkitt and Hunter (1922).

Norma frontalis (fig. 3 and pl. xv, fig. 10).—The whole of the facial skeleton
is missing. Seen from the front the most salient features are the form of the supra-
orbital region and the width of the anterior frontal region. The minimum frontal
diameter measures 105 mms., which is unusually great for the Australian.

The supraorbital region shows a moderate degree of prominence in well
developed superciliary eminences. It conforms to type 11 of Cunningham (1908) in

172 MEMOIRS OF THE QUEENSLAND MUSEUM.

which the superciliary ridge coalesces with the part of the supraorbital ridge which
lies medial to the supraorbital notch. On the outer side of the notch the superciliary
eminence or ridge (b) extends outwards and fades into the flattened supraorbital
trigone (d). On the inner side the superciliary eminence turns sharply under the
glabellar region so that there must have been in this skull a very depressed root of the
nose. Unfortunately there is no trace of either a fronto-nasal or fronto-maxillary
suture. The glabellar region (g) is slightly depressed and extends upwards into a
well marked fossa supraglabellaris (s). |

The supraorbital region in this skull is almost identical with that of skull
340 of the University of Sydney series. There is, however, a significant difference
between the Stradbroke skull and those which have been sectioned in the
University of Sydney. Cunningham (1908), in discussing the relation of the
superciliary eminence to the frontal air sinuses, referred to the fact that the extent
of the air sinuses appeared to have no relation with the size of the superciliary
eminence ; and, further, referred to the observation of Schwalbe (1901), that in the
Neanderthal cranium the air sinuses lie well back and that a thick layer of condensed -
bone forms their anterior wall. Burkitt and Hunter (1922) found the maximum
thickness of this condensed outer table to be 7-5 mms. at the glabella in skull 792.

In the Stradbroke skull the air sinuses are well developed (figs. 6 and 14) ;
they are practically coincident in their extent with the superciliary eminences and
extend also to the roof of the orbit; but they differ from the other skulls in the
Sydney collection in that the condensed outer table only measures 2 mms. as
contrasted with 7-5 mms. in skull 792. .

Norma lateralis (fig. 4 and pl. xv, fig. 11, pl. xvi, fig. 13).—The prominences
produced by the torus occipitalis and arcus superciliaris are well shown when seen
from the side. The torus occipitalis is clearly defined by the depressed nuchal area
below and by the fossa supra-toralis above. The fossa supra-toralis is somewhat
accentuated by the flattened surface of the supraoccipital region as high as lambda.

Although the mastoid processes are damaged on both sides it can be seen that
they are of small size. On the left side, where only the tip of the process is lost, the
mastoid process appears to confirm the opinion of Klaatsch (1908) that it is only
the enlargement of the crista mastoidea (fig. 13). The crista mastoidea is continuous
with the superior nuchal line and is here represented by a well marked ridge. The
supramastoid crest—the continuation of the upper border of the zygomatic process
—is separated from the mastoid crest by an interval of 30 mms. In many of the
Australian skulls this interval is occupied by a furrow which passes to the postero-
inferior angle of the parietal bone. In the Stradbroke skull there is no furrow on the

THE SKULL OF AN AUSTRALIAN ABORIGINAL. 173

petrous part of the temporal bone; but there is a definite depression where the
anterior part of the postero-inferior angle of the parietal bone fits into the petro-
squamosal angle. The region of the mastoid process conforms to the Australian

type.

The inferior temporal crest passes continuously from the external angular
process to the suprameatal crest. It is well defined throughout with only slight
deformation where it passes the coronal suture. The temporal region bounded by
this line is relatively flat. -Anteriorly it is somewhat fuller than in the majority of
Australian skulls, in harmony with the fact that its minimum frontal width is 105

mms.

The area of the skull above the temporal crest forms a continuous surface
extending from the arcus superciliaris anteriorly to the torus occipitalis posteriorly.
It is divisible into frontal, parietal and occipital sections. The frontal portion is
somewhat full and rounded as far as the coronal suture ; so that the general form
is smoothed off and there is no frontal eminence. In No. 752 and many other
Australian skulls the frontal region immediately anterior to the coronal suture is
depressed and flattened ; and in the majority of the Australian endocranial casts
which I have examined there is a corresponding precentral depression of the frontal
lobe. The precentral depression is a marked feature in the endocranial cast of the
Wynnum skull. The full and rounded frontal region externally associated with a
depressed frontal lobe internally is one of the most interesting morphological features
of the skull. In the midline section of the skull (fig. 6) the bone is unusually thick ;
at the torus occipitalis it measures 15 mms. ; above this it thins for a short distance
to 7 mms., increasing again in the region of lambda—where the sutures are completely
closed—to 14 mms. ; anterior to this the cranium in the mid-parietal region thins to
9 mms. to widen out again at bregma to 12 mms. ; and in front of this the thickness
in the frontal region increases to 13 mms. The area of the frontal bone in which this
thickness occurs extends laterally to include the whole of the frontal vault medial
to the temporal crest. This thickening produces a flattened area of the inner table
of the frontal bone which causes the precentral depression of the endocranial cast.
The thickness of the skull exceeds that of any Australian skull which I have seen,
and is slightly greater than that of the Piltdown fragments as determined from the
casts.

Figure 15 shows the X-ray picture of the left side of the-skull, for which
I am indebted to my brother Dr. K. E. Shellshear. It shows two interesting
stages in the age changes ‘of the human skull. Harris (1933) has shown that
after fifty the venous sinuses in the diploe gradually disappear, that new bone

174 MEMOIRS OF THE QUEENSLAND MUSEUM,

is deposited in the diploe, and that the radiographic appearances change. Further, he
refers to the laying down of dense new bone deposited between the dura and the
inner table in old men at the site where there is atrophy of the frontal lobes. -Now,
in the radiographic picture of this skull the pattern of the diploic veins is well shown
over the parietal area and over the lower part of the frontal area ; but at the borders
of the thickened inner table of the frontal bone the diploic veins are disappearing
and there are none over the thickness area itself. In dealing with the endocranial
cast this question will be further discussed.

The parietal portion of the superior surface shows a depressed surface
immediately posterior to the coronal elevation. This depression on the surface
corresponds to the post-central area of ill-fillimg shown on the endocranial cast ;
on this, however, it is more marked because the bone is somewhat thinner over the
parietal eminence. .

Norma basilaris (fig. 5 and pl. xvi, fig. 12)—The greater part of the base of the
skull is missing. It is sufficiently described in the figures referred to above.

The endocranial cast (pl. xviii, figs. 16 to 19).-In the study of this cast I
have the advantage of having for comparison a series of 25 casts of Australian skulls,
as well as a large series of Chinese and most of the prehistoric casts of fossil skulls.
The importance of the study of the endocranial cast is emphasised in the work of
Elliot Smith ; and furthermore he has simplified the task of their examination by
clearly setting out the principles involved. He writes (1926): “In approaching
the study of the cast of the cranial cavity and attempting from it to interpret the nature
of the brain that originally occupied a particular skull, our aims to-day are of a different
nature than those of our predecessors. It is not so much the attempt to identify
certain definite fissures and convolutions of the brain, as to determine the relative
state of development of different functional areas of the brain. ...... Hence the
chief object is to base our conclusions not so much upon the size of any particular
brain, as upon the relative size of those particular regions which are of significance in
phylogenetics.”

The distinctive features which distinguish the endocranial cast of Homo sapiens
from those of lower types of human species are so clearly seen in this cast that no
discussion is necessary on that aspect of the interpretation ; and so we may confine
our attention to those features which distinguish the cast as Australian.

In prosecuting the task of determining the difference in the brain of race, I
have collected the endocranial casts of 35 Chinese skulls, 25 Australian aboriginal
skulls, and individual casts of other races. A study of this material makes it clear

THE SKULL OF AN. AUSTRALIAN ABORIGINAL. 175

that the features of the casts indicating racial difference enable one to distinguish
racial difference with even more facility than the skulls themselves can be.
distinguished. The Chinese cast stands out with its large rounded frontal lobes
bearing a foreshortened appearance, with its ‘salient parietal eminences, its flattened
and foreshortened occipital region, and finally with its cerebellum extending nearly
as far posteriorly as the occipital pole of the cerebrum. Whereas, the Australian
cast is recognised, amongst other things, by its narrowness accentuating its height,
by its narrow and ill developed frontal lobes, and by the small amount of overhang
of the cerebellum by the tentorial surface of the cerebrum. 5

The general form of the Stradbroke cast conforms to the Australian type.
It is dolichocephalic, but the type of dolichocephaly differs from the European type
in that the occipital region looks diminutive. Looked at from above the edges of
the cerebellum project beyond the contour of the parieto-occipital region of the
cerebrum laterally. The angle formed by the anterior border of the cerebellum with
the inferior border of the temporal lobe is almost 75 degrees and indicates lack of
development of the post-temporal (T) and parieto-occipital regions (PO). Immediately
anterior to the parietal eminence there is a well marked post-coronal depression
(PCD) corresponding with the postcentral and supramarginal regions of the brain.
The post-coronal depression is a salient feature in the Australian casts, whereas in the
Chinese casts the post-coronal region is almost without-exception full and rounded.
This post-coronal depression is well seen in the endocranial casts of Homo rhodestensis
and Homo soloensis and is associated with lack of development of the parietal areas
of the brain. But, in attributing the depression to lack of development, the work
of Keith (1931) on the effect of the streams and pools of cerebro-spinal fluid on the
form of the cast must not be overlooked. Whatever the cause the presence of a post-
coronal depression is common in the endocranial cast of the Australian. Immediately
anterior to the post-coronal depression there is a fairly prominent coronal elevation
(C) which widens out superiorly to blend with a wide sagittal elevation. In the majority
of European and Chinese casts the transverse contour of the parietal region displays
and even dome-shaped curve. Any irregularity in the curve above is limited to the
size of the arachnoid granulations. In the endocranial cast of the aboriginal,
however, the general transverse outline is more rugged. Passing along the vertex
antero-posteriorly—and even extending forward into the frontal region—is a wide
raised plateau lifted up from the‘ general contour. The post-coronal depression
ceases at its outer border superiorly and becomes continuous with a depression (P)
in the parietal region which lies on the outer side of the plateau as far posteriorly
as the parieto-occipital region. On the midline plateau there may be secondary

176 MEMOIRS OF THE QUEENSLAND MUSEUM.

localised elevations corresponding to those seen in the European and Chinese casts
produced by the arachnoid granulations.

The Stradbroke cast clearly demonstrates the features described above.
The European and Chinese casts indicate the development of a higher type of brain
in the filling up of the sagittal parietal depressions (P), bringing, as it were, the general
outline to the level of the midline sagittal elevation.

The frontal lobe, looked at from above and anteriorly, shows a depressed area
(F) lying in the angle formed by the coronal and mid-sagittal elevations. This
depression of the frontal lobe is seen in most of the Australian casts and is very similar
in appearance with the frontal depression seen'in the cast of Homo rhodesiensis. In
the Stradbroke cast, however, the depression is somewhat unusual in that markings
for the sulci and gyri of the frontal lobe are not present over the depressed area. In
most casts of primitive and modern races, whether a depression is present or not,
the gyri and sulci of the frontal lobe leave some impression on the frontal lobe.

Harris (1933) has pointed out that “ certain changes in the skull bones during
old age seem to be determined by concomitant changes in the underlying brain,”
and that dense new bone is laid down between the inner table and the dura mater
where senile atrophy of the frontal lobe has taken place.

The absence of cerebral markings on the frontal depression, the great thickness
of the frontal bone, particularly in this area, and the advanced age of the skull all
indicate that, whether a depression was present before the age changes took place
or not, the frontal depression in the Stradbroke cast cannot be used as evidence for
lack of expansion. Other evidence for lack of expansion is, however, present in the
lack of fulness of the orbital margin (O) in the region of the inferior frontal gyrus
and in the relatively well marked frontal keel.

The occipital region, looked at from behind (fig. 18), presents an appearance
which is significant in determining its Australian status. The occipital poles are
almost blended together in one large and rounded swelling standing out from the
rest of the cerebrum. At no place does the contour of the parietal lobe run smoothly
into that of the occipital lobe. The occipital lobe is everywhere separated by a salient,
parieto-occipital groove which passes from tentorial border to tentorial border.
The prominence of the supra-occipital region—a feature of many Australian skulls—
thus leaves its impression on the form of the endocranial cast. The projection of
the supra-occipital region is not due to. bony thickening but to the occipital projection
of the brain; and actually the midline section of the skull shows the skull to be
thinnest in this region.

THE SKULL OF AN AUSTRALIAN ABORIGINAL. 177

Above the occipital projection the surface slopes upwards to the midline
sagittal plateau (fig. 18) bounded on either side by the parietal depression.

Below the occipital projection the cerebellum is unduly prominent and spreads
laterally beyond the level of the parieto-occipital region. Thus undue prominence of
the cerebellum, placed relatively far back, is an almost constant feature in the
endocranial cast of the Chinese ; and is also seen in the cast of the Bushman D. 709
of the Museum of the Royal College of Surgeons. The cerebellum of the European
cast, on the other hand, is, as it were, partly buried under the tentorial surfaces.
The occipital region has the appearance of having spread out posteriorly and laterally
and thus overflowing the boundaries of the cerebellum.

The relations of the cerebellum to the cerebrum cannot be attributed in the
Australian to any retention of primitive features, because in Simanthropus, Pithe-
canthropus, H. rhodesiensis, H. neanderthalensis and H. soloensis the position of the
cerebellum accords with the European type. Furthermore, the condition in the
Chinese is associated with a fairly high degree of brachycephaly and is also present
in the only brachycephalic European cast which has been available.

The significant feature, then, in the Australian cast and in the cast of the
Bushman D. 709 is that the undue prominence of the cerebellum is present in
association with dolichocephaly.

DESCRIPTION OF FIGURES.

Figs. 1 to 6. Orthogonal projections of the dorsal, posterior, anterior, lateral, inferior and medial
views of the skull. The skull was mounted in the Frankfort plane.

Fig. 7. A centimetre contoured map of the skull using the Frankfort plane as the base plane.

Pl. XV. Figs. 8 to 11. Photographs of the normae verticalis, occipitalis, frontalis and lateralis,
respectively.

Pl. XVI. Figs. 12 to 14. Photographs of the normae basalis and lateralis and of the median
section of the skull.

Pl. XVII. Fig. 15. X-ray photograph of the skull taken from the lateral aspect.

Pl. XVIII. Figs. 16 to 19. Photographs of the endocranial cast taken from the anterior, dorsal,
posterior and lateral aspects.

C. coronal elevation, F. frontal depression, P. parietal depression, PCD. post-coronal depression,

PO. parieto-occipital depression, O. orbital elevation, T. post-temporal elevation.

178

MEMOIRS OF THE QUEENSLAND MUSEUM.

REFERENCES.

Burkitt, A. N., and Hunter, J. I. 1922, Jour. Anat. Vol. LVII, Pt. I, pp. 23-45.

Cunningham, D. J. ge
Duckworth, W. H. L.

Hale, H. M., and Tindale, N. B.
Harris, H. A.

Keith, A. ta
Klaatsch, H.

Longman, H. A.
Martce: R. ie
de Miklucho-Maclay
Smith, G. Elliot
Smith, 8. A.

Todd, Wingate, and Lyon, D. W.
Turner, W.

1908, Trans. Roy: Soc. Edin. Vol. XLVI, p. 304.
1894, Jour. Anthr. Inst. Vol. 23, No. 4, p. 298.
1930, Rec. Sth. Aust. Mus. Vol. IV, No. 2, pp. 145-218.

1933, Bone Growth in Health and Disease. Oxford Univ.
Press, London

1931, New Discoveries Relating to the Antiquity of Man.
London.

1908, Rep. Path. Lab. Lunacy Dept., N.S.W. Vol. I,
Pt. IIT, pp. 43-167.

1918, Mem. Queensland Mus. Vol. VI, p. 4.

1928, Lehrbuch der Anthropologie, Bd. II, fig. 370.
1883, Proc. Linn. Soc., N.S.W. Vol. VIII, p. 401.
1926, Natural History. Vol. XXVI, No. 3, p. 295.
1918, Phil. Trans. B., Vol. 208, p. 382.

1925, Amer. Jour. Phys. Anth. Vol. VIII, p. 23-45.

. - 1884, Report of the Challenger, Zoology. Vol. X.

f.

Fi

auiriest fat fase = Suazl aeaiceys Seaeaseeny (RUREatgU Speaeraey estes
Ht ana) Cannenness rae = ei runwaeuent SEES RRaES (SRBC CER TS (elas Sense eenerece
EEE i} Sduzd SEU SSUUNEEOG) (SEEPSRSEE (Rest neanl (eree cane
euieH e PEE EE Gad SHSSaszaus SEdEEFadil (aPEUiSEUs HEISSSIEL = ne
susstarass Svat Sgstsuzn: fevezsaees goyedeaes iovaveseey iassaeaiat fast
i Susi eegsceu! sedeusvent coesestie: asteasetseemtag?
an austeaial ee Sepa Gg VaR SERRE ST SE
sabastbae; faaat insej Sgsuaeeser taaagaanes sarsassbdl jRizezezai
aeaeh Sesbsonbee benauuuss tuuee bones Seume twee Sos
i seats seavaracel pesstatats saiaesaais ievibedia
i baer tte ee " fbi SHRieG Gane ei ENE
ii ESHER ta Huue aeeed ett inten eet
Stiueesascussears ve Lal TSICRIGKES asitteai; Hite taebeurast tesesacel
SasceasueRecatasttrsstss HEI Si 150508 Ergis SeEUSSESEOF SbCEancEs
rt el 2 s

ene t] ty
. = Sondh osdaoee
Sack eartaeeecaee

ie yaeentdee
se 7 = eee eee

G8: Eaesteded

as +1 Grasceuceuss Seem * eee ewer Seer ee

—
a

tc foes Fest EES Be

pataetal F os eat ain is Tae
ESTE | SRTeET Au eee eae a AEE een at el ee ELE
Sea acre aaa Get ae

+4 te
ry
it
odes

saaseverattatsisi

=rere
See hee ee eee
SUUSSCRTER seseece

yaynE an 5
ce

{ES aes Se sted ieee iti etins sis 2 eae
ee ere Sei HSE Hare eteiat cette ccstial aria | ever aviv] austen

EH ae ie stieineie eH Heeeeiet See iene, ern PRE iseied

site cl cont Coen a
eae
sistas sassesaaed leseeaseue ceatetenca conte eutad onseanseeeae

af

iesguecenansne petpeeee ngdaauen qeeilit Hy
EERE a eee HEE aeE SESE iia a
Ht Hae eee Heine ae BEE ae ca

Pe ane wae wre He ae ice \< as a =
tH Here H eae nine eae ius aur ean suaes eee seeieuesi init * He
He iJ ann eneas Pi Lissitri iri an b+;
se SGU Acer certeeeegPaulaten| evant ett eee tate tobe ue epee peceerraper rT)

5 Culsagon (heaeaaned

SEpesaAStnceaeect tacees
Beet ieeiieat

eee ese i

seas dia eeaie a HE 78

ao se einiitiitaail Hn st Eve (RENGH nit ECR EE
2 ee
isisiesiatl aeistnenisesnist ifasa naitt S Prete oH : ae ett seit

= rirt} aeasest ina: cube fensee
i

oe ee eae ee EL EEE as

idneeseees ae Soves Wuessesees ecweeverss Saveseeapee] i js
Se eee snr pte EER a seisrecapieos ties Poe BUgueapeeayy
ieeaeeaei td abs DEVESHCREsSCURTEEa! Hine Batees SUBSE EEaaaes AH
Eueunee

ieee

Seeeasucnsteatasecad veategaa aaazqaiid satni20i fet

Saat eoneee oteaacers Serre fee ee ears ee

Sevigaean iarepeni sete jase eee ae PTT cee atas
serie elie

cor seenan seit
“pS ait

suiieet rigeeaiseeaseecsttais IE sini REE gueasiesaststcas ee rita nie
Sie Ua faen tebe eeadeanatenit ts eat deae HEH stots to A tediedsta| cecaeaiel HaHa

ai tyeubaccassnnenete saanseaea iat Sa edauanon er pat Betiren! Bazagsenss ta art
a sgerseageecess F eucdet f tesseaestd ieeatvitai

GESiScuPsosss sesstsapsszs0s0as2s 700% getgiatis PH Hs ‘ga ie iat
cEEaES See GEER aE Seae aL TTareee ast Pease oot eee ivbad tapeaietae uaeeengt Heestsidh

Heer ai dat ceeuuns Sereessess fs Lak) H+ 4 spearaztt:
Suet isezsssent senaseeett

: ceria area seasiatas
pbs ni i

=
=
brass
SOSSE Siasos
f see eee .
HH inaapenune *
=ceeeunageet

EeSaseseet ‘an

aaa ee
Rept

7 =

e+ «4
2h tanee

snare eee

-
bad

gaunt ranaies i tates
Li sazes
aeece
sean
NeencGteeeeendan tives

FSCGes Gees eee Seseneus:

PARR IDS ALS real
Se a

#. 7 (ROSS SS0 B68 oe

s2eat
pe CEueeRecvE see
atacce .atee
anaes ceeme saan
S258 SSESU OEE ETS
mel bt =
- = *amEe

reneegaegs ip

=
ae ore .

rare. | MSzrose. :

a

EG! RERGUESLGE TERS STE RE F

St assueeedes Secsemenct
tsa

tt
333
sitaiala

oe
=
=
aan
atl
=
5
ét
a

EEE

J. L. Shelishear,

Fig. 2.

ee

AB BSOEE Re eeEe:

S2haces (ERRBEEES
IESSSCEEE! SORES CSSA SERREREE RS Che: SECHEET SSPE Pee
_- Sh eee S2s Reb EeEs

as

cis

ce
HE

ii

4
z

aesesaet fat
re
te

eae Ht

Bai cane

—
a cf
nae

i

ES "SEEEEES (panes

(enna

ceeeen
SESUEREREZE

tf
1 HEE
\ Lil
Lia

. ss

fede ci
Pete tree ae ee
SNEAGGaNUS SEGRE RWERESESESAERER EDS

AUEBE ERSESTEAEE
SSSGCTR - (BEES ESERR OBES
SEEp_« (SEER RECEREESES
GEDE 7 EESEREPEES EEA
SEERU ESEERSERESE
FESPA PERE EOD ESS EES
SEF PSUS EeE BESES Sess Ee

TePreti sr ri titres tip ili vistrulry igi tir big Tey
Rasaee BS SSRS0 ORREE CRP E SERS S Pee Res

REREA CARD REE RED PON EEE SR eee

SEERA AGE ERASSRS (LAESEZER ES AESRESERSS

i} SEESEARRER CROTE SEAS LOSE EST Ss eeseE 4,

= SEAen Lew
Fi receue:
eee

Li
Pi
Lh

J
=
Hr

peaniiel
set |
BuE
see

be
*

amas |

i
i

:
_

agg: Sc8tH2
cd
mee
HE

Saeerens
[SES Ee CRSP RSE Ee
REESE GUSREEEREE

F
FH
aHtiie: *
gait
He
I"
Hi

in ©
SECESBESE
SkBSER SUSE ETS Peles
SERGE EAMES ESE ES PERS ee
SSEen.
PPT ii
ity

ait
nee

é
fit

ie

ttt
:
i

ratty

=
ry
fi
CI
H

ae}
ae
ty

P|

ade tenvasseas iets

=
Lae a
Pe a
4 as

an

Speke Sevisern: fetees
ret she ra 4 H aneceres
me esneeat seen deceriiea ne

He
bt
ae

WEES GSEES (SagyeeEst
me angueann:

' Ht

Lal

aves
:

:

*SSEREEE EESSE

aac. oe
s

Eu

em
avon:

¥
Ei
Ae
(SSRRSERSS iB
ISSUSREEDE IBORER BES

+}
a
an

me
nausea tl

EL aT) CS
SUEBS [SRESRECED LEBER.
MGEDE [MERGER EEE BSR ABes
SEESS ISSER Pe (SPS eSSes te SERS SESSe eee
Unaee LORESE: 2 180 Re BEAee “SS SU Shs SESE ERS a

REGS CARN EREES RAS ESSE
H Sees RSeUE (ew
ge ee eens (tigtiy ienaeeenan tam [jad
Zz ‘= 1m pat

\
EesacEna iets (eae ul Cz
ZO SeSeRi ISBEASEERE CHEE EEE

H

4
:

:
ri
H
E
ie
=

wa
ze

aa
4
te

S28 (ROR SEe Seow

-

Hi

H
ae ome
5 i

ee
a es SSERRZE Saene
GRRESSSREE Vee E
esse i tt Ti tte

creas | oe ee

ro 2
La

2
aed
iivaned

J. L. Shellshear,

©) FRSRE gums

hall — se
Sethe, TP eae Perri eaters eae Ja eeee ee
TT ee Scans e a

itis
SeSESSR™y FEE

Saket evened Oe we
Edt eee

"ing rH
"> (o* se een eae

>| Frans anes oe

~ 7a reane
. eT |

stoee
HE HSE

‘ar seus:

eae

om ee lhe nee ere

. : fam ry
pat -* inane mat cee
a; seoULEeRe HEIEESie dete

Ht

ath
E

te

i
ett
Fenn

i

!
ie

i

it
i]
au
it
babha
LF
+ J

Bynes tyne

fai

fits
i
=

Hee

ee

sineceeee ss

J. L, Shetlahear.

Ame ate hae et mom,

peear eneaee
seeere abe
Santer seees ee

Tes
ieiee

SS PACKS CREE CSRS SEES Ss SSeS ee
REAGGER EGO KeEe Se eee. ee

=a -
ae i]
im Ee.

see ett
stnistoai aettian iettteate
saseseoats iepsdsaect cvesssiet ielgteaaas ‘hrs

pare
fa;
vs) ae
shea

eEEey
tH

on
3 =
eae t
. aes es eee

ey igeliaiee | ; | su arn ait

Si Hi ieee iereties : ifr ent
SE eat ee eee Mn TY eed

RES facet 1 ie eater Ha ae Se
i istsittoey neesavadz il sesezes goatee ais | . i GGHsn feuniiees: etaaaia

ee Pl

fa eeet | sete

‘ H sbusaebsassanas feast H
et rakes ox iB

tbe
roy ceteetblce | terete ete ete lesteeere® ae Sere
4 LI]

gee eeeeeeeeeeleeereceeeteeete | = ns
eee eee | a

soetrern
Sheers 1
=—~eaneh
se oee
.

.
Sede

r eal +H
vps!

iiasebesss
if = °
Scbisoass gaeaioset ese

Seetnua

aoe
=

a
r]

sae
E

re

in

i ne eee oe =

E ineue
se
cH
CJ
a

i

cone
ea

i

Hi

ales sue
ee

:
4
si

nce
i258
HH

poh
-
{

i

bee

i
ib
fs

'

ma
'
]
HI

= fn
Ls.

Tet lead ; Pree

[rasetemss ‘sgualdtee
ipeeaas (eaease
ESE yese: :

i gues creer 9

aH
i

Sunes Lan
Suen :

HH
Ee

Hal
ii

Lio
Dased
ase

sen GEE. coReEs Su
as , Eat thoceuneat Es padhoe peare
Shean ances aH

= oo ions. Hs} Sa bee er oS 7 43-74

Sane (ewe eaert 12S oe ee (PVCS SVES! SCHSTR GOSS Coes sere J 7 set rere) & = s

L
3

J. L, Shelighear,

2 etl
a — ae

a a ea

aGG; seas =a mR)

ae ae

ssi? tine
a ier
aL

Pi
eanne

LC ;

.

aa
a
Liu
\WUN

oC
:
af

fi

ae

Be de a
HL

oo
aa: Coaae

at taut

pocey died gare
fans le a

a

fe
Ssors tail (essasiid

mst
iH
rH

iH

icyaet menage
4
se

cit

a

ss
Sees

ae

ecco
Storze

a “: .
eee Bf
re i] vee

; te 3, ,
Ae TAR
wa ‘
=

ee

a a
clei 6 Le oe
eu = | ,
genta HEE ane oe ee

ty ara eg 1a
a Se

, a is
fy

aa

eH

Sioa

SS SSEEE BESTE a

EE ad

Se

aaGE afiEH Siena

ite ee
Biel

J. I. Shelishear.

‘ae
a

:
Bees

penans HI
Geer ie

a

Banans : marppess

! \. ge : 8 ce

. | See it . + 3 = a5 ist esis

Se i 2d ee ee

came eae a itt eae aE : Eee ae Hg asi

‘it sania = = Pie etna d Hearth ciety © ete ee Ee ee
se ment a it

ce coe ee ae

Snes

i

rT
pyi

Hate

ai bane ae ie
Pi
t Sguoneeses feuean

Se coe oe
= poop
2s of

see ASS ae cae ata
: tee a eee
Hissar ee

Sritsuiattt
ieistis

Beet Bade
Hae

yey oy ini

feed i
: o
=e

7 au

al
=

ie ate an
SEGPes ah. +8 i. Pil
Rees Pee ‘oe
SRRESEA SRGue

J. L. Shelishear

ar
9
¢
&

9

J. L, Shelishear.

MEMOIRS OF THE QUEENSLAND

Shellshear.

MEMOIRS OF THE QUEENSLAND MUSEUM, Vot. XI, Prarn XVI.

J. L. Shellshear.

MEMOIRS OF THE QUEENSLAND MUSEUM, Vou. XI Prate XVII.

Fig. 15, J. L. Shellshear,

MEMOIRS OF THE QUEENSLAND MUSEUM, Vou. XI, Pirate XVIII.

Fig. 17.
Fig. 19.

ig. 16,

J. L. Shellshear,

THE CAMBRIAN FAUNAS OF NORTH. EASTERN
AUSTRALIA.

PART 3: THE POLYMERID TRILOBITES.
(With Supplement No. 1.)

F. W. WHITEHOUSE, PH.D., M.Sc.
(DEPARTMENT OF GEOLOGY, UNIVERSITY OF QUEENSLAND).

(Plates XIX—XXYV.)

. On Classification,
. Some morphological features.
. Systematic Descriptions.

me OF bo

. Supplement No. 1—
A. Explanation.
B. Some new Records of Agnostids.
C. Faunal and Stratal Correlations.
5. Localities.
6. Bibliography.
7. Explanation of Plates.

1. ON CLASSIFICATION.

The trilobites appear suddenly, a highly organised, considerably differentiated
order in the Cambrian, the earliest period in which fossils are adequately known.
In Lower Cambrian beds they are represented by seven distinct groups—the
agnostids, the eodiscids, the mesonacids (with their relatives the Redlichiidae and
Zacanthoididae), the conocoryphids, the ptychopariids and the group of large-tailed
forms belonging to such related genera as Bonnia and Kootenia.

Slightly later in the Cambrian two other new groups appear, the Burlingiidae
and the Menomoniidae, without any apparent links between them and the other
groups. The normal, post-Cambrian, proparian trilobites—the Phacopidea of Richter
(1932, p. 852)—may represent yet another distinct assemblage, making its first
appearance early in the Ordovician. There may be other isolated groups; but it
is possible that all known trilobite families may be traced back, more or less satis-
factorily, to one or other of these early (Cambrian and Ordovician) units.

180 MEMOIRS OF THE QUEENSLAND MUSEUM.

Some of the Lower Cambrian groups, the agnostids and the eodiscids.
particularly, are so different from the other stocks that their lineages clearly must:
have departed from the others at a much earlier period? But no earlier genera are:
known. We may have overlooked a number of Pre-Cambrian fossils ; but we cannot,
surely, have failed to collect some of the earlier faunas if all the immediate predecessors.
of the Cambrian trilobites had hard parts. .

These things strongly suggest that the ancestors of the trilobites, and even.
the earliest trilobites themselves, were soft-shelled or testless forms, a conclusion
that may gain some support from the study of the segmental structures of Redlichia,
described below.

Future collecting may close the gaps between some of these groups but it is
hardly likely to bridge them all. Such evidence suggests to me that the trilobites
already had diverged into a number of separate groups before the habit of strengthening
the dorsum with lime developed; that independantly, though not concurrently,
calcification began in a number of lineages ; and that, probably, many lineages never
developed a crusted test. Never, of course, did they evolve a hard, ventral integument.
It is doubtful if any new calcified groups arose after the Ordovician ; but it is likely
that soft forms persisted. As previously I have suggested (1936, p. 79) such strange
creatures as Marella and Mimetaster may be end points of lineages, with but partly
calcified dorsa, that arose from soft, near-trilobitic stocks.

If this be a correct explanation of the early history of the trilobites it will be
rather difficult to construct a perfectly natural classification of the order, since many
of the lineages are lost to us. The most adequate classification will be that which
clearly expresses the relationships of the families of the trilobites to one another and.
to the several discrete units that independantly appeared.

Several workers have classified the trilobites by means of the variations of
some single feature of the test. For such purposes Beecher (1897) used the type
of facial suture, Giirich (1907) the number of thoracic segments, Poulsen (1927) the
presence or absence of facial sutures. There is always a grave danger, in classifications.
based on unit characters, that homoeomorphs may be grouped together and cognate
groups divided. Disregarding the polyphyletic Hypoparia, Beecher’s, grouping into.
Pro- and Opisthoparia seems to express a fundamental division even though there is.
an. exception in the proparian Jujuyaspis, a member of the otherwise opisthoparian
Olenidae. For such a grouping there is one debated section—the morphologically
intermediate forms with sutures extending through the genal angle. These, including
the families Menomoniidae, Triarthridae, Calymenidae and Homalonotidae, may
be a natural, intermediate group (the Gonatoparia of. Giirich) or the last three may

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 181

be offshoots from Olenidae which, considering the genus Jujuyaspis, would then be
a rather insecurely opisthoparian group.

However in the classifications of the majority of recent systematists, for
example Swinnerton (1915), Warburg (1925), Richter (1932) and Kobayashi (1935),
there is a tendency to group the families into named genetic units and to pay less
attention for the purpose of classification, to the variations of some single characteristic.

Such, it seems to me, is the logical method with the trilobites as now we know
them; and I suggest that the seven, discrete Lower Cambrian groups, with their
descendant families, be treated as seven suborders as follows :

Mesonacida (Swinnerton), including Mesonacidae, Redlichiidae, Paradoxididae and
probably Zacanthoididae.

Conocoryphida (Swinnerton), a small suborder restricted to the eyeless forms, recently
reviewed by Resser (1936, pp. 13 et seq.).

Ptychopariida (Richter), a large suborder for Ptychopariidae, Anomocaridae, Olenidae
etc.

- Ellipsocephalida (nom. nov.), a small suborder for Ellipsocephalidae, Agraulidae, etc.

Corynexochida (Kobayashi), a large suborder, for families such as Corynexochidae,
Asaphiscidae, Bathyuriscidae etc,, leading to such groups as Asaphidae, Illaenidae,
etc.

Agnostida (Kobayashi), for the many families recognised by recent division of the old
Agnostidae (sensu lato).

Eodiscida (nom. nov.) for the blind eodiscids and the sutured forms such as Pagetia,
Delgadoia and Hebediscus.

Little need be said about the Mesonacida, for this grouping has been used by
many trilobite workers in recent years.

The Conocoryphida is a small group, similar to many of the ptychopariids
in most features other than the presence of eyes and facial sutures!. It seems
reasonable to suppose that they are the descendants of the blind stock from which
Ptychopariida arose, shortly before the Cambrian. It is perhaps doubtful whether
they should be treated as a separate suborder.

To Ptychopariida may be traced many families (Solenopleuridae, Olenidae,
Anomocaridae, etc.) with many thoracic segments, small pygidia, and cephala that
have prominent palpebral ridges and wrinkled pre-palpebral surfaces. Later they
seem to have given rise to more specialised groups, notably the Harpidae.

1 Whatever be the nature of the marginal suture in Conocoryphe it is hardly a true facial
‘suture. I prefer Poulsen’s suggestion (1927, p. 313) that it is a rostral suture. I find it difficult
to accept Stubblefield’s suggestion (1936, p. 419) that Conocoryphidae is a polyphyletic group
-of secondarily blind forms.

182 MEMOIRS OF THE QUEENSLAND MUSEUM.

Ellipsocephalida usually have been included in Ptychopariida; but in the
“Lower Cambrian they are sufficiently distinct to warrant separation. The simple
form, the limited number of thoracic segments, the small pygidium and the large,
smooth or little furrowed glabella, give them a characteristic appearance. In
number of families Ellipsocephalida is probably not a large suborder; but, if .
Proetidae rightly is to be placed therein it is an important group.

The opisthoparian trilobites with large pygidia are a very varied assemblage,
even in the Cambrian. Many of the families appear to be descendants of such groups
as Bonnia and Kootenia ; and for the time being I would unite the majority of such
large-tailed forms (Corynexochidae, Dorypygidae, Dikelocephalidae, Asaphidae etc.)
in this suborder. Zacanthoides and Albertella, from community of features, would
seem to be offshoots from Mesonacida that early began to develop large pygidia.
No doubt there are a number of offshoots from that and other suborders with increased
pygidia ; but until such relationship is proved it may be convenient to group these
morphologically similar forms.

With the proparian suborders and the development of post-Cambrian families,
I am not in this paper concerned.

Three things particularly have concerned morphologists and systematists in
their studies of the dorsal structures of the trilobites—the manner of fusion of the
cephalic segments, the facial sutures (whether they are present or absent, and whether
they are pro- or opisthoparian) and the number of post-cephalic segments (the
relative number of the segments and the manner of their fusion into a pygidium).
There are differences of interpretation of all these things ; but as a working hypothesis
the following generalisations may be stated.

1. The fusion of the segments to form the cephalon took place, in Pre-Cambrian
times, when the Trilobita was still a softly enveloped stock. The community of
structure of the head of the diverse groups of trilobites indicates this.

2. The primitive groups of all natural sections of the trilobites were without
facial sutures. From several such stocks the normally sutured trilobites developed.

The cephalic features of the new genus Mesodema, and the rostrum of
Xystridura saint-smithi strengthen the belief that the families Redlichiidae and
Paradoxididae arose from Mesonacidae that had no facial sutures. The eodiscids
also seem to have produced several offshoots with eyes and facial sutures?. As

2 Poulsen (1927) holds that the sutured Pagetia, so similar to the blind Calodiscus, is not
related toit but is a homoeomorph; but, as I have pointed out (1936, p. 81) the case for the
derivation of sutured forms from blind eodiscids does not rest on Pagetia alone but on several

genera (Pagetia, Hebediscus and Delgadoia), related to different sections of Eodiscida. -

+

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 183

already noted some relatives of the blind Conocoryphida probably were ancestral
to Ptychopariida.

3. The division into proparian and opisthoparian stock antedated the formation
of facial sutures.

The function of the facial sutures, other than that it is in some way concerned
with the visual organs, is problematical. They are the weakest sutures of the test
as shown by the very common occurrence of trilobites complete except for the free
cheeks. That they were necessary for ecdysis seems rather unlikely, since there was
no hard ventral integument. The differentiation into pro- and opisthoparian groups
early in the Cambrian and the apparent derivation of pro- and opisthoparian lineages
from diverse sections of crusted, sutureless forms, suggests that the segregation into
two such types occurred very early in the history of the order and so makes Beecher’s
systematic division, based on this feature, more satisfactory than any other unit
classification.

4. In the earliest of the soft trilobites there was no fusion of the posterior
segments to form a pygidium, which is a feature developed independantly by fusion
in several groups.

In so many lineages there is some increase in the size of the pygidium that
such a generalisation seems warranted. The unfused posterior of the Mesonacidae
is strong evidence in support. Swinnerton (1915) has postulated a general increase
in trilobite stocks from micropygous to heteropygous to isopygous forms.

Using these generalisations to obtain a plan of the development of the order
it would appear that fusion of posterior segments affected some lineages more than
others before the advent of a crusted test. In Mesonacida there was no such previous
fusion; in Conocoryphida, Ptychopariida and Ellipsocephalida there was fusion to
form a small pygidium ; in Corynexochida a large pygidium had been so formed ;
while in Agnostida and Eodiscida the acme of prea formation had been reached
before the test was calcified.

The formation of facial sutures was accelerated in some lineages, retarded in
others. In Agnostida it was, apparently, always inhibited.

Diagrammatically, in text figure 1, I have sought to indicate, from such
premises, some of these trends in trilobite morphology and to show how, in such a
review of the evidence, they are related to developing lineages and the problem of
primary classification.

184 MEMOIRS OF THE QUEENSLAND MUSEUM.

Were all lineages, soft and crusted, known to us the agnostids and eodiscids
would not be so far removed from the other groups ; but as it is they are so different
that I regard Jaekel’s division of the known trilobites into Miomera (the Agnostida
and Eodiscida) and Polymera (the remaining groups) as most pertinent. Schuchert

r

MIOMERA POLYMERA
PROPARIA ? GONATOPARIA OPISTHOPARIA
POSED a Ee TT ee eS

J

- ne
oan

La, fe

Text-figure 1.—Some Suggested Trends in Trilobite Evolution. _ Unknown lineages of forms with
uncalcified dorsa are suggested by broken lines. Lineages of hard-crusted forms are indicated,
diagrammatically, by continuous lines. Within the lined area are only those forms that,
primitively, have no facial sutures. Within the stippled area are only the isopygous and
sub-isopygous groups. A Agnostida; B Eodiscida; CC Burlingiida; D Phacopida ;
E ?Calymenida; F Mesonacida; G Conocoryphida; H Ptychopariida; J Ellipsocephalida ;
K Corynexochida. X is the collateral branch leading to Marella.

(1936, p. 398) has criticised the division as “ too simple to be acceptable’; but
Resser (1938b, p. 47) has even suggested removing the agnostids from the trilobites
into a separate sub-class. Such a step I think unnecessary, Jaekel’s grouping
expressing well what seems to be a fundamental division.

2. SOME MORPHOLOGICAL FEATURES.
(a) CEPHALIC SUTURES.

The type of rostrum now described for Xystridura saint-smitht (No. 2 in text
figure 2) throws some light on the nature of the cephalic sutures. This rostrum
extends completely around the inner edge of the cephalon, into the corners of the
doublure beside the genal angles. That is, it is of the type previously known only
in the Mesonacidae, and gives some additional support to the view that Paradoxididae
arose from mesonacid stock. It throws light also on the vexed question of the
homology of the cranidial sutures.

Barrande, in 1852,3 suggested that the two branches of the facial suture, the
rostral suture and the two ventral branches connecting these, together formed one
unit, la grande suture, and that the hypostomal suture and the connective sutures
were independant units. However, since in so many genera (Nos. 3, 4, 5 and 6 in text

3 For a clear, diagrammatic and bibliographic presentation of these several theories see
Stubblefield, 1936.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 185

figure 2) the connective sutures link up in position with the facial suture and bound
the rostrum, many later workers have held that the facial suture and the connective
sutures.are part of the same unit. Beecher believed that the rostrum was the axial
part of an anterior segment of which the remainder of the doublure was the pleura.
Kiaer has suggested that the rostrum originally was a segment anterior to the doublure
and that it has been reduced in varying degrees in different genera as facial sutures
have been modified. This idea has been developed further by Warburg, and the
sections 1 and 3 to 7 in text figure 2, copied from Richter, show such a morphological

suite of variants.
mig ia Gy Zi

FRAN DED ds

Text-figure 2.—Cephala of trilobites showing types of rostra and sutures (after Richter and
Stubblefield). 1 Ayjerulfia, 2 Xystridura, 3 Paradoxides, 4 Calymene, 5 Homalonotus,
6 Isotelus, 7 Dalmanitina. Tostra are shown by a stippled convention.

From a study of the rostrum of Xystridura saint-smithi the following conclusions
may be drawn :

i. The rostrum in the sutured trilobites is of the same origin as that in the
Mesonacidae.

ii. The rostrum is a segment anterior to the doublure and not the axial portion
of it. .

iii. The facial suture may be independent of the rostral suture, and la grande
suture of Barrande is apparently a natural unit.

iv. Although in most genera the rostrum, where present, has a position within
the doublure related to the type of facial suture,* such a correspondence is not without
exception.

(6) SomE SEGMENTAL STRUcTURES IN REDLICHIA.

One remarkable specimen of Redlichia idonea has been found, an interior with
imprints of what, at first sight, appear to be appendages. Photographs of an artificial
internal mould, a plaster cast, are reproduced as figs. 5a, b of plate XIX.

4 Even in the mesonacids, without ordinary facial sutures, there may be such an agreement ;

for the lines that have been interpreted by some workers as anchylosed facial sutures, and by
others as incipient sutures, sweep round, in their anterior branches, towards the genal angles.

N

186 MEMOIRS OF THE QUEENSLAND MUSEUM.

Portion of the cephalon and nine thoracic segments are preserved on the
specimen. Each thoracic segment has a structure that arises, at the intersegmental
suture, from the ridge that divides the axis from the pleuron, and lies along the
suture between adjacent pleura. These things are very irregular in thickness.
Periodically they are thickened into sharp nodes, the internodal slope being gentle
proximally and abrupt distally. The first of the nodes, in some segments, arises at
the place of union of the structure with the test. In the posterior segments the nodes
are very oblique, sloping backwards to the axis; but on the anterior segments they
become transverse or even slope the other way.

Comparing the proportions of the specimens illustrated by figures 5a and 2b
of plate XIX it will be seen that the pleura are practically complete to their outer
ends. Yet, while some of these thinner structures taper towards their distal ends, the
majority do not decrease distally. Unless, when complete, some of them are abruptly
trucated, then either they protruded beyond the ends of the segments or else, if
internal, they were deflected within the doublure.

The maximum number of nodes (apart from a basal thickening) is eight,
dividing the structure into nine sections. The minimum number is five.

The two comparable structures on the head are close together at the rear.
The posterior member is similar in size to those of the thorax and three nodes are
shown on it. The anterior member is thick, with two widely spaced nodes visible.

The following features need to be emphasised :

1. These structures are uniramous, occurring on both thorax and cephalon.

2. Although based on a common.plan they vary in thickness and in the number
of nodes.

3. Each structure is an entity—that is the structures of adjacent segments
are not connected.
4, There is no indication that such structures were jointed appendages, formed

of articulating segments. Each is an unsegmented unit periodically thickened .into
nodes.

5. There is no trace of an extension centrally beyond the point of attachment.
6. Each of these thoracic structures is a process not confined to one segment,

but attached to the adjacent edges of two segments and lying along the sture between
them.

7. There are no offshoots in the nature of gill blades or setae.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 187

8. The fact that all thoracic structures are in the same corresponding position
suggests that these processes were in contact with and probably attached to the
inside of the test, so that they were internal. Thus they did not‘function as limbs.
This conclusion is supported by the mode of attachment, so different from that of the
known limbs of trilobites.

What, then, are these structures ? Apparently they are not limbs but some
internal processes. They arise at the places where the flexor and extensor muscles
most probably would be attached. They may be a curious type of muscle, extending
along each pleural suture of the thorax; but if so their presence on the posterior
portion of the cephalic shield is somewhat difficult to explain. That many of them
have nine internodes raises the question whether they are related to the endopodites
that have seven segments (including the basipodite) ; but I find it difficult to believe
that there was any type of correspondence, in linear detail, with a limb, which is
external, and these features, that appear to be inside. If they were muscles they
were strong muscles; and since Redlichia is older than any known trilobite with limbs
the question may be raised: did the primitive trilobites move by some means other
than, or as well as jointed appendages—a legacy, possibly, from a crustless ancestor ?

It will be interesting to see if future collecting establishes similar structures
in other trilobites or jointed limbs in Redlichia.

8. SYSTEMATIC DESCRIPTIONS.
Subclass TRILOBITA Walch, 1768.
Order PotymMERA Jaekel, 1909.
Suborder Mrsonactpa Swinnerton, 1915.
Family REDLICHIIDAE Poulsen, 1927.

Genus MESODEMA gen. nov.’
Genotype: Mesodema venulosa sp. nov.

- Diagnosis: Trilobites in which the cranidium agrees closely with that of
Redlichia, except that the palpebral lobes arise directly from the frontal lobe of the
glabella and the second glabellar lobe is continuous with the post-ocular portions
of the fixed cheeks.

Remarks: The resemblance of Mesodema to Redlichia is particularly close.
Were it not for certain structures, possibly vestigial, that are discussed below there
would be no reason to separate the genotype from Redlichia. In general features of
the cranidium it closely resembles Redlichia nobilis Walcott (1913, p. 105, pl. 7,
figs. 12).

5 wecdouyn something built between.

188 MEMOIRS OF THE QUEENSLAND MUSEUM.

Three other genera are included in the family—Redlichia Cossman 1902
(Genotype Hoeferia noetling: Redlich),® Neoredlichia Saito 1936 (genotype Redlichia
nakamurai Saito) and Redlichaspis Kobayashi 1936 (Genotype Redlichia (?) finalis
Walcott). Redlichia is the first genus to appear, followed quickly by Neoredlichia
and slightly later by Redlichaspis. Possibly in this order they represent a lineage.

Mesodema is closely allied to Redlichia (sensu stricto)—that is to the group
of species with wide anterior limbs, including such forms as &. noetlingi (Redlich),
R. nobilis Walcott and R. chinensis Walcott.

Since Walcott stated the proposition in 1910 (p. 53) it has been widely accepted
that Redlichia is intermediate between the Mesonacidae (without facial sutures) and
the Paradoxididae (typically opisthoparian). It is thus one of the most significant
of trilobite genera. The thorax and pygidium are like many mesonacids. The head
has normal facial sutures and, like all Opisthoparia with such sutures, the palpebral
lobes are not confluent. with the frontal lobe of the glabella. They do, however
impinge upon it; and by so doing resemble the mesonacids more closely than any
later forms.

Although Redlichia is transitional it does not adequately cover the gap between
the two families. The new genus Mesodema is now another link in this chain, being
intermediate between the mesonacids and Redlichia.

There are four features whereby the heads of mesonacids differ from those
of all other trilobites. These are:

i. No facial sutures are present ; but instead there are lines, usually raised but sometimes
depressed, in positions where one would expect elements of the facial suture. These
lines are almost invariably present in the post-ocular position ; "and in three genera
at least (Holmia, Kjerulfia and Callavia) 7 they are known with pre-ocular branches
as well. I propose the term “‘ facial lines ”’ for these structures.

ii. In the genera that have complete facial lines the anterior limb (i.e. the portion of the
dorsal shield in front of the glabella and the anterior branch of the facial line) sweeps
round towards the genal angles.®

iii. The palpebral lobes arise as pleural expansions of the frontal lobe of the glabella.

iv. In a number of genera (Paedeumias, Olenellus and Holmia for instance) the second
lobe of the glabella is often laterally continuous with post-ocular portions of the
cheeks.

6 Redlich (1899, p. 3, pl. 1, figs. 1-8) recognised two varieties and gave each a name (angusta
and lata). One of these must be regarded as the type section and so lose its varietal name. Reed
(1910, p. 7, pl. 1, fig. 14) subsequently figured a specimen of the narrow form as Redlichia noetlingi,
this being the only other specimen that has been illustrated. Consequently I propose that the
original of Redlich’s plate 1, figure 1, be the lectotype and the name angusta accordingly, be
abandoned.

? For figures see Kiaer, 1916, and Raw, 1936.

8 Raw. (1936, p. 239) has expressed this by using the term ‘‘ metaparial ” for facial structures
that sweep round towards the genal angles.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 189

Until recently a fifth difference could have been recorded—the rostrum of the
mesonacids extends almost to the extreme posterior position, whereas in other genera
the rostra are short. However later genera with rostra of maximum lengths
are recorded in this paper.

Features li, iii and iv are most strikingly expressed in the ontogeny of the
species. For instance in the protaspis stages of Holmia, Elliptocephala, Paedeumias
and Olenellus® the palpebral lobes are seen clearly to be the pleura of the frontal
glabellar lobe while the post-ocular portions of the cheeks are just as evidently the
pleura of the more posterior lobes of the glabella. The backwardly sweeping course
of the anterior limb is well shown on these same figures.

The change from the mesonacid type to forms with facial sutures must have
been particularly abrupt and revolutionary, one of the most complete and. sudden
changes in the known history of the trilobites. Not only are these four features
suddenly removed from the heads of later adults but also they have been eliminated
from the ontogeny of all later forms whose larval stages have been figured. The
divorce is such that one might wonder whether the mesonacids left any descendants.
Yet in Mesodema, with true opisthoparian sutures, there are vestiges of all these
features. ' ;

The palpebral lobes in Mesodema arise from the frontal lobe of the glabella
as perfectly as they do in the mesonacids. Similarly the second glabellar lobe grades
no less definitely into the post-ocular portion of the cheeks. The greatly expanded
anterior limb in Mesodema and the related species of Redlichia is morphologically
intermediate between the sweeping mesonacid type and the much reduced form in
other trilobites. This is particularly interesting in view of the reduction of the anterior
limb in other lineages of Redlichia.1° :

Finally there are, on Mesodema venulosa, a pair of fine raised lines parallel
and very close to the anterior branches of the facial sutures. On certain trilobites
(ptychopariids and olenids particularly but also in some mesonacids), numerous,
raised, radial lines occur on the anterior limb. Such structures have little more than
ornamental significance and appear to be somewhat in the nature of wrinkles in the
epidermis. The two definitely placed and widely spaced lines of Mesodema are not,
I suggest, of this type. Rather, arising from the eyes, they appear to be vestiges of
the facial lines. In passing it is worthy of note that in the mesonacids the anterior
branches of the facial lines never extend across the marginal rim. The lines on
Mesodema likewise stop at the rim.

® For figures see Walcott, 1910, and Kiaer, 1916.
10 Progressive reduction of the anterior limb is noticeable, to a slight degree, in the long-
fronted Redlichias also (based on Saito’s stratigraphical series).

190 MEMOIRS OF THE QUEENSLAND MUSEUM.

Mesodema is thus a trilobite on which, apparently, both facial sutures and facial
lines occur. If, as seems generally to be conceded, facial lines are related to facial
sutures, two interpretations of the mesonacid lines are possible. Either they are
rudimentary and represent an incipient development, a forecast as it were of facial
sutures, or they are vestigial and represent the lines of anchylosis of ancestrally free
and fixed cheeks. Each interpretation has its adherents. Even if they are vestigial,
separation of the cheeks might conceivably occur again along the course of the former
sutures. The evidence now available suggests that, whatever be the origin of the
facial lines, the separation of the cheeks to form facial sutures took place almost
coincident with them.

MESODEMA VENULOSA sp. nov.
(Pl. XIX, fig. 1.)

Diagnosis: The cranidium is widest at the anterior margin, the width being
slightly greater than the length. The glabella is long, narrow and subfastigate and
it tapers uniformly but very gradually towards the anterior. On it four pairs of
prominent but narrow furrows are impressed, each pair most deeply incised at the
outer limits. Each pair of furrows is deflected posteriorly towards the axis of the
glabella. The posterior pair, the occipital furrows, are continuous although they are
but faintly impressed in the centre. The anterior pair are faintly impressed and short.
The two intervening pairs of furrows are prominent and almost continuous. The
occipital ring has no median spine but axially it is obtusely pointed. Anteriorly
the centre portion of the glabella merges into the sinus behind the anterior rim.

The palpebral lobes are long and converge tangentially towards the front of
the glabella. Together they limit a broadly pyriform area. These palpebral lobes
arise, without intercepting furrows, from that portion of the
frontal glabellar lobe immediately in advance of the anterior
glabellar furrows. Their rims remain of constant length.
Laterally beyond the anterior portions of the palpebral
lobes furrows arise which, towards the front, bound the
antero-lateral portions of the glabella.

: Immediately behind the palpebral lobes a portion of

Raatoration. the fixed cheeks arises on each side as a lateral expansion

of the second glabellar lobe. That is, the axial furrows are

here discontinuous. These furrows reappear to bound the third glabellar lobe

but practically disappear again between the fourth lobe and the adjacent portions
of the cheeks.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 191

The anterior branches of the facial suture are widely deflected laterally,
producing a wide anterior limb. On this, on either side of the glabella, there is a
broadly convex fold radiating from the centre of the anterior margin. The anterior
rim is long, wide, and almost straight. The facial sutures, in traversing the anterior
rim and the fold behind it, have a “ stepped ” course.

On each side of the glabella a thin raised line arises near the anterior limit of
the palpebral lobe and continues parallel to the facial suture, and about 1 mm. from
it, as far as the posterior edge of the anterior rim. Here it tends to bend towards the

axis and-disappears.

Remarks: The significance of the several cephalic structures have been
discussed in the remarks upon the genus. No other species of Mesodema is known
_and only the cranidium of the genotype has been found.

As previously indicated Redlichia nobilis Walcott is a very similar form
among the Redlichias and the resemblance is close indeed. Among the mesonacids
attention may be drawn to Nevadia and Nevadella“ which alone of all those genera
have conical, tapering glabellae. The resemblance in glabellar' features to Nevadella
is particularly striking. Callavia also is a genus whose glabella may be compared -
with that of Mesodema. Although not tapering it has similar types of glabellar
furrows. ©

Locality and horizon : One specimen only, from the Redlichia Stage at Yelver-
toft homestead. (University of Queensland Collection).

Genus REDLICHIA Cossman, 1902.
Genotype: Hoeferia noetling: Redlich.?
REDLICHIA IDONEA sp. nov.
(Pl. XIX, figs. 2-11; pl. XX, figs. 1-5.)
Diagnosis: Test with a sublanceolate outline, tapering posteriorly.

Cranidium with a long, tapering, straight-sided glabella that reaches almost
to the anterior rim. The occipital furrow and the two posterior glabellar furrows are
continuous, and there is a third, faint glabellar furrow (anterior). These four furrows
divide the glabella (and occipital ring) into four subequal parts. The axial furrow is

11Jn erecting the genus Nevadella Raw (1936, pp. 243 and 249) did not specifically name
a genotype. From his remarks it would appear that he regarded Callavia eucharis Walcott as the
type species; and I take that as genolectotype.

12 See footnote 6 on page 188 of this paper.

192 MEMOIRS OF THE QUEENSLAND MUSEUM.

‘continuous but it becomes shallow at the points where, in Mesodema, the two anterior

lobes of the glabella are extended into the palpebral and post-palpebral structures.
The axial furrow widens in the antero-lateral regions. The fixed cheeks are narrow,
bounded by the long, slightly curving rim of the palpebral lobe. The anterior branch
of the facial suture extends from near the axial furrow almost transversely to the
margin. On the two lateral wings of the anterior limb there are, in some specimens,
faint traces of the facial lines. The free cheeks are wide, little inflated with marked
genal angles. The spines, which continue the curved line of the rim, are anterior to
the genal angles. |

There are fifteen segments in the thorax, the eleventh segment bearing a
long, median spine. Posterior to this the segments become less prominent. The
pleura, which have narrow, oblique grooves, terminate in short, backwardly directed
spines.

The pygidium is very small, sub-globular, with a truncated posterior rim. The
axis, which occupies most of its area, is bulbous, with three or four tiny axial rings.
The pleura are short, merging into a posterior rim. |

The rostrum is long, perhaps extending to the area of the genal angle. The
hypostome is globular and is either fused to the rostrum or tightly attached across
the suture.

The segments of the thorax and the posterior portion of the Cephalon bear
uniramous structures, attached at the axial margin, and consisting of cylinders
periodically thickened.

Remarks : Over one hundred fragments of this species have been examined.
Most of the cranidia are approximately 15 mm. long, but cranidia up to 60 mm.
in length (pl. XX, fig. 1) have been obtained. Other structures occur proportionately.
The total length of such a large specimen would be approximately 180 mm. (7 inches),
by far the largest Redlichia yet discovered. There does not seem to be, in this matter
of size, a division into two species or varieties, for adequate specimens of inter-
mediate size also occur. The same features in size (an average, common size, with
some specimens very much larger) has been noted also in the species Xystridura
saint-smitht.

In the shallowing of the axial furrow near the first and second lobes of the
glabella, and in the presence, occasionally, of traces of facial lines, a relationship is
indicated with Mesodema.

—

13 These have been described in another section of this paper (pp. 185-187).

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 193

Redlichia idonea is most similar to R. noetlingi (Redlich) and R. nobilis Walcott
(see 1913, p. 105, pl. 7, figs. 5); particularly to the Korean forms figured by Saito
(1934, pl. X XVI, fig. 18) as R. nobilis. These species morphologically are intermediate
between the long-fronted species such as R. chinensis Walcott (see 1913, p. 104, pl. 7,
figs. 11), R. manchuriensis Endo and Resser (1937, p. 280, pl. 18, figs. 11-24), and
f. murakamii Endo and Resser (1937, p. 281, pl. 19, figs. 7), and the forms with
short anterior limbs, such as R. coreanica Saito (1934, p. 223, pl. X XVI, figs. 11-14)
and R. nakamurai Saito (1934, p. 224, pl. X XVI, figs. 15-17). The outline tapers
much more than in R. mansuyr Endo and Resser (1937, p. 282, pl. 19, figs. 1-3) and
R. yunnanensis Endo and Resser (1937, p. 283, pl. 19, fig. 4). Two of the species from
Yunnan are rather similar—R. walcotti Mansuy (1912, p. 26, pl. 3, figs. 4) and R.
carinata Mansuy (1912, p. 27, pl. 3, figs. 5). However they are rather broader forms
with slightly different facial sutures.

Redlichia has been recorded from Australia on several occasions. Ptychoparia
thielet-Chapman (191], p. 316, pl. lviii, figs. 2, 3, 5, 7 and 10), from the Cambrian of
Victoria, was placed in Redlichia by Walcott (1914, p. 62). Ptychoparia minima
Chapman (1911, p. 318, pl. lviii, fig. 1) was referred doubtfully to the genus by
Etheridge (1919, p. 389). Neither of these species is a true Redlichia. True species
of Redlichia are represented by Olenellus (?) forrestt (Etheridge fil. ms.) Foord (1890,
p- 99, pl. iv, figs. 2a, b), which was placed in Redlichia by Walcott (1913, -p. 104),
the Olenellus (?) sp. figured by Etheridge from South Australia (1919, p. 382, pl.
xxix, fig. 1), and the Kimberley specimens referred to by Chapman (1924).* Some
of these latter forms were figured by Wade (1924, pl. 45). The heads are different
from those of R. idonea, and suggest R. forresti (Foord) ; but the long tapering body
is interesting, suggesting some relationship with the present form. Through the
courtesy of the British Museum (Natural History) I have had the opportunity of
studying a plaster cast of the holotype of R. forresti. It is not well preserved, but
differs from #. tdonea in being more inflated and with less well defined glabellar furrows.
The South Australian specimen already referred to was loaned to me by Sir Douglas
Mawson from the University of Adelaide Collection. It is a flatter form with faint

furrows, quite distinct from F&. idonea.

Localities and horizon: All figured specimens and many others are from the
Redlichia Stage at Yelvertoft Homestead. (University of Queensland Collection
and also Schmidt Collection, on loan to the University of Queensland.) A few
specimens have been obtained also from. beds at the base of a hill, about 16 miles
from Yelvertoft dip on the road to the Paradice Goldfield.

* Since the above was written I have examined Chapman’s types. All of his Cambrian
trilobites from the Kimberley district appear to be species of Redlichia.

194 MEMOIRS OF THE QUEENSLAND MUSEUM.

Family PARADOXIDIDAE Emmrich, 1839.14
Subfamily ParapoxIpINAE Howell, 1933.4
Genus PARADOXIDES Brongniart, 1822.
Genotype: Paradoxides tessini Brongniart.

PARADOXIDES PEREGRINUS sp. nov.
(Pl. XX, fig. 6.)

Diagnosis: Cephalon and anterior thoracic segments unknown.

The posterior portion of the thorax (excluding the spines) has a semicircular
outline. The axis is wide, tapering rather quickly in the last three segments. Pleura
are obliquely grooved, each groove slowly widening and deepening until the base of
the spine is reached when it terminates rather suddenly. Where the doublure begins
the pleura are prolonged,’ without contracting in width, as long, backwardly sloping
spines, each with a channelled centre. The outline of a complete pleuron is subfalcate,
with the fulcrum situated just before the doublure begins.

The pygidium consists of a tail plate and one fused anterior segment. , This
tail plate is long, with parallel sides at right angles to the anterior furrow, and
terminates in two long, thick spines contiguous where they leave the margin. This
tail plate shows no differentiation into axial and pleural portions. The fused segment
is similar in all respects to the adjacent thoracic segments, but the pleura are reflexed
immediately they leave the axis.

Remarks : Trilobites with long, pleural spines and with the pygidium composed
of little if anything more than a tiny tail plate are characteristic of the three early
groups of the Mesonacidae, Redlichiidae and Paradoxididae, the families that Richter
(1932, p. 855) united as the superfamily Redlichiidea. The type of tail plate, without
transverse furrows or defined pleural zones, is most similar to that of certain
mesonacid. genera (Wanneria for instance) to which it shows other resemblances in
the type of posterior spines. But the presence of a fused anterior segment is some-
thing foreign to the mesonacids and allies the species to the other groups. Clearly
on general features it is not a Redlichia ; and since in certain forms of Paradozides,
for example P. davidis Salter, all its essential features are repeated, I have little
hesitation in referring this new species to Paradoxides. The bed from which it is
recorded (on the Templeton River) is well within the Middle Cambrian and. so
later than all known deposits with Mesonacidae.

14Tn conformity with the decisions of the International Zoological Congress I have used
the spellings Paradoxididae and Paradoxidinae instead of the forms, Paradoxidae and Paradoxinae,
adopted by the authors who first named these divisions,

18] may say that the specimen is preserved in a fragment of smooth, buff-coloured chert
of a type that is common at many horizons in the Queensland Cambrian sequence. Cherts of this
‘type occur in the beds on the Templeton River but are not very common there in the fossiliferous
portion. I have wondered, sometimes, whether this specimen is rightly recorded for its locality.
The late Mr. Dunstan who collected the specimen for the Geological Survey cabinets, collected also
at Yelvertoft Station where such cherts (in both Redlichia and Dinesus Stages) are very common.
There may be a confusion in localities, although I would not like to affirm this.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 195

Only one specimen is known, a posterior fragment 10 mm. long, on which the
anterior pleura are telescoped over some of the later segments. Nevertheless in its
features it is sufficiently distinct from all other described species so that I have thought
it worthy of a name.

Species of Paradoxides with long, delicate, pleural spines are known from many
horizons, from the earliest hemerae in the range of the genus to near its close. Well
known species of this type include P. oelandicus Sjogren, P. pinus (Holm ms.) Wester-
gaard, P. bidentatus Westergaard, P. tessini Brongniart, P. bohemicus (Boeck), P.
spinosus (Boeck), P. barrandet Barrande and P. davidis Salter. However in the
features of the pygidium P. peregrinus differs from all such forms although it is most
similar to P. davidis. It has a spinose segment, generally similar to the thoracic
segments, fused anteriorly to the tail plate. So far as I am aware, P. davidis Salter
(see Lake, 1935, p. 203, pls. 27 and 28) is the only other spinose Paradoxides that
has this feature. The pygidial spines of P. peregrinus are stouter and longer than in
any other species of the genus. There are many species of Paradoxides with simple,
non-spinose margins to the pygidium. Other forms have spines, some a pair of
marginal spines, others two pairs.1° But the spines are never massive and between
them there is generally a portion of the normally curved margin. In this species
the spines diverge in a V-shaped manner without the intervening curves. The outer
edges of the spines continue the straight, lateral lines of the tail plate. Yet another
pecular feature is the uniform surface of the tail plate. These plates vary considerably
in this genus. In some forms, for instance P. bohemicus (Boeck),!? there is clear
evidence on the axis that the plate is composed of fused segments. In others evidence
of segmentation is suppressed and the axis is smooth, without transverse furrows. But
even in these there is differentiation between the axial and pleural regions. In P.
. peregrinus elision of the surface detail is further developed and the surface of the
tail plate is not so divided. One of the specimens of P. davidis Salter figured by
Lake (1935, pl. 28, fig. 3) has a tail plate in which there seems to be a similar lack of
division between axis and pleura.

The variation within the genus Paradoxides, as it is understood at present,
is such that, as Howell (1933, p. 216) has suggested, it may be expedient before long
to make some subdivision. But the group is a unit, whether it be interpreted as a
genus or a subfamily, and as such was known hitherto only in the atlantic province
of western Europe, northern Africa and eastern North America. To record a species

16 In some curious variations of P. bidentatus Westergaard (1936, pl. 3, figs. 10 and 11)
figured specimens with three and possibly four pairs of pygidial spines.

17 See Barrande (1852) pl. 10, fig. 22.

196 MEMOIRS OF THE QUEENSLAND MUSEUM.

as far distant from the province as Australia is of more than ordinary interest..
Paradoxides must have been a rare form in these southern waters ; for only the one
specimen has been obtained from beds in which other trilobites abound.

Locality and Collection : Holotype (in the collection of the Geological Survey
of Queensland) from the Dinesus Stage of the Templeton River (probably from
Beetle Creek).18

Subfamily XySTRIDURINAE nov.

Taken generally we may recognise three main morphological, probably genetic,
groupings of the genera that usually are placed in Paradoxididae. The type section.
to which Howell in 1933 gave the name Paradoxinae (here transcribed as.
Paradoxidinae)!® included Paradoxides Emmrich and Plutonides Hicks. No doubt,
as Howell (1933, p. 215) has suggested, it may be expedient before long to divide:
Paradoxides into a number of genera,?° all to be grouped in Paradoxidinae. The
second division, including the genera Centropleura Angelin, Anopolenus Salter and.
Clarella Howell, comprises the subfamily Centropleurinae.2!_ Metadoxides Bornemann,
Catadoxides Matthew and Anadoxides Matthew form a third group for which I suggest:
the new subfamily name of Metadoxidinae. It is possible that the family begins in.
the late Lower Cambrian, with the incompletely known genus Strettonia Cobbold
(1931, p. 471).

Paradoxidinae have expanding glabellae, palpebral lobes that are not in
contact with the glabella, and tiny pygidia usually with unbroken margins but
occasionally with four spines. Centropleurinae differ in having the anterior section
of the facial suture extended considerably at either end, a pair of anterior glabellar
furrows that converge posteriorly, and a relatively large pygidium on which the four
spines are permanent, thorn-like structures. Metadoxidinae differ from the other
groups in the narrowing glabella.

Xystridura, as Kobayashi (1935, p. 126) has noted, has characteristic parad-
oxidid features. But it is difficult to place it in any of these three groups. Clearly,
with its expanded glabella, it has no close relationships with Metadoxidinae. It has:
a cephalon precisely of the same type as Paradoxidinae but a pygidium of the same
form as Centropleurinae. From the stratigraphical position of the Templeton Series,

18 See footnote on page 194.

19 Kobayashi (1935, p. 127) independently has made the same suggestion.

20 If Paradoxides is to be subdivided then Hicks’ name Plutonides ro doubt will be retained
for one section. For that reason I have listed the name in this paragraph.

21 Howell, in 1933, proposed Centropleurinae as a new subfamily. But in 1854 Angelin had
used the name Centropleuridae ; so that it is preferable, as I see Kobayashi (1935, p. 127) recently
has done, to credit Angelin with the name when using Centropleurinae as a subfamily.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 197

where it occurs, Xystridura must belong to the early part of the range of Paradoxididae.
‘Centropleurinae, as at present we know the group, occur only late in this range.
It may be, therefore, that Xystridura is an early form of Centropleurinae, developed
from Paradoxidinae, in which the pygidium has assumed the new features but not
the cephalon. I do not think so. The pygidial features are even more strongly
developed than in the three genera of Centropleurinae, whereas the cephalon is more
remote from the centropleurid type than is the average species of Paradoxides. For
instance there are no reflexed glabella furrows and the facial suture shows not the
slightest tendency to the incised pre-palpebral form of Centropleurinae. Actually
the facial suture is further removed from the glabella than in the majority of species
ot Paradoxides. From these premises it seems to me improbable that Xystridura is
-on the lineage leading from Paradoxidinae to Centropleurinae but rather that it is a
separate offshoot with a Paradoxides-like head, and a relatively large, quadrispinose
pygidium.” I therefore have placed it in a separate subfamily.

Genus XYSTRIDURA Whitehouse, 1936.8
(= Milesia Chapman 1929, non Latreille 1804)
Genotype: Olenellus Brownii Etheridge fil.
Diagnosis: Dorsal shield regularly ovate, little inflated.

The cephalon is large and semi-circular. The glabella expands anteriorly,
reaches to the anterior furrow, and is marked by an occipital furrow and three other
discontinuous glabellar furrows. The fixed cheeks are wide, flat, and generally give
a sub-quadrate outline to the cranidium. The palpebral lobes are regularly arcuate,
not in contact with the glabella and with a marked impression (?muscle scar) at each
‘posterior extremity. Facial sutures have divergent anterior branches and very
short posterior branches. The anterior branch of the facial suture meets the palpebral
lobe at some considerable distance from the glabella. Free cheeks have moderately
long genal spines.

The thorax consists of thirteen segments. The axis is about one fourth the
total width and is. not indented. The pleura end in obtuse, reflexed spines, the
‘fulcrum being near the inner end of the doublure. Each pleuron is obliquely grooved
and, adjacent to the axis, has a prominent impression (muscle scar).

22 If the late Lower Cambrian genus Sfrettonia 1s indeed a member of Paradoxididae it,
rather than Xystridura, might have claims to be an early form of Centropleurinae. In that genus
both cephalon and pygidium have such incipient centropleurid tendencies. This raises the question
-of whether Paradoxidinae or Centropleurinae is really the parent stem or whether, perhaps, a
Strettonia-like lineage may have persisted and given rise to all subfamily groups.

23 As a nomen nudum the name Xystridura dates from 1932 when David (1932, p. 37), in
using some of my manuscript notes, quoted it in a list of genera of the Templeton Series.

198 MEMOIRS OF THE QUEENSLAND MUSEUM.

The pygidium is about one sixth the length of the test. Its axis has two
anterior furrows and a relatively large, unfurrowed extremity. The pleural portions
of the pygidium are wide, with four widening furrows on each side and with a flat
margin ending in four wide, short and recurved spines.

The rostrum is a narrow semi-circular band agreeing in shape with the anterior
border of the cephalon but completely separated from it by a suture.

The hypostome is semicircular in the anterior portion. It narrows posteriorly,
ending in a straight margin. The hypostome is not permanently attached to the

rostrum.

Remarks: The forms from the Templeton River that here are recorded as
Xystridura were placed by Chapman (1929) in the established genera Bathyuriscus,
Marjumia and Dikelocephalus and a new genus (of preoccupied name), Milesia.
Recently Kobayashi (1935, p. 126) noted the resemblance to Paradoxididae and in
doing so drew attention to the extreme improbability of the forms having any
relationships to the three established genera that were quoted. The agreement
with Paradoxididae is strong. As mentioned above the head is precisely of the type
seen in many species of Paradoxides; the thorax has the normal features of the
family ; and the pygidium is of the same type as that of the subfamily Centropleurinae.

XYSTRIDURA BROWNI (Etheridge fil.)
(Pl. XXI, fig. 1.)

1897 Olenellus Brownii Etheridge Jr., p. 13, pl. 1, fig. 1.
1919 Olenellus (?) browns Etheridge Jr., p. 381.
1936 Xystridura brownt Whitehouse, p. 74.

Etheridge originally described the species in the following words :

Sp. Char.—General form unknown. Cephalon, or cephalic shield, subsemicircular ; width
all but double the length ; frontal and lateral borders thickened and raised ; posterior border to
all intents and purposes straight. Glabella rather pyriform, in front abutting against the frontal
border; axial grooves well marked and deep; basal furrow continuous across the glabella, and
very faintly V-shaped ; the second and third furrows horizontal, leaving between their inner ends
a slight median longitudinal convexity ; frontal lobe subsemicircular and largest, the second lobe
next largest, the third next, and the basal lobe smallest, the two halves of the latter somewhat
wedge-shaped. Neck segment narrowly oblong, with a central tubercle ; neck furrow deep and
horizontal. Eye lobes narrow, semilunate, extending from opposite the foremost pair of glabella
furrows to the posterior margin ; palpebral lobes appear to be deltoid ; genal spines short extending
to the third and perhaps the fourth pleura. Two anterior thoracic segments of equal width, slightly
convex, of equal width with the pleurae, or if anything very slightly wider, devoid of median spines
or nodes ; two anterior pleurae flat, slightly inclined from the axis, the geniculation apparently
feeble, the anterior margins strong and very marked. Sculpture anastomosing lines faintly visible

on the cheeks.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 199

Only the one specimen, the holotype, is known. On the left side of this
specimen the facial suture is clearly shown, the anterior branch making an acute
angle with the palpebral lobe and a right angle with the lateral margin. The posterior
limb is very short, because of position of the eye. Etheridge was doubtful whether
this was a true facial suture or a crack, and so placed the species in Olenellus. It is,
however, a regular suture of the type met with in the Paradoxididae.

I have not been able to see, on the specimen, the ‘‘ anastomosing lines faintly
visible on the cheeks ” ; but the specimen is preserved in an earthy limestone, with
a friable surface, so that it is possible that fine detail which was visible in 1897 has
since disappeared. |

Apart from these features Etheridge’s description gives an adequate account
of the holotype. Two notable characters are the length of the cephalon and the
shortness of the thoracic pleura.

Locality and horizon : From beds of the Dinesus Stage in “ a well, 200 feet deep,
five miles north of the station [Alexandria]. (Eth. fil.) The holotype is in the
Australian Museum Collection.

XYSTRIDURA SAINT-SMITHI (Chapman).
(Pl. XXI, figs. 2-16; pl. XXII, figs. 1, 2.)

1929 Bathyuriscus saint-smithii Chapman, p. 209, pl. X-XI, figs. 7, 8.
1929 Bathyuriscus nitidus Chapman, p. 210, pl. XX, fig. 9.

1929 Bathyuriscus olenelloides Chapman, p. 210, pl. X XI, fig. 10.
1929 Marjumia milesi Chapman, p. 211, pl. XXI, fig. 11.

1929 Marjumia conspicabilis Chapman, p. 212, pl. X_XIT, fig. 13.
1929 Marjumia elegans Chapman, p. 212, pl. X_XIT, figs. 14-16.

1929 Dikelocephalus dunstani Chapman, p. 213, pl. XXII, figs. 17, 18.
1929 Milesia templetonensis Chapman, p. 214, pl. XXII, fig. 19.
1931 Unnamed trilobite, Anderson, text. fig. ‘

1935 Unnamed trilobite, Fletcher, text. fig.

1936 Xystridura saint-snuthi Whitehouse, p. 62.

[non 1936 Xystridura sp. indet. Tillyard, pl. D].

Remarks: This trilobite occurs in colossal numbers in the shales of Beetle
Creek, 14 miles north-west of Mt. Isa and in other less known localities. From these
beds I have examined a large number of complete specimens and many fragments.
Also I have had the opportunity of studying the specimens described by Chapman
in 1929. As a result I have no hestitation in grouping as the one species the forms,
listed above in the synonymy, which he placed in four genera and eight species.
The holotypes of Chapman’s species are refigured on plate XXI. A number of his
specimens were fragments. One was a very immature specimen and another was
badly worn. But when they are compared with a suite of complete and undistorted
specimens the specific identity of these eight forms is apparent. Some of Chapman’s

200 MEMOIRS OF THE QUEENSLAND MUSEUM.

statements are in need of correction. For instance, six of his holotypes have a complete
thorax, each with thirteen segments (the normal number); but the number of
thoracic segments that he quotes for these is as follows: Bathyuriscus saint-smithi
(12), B. olenellordes (10), Marjumia miles: (not stated), M. conspicabilis (14), M.
elegans (14) and Milesia templetonensis (12).

The first species described by Chapman was Bathyuriscus saint-smithii ; and
accordingly I regard the other seven names as synonymous with this. In conformity
with the international rules I have used the spelling saint-smithv.

The specimens that I have examined range in length from 3,mm. to 180 mm.
Specimens from 60 to 100 mm. long are the most common.

The species may be re-described as follows ::

Outline regularly oval. The cephalon is subsemicircular. The glabella widens anteriorly,
after a slight constriction at the posterior third, and anteriorly impinges upon the rim. The
occipital furrow is continuous, although faintly impressed at the
centre. There are three pairs of glabellar furrows proper, each
pair narrow and discontinuous ; the posterior pair are directed
posteriorly, the others are strictly transverse. Fixed cheeks are
flat. The palpebral lobes are long, with muscle scars at the
posterior ends. They reach posteriorly almost to the border.
The anterior branch of the facial suture makes an acute angle
with the palpebral lobe and is subtransverse. Free cheeks bear
prominent genal spines. The rostrum is semicircular, extending
to the posterior angles of the doublure of the cephalon. The
hypostome is very wide anteriorly, with a short, narrow,
posterior extension that has subparallel sides. The anterior
lobe extends to the anterior and lateral margins. Maculae are
transversely elongate. The posterior lobe is practically non-
existant. Anterior and posterior wings are small.

The thorax consists of thirteen segments. The axis is
about one-quarter the total width and simply arched. The
Resiiration: pleura are relatively flat, with prominent, oblique furrows and
short, wide spines. The doublure is relatively wide. }

The pygidium has a short axis transversely divided by three furrows into three narrow
rings and a sub-globose posterior lobe. Four arcuate pleural grooves occur on each side, terminating
in shallow pits a short distance from the margin. There are four, prominent, equally spaced,
marginal spines, thorn-like in appearance.

From X. browni (Eth. fil.) the species is distinguished by a shorter cephalon
and longer thoracic pleura.

A feature of considerable interest is the length of the rostrum, which extends
into the postero-lateral angles of the doublure. That is, it is not limited by the
facial suture. Several isolated rostra have been collected, while many specimens

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 201

of complete individuals are internal moulds showing the rostrum in place, or slightly
displaced, often with the hypostome in position. For instance, on the holotype
(pl. X XJ, fig. 2) the rostrum can be seen slightly displaced. On the specimen shown
in pl. XXII, fig. 1, the rostrum and the hypostome are in situ.

The smallest specimens (pl. XXI, fig. 16) have very long, needle-like, genal
spines and long meta-cranidial spines. From that stage onward the length of the
genal spines gradually decreases. The small specimen chosen by Chapman as-the
holotype of “‘ Bathyuriscus nitidus”’ (pl. X XI, fig. 4) shows the long genal spines of
an immature individual.

Localities and horizon: From beds of the Dinesus Stage. The holotype, in the
collection of the Commonwealth Geological Survey at Canberra, was recorded by
Chapman from “twelve miles west of Mount Isa, at the head of the Templeton
River.” This (fide local opinion at Mount Isa) is apparently the locality quoted else-
where in this paper as Beetle Creek (a branch of the Templeton River), about twelve
miles north-west of Mount Isa. The same locality is quoted by Chapman for the
holotypes of Bathyuriscus nitidus, B. olenelloides, Marjumia milesi, M. elegans and.
Milesia templetonensis. The holotype of Marjumia conspicabilis is from “ Shepherd
Creek, near Miles Creek, north branch of the Templeton River” and the holotype of
Dikelocephalus dunstani from the “ Thornton River, N.W. Queensland.” The types
of these seven species, here included in X. saint-smithi, are in the collections of the
Geological Survey of Queensland. Abundant material from Beetle Creek is contained
in the collections of the University of Queensland, Queensland Museum, Geological
Survey of Queensland, Australian Museum and others. In the University of Queens-
land Collection there are also specimens from Mingera Creek (on the road from Mount
Isa to May Downs) from No. 1 bore, Alroy Downs, and from a hill top 16 miles north
east of Yelvertoft Dip.

XYSTRIDURA SP. |

An imperfect cranidium, with fainter furrows than X. brownt and X. saint-
smithi has been collected from the Hurostina Stage at_the No. 4 bore on Alroy Downs
(University of Queensland Collection).

Suborder PrycHoparimpa Richter, 1932.
Family PTYCHOPARITDAE Matthew, 1887.

Many genera of ptychopariids have been established recently by Walcott and
others on species whose dorsal shields were well known. Many others have been
based only on cranidia. It is a matter of considerable difficulty to interpret some of
these latter genera. Similar types of cranidia appear at so many horizons in the

O

202 MEMOIRS OF THE QUEENSLAND MUSEUM.

Middle and Upper Cambrian that, in the absence of complete specimens, it is some-
times scarcely possible to decide whether simple types (such as they are) persisted or
whether such groups are homoeomorphs of several lineages following similar trends.
Kobayashi (1935) recently has introduced an elaborate sub-family classification of
these forms ; but such classifications are so apt to be changed when complete specimens
come to light that I have preferred not to go into such details.

Two forms that occur on many horizons particularly concern us here—the
square type of cranidium (the Elathria type) and the triangular form (the Inouyella
type). It is of some interest to note how very similar are the early Middle Cambrian
Eurostina trigona (Pl. XXII, figs. 1la, b) and the Upper Cambrian Protemnites elegans
(Pl. XXII, figs. 12a, b, and 18).

' Genus LYRIASPIS gen. nov.*
Genotype: Lyriaspis sigillum sp. nov.

Diagnosis : Cephalon with narrow convex rim. The glabella narrows anteriorly
and is marked by the occipital furrow and two pairs of discontinuous, glabellar
furrows. The fixed cheeks are relatively wide and inflated, with small, palpebral
lobes and narrow, palpebral ridges. The anterior limb is of moderate width and
convex. The anterior margins of the facial sutures are subparallel and the posterior
margins slope obliquely. The free cheeks have genal spines.

There are thirteen or fourteen segments in the thorax. The pleura have wide
furrows with subparallel sides and end obtusely.

_ The pygidium is small, with a prominent axis, and has the pleura marked with
about three furrows on each side. The rim fringing the furrowed area is either very
narrow or absent.

Remarks: Lyriaspis is very similar to Hlathria Walcott (1924). The dis-
tinguishing features are the slightly wider fixed cheeks, the broadly furrowed pleura
of the thorax and the absence of a prominent rim on the pygidium. Ptychoparia
Corda, as restricted to the group of P. striata Emmrich, is another very similar form
but it has a markedly larger. pygidium. Elathriella, which is discussed below, has
narrower fixed cheeks, larger eyes and more slender posterior lobes on the cranidium.
A very close comparison may be made with Proaulacopleura Kobayashi (1936a). This
genus from the early Upper Cambrian of Alabama” differs only in having a smooth
glabella. None of the many other ptychopariid genera that have been defined in
recent times are quite so similar as these four.

24 XU'pa a lyre, do7is a shield.
25 It is associated with Glyptagnostus.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 203

The genus is represented by Middle Cambrian species in Australia and Asia.
I would place in Lyriaspis the Chinese Ptychoparia kochibei Walcott (1913, p. 132,
pl. 12, fig. 5) and a group of Indian species described by Reed in 1910 (P. spitiensis
Reed, P. admissa Reed, P. stracheyi Reed and possibly others).

There is a considerable variation in the genal spines in this genus. On L.
sigillum they are very small. They are very large on L. kochibei ; while if Ptychoparia
maopoensis Reed (1910, p. 28, pl. 3, figs. 20-25) is a Lyriaspis, the spines are taken
to extremes in that member.

Frequently in species of Lyriaspis the impression of muscle scars are seen on the
thorax. Generally they lie in the axial furrows or on the axis itself and are similar
to those seen on Ptychoparia (s. str.).

The Australian, Indian and Chinese beds with Lyriaspis are all of Middle
Cambrian age—and apparently early Middle Cambrian.

LYRIASPIS ALROIENSIS (Etheridge fil.)
(Pl. XXII, fig. 6.)

1915 Undescribed trilobite, Jensen, pl. 4, figs. 2, 3.
1919 Ptychoparia alroiensis Etheridge Jr., p. 385, pl. 40, fig. 8.
1922 Ptychoparia merrotskii Mitchell, p. 539, pl. 54, figs. 11, 12.

Etheridge’s original description of this species is as follows :

‘“Cephalon semicircular (when perfect). Glabella obtusely conical, rounded in
front, separated from the fixed cheeks and anterior limb by well-marked deep axial
grooves; two pairs of furrows, the basal pair circumscribing
prominent basal lobes ; fixed cheeks, comparatively large, but less
convex than the glabella; palpebral lobes small, the connecting
eye-lines, or ocular ridges, situated just in advance of the anterior
pair of glabella furrows, anterior limb like the fixed cheeks gently
convex, in the same plane as the glabella, separated from ‘the
anterior margin or fillet, which is cord-like and prominent, by a
shallow groove; neck-ring .in its median portion comparatively
thick, its groove well defined. Facial sutures in front of the palpebral
lobes almost longitudinally straight, really very slightly convex,
posterior to them curving downwards with a concave sweep and
sharply outwards in the direction of the genal angles.

Restoration.

Thoracic somites fourteen ; axis elongately and narrowly obconical, gently convex ;
axial grooves wide and open. Pleurae arched, angular in the middle line, each strongly
grooved or furrowed, the proximal half horizontally so, the distal obliquely bent. Pygidium
small, of two (or perhaps three) coalesced segments, and a small terminal appendage ;
those of the pleurae deflected backwards to a slight degree ; posterior margin truncate
and nearly straight.” *

204 MEMOIRS OF THE QUEENSLAND MUSEUM.

Mitchell, apparently, was unaware of Etheridge’s work ; for the same specimen
was chosen as holotype for his Ptychoparia merrotskii described three years after
Etheridge had established the species as Ptychoparia alroiensis.

The sharp convexity in the mid-pleural regions of the thorax gives this form
a somewhat peculiar appearance. Perhaps it is best compared with L. strachey: Reed
sp. (1910, p. 21, pl. 2, figs. 8-13) in which, however, the courses of the anterior and
posterior extremities of the facial suture are rather different. The Chinese L. kochibex
Walcott sp. (1913, p. 132, pl. 12, fig. 5) is also somewhat similar but it is a flatter
form with a wider cranidium. In many features L. admissa Reed sp. (1-10, p. 25,
pl. 2, figs. 22, 23; pl. 3, figs. 1-4) may be closely compared ; but that species has
an extra pair of faint furrows on the anterior portion of the glabella.

An interesting feature is to be noted about the locality of the holotype. This
was stated by Etheridge to be eight miles east of Alroy Downs, in the Northern
Territory. Some years ago I visited that locality but found no outcrops of Cambrian
beds. -Mr. Merrotsy, who found the specimen, had left the Northern Territory and
I had no means of checking the occurrence. However recently Mr. Merrotsy has
been in communication with me and has lent me the specimen. It is an external
mould, impressed upon a piece of chert that has been chipped into an artefact by an
aboriginal. Consequently the locality reference has little significance. Outcrops of
similar cherts with cranidia of this species I have found at no great distance—eleven
miles E.N.E. of Alroy Downs homestead at the crossing of Buchanan’s Creek on the
road to Alexandria.

Neither the figure given by Etheridge nor that in Mitchell’s paper gives an
adequate idea of the holotype. The figure published by Jensen is considerably
better ; but since Jensen’s paper is hard to obtain I have refigured the holotype.

Localities and horizon: The holotype came from eight miles east of Alroy
Downs in the Northern Territory where it was not in situ. The specimen is in the
private collection of Mr. A. L. Merrotsy. Other specimens, cranidia and portions of
thorax, are in the collections of the University of Queensland from the Dinesus Stage
at No. 1 Bore Alroy Downs (on the banks of Buchanan’s Creek, 11 miles E.N.E. of
Alroy Downs homestead).

LYRIASPIS SIGILLUM sp. nov.
(Pl. XXIT, figs. 3-5.)
1936 Xystridura sp. indet. ‘Tillyard, pl. D.
Diagnosis: Dorsal shield ovate. The cephalon is semicircular with a
regularly rounded front and a narrow convex rim. The cranidium is rhomboidal in
outline with a glabella that is about three-quarters of its length. The glabella

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 205

converges anteriorly, has a rounded front and bears an occipital furrow and three
other pairs of discontinuous, backwardly-sloping furrows. Around it the axial furrow
is prominently incised. The fixed cheeks are inflated and of moderate width. The
palpebral lobes are small and are continued by slightly converging palpebral ridges.
Anteriorly the facial sutures converge very slightly. Behind the
eye they extend obliquely in a fairly straight line to near the
genal angle. The anterior limb of the cranidium is narrow. The
free cheeks are small, subtriangular and end in short genal spines.

There are fourteen segments in the thorax. The axis hag
well marked and well spaced rings and the pleura end very

bluntly. In the axial furrows the muscle scars are often very
prominently impressed. These muscle scars form almost con- Restoration.
tinuous lines.

The pygidium is small and relatively smooth. The axis ends abruptly, is
sharply truncated, and occupies more than one-third of the width of the pygidium.

Remarks : The most similar form is undoubtedly LZ. alrovensis which, however,
is slightly wider, has broader fixed cheeks and a wider anterior limb.

In 1936 Tillyard figured a topotype of this species under the name Xystridura
sp. indet. It is a most excellent figure but I suggest that a mistake has been made
in the explanation. The figure which is about four times the size of the species is
stated to be magnified two diameters.

Locality and horizon: The holotype and many other forms (in the University
of Queensland Collection) come from the Middle Cambrian (Dinesus Stage) of Beetle
Creek, 12 miles N.W. of Mount Isa.

Genus CHANCIA Walcott, 1924.
Genotype : Chancia ebdome Walcott.

CHANCIA VICENALIS sp. nov.
(Pl. XXII, fig. 7.)

Diagnosis: Cephalon unknown. The thorax consists of twenty segments.
In the anterior portion the axis is one-fifth the width of the thorax and has well
marked and regular furrows. The pleura are flat and straight as far as the fulcrum
which is near the margiri. Each pleuron is obliquely grooved by a very wide furrow
that limits a narrow anterior rim of even width. The pleura end in short spines.
The muscle scars of the thorax occur in the axial furrows.

206 MEMOIRS OF THE QUEENSLAND MUSEUM.

The pygidium is small with a narrow convex rim. The axis is not strongly
defined and has a number of very crowded furrows. Three wide, curving pleural
furrows are arranged on each side, so disposed that they leave between each pair a
narrow raised portion.

Remarks: Only one specimen, with typical ptychopariid features, is known
and from it the cephalon is missing.

I have referred the form to Chancia which also has this unusual number of
thoracic segments and which has very similar thoracic and pygidial structures.
Chancia has been recorded from the late Lower Cambrian beds of Greenland (Poulsen,
1924) and the early Middle Cambrian of North America. No comparison can be made
with the Greenland forms for they are known only from cranidia. Two of the
american species however, the genotype (C. ebdome Walcott) and C. evax Walcott
(1924. and 1925), are known from complete specimens. In C. evaa there are 24 segments
in the thorax and the pleura end more bluntly. But C. ebdome is very similar to
C. vicenalis. It has twenty thoracic segments and their extremities are quite like
those of the Queensland form. However the thorax is perceptibly wider in the
american species. In all three species the course of the thoracic muscle scars is
similar. |

Locality and horizon : One specimen is present (the holotype) in the collections
of the Geological Survey of Queensland from the North Templeton River. It was
not found in situ. However, no fossiliferous beds have been found in this area other
than those of the Dinesus Stage of the Middle Cambrian, which are locally well
developed. It is probable therefore that the specimen came from such beds.

cise ELATHRIELLA Poulsen, 1927.
Genotype: Elathriella obscura Poulsen.

Elathriella was established by Poulsen (1927, p. 276, pl. 17, fig. 35) for a species
from Greenland that was based on two cranidia. No other forms were found associated
with it so that the age of the genotype is uncertain. The beds in which it occurs are
thin and overlie a formation with Middle Cambrian fossils.2° Some considerable
distance above are beds with a very late Cambrian fauna. Poulsen tentatively placed
the Elathriella beds as Upper Cambrian or Lower Ozarkian, apparently basing his
conclusions on his statement (p. 341) that “a very closely related species [to
Elathriella obscura] occurs in Montana in strata, the age of which has not yet been
fixed (Upper Cambrian or Lower Ozarkian).”” However I understand from Professor
Howell that these beds in Montana are now believed to be of Middle Cambrian age.

93

76 Poulsen correlated these latter beds with the Stephen Formation.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 207

In Upper Cambrian beds of the Pituri Sandstones in Queensland there are
trilobites whose cranidia are of the same type as Elathriella obscura. The agreement is
so close that there is no valid reason for removing them from Elathriella, This raises
the question of what is the range of the genus. It is possible, of course, that when more
complete specimens of the genotype are known the Queensland forms might prove
to be generically distinct. Thisis possible, but not indisputable; for the range involved
is really not great. Even if the age of the Greenland beds should be Middle
Cambrian (which remains to be proved) they must be in the upper half of the Middle
Cambrian. It is not, then, a very great range for the genus to extend to the horizon
in Queensland which I have placed, on available evidence (Whitehouse 1936), as
approximately equivalent to the zone of Orusia lenticularis. I stress this advisedly ;
for Hlathriella is a simple type of trilobite that might well have a range even greater
than this.

%

Elathriella, interpreting, it in light of the more complete Queensland specimens,
is a typically ptychopariid genus. It may even be compared with some of the
Olenidae, a family that, as I have indicated elsewhere (p. 217) I believe arose from
Ptychopariidae. However the size of the palpebral lobes, the converging palpebral
ridges, the type of posterior lobes on the fixed cheeks and the blunt pleura of the
thorax all indicate the Ptychopariidae.

ELATHRIELLA PLEBEIA sp. nov.
(Pl. XXII, figs. 8, 9; pl. XXV, fig 16.)

Diagnosis: Cranidium subquadrate ornamented with very small scattered |
tubercles. The glabella is conical with rounded front, and extends. for about two-
thirds the length of the cranidium. On it the occipital furrow is prominently incised ;
and anterior to this there are two pairs of short, discontinuous glabellar furrows that
bend backwards. The axial furrow is relatively deep. The fixed cheeks are narrow
and convex. The palpebral lobes extend for-rather more than one-third the length
of the cranidium. They are united to the anterior glabellar lobes by converging,
palpebral ridges. The anterior limb is convex and about one-sixth of the length of
the cranidium. Separated from it by a furrow is the prominent, convex, anterior
rim. The posterior lobes of the fixed cheeks are long and narrow. The anterior
portions of the facial suture diverge slightly as they approach the anterior margin
of the shield. The free cheeks are imperfectly known.

When the dorsal shield is of a length of 4-5 mm. the thorax has thirteen rather
, narrow segments. In the anterior portion of the thorax the axis (in the young stage

208 « MEMOIRS OF THE QUEENSLAND MUSEUM.

at least) is rather less than a quarter the width of the test. The axial portion of each
segment has a wide and long articulating portion. The pleura are widely furrowed,
the furrows having parallel sides, and end rather bluntly.

The pygidium is of moderate size with the axis of the pleura faintly furrowed.

Remarks: Cranidia have been collected showing a considerable variation in
size from young forms (pl. XXII, fig. 9) to mature types (pl. XXII, fig. 8). Two
specimens were obtained with the thorax attached, one of these having the pygidium
present also. Other thoracic fragments were found that tentatively have been placed
in this species.

The pygidium on the immature form figured has the sides steeply declined. °
The shape of this pygidium is triangular. Four furrows are shown on the somewhat
abraded axis. The pleura have, on each side, four furrowed ribs extending completely
to the margin.

The cranidia (such as that shown on pl. XXII, fig. 8) that are of the same size
as the genotype show a remarkable agreement with it.. It is difficult to select points
of difference. There is a slight difference in the angle of the glabella and in the width
of the anterior rim. Such distinctions, small as they are, appear to be sufficient to
warrant separate specific recognition. |

Locality and horizon: From the Elathriella Stage ot the Pituri sandstones
at the base of a hill immediately west of Tyson’s bore on Glenormiston (University
of Queensland Collection).

ELATHRIELLA SP.
(Pl. XXII, fig. 10.)

Associated with Elathriella plebeia is a form with slightly wider fixed cheeks.
The glabella is similar and the angle of the palpebral ridges. The form is not
sufficiently known to warrant a new specific name.

Locality and horizon: From the Elathriella Stage of the Pituri Sandstones
at the base of a hill immediately west of Tyson’s Bore, Glenormiston (University of
Queensland Collection).

Genus EUROSTINA gen. nov.”
Genotype: Hurostina trigona sp. nov.

Diagnosis: Cranidium subtrigonal to subquadrate. The glabella is conical,
with three pairs of discontinuous furrows anterior to the occipital furrow. Occipital
ring wide. Axial furrow well incised. Free cheeks and anterior limb inflated, with a
pair of faint antero-lateral depressions. Palpebral lobes small and prominent,
continued by converging palpebral lobes. There is a prominent anterior rim.

27 e¥pworos robust, in reference to the well moulded cranidium.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 209

Remarks: Eurostina is apparently related to the other early Middle Cambrian
genera, Jnouyia and Lorenzella, from China. Of these Lorenzella is the more similar,
but it does not have the prominent rim of Hurostina. Ptychoparia tatiana Walcott
(see Walcott, 1913, p. 155, pl. 14, fig. 9), from the early part of the Kiu Lung and
Ch’ang-hia Formations (basal Middle Cambrian), is another member of the genus.

EUROSTINA TRIGONA sp. nov.
(Pl. XXII, figs. lla, b.)

Diagnosis: Cranidium subtriangular in shape. The glabella is plump and
narrows anteriorly. In front of the prominent occipital furrow there are three pairs
of discontinuous glabellar furrows, bending backwards towards the centre. The
neck ring is wide. Fixed cheeks and anterior limb are inflated, the maximum convexity
being immediately in front of the glabella and, laterally, opposite the centre of the
glabella. Adjacent to the antero-lateral angles of the glabella there are small, slightly
depressed areas in the fixed cheeks. The palpebral lobes are small, arcuate and raised
into prominent rims, separated from the rest of the fixed cheeks by shallow grooves,
and continued to the glabella by slightly arcuate, converging palpebral ridges. The
anterior rim of the cranidium is a strong, rounded structure separated from the main
part of the anterior limb by a furrow of equal width.

Remarks: This is the species that I referred to previously (1936, p. 74) as a
Lorenzella, and possibly also as an Inouyella. From E. tatiana (Walcott), the only
other species that seems to belong to the genus, it differs in the proportions of the
glabella and in the more prominent groove behind the anterior rim. Also the palpebral
lobes seem more definitely raised into rims.

Locality and horizon: One complete cranidium and three fragments have
been collected from the Hurostina Stage of No. 4 bore on Alroy Downs (University of
Queensland Collection).

’ Genus PROTEMNITES gen. nov.*®

Genotype: Protemnites elegans sp. nov.

Diagnosis: Cranidium subtriangular in outline. The glabella is subconical
in shape with an occipital furrow and two pairs of discontinuous glabellar furrows.
The fixed cheeks are narrow, with small, centrally set palpebral lobes and converging
palpebral ridges. The anterior limb is flat or slightly convex and the anterior rim is
thickened centrally.

The pygidium is of moderate size with a prominently furrowed axis and faintly

28 zpoTé“vw to cut off in front.

210 MEMOIRS OF THE QUEENSLAND MUSEUM.

Remarks: Trigonal forms of ptychoparian aspect occur on several horizons
in the Cambrian. They are rather common in the Middle Cambrian, e.g. Inouyella
and Hurostina. ‘‘ Conocephalites”’ vulcanus (Billings 1861, p. 952) also is of this
type and one of the Chinese forms figured by Walcott (1913, pl. 14, fig. lla) as Inouyia
capax. The genus Pesaia of Walcott and Resser (1924) continues the type into the
Upper Cambrian. The peculiar medianly-thickened anterior rim and the rather wide
and little inflated anterior limb render the present form rather distinct from any

of the generically named groups in this series.

PROTEMNITES ELEGANS sp. nov.
(Pl. XXII, figs. 12a, b, 13.)

Diagnosis: Cranidium trigonal. The glabella is regularly conical in shape
with a very narrow, rounded front. On it are the occipital furrow and two pairs
of discontinuous, backwardly sloping furrows. The fixed cheeks are convex. They
are widest at the posterior margin and narrow progressively forward. The palpebral
lobes are small but prominently raised and are continued to the glabella by converging
palpebral ridges. The anterior limb is very slightly convex (more prominently
convex in the young stage). The anterior rim is short, very convex (particularly

towards its centre) and slightly elevated.

The pygidium is transversely elongate. The axis has four prominent furrows
and a small hemispherical termination. The pleura are flat with four faint curving

furrows on each side. There is no differentiated margin.

Remarks: In the list of genera published in 1936 I recorded this form as
Pesaia (?). It is clearly not a member of that genus which, while somewhat similar,
has a narrower anterior limb and no glabellar furrows. I know of no other species
that I would place in Protemnites.

Locality and horizon: An adult and a youthful cranidium and also an adult
and a youthful pygidium are preserved on the one specimen of chert. This specimen,
from the Elathriella Stage of the Upper Cambrian, was found in beds at the base of a
hill immediately west of Tyson’s bore on Glenormiston (University of Queensland
Collection).

Family NEPEIDAE nov.

"The genera Acrocephalops, Alokistocare, Eldoradia, Bolaspis and Nepea, which
are discussed below, are apparently closely related to Ptychopariidae, in which family
most of them previously have been placed. They seem to form a natural unit, with
a community of morphological structures and occupying a limited portion of the
Middle Cambrian. I propose therefore to group them as a separate family.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 211

Genus NEPEA gen. nov.??
Genotype: Nepea narinosa sp. nov.

Diagnosis: Cranidium semicircular, with steeply descending sides. The
glabella has three pairs of lateral notches. The frontal limb is highly inflated medially,
the inflation being bounded laterally by grooves continued from the axial furrows.
The anterior rim is narrow. The fixed cheeks are inflated, have small palpebral lobes
and long, very narrow, posterior lobes.

Remarks : Nepea is related to a group of genera, in the early Middle Cambrian
deposits of North America, that have a median boss in the frontal limb. Acrocephalops
Poulsen (1927), Alokistocare Lorenz (1906), Hidoradia Resser (1935) and Bolaspis
Resser (1935) are the other members of the group. Many of the North American species
have, at times, been referred to Acrocephalites Wallerius (1895), of which the genotype
is the scandinavian Solenopleura (?) stenometopa Angelin, a genus which, as Resser
has suggested, may not be related.

In the inflation of the median boss and in the reduction of three features—
the anterior rim, the palpebral lobes and the posterior lobes of the cranidium—
Nepea reaches the acme of development for the group. The median boss is so
pronounced and so definitely bounded by grooves developing from the axial furrows,
that virtually it has the appearance of an inflated anterior lobe of the glabella. In
this way superficially it recalls the glabella in the nepionic stages of certain olenids.*°
Acrocephalops, Eldoradia and certain new species that Resser (1938a) recently has
placed in Alokistocare (notably A. blainense) are in this way the most similar. It
may be that Alokistocare blainense should even be transferred to Nepea. The filament-
like form of the posterior cranidial lobes is not matched by the type species of any
of the genera listed above ; but certain species, for instance Alokistocare ? americanum
Walcott sp. (1916, p. 177, pl. 24, fig. 2), have a similar tendency. The wire-like
anterior rim is not pecular to Nepea. Bolaspis, Eldoradia and Acrocephalops have.
this also.

Perhaps the most similar genus is Eldoradia ; but Nepea differs from it, and
indeed from all these other genera, in the filament-like form of the posterior lobes of
the cranidium.

Most genera in this group have a tuberculate ornament. The specimens of
Nepea narinosa have been found in sandstone ; so that the surface ornament is not
known.

29 yytos childish.
30 Compare, for example, the figures given by Lake (1908, pl. 6) of Olenus mundus. It is
for this reason that the name Nepea has been chosen.

212 MEMOIRS OF THE QUEENSLAND MUSEUM.

Alokistocare, Bolaspis, Eldoradia and Acrocephalops seem all to cccur in the
early Middle Cambrian of North America. The stratigraphical position of some
species is still apparently in doubt; but from my reading I have the opinion that
the group generally is typical of the Glossopleura faunas of North America.

NEPEA NARINOSA sp. nov.
(Pl. XXII, 15a, b and 16.)

Diagnosis: Cranidium semicircular. The glabella is narrow and slightly
convergent, with a straight front. The occipital furrow is continuous. Three glabellar
furrows immediately anterior to this are deeply incised but discontinuous. The
frontal limb is swollen to a subcircular bulbous structure immediately in front of
the glabella, the lateral grooves bounding the structure being continuous from the
axial furrows. Immediately in advance of this is the marginal rim, reduced to a
very narrow ‘border. Fixed cheeks are wider than the glabella, inflated and
prominently deflexed on their lateral margins. Palpebral lobes are small, raised
very prominently from the surface of the fixed cheeks, and situated on the anterior
half of the cranidium, distant from the glabella. The posterior limb of the facial
suture is embayed. The palpebral ridges are faint, extending horizontally from the
anterior glabellar furrow to the palpebral lobes. The occipital ring is narrow, with a
faint median boss. The posterior borders of the cranidium are the same width as the
occipital ring and extend laterally to form very narrow postero-lateral lobes. Free
cheeks and hypostome are as yet unknown.

The thorax had at least ten simple, narrow segments, the pleurae on each
side being almost three times as wide as the axis.

Pygidium unknown.

Remarks: No other species has been described that I would refer to this
genus. However in Victoria a form, generically if not specifically identical with NV.
narinosa and not yet described, also occurs in beds containing Amphoton.21 Many
of the American species are still not sufficiently known for a correct generic placing.
As noted above Alokistocare blainense Resser (of which the posterior cranidial lobes
have not been figured) may be a member of this genus.

Locality and horizon: From the Amphoton Stage at “‘ Split Rock ’’ on Waroona
Creek, at the crossing of the main road from Camooweal to Mount Isa. Holotype
and other specimens in the collections of the University of Queensland.

31 T have referred previously to this association (Whitehouse 1936, p, 73).

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 213

Family SOLENOPLEURIDAE Angelin, 1854.
Genus ASTHENOPSIS gen. nov.**
Genotype: Asthenopsis levior sp. nov.

Diagnosis : Cephalon wide, with a short, conical glabella that is faintly marked.
with two pairs of glabellar furrows. Occipital ring well defined. The fixed cheeks
are wide, with narrow palpebral ridges leading to small palpebral lobes situated either
centrally or slightly behind the median line. Pre-glabellar field and marginal rim
well defined. The free cheeks bear genal spines rising rather abruptly from the genal
angles.

There are fourteen segments in the thorax, the pleura of which are well furrowed
and terminate either bluntly or in obtuse spines.

The pygidium is small and non-spinose, with the axis prominently segmented
and at least two pairs of prominent, rounded, pleural ribs. ~

The test is ornamented with tubercles, usually of two sizes (larger tubercles
scattered among smaller), that sometimes may be minute.

Remarks : Solenopleura, in the common usage of the name, is a genus of many
species. The genotype (selected by Vogdes, 1925) is Solenopleura holometopa Angelin.
This is a form in which the glabella, relatively, is large and the fixed cheeks and
pre-glabellar field are accordingly narrow. More abundant than such forms are
species with a small, conical glabella, wider fixed cheeks and prominent pre-glabellar
field. Among European species are Solenopleura brachymetopa Angelin and S. applanata
(Salter). It is for this group that the name Asthenopsis is proposed. Only a few of
the many species that may be included are known from complete tests. A. levior
is chosen as the genotype, rather than one of the European species, since it is known
from whole and uncrushed specimens.

Little can be said at present about the variability of the thorax and pygidium ;
but the following variations of cephalic structures may be listed :

1. There is a papillate ornament on the test,** particularly on the cephalon,
that often is resolved into papillae of two sizes. This ornament varies from relatively
coarse (A. canaliculata Angelin sp.) to microscopic (A. levior).

2. Glabellar furrows vary from moderately incised (A. canaliculata) to very
faint (A. levior).

32 dg bevijs feeble, ots sight—in reference to the small eyes.

83 Among the solenopleurids the papillation seems to be more strongly marked on the test
than on the internal mould, which may account for the differences between the figures given by
Grénwall (1902) and Walcott (1913) for the ornamentation of Solenopleura holometopa.

214 MEMOIRS OF THE QUEENSLAND MUSEUM.

3. The palpebral ridge is narrow. In some forms (A. canaliculata) it is
prominently raised ; on others (A. levior) it is so faint that usually it can be seen only
with a lens.

4. The palpebral lobes may be central (A. applanata Salter sp.) with palpebral
ridges normal to the axis, or the lobes may be slightly posterior (A. brachymetopa.
Angelin sp.) with the ridges converging obtusely upon the glabella. These ridges are
sometimes straight, sometimes slightly arcuate. |

5. The genal spines vary from prominent (A. levior) to small structures (A.
applanata), but they arise rather abruptly from the genal angles.

6. On some forms (A. Jevior) there is a slight tendency to develop radial
wrinkles on the pre-ocular regions.

Kobayashi (1935, pp. 258-289), who seems to regard all the Asiatic species as.
generically distinct from those of the Atlantic Province, transferred many of the
Chinese species that Walcott had placed in Solenopleura to a new genus, Solenoparia.
Endo and Resser (1937) have followed him in this usage. The genotype that he
chose for Solenoparia was Ptychoparia (Liostracus) toxeus Walcott, a form with large
palpebral lobes that Walcott, I believe correctly, excluded from Solenopleuridae.
Most of Walcott’s Chinese species of Solenopleura, it seems to me, agree with or are
very close to Asthenopsis and are distinct from Solenoparia.

In passing it may be noted that Kobayashi (1935, p. 265) has referred the South.
Australian species Conocephalites australis Woodward (1884, p. 344, pl. xi, figs. 2a, b)
to Solenopleura. That species, however, is a Lower Cambrian form more akin to
Protolenus.

Asthenopsis, as defined above, is a Middle Cambrian genus known in the
Atlantic Province of Europe and North America and also in Asia and Australia. In
Europe it is found from the zone of Paradoxides hicksi to the zone of Paradoxides
forschammerv.

ASTHENOPSIS LEVIOR sp. nov.
(Pl. XXII, figs. 17-20.)

Diagnosis: Outline regularly oval. The surface of the test is ornamented
with microscopic tubercles of two sizes (large tubercles scattered among small).
These tubercles, toward the posterior end of the thorax, tend to be restricted to the
axial region.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 215

The glabella is short and conical, extending for about two thirds the length
of the head. On it the occipital furrow is well marked and continuous, and there are
two pairs of extremely faint, discontinuous furrows extending only a short distance
from the axial furrow. The fixed cheeks are wide and _ inflated,
with small, post-central palpebral lobes and extremely faint palpebral
ridges. The facial sutures converge slightly anteriorly, while
posteriorly they reach the margin a little behind the genal spines.
The pre-glabellar field is slightly inflated and, like the free cheeks,
bears faint radial wrinkles. The anterior rim is moderately flat and
wide. Free cheeks are small, with prominent, stout, genal spines
that do not continue the curve of the anterior margin but extend

|
ji

a)

f Restoration,
obtusely to it.

There are fourteen segments in the thorax. The rings on the axis are cord-
like. The pleura are deeply furrowed, markedly geniculate and terminate in obtuse
spines.

The pygidium is approximately one seventh of the total length. On it there
are three prominent axial rings and a posterior axial lobe. On the pleura there are
four pairs of arcuate ribs, increasing in size outwards but terminating some distance
before the margin. The posterior margin is straight or slightly concave.

Remarks : As noted above A. levior, in many of its features, is at one extreme
of the variations shown by the genus. Its papillose ornament is remarkably fine,
its glabellar furrows are extremely faint, the genal spines are stouter than in most
species, the eyes are in a post-central position and there is a slight suggestion of
radial wrinkles on the pre-glabellar field and the free cheeks. In this combination
of characters it is distinct from the other species of the genus.

Locality and horizon: Very common in the limestones of the Papyriaspis
Stage at the road crossing of V Creek, between Undilla and Thorntonia (University

of Queensland Collection).

Family OLENIDAE Burmeister, 1843.
Subfamily PapyRIASPINAE nov.

The four genera Pianaspis Saito and Sakakura (1936, p. 114), Papyriaspis
gen. nov., Rhodonaspis gen. nov. and Hedinia Troedsson (1937, p. 56) have the
essential features of the Olenidae and possess also certain features in common that
stamp them as a group distinct from the several subfamilies that have been proposed.
For them I suggest the subfamily name Papyriaspinae. Through this group, at

216 MEMOIRS OF THE QUEENSLAND MUSEUM.

present known only in the Australian-Asiatic Province, the Olenidae may be traced
back to the early part of the Middle Cambrian. The four genera are characterised
particularly by the flatness of the test and by the relatively narrow axis.

Until 1936 no definite olenid was known in beds earlier than the Upper
Cambrian. The ancestors of the family had been sought in diverse groups. Of
recent writers Swinnerton (1915) derived them from Conocoryphidae. Westergaard
(1922), after careful zonal studies, traced them back to Jvostracus costatus Angelin,
a species whose relationships still are not well defined.#* Warburg (1925) has made
suggestive remarks about a relationship between Olenidae and Ptychopariidae without
committing herself to a belief in any direct derivation. Poulsen (1927) considered
an origin in Ptychopariidae but only doubtfully. Richter (1932) supposed
Ellipsocephalidae to be the parent family. Finally Kobayashi (1936a) has revived
the idea of an origin in Protolenus and has suggested Annamittia as a possible link.
The very name Protolenus that Matthew gave to a Lower Cambrian trilobite in
1892 has tended, I think, consciously or unconsciously, to draw undue attention to
that genus whose test still is imperfectly known.

Most writers on the olenids, no matter what their beliefs have been in such
origins, have been impressed ,by the close similarity between Olenidae and
Ptychopariidae. There is hardly a feature of the test that, if one regards it as typical
of the one family, cannot be matched in some genus of the other. Within each family
there is a particularly wide variation in characters.*° Possibly because of this latitude
most workers have been inclined to regard resemblances as homoeomorphic repetitions
and few who have written about lineages have linked the two families directly.

Hedinia occurs in the Upper Cambrian of T’ien-Shan; Rhodonaspis is from
the early Upper Cambrian in Australia; Papyriaspis occurs at the top of the Australian
Middle Cambrian ; while Pianaspis comes from the early Middle Cambrian of Korea.
Tracing the family back in this way the relationships with Ptychopariidae seem even
more to be emphasised. For instance the axis in these four genera is narrower than in
Oleninae and approximates more to the proportions of the ptychopariids. Also
the types of thoracic pleura are morphologically intermediate between typical
Ptychopariidae and Oleninae—although Rhodonaspis has long, terminal, pleural
spines, as in typical olenids, Papyriaspis has blunt pleura, as in Ptychopariidae,
while Pianaspis and Hedinia represent an intermediate stage with short spines.

——sni

34 Many authors have suggested, for instance, that Liostracus costatus is a ptychopariid
genus and not a member of Liostracidae. It may even be a true olenid. ;

35 In Olenidae this variation goes to such an extreme that in Jujuyaspis Kobayashi (1936a,
p- 89) it produces even a genus with proparian facial sutures.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 217

The resemblance of the earliest genus, Pianaspis, to the contemporaneous
ptychopariid Lyriaspis is particularly striking.*®

I would suggest that, early in the Middle Cambrian, the olenids arose from
Ptychopariidae, possibly from Lyriaspis, and, as Papyriaspinae, have not yet been
found in Middle Cambrian deposits beyond the Australian-Asiatic Province. Then,
in the Atlantic and South American seas in the Upper Cambrian, they evolved into
other subfamilies—Oleninae, Leptoplastinae, Triarthrinae.

Genus PAPYRIASPIS gen. nov.*’
Genotype: Papyriaspis lanceola sp. nov.

Diagnosis: Test flat and thin. Outline regularly subovate. The axis is
relatively narrow—about one-fifth of the total width. Cephalon is one-third of the
total length, with a short, tapering glabella marked by the occipital furrow and two
or three pairs of glabellar notches that do not reach the axial furrow. The occipital!
ring is markedly wider than the first thoracic segment. Fixed cheeks are wide.
Palpebral ridges are narrow, leading to small palpebral lobes. The pre-palpebral
region is marked by fine radial wrinkles. The free cheeks are wide, with prominent
genal spines. |

The thorax has eighteen segments with flat, narrow, widely grooved pleura
ending bluntly.

The pygidium is semicircular, with a non-spinose border. Its axis is well
segmented. The pleura are flat and marked by wide grooves.

Remarks: Hedinia Troedsson is the most similar genus. The two genera
agree in general shape, in the lateral dimensions of the axis, in the notched glabella
that is notably wider than the axis of the thorax, widely furrowed thoracic pleura
that end in short spines and in the non-spinose rim of the pygidium. Papyriaspis
has, however, a shorter and more tapering glabella, a continuous occipital furrow
and a larger pygidium. Pianaspis Saito and Sakakura has lateral notches on the
glabella, relatively a wider thoracic axis, and more deeply furrowed thoracic pleura.

36 Saito and Sakakura (1936, p. 116), when erecting the genus Pianaspis, compared it
closely with Ptychoparia kochibet Walcott, a member of Lyriaspis.

3° Tamvpos papyrus, do7is a shield—in reference to the thin, parchment-like nature of
the test.

a

218 MEMOIRS OF THE QUEENSLAND MUSEUM.

PAPYRIASPIS LANCEOLA sp. nov.
(Pl. XXIII, figs. 1-8.)

Diagnosis : Test flat, subovate, the length nearly twice the width.

The cephalon is transversely elongate. The cranidium is rhomboidal in
outline, tapering anteriorly. The glabella narrows anteriorly and is about two-thirds
the axial length of the cranidium. On it the occipital furrow is continuous and there
are four pairs of discontinuous glabellar notches. The first two pairs (anterior)
are small circular depressions not connected with the axial furrow. The third pair
consists of long, narrow pits, extending obliquely backwards but connected to the
axial furrow by curious anterior extensions that are discordant in direction with the
main pits. The fourth pair have sinuous courses and arise obliquely from the axial
furrow as if to form occipital lobes; but they are not continuous with the occipital
furrow, although immediately anterior to the occipital furrow there are two small
circular depressions, in line with these fourth furrows and probably related to them.
The postero-lateral margins of the glabella in this way are rather irregular.

The fixed cheeks are relatively wide and are crossed by very narrow palpebral
ridges that converge at an angle of approximately 150° at the anterior end of the
glabella. The palpebral lobes are small and semicircular, situated immediately
posterior to the median line. Fine, radial, anastomising wrinkles
cover the pre-glabellar field. The anterior rim is narrow and
well defined. The facial sutures converge anteriorly from a region
behind the eyes; but the posterior limbs of the facial suture
extend obliquely outwards. The free cheeks have short genal
spines. Like the pre-glabellar field they are marked by radial
wrinkles.

The thorax consists of eighteen segments. The axis is
narrow (about one quarter the total width), it tapers uniformly
to the rear, and the first segment of it is markedly narrower than
the occipital ring. The pleura are flat, very long and narrow, presenting a lath-like
appearance ; they are grooved by wide, shallow furrows with sub-parallel sides and
they terminate bluntly.

Restoration.

The pygidium is semicircular in outline and is about one eighth of the total

_length of the test. There are four narrow rings on the axis and a large posterior lobe.

The pleura are flat and are marked by five pairs of long, narrow, shallow furrows with
subparallel sides.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 219

Remarks : No other species of the genus is yet known. I have several dorsal
tests that are complete but for the free cheeks, and numerous fragments of the species.
The holotype is 39 mm. long; but from comparative measurements of fragmental
material it would seem that specimens may reach a length of 50 mm.

On plate XXIII, figure 4, a small form is illustrated that is approximately 20:
mm. long. On such forms the pre-glabellar field is smoother, and rather more inflated
than on the larger forms and there is little trace of the fourth (posterior) pair of glabellar
furrows. However, I believe these to be immature forms of this species. Comparable
differences between young and mature forms occur in the related genus Hedinia
(Troedsson, 1937).

Locality: From the Papyriaspis stage at the crossing of V Creek, between.
Undilla and Thorntonia (University of Queensland Collection).

Genus RHODONASPIS gen. nov.**
Genotype: Rhodonaspis longula sp. nov.

Diagnosis: Cranidium wide. The glabella is narrow, it has subparallel sides
and is marked by the occipital furrow and two or three pairs of discontinuous glabellar
furrows that arise from the axial furrow. The anterior limb and the fixed cheeks
are wide. Palpebral ridges extend, at right angles to the axis, from the anterior lobe
of the glabella to the arcuate palpebral lobes.

The thorax consists of about nineteen segments, with narrow axis and widely
furrowed pleura that terminate in spines.

The pygidium is small with sharp pleural ridges and marginal spines.

Remarks: The notches of the glabella and the narrow axis of Rhodonaspis
suggest a close relationship to Papyriaspis and Hedinia. The genus is sufficiently
distinct from them in the features of the glabella, the palpebral ridges and in the
long, marginal spines of the thorax and pygidium. In the development of these
spines and in certain features of the head comparison may be made with several
genera of Oleninae, notably Olenus and Eurycare ; although the assemblage of
characters indicates Papyriaspinae as the appropriate subfamily.

88 “‘oddov a rose, dois a shield—in reference to a resemblance to a serrated rose leaf.

220 MEMOIRS OF THE QUEENSLAND MUSEUM.

RHODONASPIS LONGULA sp. nov.
(Pl. XXIII, fig. 9.)

Diagnosis : Cranidium wide and flat. The glabella has strictly parallel sides
and broadly rounded front ; and, with the occipital ring, is about twice as long as it
is wide. The occipital furrow is wide, prominent and continuous. In front of ‘it
are two pairs of wide glabellar furrows extending only a short
distance on each side from theaxial furrow, thus giving the glabella
a notched appearance; anteriorly there is a trace of a pair of
circular pits, not in contact with the axial furrow. The fixed
cheeks are very wide and flat. The palpebral lobes are arcuate
and have a narrow but well defined and very uniform rim. This

Wieiier rim continues anteriorly beyond the region of the lobes. Palpebral
storation, ; ‘ . :

ridges are present, at right angles to the axis, connecting
the anterior end of the palpebral lobe to the middle portion of the anterior

glabellar furrow.

The thorax consists of nineteen segments and is lanceolate in outline. The
axis tapers uniformly. In the anterior half of the thorax it is about half the width
of a pleuron, but towards the rear it is relatively wider. Each segment of the axis
has a median transverse ridge. The pleura are flat. Each is marked by a very wide
furrow with parallel sides as far as the fulcrum. The pleura end in narrow recurved
spines.

The pygidium is very small. The axis, which has three furrows, extends
almost to the posterior margin. The pleura have one pair of faint furrows on the
interior portion but otherwise they are smooth. They have, on each side, three
reflexed, marginal spines.*

Locality and horizon: One specimen, the holotype, is in the collection of the
University of Queensland, from about two miles south of Tyson’s Bore on Glenor-
miston (Rhodonaspis Stage).

RHODONASPIS PROSECTA sp. nov.
(Pl. XXII, figs. 10-12.)
Diagnosis : Cranidium quadrate in outline. The glabella is long with parallel

sides and an abruptly truncated front. Anterior to the occipital furrow there are
three pairs of equally spaced, discontinuous and sharply incised furrows. The

- The pygidium as drawn in the text-figure is slightly incorrect. ‘The axis is too short
and two tiny posterior spines have been omitted.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 221

posterior pair slope backwards towards the axis. The two anterior pairs are at right
angles to the margin. The fixed cheeks are wide, expanding considerably in the
palpebral regions. The anterior portion of the facial suture are parallel. The
palpebral lobes are sub-semicircular with a uniform, narrow, raised rim. They are
continued to the glabella by palpebral ridges of similar width to the palpebral rim.
These ridges are in line and parallel to the anterior margin of the head. The anterior

limb is narrow and convex and is succeeded by a narrow, raised, cord-like anterior rim.

The pygidium is sub-semicircular in outline. The axis is conical, prominently
raised, and has five or six rings on the axis. The flanks bear three sharply-ridged,

radiating, ribs that terminate in spines.

Remarks: This species differs from R. longula in its shorter anterior limb and

in the greater prominence of the pygidial ribs.

Locality and horizon: From the Rhodonaspis Stage, at the western base of
Polly’s Lookout, Glenormiston (University of Queensland Collection).

RHODONASPIS SP.

(Pl. XXIII, fig. 13.)
Several fragments of a species have been found with the following features :

Cranidium imperfectly known. The anterior margin is rounded. The glabella
is long, slightly tapering and is not prominently bounded anteriorly. The two
posterior pairs of discontinuous, glabellar furrows slope backwardly towards the
axis. The anterior pair are at right angles to the margin. The anterior limb is
moderately wide and covered with fine radiating wrinkles.

It is distinct species in a number of features—the slow tapering of the glabella,
the slight obliquity of the second pair of glabellar furrows, the indefinite boundary
to the front of the glabella, and the rounded anterior margin of the cranidium.
However the fragments, if figured, would make it difficult to picture the species,

so that I postpone giving it a name until more complete specimens are collected.

Locality and horizon: From the Rhodonaspis Stage, five miles north of the
Twenty Mile Bore, Glencrmiston (University of Queensland Collection).

222 MEMOIRS OF THE QUEENSLAND MUSEUM.

Subfamily OLENINAE Kobayashi, 1935.
Genus OLENUS Dalman, 1827.
Genotype: Hntomostracites gibbosus Wahlenberg.
OLENUS (?) SP.

(Pl. XXIII, fig. 14.)

A cranidium has been obtained from the Elathriella Stage that tentatively
may be placed in the genus Olenus. It is imperfectly preserved. The wide and long
anterior limb, the fairly horizontal palpebral ridges, the presence of three pairs of
glabellar furrows (instead of two) before the occipital furrow, suggest that it is not an
Elathriella. Unfortunately the course of the facial suture behind the palpebral lobes
is hardly known. The form, from such of its features as are known, agrees fairly well
with Olenus and I leave it there for the present. Olenids are recorded from several
areas in the pacific province. Kobayashi (1936a) has recently reviewed all these
records.

Locality and horizon: From the Elathriella Stage of the Pituri Sandstones
at the base of a hill immediately west of Tyson’s Bore on Glenormiston (University
of Queensland Collection).

A somewhat similar and indefinite fragment has been found in the beds of the
Glyptagnostus Stage, 16 miles south of Glenormiston homestead.

Family ANOMOCARIDAE Poulsen, 1927.
Genus ANOMOCARE Angelin, 1854.
Genotype: Anomocare laeve Angelin.
ANOMOCARE CONFERTUM sp. nov.

(Pl. XXII, figs. 22-28.)

Description: The cranidium is of average size for the genus. The glabella
which slowly tapers anteriorly has a rounded front, and its length is about twice its
maximum width. The occipital furrow is distinct. Two other glabellar furrows are
indicated slightly on each side. The neck ring has no median
spine. The anterior limb is slightly concave, extending to a
distance equal to about one-third the length of the glabella. Fixed
cheeks are semicircular, with regular palpebral lobes impinging at

\ . each end upon the glabella. The width of each fixed cheek is

GS) slightly less than the maximum width of the glabella. The postero-
lateral limbs of the fixed cheeks are very small and rudimentary.
Facial sutures converge towards the anterior ends of the
anterior portion of the glabella and almost impinge upon it where they meet the
anterior ends of the palpebral lobes.

Restoration.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 223

The free cheeks have narrow, flattened lateral and posterior borders, and
short, acute, genal spines.

‘Thoracic segments are narrow.

The pygidium has an axis that narrows but becomes more convex posteriorly.
Axial furrows are not very prominent, while the furrows on the pleural lobes become
merely incised lines. A smooth, regular margin, inflected at the posterior apex, is
present, the uniform width of which is approximately equal to the maximum width
of each pleural lobe.

The hypostome is elongate and oval with the anterior lobe highly inflated
and. occupying about two-thirds of the total area. The posterior lobe merges into
the lateral borders. Maculae are absent.

Remarks: The genus Anomocare was originally described from Scandinavia.
Since then it has been recorded in Siberia, Bennett Island (north of Siberia), England,
Sardinia, North America, China, Tonkin, India and Persia. The forms from Bennett
Island, described by Holm & Westergaard in 1930, are close to the genotype ; but it
seems hardly likely that any of the species from countries other than this and from
Scandinavia can belong to Anomocare in its restricted sense. Kobayashi and also
Resser & Endo have recently removed many of the eastern asiatic species to other
genera. In view of this it is curiously interesting to record now from Australia a
species that 7s close to the genotype.

The specific features have been determined from a number of fragments.
A. confertum has a cranidium in which the palpebral lobes impinge upon the glabella,
and the facial sutures join the anterior portions of the lobes near these points of
contact. This feature is seen also in A. excavatum Angelin,®® A. extornatum Holm and
Westergaard (1930, pl. 4, figs. 20-23) and A. sibiricum Holm and Westergaard (1930,
p. 17, pl. 2, figs. 15-20, pl. 3, figs. 1, 2). From these three species 4. confertum is
easily distinguished by several minor features, for instance its more acuminate
glabella. Except for this aspect of the facial suture and for the somewhat shorter
frontal limb, the cranidium of A. confertum is very similar to that of the genotype,
A. laeve Angelin, which is probably the most closely allied form.

The inflected margin of the pygidium is a characteristic feature of this species.
In this respect and in other pygidial features A. laeve* is again the most similar form.

39 Angelin, 1854, p. 25, pl. 18, fig. 3. See also Hohn & sicaee ii 1930, p. 16, pl. 2, figs.
1-14, pl. 4, figs. 19-21.

40 That is, in the interpretation of Holm & Westergaard. There has been some disagreement.
about what is the pygidium of A. laeve. Grénwall (1902, p. 141, pl. 4, fig. 9), and following him
Walcott (1913, pl. 17, fig. 1) have figured a pygidium supposed to belong to this species; but
Holm & Westergaard (1930, p. 17, pl. 4, fig. 18) disagree with this interpretation and have figured
the pygidium that, according to Westergaard, is associated with the typical cranidia at the type
locality.

224 MEMOIRS OF THE QUEENSLAND MUSEUM.

The hypostome has the characteristic inflated form for Anomocare and may be
compared particularly with that of A. sibiricum.

Locality and horizon : Holotype and several other specimens in the University
of Queensland Collection from the Anomocare Stage, five miles east of Harris Creek
on the old main road from Camooweal to Burketown, via Thorntonia Station (52 miles
from Camooweal),

ANOMOCARE (?) ANGUSTUM sp. nov.
(Pl. XXIII, fig. 21.)

Diagnosis: Cranidium narrow. Glabella long and tapering, with straight
sides and rounded front. The occipital furrow is sharply incised at the sides but very
faint inthe centre. The posterior glabellar furrow proper is reduced to two depressions.
Anteriorly there is a faint, sub-continuous furrow. The frontal limb is slightly concave,
bounded by a narrow, anterior border. The fixed cheeks are narrow. Palpebral lobes
are small, near the middle of the glabella. The anterior branches of the facial suture
are divergent ; the posterior branches are short and, apparently, sub-normal to the
posterior margin.

Remarks: This species is apparently a member of Anomocaridae, but the
palpebral lobes are toossmall for a member of Anomocare itself. It occurs in beds
immediately older than those with true Anomocare. Only the one cranidium has
been found ; and since there is no properly defined genus to accommodate it, the species
may be left provisionally in Anomocare, pending the discovery of more material.

Attention may be called to the casual resemblance the species bears to the
earlier forms of Proasaphiscus, near the base of the Middle Cambrian, particularly
to P. ephori (Walcott), (see 1913, p. 190, pl. 18, figs. 5) and P. centronatus Endo and
Resser (1937, p. 259, pl. 37, figs. 17-20).

Locality and horizon: From limestones of the Papyriaspis Stage at the road.
crossing of V Creek, between Undilla and Thorntonia (University of Queensland
Collection).

Genus EUGONOCARE gen. nov."
Genotype : Eugonocare tessellatum sp. nov.

Diagnosis: Cranidium subquadrate. The glabella is subrectangular with
occipital furrow and two pairs of discontinuous furrows. The palpebral lobes are
about one-half the length of the cranidium, well separated from the glabella and
connected to it by faint palpebral ridges. The anterior limb is moderately wide and

41 ei-ywvios with regular angles, «apy the head.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 225

convex with a surface covered with very fine radial wrinkles. The anterior border
is narrow, convex, and with a parallel furrow behind. Facial sutures have the anterior
portion more or less at right angles to the anterior margin of the cephalon.

The pygidium is slightly smaller than the cranidium. Its axis has five or six
rings and the pleura are faintly furrowed. There is a smooth, flattened border.

Remarks: In its pygidial features and in many of the aspects of the head
Eugonocare resembles Anomocare ; but the squat glabella, the straighter facial sutures
and the raised anterior border are cranidial features that sufficiently distinguish it.

Poulsen (1927, p. 325) has suggested that Anomocaridae arose from
Ptychopariidae by the forms becoming “isopygous and in having fewer thoracic
segments, and furthermore, the palpebral lobes are considerably larger than in that
family.”” I also am inclined to this view. Poulsen stresses, in addition, the wide,
strongly concave rim of the head of Anomocaridae and the strongly furrowed pygidium
with its broad, concave border. It is in these features, with the exception of the
number of thoracic segments (of which I have no knowledge) and the concavity of the
anterior limb, that Huganocare differs from certain typical members of Ptychopariidae.
(It should be noted, by the way, that the anterior limb of Anomocare is not always
concave—the genotype A. laeve is somewhat convex). ;

Many of the American and Asiatic forms that have been placed in Anomocare
and Anomocarella and which recently have been grouped into other genera” by various
authors may represent similar offshoots from ptychoparian stock. The development
of isopygous forms with wide-bordered pygidia and increased eyes seems to me to be
a normal tendency with Ptychopariidae. If this be so a number of family names
will be needed in future to accommodate whatever natural groups may be recognised.
For the present, however, until more is certain of lineages, I am placing Eugonocare
in Anomocaridae. Kobayashi (1935), it may be noted, already has employed a
considerable family nomenclature for such forms, although his grouping indicates
that he does not regard them as having a common ancestry in Ptychopariidae.

Pertinent comparison may be made with Anomocarella Walcott, Lioparia
Lorenz, Tollaspis Kobayashi and Manchuriella (Resser and Endo ms.) Kobayashi,
which genera, with Anomocare, seem to me to be the most similar members of this
group. Anomocarella, in its restricted sense, is rather similar in its pre-glabellar
_ structures ; but it has no glabellar furrows and the facial sutures are more sinuous.
Inoparia has a concave anterior limb and smaller palpebral lobes. T'ollaspis** is the

4 Such genera as Anomocarella (sensu stricto), Lioparia, Coosia, Kingaspis, Eymekops,
Koptura, Manchuriella, Tollaspis and Glyphaspis.

43 As mentioned elsewhere in this paper (p. 228) it is probable that Tollaspis as defined by
Kobayashi was polyphyletic. My remarks apply to the genotype, Anomocare pavlowskii Schmidt.

226 MEMOIRS OF THE QUEENSLAND MUSEUM.

genus that is perhaps most similar to Hugonocare. There is a smaller quadrate outline
to the cranidium and glabella, similar anterior border and glabellar furrows, but the
anterior limb is narrower. The pygidium of this genus has not yet been figured unless
possibly it is represented by the specimen figured by von Toll (1899, pl. 2, fig. 11) as
Bathyuriscus howelli Walcott. Manchuriella is generally similar but it has a narrower
cranidium and a rather longer glabella.

EUGONOCARE TESSELLATUM sp. nov.
(Pl. XXIU, figs. 15-18; pl. XXV, fig. 7, b.)

Diagnosis: Cranidium subrectangular, the width slightly greater than the
length. The glabella narrows only very slightly, has an abruptly truncated anterior
and thus has a subrectangular form. With the occipital ring it is three-quarters the
length of the cranidium. The occipital furrow is rather shallow. In front of it are
two pairs of discontinuous glabellar furrows that slope backwards at a prominent
angle. The axial furrow is prominent. Beyond it the free cheeks and the anterior
rim arise convexly, The free cheeks are wide, each almost as wide as the glabella.
The facial sutures extend more or less at right angles to the anterior and posterior
margins of the cephalon, and are only slightly modified by the palpebral lobes. These
lobes are between one-third and one half the length of the glabella, are narrowly
bordered and are continued to the anterior angles of the glabella (sloping forward at
a slight angle), by faint palpebral ridges. The anterior rim is convex in its posterior
portion but changes into a groove immediately behind the anterior border. This
border is sharply defined and convex. |

The pygidium is sub-semicircular and has a raised anterior rim. The axis is
distinct but is not prominently inflated and bears four raised narrow rings (dividing
the fused segments) that decrease in prominence posteriorly. The pleura bear faint
ridges, extending from the axial rings for a short distance. The portion of the pygidium
(axis and pleura) that have these traces of segmentation is somewhat markedly
triangular. Beyond it and rather abruptly separated from it there is a wide smooth
border set at an obtuse angle to the furrowed part.

Remarks: I would call attention to the Persian form figured by King (1930,
p- 319, pl. 17, fig. 5) as Anomocare subquadratum (Dames). This species which probably
is distinct from Dames’ form, is more similar to LZ. tessellatum than any other I have
seen and may be congeneric.

Locality and horizon : Holotype (a cranidium), other cranidia and two pygidia
are in the collections of the University of Queensland from the Hugonocare Stage
about four and a half miles north of the Twenty Mile Bore on Glenormiston.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 227

EUGONOCARE PROPINQUUM sp. nov.
(Pl. XXII, figs. 19, 20.)

Diagnosis : Cranidium indistinguishable from that of EH. tessellatum.

Free cheeks subtriangular, the genal angles produced into long spines oblique
to the general length. The rim is flat and the surface is covered with fine radial
wrinkles. | ?

The pygidium is sub-semicircular. The axis has four or five rings that are well
impressed. The pleura are convex and on each side have four rounded ribs that extend
straight and radially for two-thirds of their length and then rather abruptly contract.
The margin is flat and narrow.

Remarks: At a locality a little further north from where the genotype (£.
tessellatum) was collected another species has been obtained. I can find no features
on the cranidium that will allow me to separate the two forms, but the pygidia are
different. On LF. tessellatum the pleural portions of the pygidium are sub-triangular
and are bounded by a wide margin. In H. propinquum these regions are sub-semicircular
and the margin is narrow. Several heads and tails have been found at each locality
and these resemblances and differences are constant.

Locality and horizon: From the Eugonocare Stage, five miles north of the
‘Twenty Mile Bore, Glenormiston (University of Queensland Collection).

EUGONOCARE SP. INDET.

From the Glyptagnostus Stage, about sixteen miles south of Glenormiston
homestead, cranidia of Hugonocare have been collected. No pygidia have been
obtained ; and without these it is not possible to place the form specifically.

: Genus ANOMOCARELLA Walcott, 1905.

Genotype: Anomocarella chinensis Walcott.

ANOMOCARELLA (7?) SP.
(Pl. XXII, fig. 14.)

One cranidium from the Hurostina Stage of No, 4 bore, Alroy Downs is figured.
It has some affinities with Anomocarella but is most likely distinct. It is illustrated
for future comparisons. (University of Queensland Collection).

228 MEMOIRS OF THE QUEENSLAND MUSEUM.

Suborder ELLIPSOCEPHALIDA nov.
Family ELLIPSOCEPHALIDAE Matthew, 1887.
Genus DINESUS Etheridge fil., 1896.
Genotype: Dinesus ida Etheridge fil.

In 1896 Etheridge founded the genus Dinesus on associated cranidia and
pygidia from Knowsley in Victoria. Later, in 1903, Gregory gave reason to believe
that these were fragments of different genera ; and he restricted Dinesus to the type
of cranidia that Etheridge figured, transferring the pygidia (which he associated with
quite different cranidia) to a new genus, Notasaphus. Most later authors, including
Etheridge, have concurred in Gregory’s grouping ; and the complete specimens of
species of both genera now figured give added testimony to this. A pygidium that
more likely is to be regarded as that of Dinesus ida (the genotype of Dinesus) was
figured by Chapman in 1917 (pl. 7, fig. 22) and a complete specimen of D. ida from
the Templeton Series in Queensland was figured by Kobayashi in 1935.

While the relationship of the fragments of the test was uncertain the affinities
of Dinesus could not be decided. Walecott’s remarks (1905, p. 35; 1913, p. 124),
comparing the genus to Dorypyge, were based on Etheridge’s original interpretation
of the fragments.

Gregory (1903, p. 154) who knew only the cranidium, considered Dinesus to
belong to the Asaphidae. Kobayashi (1935, p. 182) who had a complete specimen,
placed it in Dolichometopinae. Complete forms, showing the tiny pygidium, give
sufficient evidence that it is not an asaphid ; -and I find it almost equally difficult to
believe that it is a dolichometopid. The features of the cranidium, thorax and.
pygidium are, however, similar to those of Hllipsocephalus and I now suggest placing
it in the same family Ellipsocephalidae. Dinésus differs from E llipsocephalus in having
thoracic spines, basal glabellar lobes and antero-lateral cephalic furrows that connect
the axial and anterior furrows. Such features appear independantly in many groups
of the trilobites, although it is rare for antero-lateral furrows to be so deeply impressed
as they are in Dinesus. .

Dinesus is present in the early Middle Cambrian beds of Queensland and
Victoria. Elsewhere (Whitehouse 1936, p. 73) I have suggested that it is present also
in the Siberian Cambrian, being represented by Solenopleura (?) sibirica (Schmidt)
as figured by von Toll (1899, p. 36, pl. 2, figs. 12, 13, 16). Kobayashi (1935, p. 262)
included that Siberian species in 7'ollaspis. I do not agree with this; for the features
of the cranidium, particularly of the glabellar structures, are very different from those
of Anomocare pawlowski Schmidt, the genotype of Tollaspis. The relatively smooth
glabella, the small eyes, the basal lobes, the antero-lateral furrows and the general
proportions of Solenopleura (?) sibirica are all typical of Dinesus.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 229

DINESUS IDA (Etheridge fil.)
(Pl. XXIV, figs. 1, 2, 3a, 3b.)

1896 Dinesus ida Etheridge Jr., p. 56, pl. 1, figs. 1-4 (non figs. 5 and 6).
1903 Dinesus ida Gregory, p. 155, pl. 26, figs. 8-10.

1905 Dinesus ida Walcott, p. 35.

1913 Dinesus ida Walcott, p. 124.

1917 Dinesus ida Chapman, p. 98, pl. 7, figs. 19-22.

1919 Dinesus ida Etheridge Jr., p. 381.

1935 Dinesus ida Thomas, p. 92.

1935 Dinesus ida Kobayashi, p. 134, pl. 22, fig. 3.

A number of complete dorsal shields of a Dinesus have been found on the
Templeton River in Queensland. Three of these are now figured and Kobayashi has
figured a fourth. A certain range of variation is apparent in these forms. Many
cranidia of D. ida have been found in the Knowsley beds of Victoria and a number
of these have been illustrated by Etheridge, Gregory and Chapman. They too show
a range of variation, and this range is comparable to that seen in the Queensland
cranidia. For that reason I see no advantage, on present evidence, of making a
specific distinction. Placing the Queensland species in D. ida is in some degree a
tentative measure; for although the cranidia agree very well it may be that other
portions of the test, if they were known in the Victorian beds, might show some
difference. If the small pygidium that Chapman has figured from Knowsley (1917,
pl. 7, fig. 22) really belongs to D. ida it would indicate such a difference. But this
pygidium can be assigned only provisionally to D. ida—Conceivably (and I think
quite likely) it might belong to some ptychopariid genus.

A specimen (Pl. XXIV, fig. 3a) that is figured from wax squeezes of external
moulds shows that axial spines were present on the occipital ring and on each thoracic
segment. The other specimens do not show such spines. Presumably they are
internal moulds of the test, which thus would appear to have had solid axial spines.
Figures 3a and 3b on plate XXIV show respectively the dorsal surface’ and the
internal mould of the same individual; and the evidence of spines on the outer
surface but not on the internal mould is thereby clearly illustrated.

The specimen shown by plate X XIV, figure 1, has a wider and squarer glabella
than the others; but this.would appear to be due to a subsequent flattening of the
test. The specimen figured on plate XXIV, figs. 3a and 3b, has a_ perceptibly
narrower thorax than the other ; but I do not regard it as specifically distinct.

For comparative purposes I have tabulated the dimensions of the five
moderately complete specimens as follows. The first entry (1) for each specimen is
the overall length of the test in millimetres. The other entries are given as percentages

44 A wax squeeze of an internal mould.

230 MEMOIRS OF THE QUEENSLAND MUSEUM.

of this. These are, in order; the length of the cranidium (cl), the width of the
cranidium at its base (cb), the width of the cranidium across the eyes (ce), the width
of the glabella (gw), the width of the thorax at the fourth segment (tw) and the width
of the axis of the fourth thoracic segment (aw).

— I cl cb ce gw tw aw
Unfigured specimen 24-6 36 55 41 15 60 17
Pl. XXIV—fig. 3 P $7 24-6 38 55 37 13 53 17
Pl. XXIV—fig. 2 aa «| 27-0 37 59 39 14 56 15
Pl. XXIV—fig. 1 28-8 37 58 37 15 57 16
Kobayashi’s specimen 28-5 37 59 41 16 58 19

Based on this Queensland material the species may be redefined in the following
terms :
The dorsal shield is of an hour-glass shape.

The cephalon is tranversely elongate, the length approximately one half the
width. Its margin which is straight across the cranidium and arcuate in the free
cheeks has a narrow, raised rim. The maximum width of the cephalon is at the
posterior margin. The cranidium is rhomboidal in outline. The glabella is uniformly
inflated, has parallel sides and rounded front, and extends almost to the anterior
furrow.. On it there are a pair of posterior glabellar furrows that unite with the
occipital furrow, thus including a pair of narrow basal lobes. No
other glabellar furrows are present. There is a small solid median
spine on the occipital rng. From the antero-lateral angles of the
glabella a pair of short, deep, diverging furrows extend, connecting
the axial and anterior furrows. The anterior limb is very narrow.
The fixed cheeks are narrow in the anterior and median regions.
Palpebral lobes are small (about one quarter the length of the
cranidium), centrally situated and have no palpebral ridges. The
anterior portions of the facial sutures are subparallel, while the

Restoration.

posterior portions slope, fairly uniformly, at an angle of approximately 45° to the
length. The free cheeks are relatively wide, are little inflated and have narrow,
slightly recurved genal spines that arise rather abruptly from the genal angles.

The thorax is ovate in outline, with its maximum width in the region of the
fourth segment. It is not as wide as the cephalon. This ovate outline of the thorax,
combined with the relatively great width of the cephalon, gives the curious hour-
glass shape to the dorsal shield as a whole. There are eleven segments in the thorax.
The axis, in the middle region, is about one third the total width, and it bears on

THE CAMBRIAN FAUNAS. OF NORTH-EASTERN AUSTRALIA. 231

each segment a small solid spine. The transverse furrows of the axis are wide. The
pleura are widely furrowed, the furrows having subparallel sides. They terminate
rather abruptly in small and somewhat nodular structures that are continued
posteriorly as small spines.

The pygidium is very small, apparently without trace of transverse grooving.
The axis is rather more than one-third the whole width and extends to the posterior
margin. The pleura are subtriangular and, apparently, smooth.

The hypostome is not yet known.

Only one other species appears to be known—the Siberian form described by
von Toll (1899, p. 36, pl. 2, figs. 12, 13 and 16) as ? Solenopleura sibirica (Schmidt).
This form appears to have very faint glabellar furrows in advance of the basal lobes.

Such furrows have not been seen on any of the Australian specimens. Also it has a
slightly wider anterior limb on the cranidium and, correlated with this, a slightly
convex anterior margin.

Locality and horizon :- The species is represented by specimens in the collections
of the University of Queensland, the Geological Survey of Queensland and the
University of Harvard. They all come from beds of the Dinesus Stage of the Middle
Cambrian in the Templeton River area, N.W. Queensland. »

Genus ELLIPSOCEPHALUS Zenker, 1833.
Genotype : T'rilobites hoffi Schlotheim.
_ ELLIPSOCEPHALUS (7) SP.
(Pl. XXIV, figs. 11, 12.)

Several minute cranidia have been found in the beds of the Hugonocare Stage,
five miles north of the Twenty Mile Bore on Glenormiston, that may represent the
genus Ellipsocephalus. However they are so small that, possibly, they are the larval
stage of some genus yet to be determined. Two of these cranidia are now figured.

Family PAGODITDAE Kobayashi, 1935.
Genus IDAMEA gen. nov.”
Genotype : Idamea venusta sp. nov.

Diagnosis: Cranidium rhomboidal in shape, with a smooth glabella that
extends to the anterior rim. The sides of the glabella are parallel or very slightly
converging. The occipital ring widens at its centre. The fixed cheeks are narrow,
convex, with a small palpebral lobe slightly behind the central point. There are no
palpebral ridges. The anterior rim is narrow and convex.

45 T have thought to perpetuate in this way the name of the Idamea tribe. of aborigines in
whose territory these fossils were collected. Idamea was also the origmal name of Glenormiston
Station. (The pronunciation is {-dia-mé-ii.)

232 MEMOIRS OF THE QUEENSLAND MUSEUM.

The pygidium has an axis with several rings, and pleura that are obliquely
furrowed. There is a flattened margin to the pygidium.

Remarks: Kobayashi placed in Pagodiidae the five genera Pagodia Walcott,
Insania Walcott, Aojia Resser and Endo, Hardyia Walcott and Pseudolisania
Kobayashi, although for reasons that I find difficult to follow, he thought Lisania
might be excluded.

Idamea is very similar to Pagodia. Indeed my previous references to Pagodia
(Whitehouse 1930 and 1936) were based on this form. However the fixed cheeks are.
narrower, the occipital ring widens centrally, and there are no traces of glabellar
furrows such as are very faintly impressed on Pagodia. Lisania and Aojia are possibly
closer but they have larger eyes. Pseudolisania has not the sharply defined anterior
rim. Hardyia, however, is the genus that most closely resembles Jdamea. The eyes
are slightly more anterior in position in that genus and Walcott mentions the presence
of faint glabellar furrows that, however, do not show in his figures. Jdamea seems
to be devoid of glabellar furrows anterior to the occipital furrow.

Certain members of other families occasionally mimic the cranidium of Jdamea.
For instance, Menocephalites acis Walcott sp. (1913, p. 175, pl. 16, fig. 11) loses trace
of glabellar furrows and so is similar in general form to Jdamea—but it still retains
the characteristic pustulose ornament of its own genus.

Idamea has been found in the Rhodonaspis and Elathriella Stages of the
Queensland Cambrian.

IDAMEA VENUSTA sp. nov.
(Pl. XXIV, figs. 4-6.)

Diagnosis : Cranidium rhomboidal in outline. The glabella is inflated, smooth
and converges very slightly until it reaches the anterior rim. The occipital furrow
is moderately well incised and behind it the occipital ring widens slowly towards
the centre. The fixed cheeks are narrow and convex. The facial sutures converge
constantly and fairly evenly towards the front. The palpebral lobes are small,
markedly raised from the facial suture posteriorly, and have no continuation as

palpebral ridges.

The pygidium has a prominent axis that is rather less than one-fifth the width
of the shield. Five or six furrows are impressed on the axis, these furrows becoming
crowded in the posterior portion. The pleura, which are fringed by a fairly wide,
flat margin, bear three or four simple furrows on each side. The axis reaches to the
anterior edge of this marginal zone.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 233

Remarks : Three cranidia and one pygidium have been examined. Associated.
with them are fragments of free cheeks with long genal spines ; but I am far from
. certain that they belong to this species.

Locality and horizon: From the Rhodonaspis Stage of the Upper Cambrian,
fourteen miles south of Glenormiston homestead (Geological Survey of Queensland
collection).

IDAMEA SUPERSTES sp. nov.
(Pl. XXIV, fig. 7.)
Diagnosis : Cranidium short. The glabella is wide and smooth, with subparallel
sides (very slightly converging). The free cheeks are narrow and the palpebral lobes
very small. The anterior rim is narrow and convex.

Remarks: Only one cranidium of this form has been found. It is generally
similar to I. venusta; but the glabella is shorter and wider, with more strictly
parallel sides.

Locality and horizon: From the Hlathriella Stage of the Upper Cambrian at
the base of a hill immediately west of Tyson’s Bore on Glenormiston (University of
Queensland Collection).

Suborder CoryNExocHiIpA Kobayashi, 1935.
Family CORYNEXOCHIDAE Angelin, 1854.
Subfamily CORYNEXOCHINAE Raymond, 1928.

Genus CORYNEXOCHUS Angelin, 1854.
(= Karlia Walcott, 1889.)

Genotype : Corynexochus spinulosus Angelin.

Much has been written on the genus Corynexochus in recent years. Walcott
in 1916 revised the genus and published figures of all the forms that, until that. time,
had been referred to Corynexochus. Considerably earlier (in 1889) Walcott had
introduced the generic name Karlia; but in his revision of 1916 he made Karlia‘a
synonym of Corynexochus. Notwithstanding this Raymond (in 1928) and Kobayashi
(in 1935) have retained Karlia as a separate genus. However, the genotypes of the
two are so similar that, like Lake (1934), I am satisfied to regard the genera as identical.
As a distinguishing feature Raymond noted that the pygidia of the forms referred to
Karlia had no spines, whereas spines were supposed to be present on the pygidium of
C. spinulosus, the genotype of Corynexochus. But from the observations of Matthew
(1899), of Grénwall (1902) and of Lake (1934) it is reasonably certain that pygidial
spines are not present on any known form of Corynexochus so that this distinction
is invalid.

Q

234 MEMOIRS OF THE QUEENSLAND MUSEUM.

In his review of 1916 Walcott had recognised a subgenus (Bonnia) of
Corynexochus ; but recent writers agree in giving this group full generic rank.

The species that in 1916 Walcott placed in the type section of Corynexochus
included both Lower Cambrian and Middle Cambrian forms. Lake (1934) with good
reason removed the Lower Cambrian species but retained all the Middle Cambrian
forms with the exception of C. stephenensis (Walcott). He included also the Welsh
species C. cambrensis Nicholas. As a further modification I would suggest removing
C. clavatus (Walcott) which has quite different palpebral lobes.

By this reading the genus is restricted to Middle Cambrian faunas known
only in Scandinavia, Britain, France and Newfoundland—that is, it is a genus of
the North Atlantic Province. To this group I now add an Australian species.

In the Northern Hemisphere Corynexochus ranges through much of the Middle
Cambrian—from the zone of Hypagnostus parvifrons to the zone of Paradoxides
forschammeri. Whether it goes into the zone above (the zone of Lejopyge laevigatus)
which marks the top of the Middle Cambrian, I am not sure. The Australian form
in my opinion (Whitehouse 1936, p. 76) extends the range into the basal zone of the
Upper Cambrian. Grabau (1937, p. 378), who postulates a disconformity between
the Middle and Upper Cambrian, has suggested that the Corynexochus in these beds
is “ a weathered-out, residual specimen from the Middle Cambrian beds of that region.”
That cannot be accepted ; for not one but a number of specimens of Corynexochus
plumula have been collected in these beds, and their preservation does not suggest
that they are remané fossils.

CORYNEXOCHUS PLUMULA sp. nov.
(Pl. XXTV, figs. 8-10.)

. Diagnosis : The cranidium is very convex, composed of three smooth, rounded
elements due to the extension of the axial furrows to the anterior margin. Each of
these furrows curves in a graceful arc. The glabella expands towards the front, is
about twice as long a8 its maximum (anterior) width, and extends anteriorly beyond
the limits of the fixed cheeks. It has a broadly arcuate anterior margin prominently
distinct from the lateral furrows. The occipital furrow is well shown but no other
furrows are present on the glabella. The occipital ring is highly convex (almost
tubercular) and extends posteriorly well below the limits of the cheeks. The fixed
cheeks are roughly in the form of isosceles triangles with the height greater than the
breadth. At the base the width of each fixed cheek is about the same as that of the
glabella. These cheeks are inflated, are quite smooth, and at their lateral margins
curve steeply downwards. The posterior marginal furrows are broad and increase

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 235

in width towards the sides where they curve forward in sympathy with the posterior
lateral portions of the cranidium. The palpebral lobes have slight curvature, are
rather widely flattened, contiguous with the glabella and occupy about one half of
the lateral margin of the fixed cheeks.

The free cheeks have not been observed ; but from the shape of the margin
of the fixed cheeks and the course of the posterior marginal furrow they must have
been very small.

The pygidium is relatively large and moderately smooth. The axis extends
almost to the posterior margin, is highly inflated and bears three faint furrows. The
pleura bend prominently downwards in the marginal area and bear no spines. On
each side there are two faint furrows (the posterior furrow hard to discern) that are
directed anteriorly from the two anterior axial furrows. A broad, raised anterior
rim is present on the pleura, prominent but not sharply limited. The rim has its
maximum development about midway along its length.

Remarks ; Overseas only five species seem to have been described that may
be left in Corynexochus. These are C. bornholmiensis Gronwall (1902, p. 137, pl. 4,
figs. 1, 2) which ranges from the zone of Hypagnostus parvifrons to, possibly, the
zone of Paradoxides davidis in Scandinavia ; the Scandinavian C. spinulosus Angelin®
from the zone of Paradoxides forschammeri; the Welsh C. cambrensis Nicholas (1916,
p. 464, pl. 39, figs. 4-6) from the zone of Paradoxides davidis ; C. minor (Walcott)*
from the Paradoides davidis zone in Newfoundland ; and the French C. delaget Miquel
(1905, p. 481, pl. 15, fig. 4), of the zonal position of which I am not certain. I have
given reason (1916, p. 77) for my belief that the new species C. plumula is slightly
later in age, representing the zone of Agnostus pisiformis.

C. bornholmiensis, the earliest of the forms to appear, has faint glabellar
furrows. Such furrows are intensified on C. spinulosus and C. minor, but disappear
from C, cambrensis and C. delagei. With these latter species it is most apt to compare
C. plumula since it, too, has a smooth glabella. Both C. cambrensis and C. delager
are known only from distorted fragments. In both those species, however, the
pygidium is clearly different. In-C. cambrensis the axis and in C. delagei both axis
and pleura are more deeply furrowed. In both species also the glabella appears to
be narrower at the base than it is in C. plumula.

Locality and horizon: Five cranidia and three pygidia are in the collections of
the University of Queensland from the Bugonocare Stage, about four = a half miles
north of the Twenty Mile Bore on Glenormiston.

46 Angelin, 1854, p. 59, pl. 33, fig. 9 only ; See also Grénwall 1902, p. 139, pl. 4, figs. 3, 4.
47 See Walcott 1916, p. 324, pl. 36, fig. 7.

236 MEMOIRS OF THE QUEENSLAMD MUSEUM.

Subfamily DoticHomEtToPiInaE Walcott, 1916.
Genus AMPHOTON Lorenz, 1906.
Genotype : Dolichometopus deois Walcott. —

The name Amphoton was proposed by Lorenz (1906, p. 75) for a species that
he named Amphoton steinmanni. However, in the previous year Walcott (1905, p. 94)
had described a species, Dolichometopus deois, that, as he pointed out later (1913,
p. 216), is identical with Lorenz’s genotype. In this latter paper Walcott retained
the species in Dolichometopus and thus regarded Amphoton as a synonym. This
procedure later was followed by Sun (1924, p. 81).

Walcott described five species from the Middle Cambrian of China under the
name Dolichometopus. These are uniformly different from the genotype, D. svecicus
- Angelin (1854, p. 72, pl. 37, fig. 9), in the characters of the pygidium, as Walcott
(1916, p. 359) had noted. They form a related group and I agree with Kobayashi
(1935, p. 137) that Amphoton should be revived as a generic name for it. There can,
be little doubt, as Walcott points out, that Dolichometopus deois Walcott is identical
with and has precedence over Amphoton steonmanni Lorenz. The genotype thus
becomes Amphoton deois (Walcott).

In Asia Amphoton occurs in the early Middle Cambrian deposits (Mapan
Formation) of North China, Manchuria and Chosen and reappears, somewhat later in
the Middle Cambrian, in the Taitzu Formation. The genus has a similar, discontinuous
range in Queensland. The agreement between early and late species is very close,
extending even to microscopic features. For instance Walcott states that A. alceste
Walcott has a punctate test. Whether or not such a relatively unusual feature as
punctation is general for these early forms is unknown ; for some of the known species
are not well described. But it is interesting to note that the late species here recorded
(A. serotinum) is also punctate. It would seem that species of the genus confidently
may be expected to be found in intermediate zones of the Middle Cambrian.

¥

AMPHOTON SPINIGERUM sp. nov.
(Pl. XXIV, figs. 13-20.)

Diagnosis: The cranidium is plain. The glabella is long and narrow (length
about three times the width), with parallel sides and broadly rounded front. The
occipital furrow is well marked. A pair of posterior glabellar furrows are present,
extending backwards almost to isolate basal lobes. On some specimens. other
(anterior) glabellar furrows are faintly indicated on the sides. The neck ring has a
median spine. The anterior limb is narrow, flat, and rounded in front and is about

THE GAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 237

one-sixth the length of the glabella. Fixed cheeks are narrow, the width at the middle
of the palpebral lobes being about two thirds the width of the glabella. The palpebral
lobes impinge upon the glabella near the anterior end, from which points the facial
sutures extend to the frontal margin keeping subparallel. The postero-lateral lobes —
are long.

The free cheeks are wide, with long genal spines diverging prominently.
Thoracic segments have the axis almost equal in width to the pleuron. Long median
spines are present on this axis. The pleura have prominent ventral curvature and
well marked furrows.

The pygidium is imperfectly known. Its axis has three or four segments, the
axial furrows being continued as pleural grooves. The margin is relatively narrow.

The hypostome is elongate, tapering posteriorly, with the extremities
prominently rounded. Its anterior lobe is large, occupying about two thirds the total
area. The posterior lobe is flat. Anterior wings are very prominent on the hypostome.
Maculae are absent. |

Remarks: This species, in the features of the cranidium and the thoracic
segments, is indistinguishable from the genotype, A. deows (Walcott). However the
free cheeks and the hypostome that are associated with the genotype are each distinct
from those of A. spinigerum. The free cheek of A. deois has a very blunt genal spine
(almost none at all) and the hypostome has not the characteristic anterior wings of
A. spinigerum. Two other species—A. dirce (Walcott) and A. hyrie (Walcott)—
have long genal spines; but in the general shape of the free cheeks neither agrees
quite so closely as does A. deois. The cranidial features distinguish still further these ,
two forms.

Endo and Resser (1937, pp. 206-207) have described two other species from the
Mapan Formation of Manchukuo—A. parallela and A. alia. Though similar in many
respects neither of these species has a prominent neck spine like A. spinigerum.

The specimens are not sufficiently well preserved to determine whether the
surface was finely punctate or not.

In the collections of the University of Queensland there are specimens of a
crushed form of Amphoton from Knowsley in Victoria. They are associated with a
species of Nepea and come from the horizon called by Thomas (1935, p. 92) the
“ Dolichometopus Band.’ The species possibly is identical with A. spinigerum.

Locality: From the Amphoton Stage at ‘Split Rock” on Waroona Creek,
at the crossing of the main road from Camooweal to Mount Isa (University of Queens-
land Collection).

238 MEMOIRS OF THE QUEENSLAND MUSEUM.

AMPHOTON SEROTINUM sp. nov.
(Pl. XXIV, figs. 21-23.)

Diagnosis: Cranidium with a long, narrow glabella that expands slightly in
the anterior quarter. On it there is a slight tendency to form a keel and there are
very faint traces of glabellar furrows. The occipital furrow is shallow and there is
a blunt occipital spine. The anterior rim is narrow, slightly curved and of uniform
width, and it impinges upon the glabella. The fixed cheeks are narrow with long
arcuate palpebral lobes that almost impinge upon the glabella. The posterior lobes
of the cranidium are long and extend almost to the genal angles. The free cheeks
are wide, rather tumid and have short genal spines.

The pygidium has a prominent raised axis, the posterior lobe of which is
markedly elevated. On the axis there are three prominent rings. The pleura are
relatively narrow with two pairs of short, rather vaguely defined ribs that end abruptly.
The margin is flat with a slight posterior concavity.

The surface is marked by widely spaced microscopic punctations.

Remarks: Amphoton deois is the most similar of the early species, but it
differs in the absence of genal spines and in the rather more prominent pygidial ribs.
A. alia Endo and Resser (1937, p. 207, pl. 38, figs. 14-18) is also closely to be compared.
It has, apparently, a slightly wider glabella and a more prominent occipital furrow.
A. alceste Walcott sp. (1913, p. 215, pl. 22, figs. 3a, b) which is similarly punctate
has prominent glabellar furrows. Since A. serotinwm occurs late in the sequence it
is interesting to compare it with the late forms in Manchukuo. Endo and Resser
(1937, p. 208, pl. 48, figs. 31-33) have recorded one species, A. divergens, from the
Taitzu formation ; but it has a wider glabella than A. serotinum. Their other late
species, A. (?) subhorrida Endo and Resser (1937, p. 208, pl. 48, fig. 34), almost

certainly is generically distinct. The short nuchal and gene spines of A. serotinum
distinguish it readily from A. spinigerum.

Locality and horizon: From the Papyriaspis Stage at a locality a mile and
a half west of the road crossing of V Creek, between Undilla and Thorntonia
(University of Queensland Collection). ;

Family ASAPHIDAE Burmeister, 1843.
Genus CHARCHAQIA Troedsson, 1937.
Genotype : Charchaqia norini Troedsson.

Charchaqia Troedsson is very similar to Hoasaphus Kobayashi 1936 (= Anorina
Whitehouse 193648). It is poneee debatable whether the two genera should not be

48 Tiostracus (7) superstes Pie ieoasan was independantly chosen by astavachs and by
myself as genotype respectively of Hoasaphus and Anorina. Kobayashi’s name Hoasaphus (March
1936) takes precedence of Anorina (April 1936).

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 239

merged under the older name Hoasaphus. Eoasaphus superstes has a faint occipital
furrow and light traces of other glabellar furrows, while the glabella has slightly
convergent sides. In the species of Charchaqia described by Troedsson the sides of
the glabella are parallel and the occipital ring merges into the glabella without any
trace of an occipital furrow. There are no other notable differences. In other features
of the head, and in the details of the thorax and pygidium, there is close agreement.
However the differences quoted may be sufficient to allow generic separation. It is
unfortunate for such considerations that Hoasaphus is known only from one specimen—
the holotype of #. suwperstes Linnarsson. 7

On this division the form now to be described would be placed in Charchaqia.

The resemblance to the asaphids, as Kobayashi and Troedsson have noted,
is very close and it is probably correct to regard the genera as early members of
Asaphidae. However I would call attention to the marked similarity to many
dolichometopid genera, for instance Anoria Walcott and Glossopleura Poulsen. The
chief difference is in the number of thoracic pleura—eight in Hoasaphus and Charchaqia
(as in Asaphidae) and seven in Dolichometopinae. Bearing these things in mind
it is permissible to suggest that Asaphidae may have arisen from Dolichometopinae.
The range in time would allow this. The chief difficulty in such a suggestion, it seems
to me, is the systematic position of the early Ogygopsis Walcott, a genus that frequently
is referred to Asaphidae.

Charchagia has been recorded hitherto only in the Upper Cambrian of T’ien-
Shan. Troedsson has not determined to which part of the Upper Cambrian it should
be referred. The related Hoasaphus has been found, in Sweden, in the early part of
the Upper Cambrian—either the top of the Olenus zone or else from the zone of
Orusia lenticularis.

CHARCHAQIA ERUGATA sp. nov.
(Pl. XXYV, figs. 6, Ta.)

Diagnosis : Dorsal shield small with parallel sides and with» rounded anterior
and posterior extremities.

The cranidium is relatively smooth. The glabella is long and inflated, with
subparallel sides and rounded front. It is but faintly marked by the occipital furrow.
No other furrows are present on the glabella. Although there is a marked difference
in convexity between the glabella and the fixed cheeks, the axial furrow is not
prominent. The fixed cheeks are narrow. The palpebral lobes are less than half the

240 MEMOIRS OF THE QUEENSLAND MUSEUM.

length of the glabella, are semicircular in outline and have small posterior impressions,
possibly muscle scars. The posterior portion of the facial suture sharply isolates the
posterior part of the palpebral lobes and proceeds obliquely to the

‘margin of the head.

There are eight thoracic segments. The axis is about two-thirds
the width of a pleuron. The axial rings are smooth and the pleura
are widely but shallowly furrowed. These pleura bend rather sharply

Restoration. downwards after about two-thirds of their length.

The pygidium is large, with about six rings discernible on the axis, these rings
decreasing in prominence towards the posterior. The axis reaches to the end of the
major portion of the pygidium. On the pleural regions only very faint traces of
furrows can be discerned. There is a relatively wide, concave margin to the pygidium.

The hypostome and the free cheeks are not yet known.

Remarks: Troedsson (1937) has described three species of Charchaqia (C.
norini T., C. lata T. and C. curvata T.). Westergaard (1922, pl. 2, fig. 20) has refigured
the holotype of Hoasaphus superstes. This is the only strictly comparable material.
C’. erugata is a smoother form than any of these. The glabella is unfurrowed (as in the
other species of Charchaqgia) but, because of the faintness of the axial furrow, it merges
into the fixed cheeks. The thorax similarly is smoother, and the pygidium has only
very faint furrows on the axis and none at all on the pleura.

‘Locality and horizon: One specimen (the holotype) in the University of
Queensland Collection from the Hugonocare Stage, about four miles north of the
Twenty Mile Bore on Glenormiston Station.

. CHARCHAQIA SPP.

Relatively smooth asaphid-like pygidia have been collected at two other
localities in Queensland. From the beds of the Hugonocare Stage, five miles north of
the Twenty Mile Bore on Glenormiston, some small, flat tails were obtained. From
other beds, tentatively equated with the same stage, at the mouth of Lily Creek on
Chatsworth Station, another species was obtained, larger and more transversely
elongated than th@ others. No pertinent cranidia have yet been found associated.
The specimens are in the University of Queensland Collection.

Family DORYPYGIDAE Kobayashi, 1935.

The genera Bonnia, Bonniella, Notasaphus, Kootenia, Olenoides, Dorypyge
and Holteria form a natural unit. Kobayashi (1935, p. 145) has grouped the last
four genera as Dorypyginae, a subfamily of Oryctocephalidae, and has suggested

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 24]

(p. 131) that Bonnia also might be included. Resser (1937, p. 4), in establishing the
genus Bonniella, also noted the resemblance of Bonnia to Kootenia and its allies.
The group is essentially a Middle Cambrian unit, ranging through the whole of that
division ; but Bonnia, Bonniella and Kootenia (sensu lato) are members that occur
also in the Lower Cambrian.” I suggest that the group be given full family status.

Genus NOTASAPHUS Gregory, 1903.
Genotype: Notasaphus fergusoni Gregory.

In 1896 Etheridge figured certain heads and tails of Cambrian trilobites from
Heathcote in Victoria as Dinesus ida. Gregory (1903, p. 155) pointed out that these
belonged to two different genera, and he proposed the name Notasaphus fergusoni
for the tails and also for certain other heads that he found in the same bed. Since
then the name Notasaphus has been used consistently in Australia for such forms.

Notasaphus ferguson is a very ill-defined species. The specimens from the
type locality consist of moulds and casts of isolated cranidia and pygidia, all of them
more or less crushed, while more than one species of the genus is present and has been
recorded under thisname. Also most of the published figures are very poor. Gregory’s
original figures are crudely drawn outline sketches. Three of his syntypes have been
identified in the collections of the Geological Survey of Victoria. The original of
plate 26, fig. 13 in Gregory’s paper (a tail, here refigured as Plate X XV, fig. 5) is
chosen as lectotype.

Two types of pygidia have been figured from the type locality as Notasaphus
fergusoni. One of these is a relatively long form, with widely spaced ribs. The
lectotype, which is of this form, has five spines on each side. Such tails are the
commonest in the collections that I have seen. A rather similar form was figured
by Etheridge (1895, pl. 1, figs. 5, 6) as Dinesus ida and later included by Gregory as
Notasaphus fergusoni. These specimens, however, probably have six spines on each
side. This may be a distinct species ; but a form figured by Chapman (1917, pl. 7,
fig. 29) with some rudimentary spines, may serve to link the two. For the present
all such types may be regarded as NV. fergusoni.

Distinct from these is a pygidium figured by Chapman (1917, pl. 7, fig. 28).
This is a shorter, broader form, with shorter spines. Chapman’s figure shows six
spines on each side ; but another specimen of this type in the collection of the Geological
Survey of Victoria from the same locality has at least seven on each side. The number
probably is variable in the species.

49 Chapman (1917, p. 91), in recording Notasaphus fergusoni described it as “ almost
essentially of an Upper Cambrian type ”—a statement that appears to be curiously incorrect.

242 MEMOIRS OF THE QUEENSLAND MUSEUM.

Correspondingly, apparently, there seem to be two types of cranidia. Those
figured by Gregory are relatively narrow forms with rectangular glabella. They are
the commonest type and no doubt belong to N. fergusoni as interpreted above from
pygidial data. A specimen of this type is shown in plate XXV, fig. 4.

Chapman (1917, pl. 7, fig. 27) figured a different type of head—a broader form
with expanding glabella. I have not seen this specimen or any other like it; but
possibly it is conspecific with the broader tail referred to above.

Interpreting NV. fergusont in this way it may be defined as a form with the
following features.

The cranidium is approximately equidimensional, with slightly rounded raised rim. The
glabella is rectangular, with rounded front, and is marked only by the deep occipital furrow. It
extends forwards right to the anterior rim of the head. There is a short, thick occipital spine.
The fixed cheeks are smooth and relatively broad with faint, converging, palpebral ridges. The
anterior portions of the facial suture appear to be slightly convergent. Palpebral lobes are long.
The posterior branches of the facial suture reach the posterior margin of the head at an acute
angle.

The pygidium is sub-equidimensional. There are four or five rings on the axis, the last
being usually longer than the others. Median spines are not present. The pleura have three stout,
rounded and widely spaced ribs, in addition to the raised anterior rim. Beyond these there is a
flattened rim that bears four or five recurved spines on each side. The margin, between the two
central spines, is sometimes straight, sometimes with rudimentary spines.

The task of framing a definition is made most difficult by the crushed nature
of the specimens. Because the specimens are so poor and because of the possibility
that several species are represented by such fragments, it is difficult to interpret
Notasaphus adequately. Until complete dorsal shields are found as topotypes it may
be advisable to use the closely related and complete Queensland species, N. modicus,
for comparative purposes.

In 1889 Walcott®® erected the genus Kootenia basing it on a specimen figured
by Rominger (1888, pl. 1, fig. 8) which he named Bathyuriscus (Kootenia) dawsoni.
I am not aware that the holotype has since been figured. The name Kootenia has
been widely used of late notably by Kobayashi (1935) and by Resser (1937). In the
sense of such recent usage Notasaphus fergusoni falls well within the limits of Kootenia ;
and recently (Whitehouse 1936, p. 74) I recorded Notasaphus as a synonym.

°° The date of publication of this name usually has been misquoted. It was erected on
page 446 of Walcott’s paper “* Description of New Genera and Species of Fossils from the Middle
Cambrian,” published in volume 11 of the Proceedings of the United States National Museum.
Although this was the volume for 1888 Walcott (1913, p. 13) gives the date of publication as
September 3rd, 1889. Vogdes (1925, p. 103) quoted the date as 1888. Kobayashi, on two pages
of the same paper (1935, pp. 145 and 146) gives it as 1888 and 1899. Resser (1937, p. 15) stated
it to be 1908.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 243

Had not the name Notasaphus been proposed I would have been content to
leave the species now to be described in Kootenia. However, opinions on the morpho-
logical limits of genera vary, and it is necessary to enquire whether, on finer delimita-
tions, Notasaphus may yet be retained as a valid genus. Certainly among the species
at present referred to Kootenia, a suite that occurs in both Lower and Middle
Cambrian beds, there is marked variation. Glabellar furrows are present on some,
absent from others. There is a considerable range in palpebral structures. Axial spines
may occur on or be absent from the pygidium. The number of thoracic segments varies,
although in most species the thorax has not been described. On the pygidium there
is a variation in the number of axial rings, pleural furrows and marginal spines.
Such variation conceivably may lead to generic subdivision. Two major groupings
may be noted—a series in which axial spines, present on the thorax, are continued
on the pygidium and another group in which such pygidial spines do not occur.
Within each such group there is considerable variation. For instance, in both sections
there are forms with smooth glahella and others that have glabellar furrows. However
to make a complete generic analysis of the suite it would be necessary to have access
to at least the American specimens.

The genotype of Kootenia has axial spines on the pygidium. The genotype
of Notasaphus has no such spines.*! The eyes of Notasaphus are larger. Furthermore,
seven segments are present in the thorax of the genotype of Kootenia whereas in
Notasaphus modicus (the only member of the Notasaphus group in which a complete
thorax is known) there are eight segments. It is permissible, therefore, to retain
Notasaphus as a distinct genus. :

To what extent it may be independent of the American forms without axial
spines on the pygidium—forms with smaller eyes, stronger palpebral ridges and,
sometimes, faint glabellar furrows—it would be injudicious to say at present. Perhaps
by examining the material, more than one generic unit may be recognised in this
assemblage.

NOTASAPHUS MODICUS sp. nov.
(Pl. XXV, figs. 1-3.)
Diagnosis: Dorsal shield regularly oval.
The cephalon is transversely elongate, the width being twice the length. The
anterior and lateral margins form a uniform arc. The glabella is wide, with parallel
sides and semi-circular front and extends to the anterior rim. On it the only marking

is the broad occipital furrow. The occipital ring bears a median spine. The fixed

51 Care must be taken in determining whether such spines are present or not. An internal
mould may not show traces of spines, if these were solid structures.

244 ; MEMOIRS OF THE QUEENSLAND MUSEUM.

cheeks are narrow. The facial sutures converge slightly towards the front, are only

slightly arcuate in the palpebral lobes, and posteriorly extend outwards and backwards

in semicircular curves. The palpebral lobes are continued to the glabella by converging

processes that are hardly ridges but really changes in the declivity of the fixed cheeks.

The area in front of these processes is depressed ; behind them the cheeks have the.
normal inflation. The free cheeks are slightly wider than the fixed cheeks and bear’
prominent genal spines that diverge posteriorly.

The thorax has eight segments. The axis, which is about one third the total
width, is simply arched and bears a median spine towards the posterior edge of each
segment. The pleura are widely furrowed ; beyond the fulcrum they contract in
acute angles and terminate in recurved spines.

The pygidium is slightly shorter than the head. It is semicircular in outline.
The axis tapers posteriorly but ends bluntly. Five segments are defined on it by
rather broad grooves, the final segment being longer than the others. Four pleural
furrows are present on each side, the ribs between them being
uniformly convex and not otherwise furrowed. Beyond these
ribs there is a narrow, flat margin fringed on each side with five
slightly recurved spines with interspaces about twice their
own width.

The hypostome consists of an ovate lobe surrounded by
a very narrow rim. This rim is concave anteriorly, with small
anterior wings. Laterally and posteriorly it is a raised structure.
The central, ovate lobe slopes gradually to the posterior rim
but descends rather steeply on the sides. Maculae are present on
the posterior portion of the lobe. They have well defined margins in the postero-
lateral regions ; but anteriorly they merge into the lobe.

Restoration.

Remarks : Two complete dorsal shields (one of an immature individual) and
several isolated cranidia, pygidia and hypostomes have been found.

Generally it is very similar to N. fergusoni, but the fixed. cheeks are rather
narrower, the occipital spine more slender and the number of marginal spines on the
pygidium is constant. There are five such spines on each side, whereas in NV. fergusons
the number seems to vary from four to six. Comparison with overseas species is
difficult, again by reason of the lack of complete specimens.

Locality and horizon: From the Dinesus Stage of Beetle Creek, 12 miles north-
west of Mount Isa (University of Queensland Collection).

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 245

NOTASAPHUS (?) SP.

Two fragmentary cranidia, belonging to some genus of Dorypygidae have
been obtained from the Hurostina Stage at No. 4 bore, Alroy Downs.

4

Family CERATOPYGIDAE Raymond, 1913.

Three genera from the Cambrian of Scandinavia, Proceratopyge Wallerius,
Hysterolenus Moberg and Ceratopyge Corda,” are so similar that no one has yet
questioned their relationship and they have been grouped in the one family. Procera-
topyge is found in the uppermost zone of the Middle Cambrian and in the three
succeeding zones—that is, in the early part of the Upper Cambrian. Hysterolenus
is a Lower Tremadocian and Ceratopyge an Upper Tremadocian genus. All three
are known in Scandinavia from dismembered fragments only. They agree well in
details of the cranidium and of the pygidium.

The cranidia of these forms have narrow fixed cheeks, small but prominent
eye lobes and long posterior limbs. The glabella is long, often with grouped anterior
furrows and typically with the posterior pair of furrows isolated and extending more
or less parallel with the axis. In the earliest genus (Proceratopyge) the glabellar
furrows are much less sharply incised than in the later genera.

The pygidia have a pair of prominent spines that arise from fused anterior
pleura, these particular pleural portions being markedly distinct from the rest of
the tail.

In 1912 Walcott described a species from British Columbia as Ceratopyge
canadensis. Later (1916, p. 374, pl. 65, fig. 1) he proposed the generic name Housia
for another cordilleran species, Dolichometopus (Housia) varro Walcott. In 1925
he transferred Ceratopyge canadensis to Housia. The resemblance of Housia to
Proceratopyge is most marked. The American forms have the same type of pygidial
spines, while the cranidium, in H. canadensis at least, has the peculiar glabellar
furrows of the family. Compared with the genotype of Proceratopyge the American
species have longer and more concave anterior limbs and narrower fixed cheeks.
Unfortunately the genotype is the only Scandinavian species whose cranidium is known,
the other species being based on pygidia only. However the cranidia of the species
that now are described from Australia show a gradation from forms like the genotype
to others that are closely to be compared with the American species. On P. lata for
instance, like the genotype, the fixed cheeks are relatively wide and the anterior limb
is comparatively short, with shallow, diverging furrows. P. rutellum, on the other

52 For good figures of the three genotypes see Westergaard 1922, pl. 2 (Proceratopyge), Moberg
1898, pl. 17 (Hysterolenus) and Moberg and Segerberg 1906, pls. 4 and 5 (Hysterolenus and

Ceratopyge).

246 MEMOIRS OF THE QUEENSLAND MUSEUM.

hand, has a longer and a smooth anterior limb as in the American species and, like
them, the fixed cheeks are so reduced that the eye lobes practically impinge upon the
glabella. P. nectans and P. polita are intermediate in their characters. The gradation
is such that a generic separation is not warranted on cranidial features alone.

I have not handled specimens of the Scandinavian and American forms ;
but by comparing figures of the pygidia it would seem that on the American species.
the spinose pleural portions are more markedly distinct from the rest of the tail
than they are on the Scandinavian species while, in addition, the posterior border is
possibly wider. This latter feature I would not stress; for the Scandinavian
P. nathorsti Westergaard has a rim as wide as the genotype of Housia. The degree of
distinction between the spinose and non-spinose pleural regions may be more apparent
than real. Most of the published photographs from America have been retouched
and. these pleural boundaries outlined. Those that seem to be the least retouched
(Walcott 1925, pl. 22, figs. 10 and 11) are least suggestive of this sharp distinction.
Possibly a direct comparison of the Scandinavian, American and Australian material
may show some grounds for separation ; but from the evidence that I have at present
I find no reason to recognise two genera. The American species, occurring in the
Mons and Goodsir Formations, are apparently somewhat later in age than the
Scandinavian forms but not, I think, sufficiently later to warrant generic separation

purely for such a reason.

In Proceratopyge (including Housia) the spinose pleura of the pygidium are the
anterior members of the tail. In the later genera, Hysterolenus and Ceratopyge, a
non-spinose segment is fused to the pygidium in advance of these. A curious form
described by Moberg®® as Hysterolenus (?) levicauda has three non-spinose segments
in front of the spinose pleura of the pygidium. Unfortunately the cranidium of this

form has not been found.

The fusion of a spinose segment on the anterior portion of the pygidium has
happened in many families of the trilobites. Redlichia has it. So too have some
species of Paradowides. Of itself this feature is not of genetic importance. But when
to this is added other curious features (in this group the peculiar glabellar furrows
on very similar cranidia) we reasonably may combine the genera into a family.
Nevertheless Kobayashi (1935, pp. 269-273) recently has placed Housia in a very
different family, the Marjumiidae.* He regards it as having a cranidium and a.

53 Moberg and Segerberg, 1906, p. 84, pl. 4, fig. 40.

*47 cannot regard Housia and Marjumia as related genera. ‘All portions of the test,
cephalon, thorax and pygidium, are different in the two groups. The pygidum of Marjumia callas
Walcott (1916, p. 404, pl. 65, fig. 3) is similar to that of Housia ; but in this way M. callas differs.
markedly from the genotype of Marjumia (M. typa, Walcott sp.).

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 247

pygidium very different from typical Ceratopygidae. I fail to see this difference in
cranidia. As I have already indicated, the Australian forms provide a link between
the two. Also Kobayashi’s remarks about the pygidia are, I venture to suggest, not
in accord with the facts. According to him the spines in Housia arise from the anterior
pygidial segment but in Ceratopygidae (and he includes Proceratopyge) they are on the
second segment. Certainly in Hysterolenus and Ceratopyge they are on the second
segment ; but in Proceratopyge they occupy the same anterior position as they do
on Housia. . 7 |

Kobayashi (1935, p. 293) added a Middle Cambrian genus, Kogeniwm, to the
family. This has a tail similar to Proceratopyge ; but the cranidium has wider, fixed
cheeks and a rather different anterior rim. Nevertheless there is, on its glabella, a
sharp recurvature of the furrows that may be an incipient development of the curious
furrows of Ceratopyge ; and for this reason the genus may be retained as an early
member of the family.

Recently Troedsson (1937, pp. 35 and 39) allotted two new Asiatic genera,
Lopnorites and Diceratopyge to the family. In the latter genus the glabella is smooth ;
but Lopnorites has isolated posterior glabellar lobes comparable with those of other
members of the family. In his review Troedsson proposed making two subfamilies—
Hysteroleninae (to include Kogeniwm (?), Lopnorites and Hysterolenus) and
Ceratopyginae (for Proceratopyge, Diceratopyge, and Ceratopyge). The distinction
was based on eye lines present in the first group but absent from the second, and
the greater segmentation of the pygidium in Hysteroleninae.

The distinction seems to me to be poorly founded. As Troedsson pointed
out, in some species of Proceratopyge the pygidium is quite well segmented. Further-
more, Ceratopyge latelimbata Moberg and Segerberg (1906, p. 87, pl. 5, fig. 6), occurring
in the same bed as the genotype and a typical member of Ceratopyge, has quite definite
eye lines. I prefer, therefore, not to use the division.

Of these six genera three only (Proceratopyge, Lopnorites and Diceratopyge)
are known from complete dorsal shields. Kogeniwm is the first genus to appear
(relatively early in the Middle Cambrian) followed by Proceratopyge. Lopnorites and
Diceratopyge, possibly contemporaneous genera, are the next to arrive, followed
by Hysterolenus and then, closing the sequence, Ceratopyge.

From Kogenium to Proceratopyge the main change is in the reduction of the
free cheeks. This change continues during the range of Proceratopyge. The first
species to appear (P. conifrons) has the widest cheeks ; but in P. rutellum, P. varro
and P. canadensis the reduction of the fixed cheeks reaches the acme for the family.

248 MEMOIRS OF THE QUEENSLAND MUSEUM.

The later genera show other changes. In Lopnorites the number of thoracic
segments is reduced to nine (there are ten in Proceratopyge, as shown on P. canadensis)
and eye lines appear. The glabella retains the characteristic furrows. The pygidium
is in some species but moderately segmented, in others highly furrowed. Diceratopyge
differs in quite other ways. It adds to its pygidium, which is most poorly furrowed,
a wide zone (possibly composed of more than one fused segment) in front of the
spinose member, while the number of thoracic segments is reduced to six. Also,
following the mode of certain species of Proceratopyge, the glabella becomes smooth.
Hysterolenus retains the early form of pygidium, with spines on the anterior part,
and has no eye lines. In Ceratopyge eye lines are sometimes present, sometimes
absent, and once more the pygidium has a non-spinose anterior member. In this
genus, in contrast to all the others, the glabella widens anteriorly.

Thus instead of seeing two well-defined sections in the family, as Troedsson
did, I interpret the evidence as showing a progressively developing group in which
the several changes (furrowing of the pygidium, the addition of a non-spinose anterior
zone on the pygidium, the smoothing of the glabella and the development of eye
lines, for instance) have operated on more than one occasion, prosneme a plexus
of forms from which it is difficult to make sub-groups.

It is interesting to note that each of the species of which complete specimens
are known has rather bluntly terminated thoracic pleura. This may explain why,
when anterior additions are made to the pygidium, they are non-spinose. Also,
correlated with such additions to the pygidium, may be the reduction in the number
of thoracic segments—markedly in Diceratopyge in which the pygidium has a very
wide anterior non-spinose zone and there are but six thoracic segments.

Genus PROCERATOPYGE Wallerius, 1895.
(= Housia Walcott, 1916)
Genotype : Proceratopyge conifrons Wallerius.
PROCERATOPYGE LATA sp. nov.
(Pl. XXV, figs. 12 and 13 (?).)

Diagnosis; Cranidium narrow. The glabella narrows anteriorly with straight
sides and rounded front. It is distinct from the free cheeks by a sharp change in the
inflation of the test rather than by the presence of limiting furrows. On the glabella,
anterior to the shallow occipital furrow, there are three extremely faint pairs of
short furrows. The two anterior pairs extend approximately at right angles to the
axis, whereas the posterior pair are aligned more or less parallel to it. Actually they

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 249

are little more than shallow depressions. These furrows are difficult to see except
in oblique light. The fixed cheeks are narrow, their width at the eye lobes being a
little more than half the width ‘of the glabella. The palpebral lobes are very small,
centrally situated and they project prominently from the edges of the free-cheeks.
They are connected to the anterior lobe of the glabella by very faint, converging
palpebral ridges. The anterior portions of the facial suture are for the most part
parallel, but they converge slightly towards the anterior margin. The posterior
portions of the facial suture extend sub-parallel to the posterior margin of the cephalon,
thus developing long posterior lobes to the fixed cheeks. The anterior limb-is about
one-fifth the length of the cranidium and is concave. Diverging from the antero-
lateral region of the glabella are two areas slightly more concave than the rest of
the anterior limb. |

Remarks: At.the only locality from which I have obtained this species it is
represented by numerous cranidia. In the same rock specimens the cranidia of
P. rutellum are equally abundant. However only one type of pygidium has been
found. These tails, too, are numerous. Cranidia of P. rutellum are found at another
locality from which I have not obtained P. lata, and there they occur with the same
types of tail. For that reason I have associated this pygidium with P. rutellum ;
but I have been tempted to think that, because of the abundant specimens, P. rutellam
and P. lata may have had pygidia of similar type. It may be mentioned that of the
only other Australian form whose pygidium is known, P. nectans, the tail differs only
slightly from this type.

In its cranial features P. lata is very similar to the genotype P. conifrons
Wallerius: The glabella is conical. There are similar glabellar furrows, an anterior
limb of comparable length with similar divergent grooves, and the fixed cheeks are of
much the same width and general type. Our form may be distinguished by the
straighter lines of the anterior facial suture, by the eye being situated rather more
to the rear and by the flatter form of the anterior limb.

Locality and horizon: From the Eugonocare Stage, five miles north of the
Twenty Mile Bore, Glenormiston (University of Queensland Collection).

PROCERATOPYGE NECTANS sp. nov.
(Pl. XXYV, figs. 8a, b.)

Diagnosis : Cranidium axially elongate. The glabella narrows anteriorly with
straight sides and rounded front. On it the furrows are most faintly impressed.
The posterior pair are long and slope backwards towards the axis. The fixed cheeks
are relatively narrow. The palpebral lobes are situated opposite the centre of the

R

250 MEMOIRS OF THE QUEENSLAND MUSEUM.

glabella and are continued by faint, converging, palpebral ridges. The facial sutures
diverge slightly before the eye. The anterior limb is of moderate length and is flat
to slightly concave. Faint diverging furrows extend on it from the anterior end of the
glabella.

The pygidium is transversely elongate. The axis has four segments, the
posterior being considerably larger than the others. From the anterior segment the
spinose pleural portions radiate, increasing laterally and ending in long spines roughly
parallel with the axis. The remainder of the pleural region has two extremely faint
and narrow furrows on each side. The posterior margin is smooth, wide and slightly

concave.

Remarks: P. nectans is intermediate in its cranidial characters between
P. lata and P. rutellum. The anterior limb and the fixed cheeks are intermediate in
size. As on P. lata and P. conifrons there are faint divergent furrows on the anterior
limb. Such furrows are not seen on P. rutellum or on the American species. The
eye lobes come close to the glabella, but they do not impinge upon it as they do in
.P. rutellum.

Locality and horizon: From the Glyptagnostus Stage 16 miles south of Glenor-
miston homestead (University of Queensland Collection).

PROCERATOPYGE RUTELLUM sp. nov.
(Pl. XXV, figs. 9-11 and 13 (?).)

Diagnosis: Cranidium axially elongate. The glabella narrows very slightly
anteriorly with straight sides and rounded front. The glabellar furrows are particularly
faint. Two pairs of discontinuous anterior furrows can be seen and a posterior pair
that slope backwards towards the axis. The change from the glabella to the fixed
cheeks is not well marked. The fixed cheeks are very much reduced. The facial
sutures diverge slightly towards the front and impinge upon the glabella at the front
of the palpebral lobe. This lobe, situated opposite the centre of the glabella, has the
appearance of projecting from it. The posterior lobes of the cranidium are long. |
The anterior limb is wide and concave and is succeeded by a convex, subtriangular
anterior rim that gives an obtusely angular front to the head.

The pygidium is transversely elongate. There are five segments on the axis
separated by shallow furrows. The anterior pleural portions extend outwards from
the first axial segment, increasing laterally to a triangular shape. These portions
are not furrowed and they end in strong spines that are almost parallel to the axis.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 251

The remainder of the pleura have faint furrows extending outwards from the axial
furrows. A concave rim borders the pygidium behind the spinose segments. This
is of moderate width and is truncated posteriorly.

Remarks : In its cranidium P. rutellum is very like the American species P.
varro and P. canadensis. Like them the anterior limb is long and concave, without
the faint diverging furrows that are present on the genotype. Also, like those two
species, the fixed cheeks are reduced to a minimum, the palpebral lobes impinging
upon the glabella. The pygidium of P. rutellum is quite like that of P. nathorsti
Westergaard (1922, p. 120, pl. 2, figs. 3-5) ; but its axis terminates more bluntly and
it has a truncated posterior margin that is not present on P. nathorstz.

‘Localities and horizon: From the Hugonocare Stage at distances of
approximately 44 miles and 5 miles north of the Twenty Mile Bore, Glenormiston
(University of Queensland Collection).

PROCERATOPYGE POLITA sp. nov.
(Fl. XXV, fig. 14.)

Diagnosis: Cranidium axially elongate. The glabella is long, with sub-
parallel sides and semicircular front. The occipital furrow is well marked but there
are no other grooves on the glabella. The occipital ring is plain and of constant
width. The facial sutures diverge slightly in their anterior portions. At the front
of the eye lobes they closely approach but do not impinge upon the glabella. The
palpebral lobes are situated in the posterior part of the fixed cheeks. The anterior
limb is wide and concave and there is no raised rim before it.

The other portions of the test are not yet known.

Remarks : This species is distinguished by its unfurrowed glabella and by the
eyes being set a little more to the rear than in most other species. P. varro (Walcott)
may be another species with a smooth though narrowing glabella. Walcott makes
no comment on this matter. But his figures show no glabellar furrows. P. polita
* resembles in many ways P. rutellum in which, however, the sides of the glabella
converge slightly, the anterior limb is longer and the fixed cheeks are somewhat
narrower.

Locality and horizon: From the Rhodonaspis Stage at the western base of
“* Polly’s Lookout,” Glenormiston. (University of Queensland Collection.)

MEMOIRS OF THE QUEENSLAND MUSEUM.

a)
OL
bo

Family ORYCTOCEPHALIDAE Beecher, 1897 (emend. Raymond, 1913).°°
Subfamily ORycTOCEPHALINAE Kobayashi, 1935.°°
Genus ORYCTOCEPHALUS Walcott, 1886.
Genotype : Oryctocephalus primus Walcott.

ORYCTOCEPHALUS DISCUS sp. nov.
(Pl. XXV, fig. 15.)

Diagnosis : Cranidium roughly semi-circular in outline. The glabella is wide
with a sub-ovate outline, and extends to the anterior rim. It is marked by the
occipital furrow and four pairs of narrow, discontinuous furrows that reach neither
the centre nor the sides of the glabella. These furrows decrease in intensity towards
the front, the anterior pair being very far forward, very faint and are most nearly
at the margins. The fixed cheeks are relatively narrow. The palpebral lobes have
convex margins, are regularly arcuate, and are continued in the same arcs by palpebral
ridges of similar width. These palpebral ridges meet the glabella near the anterior
end, close to the anterior glabellar furrows. The palpebral lobes extend backwards
for a considerable distance so that the posterior sections of the facial sutures are
very short. The anterior sections of the facial sutures are also short and converge
slightly. The pre-palpebral regions of the fixed cheeks are thus small.

Remarks : I have seen only one specimen, a cranidium.*®

The width and the shape of the glabella, and also the form of the pre-palpebral
portions of the fixed cheeks are features whereby the species differs from other known

members of the genus.

,

Oryctocephalus has been described from Middle Cambrian beds in British
Columbia, Nevada, Idaho, Chosen and India (Spiti). The genotype, O. primus
Walcott (1886, p. 210, pl. 29, fig. 3) from Nevada, and O. reynoldsi Reed (1899, p. 359)
from British Columbia are perhaps the most similar forms. In the course of the
facial sutures and the palpebral structures they are closely to be compared, but the

55 Actually Beecher’s original name was Oryctocephalinae, for he interpreted the group (in a
wide sense) as a subfamily of Olenidae. Raymond (1913) defined the family in its present generally
accepted limits and, using the termination “ -idae,”’ credited the name to Beecher. Kobayashi
(1935) subdivided the family into Oryctocephalinae and Dorypyginae, a division that I have
followed. In so doing Kobayashi credited the family name (Oryctocephalidae) to Raymond who
first used it and the subfamily name (Orcytocephalinae) to Beecher. Etymologically this is of
course correct. But the usage that I have adopted here expresses more adequately the growth of
ideas about the family. ’

56 Dr. W. E. Schevill informs me that he collected Oryctocephalus from the beds of the
Templeton Series and that these specimens. (which I have not seen) are in the collections of Harvard
University. It is probably to these specimens that Kobayashi (1934, p. 296) has referred when
recording Oryctocephalus from the Templeton River.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 253

outlines of the glabellae are different. In O. primus the sides are parallel. In O.
reynoldsi the glabella slowly expands—more slowly than does O. discus. The Indian
O. saltert Reed (1910, p. 11, pl. 1, figs. 16-21) and the Korean O. orientalis Saito (1934,
p. 230, pl. 25, figs. 17-22) and O. Kobayashit Saito (1934, p. 231, pl. 25, figs. 23-25)
have deeper pits on the cranidium. Also the glabella in the two latter species does
not expand. In O. salteri the glabella is most comparable in outline to that of O.
discus.

Locality and horizon: Holotype from the Dinesus Stage of the Templeton
River (University of Queensland collection). I understand that specimens of a
species of Oryctocephalus from these beds are in the Harvard University Collections
(see footnote 56, p. 252).

4. SUPPLEMENT No. 1.

(A) EXPLANATION.

While this series of papers is being issued field work in the area is progressing
intermittently ; and so I have thought it advisable to attach to each new part that is
published a supplement—to record additions to groups that previously have been
illustrated and new facts about the stratigraphy of the Cambrian in this region.
Part I of this series was a stratigraphical outline. The correlations that I made then
with areas elsewhere in Australia and provinces overseas were stated very briefly.
No more was possible, for most of the evidence had yet to be presented. As in each
succeeding part more of this evidence appears reasons can be given more fully for the
correlations that have been made ; and in the light of new discoveries such correlations
can be criticised and reviewed. These things I propose to do in the supplements.

(B) Some New ReEcorps or AGNOSTIDS.

Some new agnostid trilobites have been found since Part 2 of this series of
papers was printed.

Family AGNOSTIDAE McCoy, 1849.

I.have already discussed the limits of this family (1936, p. 84). Within it
two groups may be recognised—the genera Agnostus, Huagnostus, Solenagnostus,
Fallagnostus and Ceratagnostus have simple glabellae; but on Goniagnostus,
Triplagnostus and. frequently on Homagnostus the posterior glabellar lobe has a pair
of notches. This feature is retained in the family Pseudagnostidae which possibly
arose from Homagnostus. However I do not think it wise to make a subfamily
_ division in Agnostidae on this feature; for on some species of Agnostus, for instance

254 — . MEMOIRS OF THE QUEENSLAND MUSEUM.

A. hedini Troedsson (1937, p. 20, pl. 1, figs. 6-8) and A. intermedius Tullberg
(1880, p. 17, pl. 1, figs. 4a, b), faint depressions occur in positions corresponding to
the glabellar notches of Goniagnostus, while notches are sometimes present, some-
times absent on Homagnostus. :

Genus AGNOSTUS Brongniart, 1822.
= Batius Dalman, 1826.)

Genotype : Entomolithus paradoxus pisiformis Linnaeus.

In describing Agnostus seminula I have pointed out that it belongs to a group
of many species, differing from the genotype and its immediate relatives, in having
a narrower axis and no pygidial spines. Also it occurs earlier than the type section.
If later researches show that this stratigraphical distinction is constant it may be
advisable, eventually, to separate this group subgenerically.

AGNOSTUS SEMINULA sp. nov.
(Pl. XXV, fig. 24.)
Diagnosis: Relatively a narrow form, only slightly inflated. The facets at the
base of the cephalon and the front of the pygidium are well developed.

The cephalon narrows slightly towards the front and has a very narrow uniform
vim. The axial furrows are well incised while the other cephalic furrows are less
prominent. The glabella is very narrow, extending forwards for about three quarters
the length of the head and occupying about. one quarter of the total width. The
anterior lobe is one third the length of the glabella. The posterior lobe has parallel
sides. Accessory lobes are narrow.

The two thoracic segments are of the usual types with tuberculate axis and
short inflated pleura.

The pygidium has a non-spinose rim that is wider than the cephalic rim and
widens slightly but progressively to the posterior. The axis is wider than that of the
cephalon. The margins of the anterior segment converge anteriorly. Posterior to
this the axis widens very slightly towards the centre and then tapers to a point close
to the rim. The anterior segment is very narrow, limited posteriorly by a prominent
furrow. The median segment has a tubercle that increases in prominence posteriorly.
Between the median and posterior segments the furrow is moderately incised,
extending posteriorly in the centre to bound the median tubercle. Pleura are con-
tinuous in the narrow space behind the axis.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 255

Remarks : A.seminula is easily distinguished from the genotype, A. pisiformis
Linnaeus sp. (see Tullberg, 1880, pl. 2, figs. 14a, b), and from the Canadian form
described by Matthew as A. pisiformis var. These are plump forms with wider axes,
wider accessory lobes and, also, with pygidial spines. The Asiatic A. hedini Troedsson
(1937, p. 20, pl. 1, figs. 6-8) and the arctic A. pater Holm and Westergaard (1930,
p- 9, pl. 1, fig. 1, pl. 4, figs. 9, 10) belong to the group of A. pisiformis and also are
markedly distinct from A. seminula. There is, however, a series of flatter forms with
narrower axes and no pygidial spines with which it should be compared. A. intermedius
Tullberg (1880, p. 17, pl. 1, figs. 4a, b) is of this type. It has a more acuminate glabella
and has pleural furrows on the head. A specimen figured by Illing (1915, p. 408, pl.
28, figs. 11, 12) from the Stockingford Shales as A. cf. intermedius is more akin to this
species and may even be identical. However some of the pertinent detail is not shown
on Illing’s specimen. The Welsh A. typicalis Nicholas (1915, p. 454, pl. 39, fig. 2) is
very similar, particularly in the glabellar features. However on the pygidium the axis
is rather shorter and the second transverse axial furrow is not so much deflected in the
centre. A. acutilobus Matthew (1886, p. 73, pl. ig fig. 10) from Canada also is most
closely allied. It is a narrower form and, as in A. typicalis, the second axial furrow
on the pygidium is straighter. These four species that closely may be compared
with A. seminula all occur in the zone of Paradoxides hicksi. The plumper forms
with wider axes and pygidial spines seem everywhere to be slightly later in age.

Locality and horizon: From limestones in the Agnostus seminula Stage, eleven
and a half miles north east-of Yelvertoft Dip, on the road to the Paradice Goldfield
(Universit y of Queensland Collection).

Genus CERATAGNOSTUS gen. nov.*’
Genotype : Ceratagnostus magister sp. nov.

Diagnosis: Forms with the cephalon of the same type as Agnostus. The
pygidium has a bispinose rim, a post-axial furrow and an axis that is prominently
constricted, laterally and vertically, at its posterior end. The posterior transverse
furrow has been effaced from the pygidial axis, and the anterior furrow is either very
faint or absent.

Remarks: There is a group of Middle Cambrian species, in the zone of
Paradoxides davidis,. with the above features. C. magister sp. nov. from Australia
and C. incertus (Brégger) from Europe are of this type, while Howell records Agnostus
cf. incertus from Canada. Since it is an easily recognised group, widely distributed
and of limited stratigraphical range, even though it is separated by only small
differences from some other agnostids, I have thought it advisable to give it a generic

5? In reference to the horn-like form (képas a horn) of the pygidial axis.

256 MEMOIRS OF THE QUEENSLAND MUSEUM.

name. Defined in this way Ceratagnostus is similar to Solenagnostus on the one hand.
(differing in the constricted termination and the rudimentary furrows of the
pygidial axis) to Huagnostus (which has a more regular pygidial axis) and to Goniag-
nostus (from which it differs in the absence of glabellar notches and pleural furrows).
As noted. below (p. 257) the genus possibly arises from the group of Huagnostus (?)
exaratus Gronwall.

CERATAGNOSTUS MAGISTER sp. - nov.
(Pl. XXV, fig. 27.)
Diagnosis: Relatively a very large form, the average length of specimen
being 13mm. Facets on both cephalon and pygidium are prominently shown.

The cephalon is moderately inflated, subcircular in outline and has a uniform,
slightly elevated rim. Circumaxial and glabellar furrows are deeply incised, while
the pre-glabellar furrow is less sharply marked. The glabella is subovate. Its length
is about three quarters of the total length of the cephalon while its width is about
one quarter. The anterior lobe is subtriangular with convex sides. The posterior
lobe tapers very slightly towards the front and has a strongly rounded base. Accessory’
lobes are very well marked, triangular in outline and are confluent behind the posterior
lobe on the deflected posterior slope of the head. There is a small, very faintly
developed tubercle in the centre of the posterior lobe.

The two thoracic segments have strongly tuberculate axes and normal,
short, bulbous pleura, these “‘ bulbs ” being adjacent to the head and the tail.

The pygidium is subcircular with a uniform, inflated rim that has a pair of
prominent postero-lateral spines. As on the head, the rim extends to the axis, thus
accentuating the facets. Posteriorly the axis decreases to the first
transverse furrow after which it remains fairly constant in width to
about the centre of the tail, beyond which it rapidly tapers to a
point. The tapering portion is considerably less inflated than the
rest’ of this axis. On the axes the first transverse furrow is faintly
_ ineised, but the second furrow has practically disappeared, only a
very slight. depression being left. The median tubercle is not
i) prominently defined. |

Remarks: C. magister is very similar to C. incertus Broégger

Restoration. sp- (as figured by Tullberg, 1880, pl. 1, figs. 6a, b), differing in the
NAc larger sizes (it is about 50°, larger in linear dimensions) and by the
slightly more obtuse termination to the axis of the pygidium. As Agnostus
incertus Lake (1907, p. 29, pl. IIT, figs. 1-3) has figured some Shropshire forms that
may represent another species of the genus. Lake has called attention to certain

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 257

.differences in the cephalon; while his figures seem to indicate a pygidial axis of
slightly different shape and that the marginal spines on the pygidium are closer
together. The tail figured by Illing (1916, p. 407, pl. XXVIII, fig. 10) as Agnostus
cf. incertus is badly preserved ; but in size it is comparable with C. magister. It is
of interest to note that Illing (1916, p. 406) records Agnostus exaratus var. tenuis

| Illing through a considerable range of beds. The last members of this lineage, in

the zone of Paradoxides davidis, develop a constricted end to the pygidial axis and
he regards them as a definite mutation. These late members may be placed in

Ceratagnostus and suggest that the group of Huagnostus (?) exaratus Grénwall is

the ancestor of the genus.

Locality and horizon: From the Papyriaspis Stage at the road crossing of
V Creek, between Undilla and Thorntonia. Also from the Phoidagnostus Stage a
mile and a-half west of that locality (University of Queensland Collection).

Genus GONIAGNOSTUS Howell, :1935.

Genotype : Agnostus nathorsti Brégger.

It is not clear, at the moment, if Goniagnostus Howell and Triplagnostus
Howell should be regarded as distinct genera. The names were proposed in the
same paper; and although Howell gave long diagnoses, these were identical in
phrasing until the closing sentences when, for distinctions, it was noted that
Goniagnostus had pleural furrows on the head and pygidial spines, while 7'riplagnostus
had smooth cheeks and no spines. But there are forms in this group, such as Agnostus
atavus Tullberg (1880, p. 14, pl. 1, fg. 1), that have deep pleural furrows but no
spines, and others, for example G. purus sp. nov., with pygidial spines but no pleural
furrows.

If the genera are to be kept distinct then apparently the only criterion for
separation is the presence of pygidial spines on Gonizagnostus but not on T'riplagnostus.
Occasionally in other groups of agnostids such a distinction, small though it is, is
valid. In Pseudagnostidae, for instance, pygidial spines‘are the only morphological
features discriminating between Pseudagnostus and Plethagnostus. There the position
is clear since the one genus (without spines) occurs slightly later than the other.
Similarly, as I have already noted, it may be considered advisable in the future,
to make a division within Agnostus (sensu stricto) since the non-spinose species appear
all to be earlier than those with spines. But in the present group there is no such

258 MEMOIRS OF THE QUEENSLAND MUSEUM.

stratigraphical distinction. Typical goniagnostids and triplagnostids commonly .
occur together. That in itself, of course, does not preclude separate generic entity.
Howell (1935, p. 109) states that T'riplagnostus ranges throughout most of the Middle
Cambrian while Goniagnostus occurs merely in the zones of Paradowides hicksi and
P. davidis in the Middle Cambrian. Goniagnostus could thus be an offshoot of more
limited vertical range, arising early from T'riplagnostus and developing spines. This
is indicated by the intermediate position of G. purus, described below.

Since the two names have been proposed it probably will reduce confusion
if both are employed, until such time as an adequate revision can be made of all
species in the group. If, eventually, only one name is to be recognised Goniagnostus
has one page precedence over T'riplagnostus and, in the absence of other reasons for
choice, should be used.

GONIAGNOSTUS PURUS sp. nov.
(Pl. XXV, figs. 21-23.)

Diagnosis: The cephalon is subcircular with a narrow rim. The glabella
tapers uniformly to the front. The axial, pre-glabellar and anterior glabellar furrows
are all narrow but sharply defined. The anterior glabellar lobe is about one third
of the length of the glabella. On the posterior glabellar lobe the paired notches are
small but distinct. Accessory lobes are narrow and acutely triangular. The glabellar
tubercle is small, sub-circular, and situated towards the rear of the posterior lobe.
Usually the pleura are smooth ; but one form has been found with faint long and
short radial furrows developed only near the rim.

The pygidium is subrectangular, the width slightly greater than the length.
Apparently it is slightly more inflated than the cephalon. The rim is fairly uniform
in width and bears two small postero-lateral spines. The axis is much inflated, has
subparallel sides in its median region and terminates bluntly. It is rather more than
two-thirds the total length of the pygidium. The two axial furrows are well impressed.
The median tubercle is prominent, on the posterior portion of the centre segment.

Remarks : No complete specimen of conjoined cephalon, thorax and pygidium
has been obtained. In the limestone from which the material was gathered heads
and tails, each of two types, occur abundantly. One such head and tail were conjoined
and have been described above (Agnostus seminula). The other type of head and
the other type of tail are thus figured here confidently as the one species.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 259

G. purus is a curious species. Although it has the marginal spines of a
Goniagnostus it is most similar among described forms, to Agnostus gibbus Linarsson
the genotype of T'riplagnostus, which species already has been recorded from Queens-
land (Whitehouse 1936, p. 84, pl. 8, figs. 6, 7) apparently from a slightly lower horizon
(Dinesus Stage). But for its spines it is hardly to be distinguished from that species.
In the blunt ending ‘of the pygidial axis and in the smoothness of the cephalic cheeks
it also suggests T'riplagnostus more than Goniagnostus. Only one specimen has shown
any trace of the radial furrows on the cheeks of the cephalon,. and on that they are
faint and restricted to the outer edge.

Howell (1935, p. 14) states that Goniagnostus begins in the zone of Paradoxides
hicksi. The only other species that occurs with G. purus is Agnostus seminula
Whitehouse, a member of a subdivision of Agnostus that is known only in the zone
of Paradowides hicksi. These things suggest that G. purus is a very early member
of Goniagnostus, having recently branched off from the group of T'riplagnostus gibbus
and not yet acquired all the typical features of a Goniagnostus.

Locality and horizon: From the Agnostus seminula Stage in limestones eleven
and a half miles north of Yelvertoft Dip (Holotype in the University of Queensland
‘Collection. Other specimens in this and in the collection of the Geological Survey
of Queensland).

GONIAGNOSTUS cf. NATHORSTI (Brégger).
(Pl. XXV, fig. 20.)

1878. Agnostus nathorsti Brégger, p. 68, pl. 5, fig. 1.
1880. Agnostus nathorsti Tullberg, p. 21, pl. 1, fig. 9.
1935. Goniagnostus nathorsti Howell, p. 14, pl. 1, figs. 1, 2.

Two Queensland specimens appear to be identical with this species, the
genotype of Goniagnostus. They agree particularly well with the figures of topotypes
recently published by Howell. They have the same shape, similar regularly tapering
-glabella, radial cephalic furrows of the same type and small pygidial spines. By
comparing figures (I have seen no topotypes) I can find no essential difference between
the Queensland and the Scandinavian forms. :

In Scandinavia the species occurs in the zones of Paradoxides davidis and
Paradoxides forschammeri.

Locality and horizon: From the Papyriaspis Stage on the road from Undilla to
‘Thorntonia, a mile and a-half west of V Creek (University of Queensland Collection).

260 MEMOIRS OF THE QUEENSLAND MUSEUM.

GONIAGNOSTUS SCARABAEUS sp. nov.
(Pl. XXV, fig. 19.)

Diagnosis: Relatively a large form (10 mm. long) with straight, parallel sides..

The cephalon is subquadrangular, about two-thirds the total length and one-
third the width. The posterior lobe has subparallel sides and the anterior lobe is:
triangular. The pair of median notches on the posterior lobe are deep and long.
Accessory lobes are subovate, constricting the base of the posterior: -
lobe. Axial, glabellar and pre-glabellar furrows are prominent. The
cheeks are marked with radial furrows in an inner and an outer zone.
In the inner zone there are three or four pits on each side of the
posterior glabellar lobe and one on each side of the anterior lobe. In.
the outer zone, on each side, there are five furrows, progressively
increasing in size anteriorly to the mid region of the head.
Anterior to this the furrows are alternately long and short.

Restoration.
(x. 3) The thorax consists of two segments with tuberculate axis and’

short bulbous pleura, the bulbs being on opposite sides of the pleura—adjacent to:
the margins of the head and tail.

The pygidium is subquadrangular with well developed facets. There is a.
narrow rim that generally remains constant in width; but at the postero-lateral
angles spines arise that continue the straight line of the sides. The axis is long,
narrow and acuminate, extending almost to the rim. Anteriorly its margins diverge:
but slightly. The transverse furrows diverge from one another in the centre around.
a tubercle.

Remarks : With its long pygidial spines, the radial cephalic furrows developed
in two zones and, to some extent, the parallel sides of the test, this species is.
sufficiently distinct from all described forms. Perhaps the genotype, G. nathorsti,
is the most similar form.

Locality and horizon : One specimen has been found in beds of the Papyriaspis
Stage at the road crossing of V Creek between Undilla and Thorntonia (University
of Queensland Collection).

Genus EUAGNOSTUS Whitehouse, 1936.
EUAGNOSTUS SP.

One pygidium of a Huagnostus (in the University of Queensland Collection)
has been found in the Papyriaspis Stage at the road crossing of V Creek, between
Undilla and Thorntonia. It is a plump, thick-shelled form like Huagnostus opimus:
Whitehouse (1936, p. 87, pl. VIII, figs. 10-12) but differs in minor details. Attention

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 261

may be called to some typical species of Huagnostus that recently have been described
by Endo and Resser (1937, pp. 157-160, pl. 30) from the Mapan Formation of
Manchoukuo under the names Agnostus damesi E. and R., A. viator E. and R., A.
zak E. and R. and A. liaotungensis E. and R. These are all well rounded forms
with wide pygidial grooves, as are the Australian forms, and presumably Huagnostus
should be restricted to such types. The Asiatic and Australian forms together extend
the range of Huagnostus through most of the Middle Cambrian.

Genus HOMAGNOSTUS Howell, 1935.
(= Oncagnostus Whitehouse, 1936.)

Genotype: Agnostus pisiformis var. obesus Belt.

In erecting the genus Oncagnostus (Whitehouse 1936, p. 84), of which the
genotype was Agnostus hoi Sun, I regret that I overlooked Homagnostus Howell
(1935, p. 15) which I regard as congeneric. The genus has been found in Scandinavia,
‘Wales, North America (New Brunswick),.China and Chosen. In all these places it
occurs in the basal portion of the Upper Cambrian. I have now to record the genus
in the same stage in the Upper Cambrian of Queensland.

HOMAGNOSTUS cf. OBESUS (Belt).
(Pl. XXV, figs. 17 and 18 (?).)

Four pygidia of a species of Homagnostus have been collected and certain
-cephala that may belong to the same species. Quite possibly they belong to H. obesus,
which species is still inadequately known. Belt’s original figures (1867, p. 295, pl.
12, fig. 4) are rather poor. His types, wherever they may be, have not been refigured,
nor yet any topotypes. Lake (1906, part 1, p. 9, pl. 1, figs. 13 and 14) has published
figures of other Welsh specimens and has suggested that Agnostus pisiformis var.
socialis Tullberg (1880, p. 25) from Sweden is a synonym. Certainly these two forms
are congeneric ; but more must be known of H. obesus before we can be sure of the
specific relationship. Howell (1935, pl. 1, figs. 11 and 12) has recently figured a
Swedish specimen®® under the name H. obesus. That form has tiny marginal spines
on-the pygidium. Neither Belt nor Lake mention these on the Welsh specimens ;
although, if present as a specific feature, they could easily be obscured by conditions
of preservation. The tails now figured have incipient marginal spines. In most
features these tails agree with H. obesus as we know it.

Two other species of Homagnostus are known—H. hoi Sun sp. (1924, p. 28,
pl. 2, figs. 2) from China and H. hoiformis Kobayashi sp. (1933, p. 97, pl. 10, figs. 1-3)

58 Possibly a topotype of Agnostus pisiformis var. socialis,

262 MEMOIRS OF THE QUEENSLAND MUSEUM.

from Chosen. The New Brunswick form which Howell has found has not yet been
figured. The pygidial axis of H. hor narrows to the rear®® and in this way differs from
the Australian, Welsh and Swedish forms. In H. hoiformis the axial furrows on the
tail are obscure. In one feature, however, the Australian form may be compared
more closely with H. hoiformis than with any other species. The cephala that are
found with the Australian tails have a pair of discontinuous, posterior glabellar furrows.
Such furrows, characteristic features of certain other genera of Agnostidae, for instance
Triplagnostus, have been recorded only on one species of Homagnostus—H. hoiformis.
That species, however, has no median, pre-glabellar furrow, such a furrow being
present on the cephala of the Australian form and the other described species.

Thus if the associated cranidia and pygidia that now are figured belong to the
same species it would be necessary to designate it by a new name; but until this
can be established®® and until more is known of H. obesus a new name is not warranted.

Locality and horizon: From the Rhodonaspis Stage at the western base of
Polly’s Lookout, Glenormiston (University of Queensland Collection).

Family PHALACROMIDAE Corda, 1847.
Genus PHALACROMA Corda, 1847.
Genotype : Phalacroma scutiforme Corda.
PHALACROMA cf. NUDUM (Beyrich).
(Pl. XXV, figs. 28, 29.)

A typical Phalacroma is present in the beds of the Papyriaspis Stage, having
the large pygidial rim characteristic of P. nudum and its allies. I have not had access
to a copy of Beyrich’s work*! or seen any topotypes. Furthermore, as I have already
indicated (1936, p. 94), there is some difficulty in equating the forms referred to
P. nudum by later workers. Because of these things, and since the Queensland
specimens, with the exception of one head, are badly crushed I leave the species for
the present provisionally referred to P. nudum.

The specimens that I have collected are, with one exception, isolated heads
and tails. One specimen (plate XXV, fig. 29) has head, thorax and pygidium
conjoined, but it is imperfect and very badly crushed. As with the typical members
. of this group the head has a very narrow and the tail a wide rim.

5® This may not be a specific character; for one of Sun’s figures (2d) does not show it.
If there is no constant narrowing then it remains to be shown how H. hoi differs from H. obesus.
__ 8° Such a head could belong also to a species of Pseudagnostus ; and tails of Pseudagnostus
oecur on the same rock specimens.

61H. Beyrich, 1845. Ueber einige béhmischen Trilobiten. Berlin.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 263

Localities and horizon : One specimen (an uncrushed head) has been obtained
from the beds of the Papyriaspis Stage at the road crossing of V Creek, between
Undilla and Thorntonia. The majority of the specimens come from beds of the same
stage a mile and a half west of this locality (University of Queensland Collection).

? Family TRINODIDAE Howell, 1935.
Genus HYPAGNOSTUS Jackel, 1909.
Genotype: Agnostus parvifrons Linnarsson.

HYPAGNOSTUS CLIPEUS sp. nov.
(Pl. XXV, figs. 25, 26.)

Diagnosis : Cephalon subcircular. The major (median) glabellar lobe is slightly
acuminate and has a rounded base. It is slightly more than one-third the length of
the cephalon, and near its base is slightly more than one-third the total width. The
accessory glabellar lobes are triangular, widening at the base to join, on the steeply
descending slope, below the median lobe. The furrows around these lobes are
prominently incised. The cephalic rim is narrow and uniform in width.

The two thoracic segments have wide axes and short pleura. The axis of each
segment has a narrow furrow that begins at the posterior angles and reaches the
anterior margin in the centre, thus dividing the surface into three
inflated regions—a central, wide subtriangular area and two small,
transversely subovate, lateral areas. The pleura are short, with blunt
ends turned slightly forward. On each there is a narrow furrow dividing
the pleuron into a narrow ridge and a broad, bulbous portion. On each ay f
anterior pleuron the furrow is posteriorly placed, and it is in the UW)

anterior position on each posterior pleuron, so that the bulbous

Restoration.
(x. 3)

The pygidium is subcircular, with bevelled antero-lateral edges. It has a very

portions are adjacent to the cephalon and pygidium.

broad axis that extends almost to the posterior rim. On it the transverse furrows
and median tubercle are but faintly marked. The circumaxial furrow is prominent.
There is no post-axial furrow. The rim is narrow, uniform and without spines.

Remarks : The most similar species is the genotype, H. parvifrons Linnarsson
sp. (see Tullberg, 1880, pl. 2, figs. 26-28). The most pronounced differences are that
in the latter species the cephalic and pygidial rims widen in the axial region while the
pygidial axis reaches right to the rim. H. brevifrons Angelin sp. (see Tullberg, 1880,
pl. 2, fig. 29), H. exsculptus (Angelin) and its variety didymus Wallerius sp. (1930,
p. 58) have shorter pygidial axes. HA. vortex Whitehouse (1936, p. 103, pl. 9, figs.
7, 8) which previously was described from Queensland, has a more quadrate outline.

264 MEMOIRS OF THE QUEENSLAND MUSEUM.

When in 1909 Jaekel erected the genus Hypagnostus he proposed also the
name Metagnostus for similar agnostids with short pygidial axes. As genotype of
the latter he chose the Ordovician M. erraticus Jaekel, and suggested that the Middle
Cambrian Agnostus brevifrons Angelin was congeneric. It is most unlikely that the
two groups, Middle Cambrian and Ordovician, belong naturally to the one genus,
since Upper Cambrian forms of this type do not seem to be known. I do not think
it necessary, on present evidence, to place in separate genera A. parvifrons Linarsson
(with pygidial axes reaching to the rim) and» A. brevifrons Angelin (with short axes).
Morphologically H. clipeus is intermediate between these two.

Locality and horizon: From beds of the Papyriaspis stage at the road crossing
of V Creek, between Undilla and Thorntonia (N.W. Queensland). This is the type
locality. The species is also present in beds of the same Stage, two miles west of this
crossing of V Creek (University of Queensland Collection).

(C) FAUNAL AND STRATAL CORRELATIONS.
(a) Summary.

Since parts 1 and 2 of this series were published a few new faunas have been
found in the Georgina Limestones and certain additions and corrections can be made
to the lists previously issued. The following is the sequence of stages recognised
at present in the trilobite beds of the region. No trilobites have yet been found in the
latest beds—the Ninmaroo Limestones.

Division. Stage. Zonal Correlation,
Upper Cambrian .. .. | Hlathriella Stage Orusia lenticularis
| Rhodonaspis Stage Olennia
i Glyptagnostus Stage
Eugonocare Stage . Agnostus pisiformis
Middle Cambrian .. .. | Anomocare Stage Paradoxides forschammert
Papyriaspis Stage : ai
Phoidagnostus Stage Pavradonides davidis
| Agnostus seminula Stage Paradoxides hicksi
Dinesus State Ctenocephalus exsulans
Eurostina Stage Triplagnostus atavus
Amphoton Stage | Paradoxides oelandicus
Lower Cambrian .. .. | Redlichia Stage

Such collecting that has been done in this vast area has been but the casual
gleanings of a few reconnaissance journeys ; but already, even from these traverses,
it has been found necessary to subdivide the zones of Paradoxides davidis and Olenus.
When the area is adequately searched there can be little doubt that the zonal sequence,
as based on the Scandinavian succession, will have to be considerably distended.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 265

Two new stages have been added as a result of recent collecting—the Agnostus
seminula Stage and the Papyriaspis Stage. Also, since the names of the new polymerid
trilobites are now available, it has been necessary to change the designation of three
stages that had been named somewhat provisionally in the previous paper. In this
way the term “ Lurostina Stage’”’ replaces “‘ Inouyella Stage,” of the earlier list,
“ Hugonocare Stage’ replaces ““Anorina Stage,” and “ Rhodonaspis Stage ”’ replaces
“ Pagodia Stage.” The name “ Solenopleura Stage ’’ has been deleted from the list.
The recognition of this Stage was based on a small collection of trilobites that I had
received from Chatsworth Homestead. Since then I have visited the Chatsworth
area, have collected from limestones on several parts of this huge property, but
have found no limestones in situ around the homestead. Instead, about the home-
stead, there are limestone slabs, used for constructional purposes, that have been
brought from several parts of the property. From which of these slabs the fossils
were obtained I could not determine. There is, consequently, little purpose to be
served by describing this small fauna, and it is better to suppress reference to it until
fossils have been collected from actual outcrops.

(b) Faunal Lists and Details of Correlations.

i. THE Repuicuta Stace contains Mesodema venulosa and Redlichia idonea.
The primitive nature of these forms suggests that the fauna is equivalent to an early
part of the Redlichia sequence, as developed in Asia. .

ii. Tok AMPHOTON STAGE contains Amphoton spinigerum and Nepea narinosa.
It is equivalent to the “ Dolichometopus ”’ band in the Cambrian of Victoria in which
both Amphoton and Nepea occur. In North America the Nepeidae and Glossopleura
(allied to Amphoton) occur in early Middle Cambrian beds, generally below those
with Pagetia.®* In the absence of other evidence the Amphoton Stage, in this list,
is placed below the Dinesus Stage (with Pagetia).* A correlation with European zones
is tentative, since neither genus is known in Europe. In China this group of Amphoton
is typical of the Mapan Formation. For correlation purposes it may be.noted that
these beds in Queensland contain a species of the rare Cambrian and Ordovician
brachiopod genus Lingulodiscina.

iii. THE Evrostrna Stace contains Hurostina trigona, Xystridura sp.,
Notasaphus (7?) sp. and Anomocarella (?) sp. It is a small and inadequately known
fauna. It is believed to be intermediate between the Hurostina and Dinesus Stages
since it contains elements of both—JLingulodiscina and X ystridura.

iv. THE Dryesus Stace contains Triplagnostus atavus, T. gibbus, Diplorrhina
elkedraensis, D. normata, Pagetia significans, Paradoxides peregrinus, Xystridura
browni, X. saint-smithi, Lyriaspis sigillum, L. alrovensis, Chancia vicenalis, Dinesus

62 In the Rennie Formation such forms occur together. Generally, however, they would
seem to be in distinct faunas. 1

* Since the above was written Mr. D. E. Thomas has demonstrated to me in the field that -
in Victoria (at Knowsley) the Amphoton beds are below those with Dinesus,

8

266 . MEMOIRS OF THE QUEENSLAND MUSEUM.

ida, Notasaphus modicus and Oryctocephalus discus. In genera, species and individuals
it is the richest fauna in the province. The type area is Beetle Creek (so named from
the abundance of trilobites, locally known as “‘fossil beetles’’), 12 miles north west
of Mount Isa, where all these species occur except Diplorrhina elkedraensis, X ystridura
brownt and Lyriaspis alrovensis. Elsewhere in the area Diplorrhina elkedraensis
occurs with Pagetia significans and it is allied to D. normata. Lyriaspis alroiensis
occurs in the Northern Territory with Xystridura saint-smithi and Pagetia significans,
and it is allied to L. sigillum. These forms, therefore, legitimately may be added
to the fauna of the Dinesus Stage. Xystridura browni is close to X. saint-smithi and
it too is regarded as a member of this fauna. This is the earliest stage from which,
locally, agnostid trilobites have been collected. It may be correlated with the
““ Dinesus beds” of the Cambrian of Victoria that contain both Dinesus and.
Notasaphus.

This is a curious fauna with generic elements that ally it to the faunas of the
Asiatic, Cordilleran, Appalachian and Scandinavian Provinces. The agnostids, and
such genera as Pagetia, Paradoxides, Oryctocephalus and Notasaphus stamp it as an
early Middle Cambrian fauna. The presence of Pagetia, Oryctocephalus and Lyriaspis
suggest that it may be correlated with part of the Stephen Formation of British
Columbia and the “ Ptychoparia Beds” of North West Korea. As noted previously
the agnostids suggest an horizon approximately equivalent to the zone of Ctenocephalus
exsulans in Europe. ;

v. THE AGNOSTUS SEMINULA STAaGE.—In this Stage, for the present, I am
grouping two faunas from the region north east of Yelvertoft Dip. Each contains
only agnostid trilobites. From one bed I have collected, but the other I have not
been able to discover. When more is known of the area it may be advisable to separate
them. In a limestone eleven and a half miles north east of Yelvertoft Dip there
occur Agnostus seminula and Goniagnostus purus, species that indicate the zone of
Paradoaxides hicksi. From some hilltop, approximately four miles further to the
north east, in an area somewhat difficult to correlate, a fauna has been collected with
Triplagnostus atavus, Solenopleura acuminatus, Cotalagnostus aff. kushanensis, and
Hypagnostus vortex.~ From erroneous information that had been supplied to me
I recorded these species, previously, as occurring in beds of the Dinesus Stage. The
coexistence of Triplagnostus atavus and Solenopleura acuminatus suggests an horizon
somewhere between the Dinesus and Phoidagnostus Stages, so that, for the time
being, it can be grouped with the Agnostus seminula assemblage. The stratal
difficulty of correlating this fauna is discussed in the following chapter.

vi. THE PHorpaGNostus STaGE contains Solenagnostus acuminatus, HEnetag-
nostus humilis, Lejopyge exilis, Phoidagnostus limbatus and Phalacroma (?)

68 It is perhaps doubtful, as mentioned elsewhere, whether this is really the locality for
Paradoxides peregrinus and Chancia vicenalis. If not obtained from this place they probably came
from near by, in an area where no other faunas are yet known.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 267

dubium. Only one rather indefinite polymerid fragment has been found. Nothing
further is to be added to the reasons previously given for correlating this with the
zone of Paradoxides davidis.

vii. THE PAPYRIASPIS STAGE contains Ceratagnostus magister, Goniagnostus
nathorsti, G. scarabeus, Huagnostus sp., Hypagnostus clipeus, Phalacroma ct. nudum,
Papyriaspis lanceola, Asthenopsis levior, Amphoton serotinum and Anomocare (?)
angustum. The coexistence of Phalacroma and Euagnostus suggests an horizon between
the Phoidagnostus and Anomocare Stages. The presence of Ceratagnostus and of
Goniagnostus cf. nathorsti indicate the zone of Paradoxides davidis. Such a correlation
accords with the presence of Asthenopsis (ranging from the zone of Paradoxides hicksi
to the zone of P. forschammeri), a species of Anomocaridae not far removed from
Anomocare and the group of Hypagnostus represented by H. clipeus. The faunas of
this Stage have been collected from two localities, in the V Creek area, about a mile
and a half apart. On field evidence these beds seem to be separated by some thickness
of strata. One fauna has yielded abundant Asthenopsis and Papyriaspis but no
Amphoton, while the other has abundant Amphoton but no Papyriaspis or Asthenopsis.
Possibly some division of this Stage may be made when more intense field work is

“carried out. The occurrence of Amphoton so far above the Amphoton Stage agrees
with the evidence in Manchoukuo, where it occurs on two horizons—in the Mapan
Formation and in the Taitzu.

viii. THE ANOMOCARE STAGE contains Huagnostus opimus, Phalacroma (?)
sp. and Anomocare confertum. The Stage is equated with the zone of Paradoxides
forschammeri because of typical Anomocare which, in Europe, appears to be restricted.
to this horizon.

ix. THE EveonocarRE STAGE contains Pseudagnostus: vastulus, Eugonocare
tessellatum, EH. propinquum, Rhodonaspis sp., Corynexochus plumula, Charchaqia
erugata, C. spp.,, Proceratopyge lata and P. rutellum. This assemblage indicates the
zone of Agnostus pisiformis at the base of the Upper Cambrian. — Proceratopyge is an
early Upper Cambrian genus, ranging down into the top zone of the Middle Cambrian.
Pseudagnotus elsewhere is early Upper Cambrian, appearing first in the zone of A.
pisiformis. Charchaqia is very close to Hoasaphus at the base of the Upper Cambrian.
Corynexochus in Europe is a Middle Cambrian genus, ending at the top of the Middle
Cambrian. Since the remainder of this fauna is so typically basal Upper Cambrian
it suggests that Corynexochus here extends one Stage further into the zone of A.
prsiformis.

x. THE GLyPTAGNOSTUS StTaGE contains Glyptagnostus toreuma, Pseudagnostus
cf. cyclopyge, Proceratopyge nectans and Olenus (?) sp. This is equated with the Olenus

zone in Europe and the Cedaria zone in the Appalachian province, to each of which
the genus Glyptagnostus is restricted.

268 MEMOIRS OF THE QUEENSLAND MUSEUM.

xi, THe RHoponasPis STAGE contains Homagnostus cf. obesus, Pseudagnostus
sp., Proceratopyge polita, Rhodonaspis longula, R. prosecta, and apparently also
Idamea venusta. Containing Pseudagnostus and Proceratopyge, and underlying beds
(the Elathriella Stage) with Pseudagnostus, this is still in the early Upper Cambrian.
It seems best to regard it as the upper portion of the Olenus zone or the lower part
of the zone of Orusia lenticularis. The specimens of [damea venusta were collected,
in the region where beds of the Rhodonaspis Stage are known, by the late Mr. B.
Dunstan. I have not found this faunule in the field ; but tentatively I leave it in
the Rhodonaspis Stage.

xii. THE ELATHRIELLA STAGE contains Pseudagnostus nuperus, Aspidagnostus
parmatus, Elathriella. plebeia, E. sp., Idamea superstes and Olenus (?) sp. The most
probable horizon for these beds is the zone of Orusia lenticularis, which seems to make
the end, in Europe, of the range of Pseudagnostus.

(c) Some Stratal Aspects.

Grabau, on the basis of his Pulsation Theory, has suggested that there is a
hiatus and disconformity in the Georgina Limestones between the Middle and Upper.
Cambrian horizons and (Grabau 1937, p. 378) he questions the occurrence of Cory-
nexochus in the: Hugonocare Stage, suggesting that it is a remané fossil from lower
beds. ‘But, as I have shown, it is an essential and contemporary member of this
fauna. For that reason, and also for the strong argument of the lithological unity
of the Georgina Limestones, I see no reason for postulating such a break.

A few words are necessary on the earliest fossiliferous Cambrian beds in the
Yelvertoft area. In that region where the Cambrian beds are more or less horizontal
they rest on quartzites, dipping at varying angles, of late Pre-Cambrian age. Around
Yelvertoft homestead, and in the area about 16 miles north east of Yelvertoft dip the
Cambrian beds consist of fragments of cherts and silicified shales resting on these earlier
quartzites. Around the homestead these cherts have yielded abundant fossils, but only
of Redlichia, Aluta and one specimen of Mesodema. They would not appear, therefore,
to be a resorted rubble. In the region north of Yelvertoft dip three faunas in these
cherts have been obtained in a small area. At the base of a hill a collection of only
Redlichia was obtained. On the top of a hill, perhaps 50 feet higher than the other
bed, a fauna yielded only members of the Dinesus Stage (Xystridura saint-smithi
and Pagetia significans). From the top of another nearby hill Mr. Dalling of Yelvertoft
has collected the agnostid fauna with Solenagnostus, mentioned above in the section
dealing with the Agnostus seminula Stage. Again there is no admixture with other
faunas. ‘There seems to be in this region a condensed sedimentary succession of
cherts, resting on these Pre-Cambrian beds, or else (since fossiliferous limestones

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 269

with chert bands occur close by) they represent a series of limestones, with chert
bands, from which the calcareous material has been removed by solution, leaving
the cherts remaining, in their original although condensed sequence, as fragments on
the older rocks. It is a curious local problem on which field work is required.

(d) A Palaeogeographic Note.

The outstanding palaeogeographical feature of this province is that the faunas
have relationships in most of the other major Cambrian provinces that have been
defined. In the beginning, in the Redlichia Stage, the faunas are entirely Asiatic in
character, agreeing with those of the Himalayan and Cathayan Provinces. As in
those regions no trilobites are known before the Redlichia Stage. Thereafter, until
the Hugonocare Stage, there are still Asiatic elements in the fauna (Amphoton, Lyriaspis
and Papyriaspidae, for instance), but genera characteristic of other provinces make
their appearance. From the Dinesus Stage onwards there is a close faunal comparison
to be made with the Atlantic Province, particularly with the Scandinavian area and its
extension into the boreal regions of Bennett Island. Some elements of the faunas
indicate relationships with the Appalachian Province and its northerly extension
into Greenland and other boreal areas. Nepeidae and Elathriella are two such
elements. The presence of a rich Ellesmereoceras horizon in the Ninmaroo Limestones
indicates a faunal analogy with that province, and with Manchoukuo, at the end of
the Cambrian.

It may be that the marine areas of the continents were geosynclines at this
period, much in the manner that Grabau (1937, pl. III) has postulated ; but if so I do
not think it necessary to believe that these fluctuating faunal relationships with the
several provinces indicate migrating geosynclines. The Australian region is one
of the most southerly Cambrian provinces ; and if it should have been (as is indicated
on so many palaeogeographical maps that have been drawn for the Cambrian) that
many of the northerly geosynclines opened into a southern sea or ocean, this would
permit of free transport of larval life by marine currents into the southern sea beside
the Australian geosynclines. Then as facies, and particularly climatic facies,
fluctuated the relationships of the Australian Cambrian faunas would fluctuate too,
having relationships now with one province, now with one or several of the others.
That, I suggest, is an explanation of these relationships in the light of our present
knowledge of Cambrian ecology.

And thus future work in extending our knowledge of the Australian Cambrian _
faunas may have two important results—it may, from community of relationships,
allow correlations to be made between other Provinces that are themselves distinct,
and it may shed an important light on the variations in climatic facies during the
period.

270 MEMOIRS OF THE QUEENSLAND MUSEUM.

5. LOCALITIES.

In the very sparsely settled region where these collections were made localities
have had to be recorded in terms that will have little significance unless one has access
to the four-mile maps of the region. Care has been taken to define the localities in
such a way that they may be found by later workers visiting the region ; while for
general purposes I append the latitude and longitude, recorded as far as possible to
the nearest minute, of a number of key localities. By comparison with this list any
locality recorded in the text may be fixed upon the map. There are few localities
in this table that can be stated in terms more accurate than one minute of latitude
and longitude. The list is as follows:

Alexandria—homestead E. 136° 42’; §. 19° 03’
Alroy Downs—homestead BK. 136° 04’; 8. 19° 18°
Alroy Downs—No. 1 bore Ei; 136? 15"; “S.19°_ 15"
Alroy Downs—No. 4 bore E. 136° 14’; 8. 19° 28’
Camooweal . me E. 138° 07’; S. 19° 55’
Cid eerie cneings “i E. 140° 18’; S. 21° 58’
Chatsworth—mouth of Lily Grok E. 140° 17’; 8. 22° 00’
Devoncourt—homestead . EB. 140° 14°; §. 21° 13’
Duchess MH. 139° 52’; §. 21° 22’
Flictian “Bowtgetaia Ee 138° 36": 8. 21°07"
Glenormiston—homestead. E..138° 48’; §..22° 55’
Glenormiston—Polly’s Lookout . . E. 138° 56’; S. 23° 05’
Glenormiston—No. 17 bore B138? 29°: §..22°°63"
Glenormiston—20-mile bore E. 138° 51.’; §. 23° 13”
Glenormiston—Tyson’s bore By US8e43%~ Ss) 2a Or"
Herbert Downs—homestead E. 139° 15’; 8. 23° 02’
Thorntonia—homestead . ee E. 138° 55’; §. 19° 30°
Warenda—Black Mt. (Ninmaroo) E. 140° 16’; S. 22° 33’
Warenda—Unbunmaroo .. E. 140° 16’ + §;. 227 38”
Warenda—-Mt. Datson he vs se i n« sed40" 25's Si 22° 47°
Yelvertoft—homestead .. : is oe a eR SS ba! 38. 2H TS"
Yelvertoft—dip (Old Yelvertoft) E. 138° 48’ ; 8. 20° 06’
Beetle Creek (on Thornton River)* E. 139° 20°;.8.-20° 37’
‘* Split Rock ” on Waroona Creek KE. 138° 29’; 8. 19° 59’ ~
Road crossing of V Creek (Undilla to Thorntonia) .. E. 138° 44’; 8. 19° 34’
Road crossing of Harris Creek (Camooweal to Old Thorntonia) E. 138° 43’; 5. 19° 26’

ACKNOWLEDGMENTS.

I wish gratefully to acknowledge financial assistance in recent field work on
these problems from the following institutions: The University of Queensland, the
Research Funds of the Commonwealth Council for Scientific and Industrial Research,
and the Australian and New Zealand Association for the Advancement of Science.

64 This locality, about 12 miles north west of Mt. Isa, was referred to previously (Whitehouse,
1936) as ‘‘ Thornton River.”

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 271

6. BIBLIOGRAPHY.

Anderson, C. 1931. Trilobites. Aust. Mus. Mag., IV, No. 5, pp. 178-179.

Angelin, N. P. 1854. Palaeontologia Scandinavica Pars 1. Crustacea formationis transitionis.
Stockholm (3rd Ed. 1878).

Barrande, J. 1852. Systeme silurien du Centre de la Bohéme. Ite Partie. Recherches
paléontologiques. 1. Prague.

Beecher, C. E. 1897. Outline of a natural Classification of the Trilobites. Amer. J. Sci., Ser. 4,
III, pp. 89-106, 181-207.

Belt, T. 1867. On some new Trilobites from the Upper Cambrian Rocks of North Wales. Geol.
Mag., Lond., IV, pp. 294-295.

Billings, E. 1861. On some new or little known Species of Lower Silurian Fossils from the Potsdam
Group (Primordial Zone). Rep. on Geol. of Vermont (Hitchcock Ed.), II, pp. 942-960.

Brogger, W. C. 1878. Om Paradoxides Skrifrene ved Krekling. Nyt Mag. Naturvid., XXIV,
Pt. 1, pp. 17-88.

Butts, C. 1926. The Palaeozoic Rocks (in “‘ Geology of Alabama ”’ by C. Butts and others). Geol.
Surv. Alabama special Rep., No. 14.

Chapman, F. 1911. New or little-known Victorian Fossils in the National Museum—Part XII:
On a Trilobite Fauna of Upper Cambrian Age (Olenus Series) in N.E. Gippsland.
Proce. R. Soc. Victoria, XXIII (NS), pp. 305-324.

1917. The Heathcote Fauna. Rees. geol. Surv. Victoria, IV, Pt. 1, pp. 89-102.
1924. The Wade Collection of Fossils (Appendix to A. Wade, 1924, q.v.).-

1929. Onsome Trilobites and Brachiopods from the Mount Isa District, N.W. Queens-
land. Proc. R. Soc, Victoria, XXIII (NS), Pt. 2, pp. 305-324.

David, T. W. E. 1928. Notes on newly discovered Fossils in the Adelaide Series (Lipalian ?),
South Australia. Trans. R. Soc. S. Aust., LITT, pp. 191-209.

1932. Explanatory Notes to accompany a new geological Map of the Commonwealth
of Australia. Sydney.

David, T. W. E. and Tillyard, R. J. 1936. Memoir on Fossils of the late Pre-Cambrian (Newer
Proterozoic) from the Adelaide Series, South Australia. Sydney.

Endo, R. and Resser, C. E. 1937. The Sinian and Cambrian Formations and Fossils of southern
Manchoukuo. Manchurian Sci. Mus. Bull. No. 1.

Etheridge, R. Jr. 1896. Evidence of the Existence of a Cambrian Fauna in Victoria. Proc. R.
Soc. Victoria, VIII (NS), pp. 52-64.

— 1897. Official Contributions to the Palaeontology of South Australia, No. 9—On the
Occurrence of Olenellus in the Northern Territory. S. Aust. Parl. Papers.

1905. Additions to the Cambrian Fauna of South Australia. Trans. R. Soc. S. Aust.,
XXIX, pp. 246-251.

1919. The Cambrian Trilobites of Australia and Tasmania. Trans. R. Soc. S. Aust.,
XLII, pp. 373-393.

Fletcher, H. 0. 1935. Trilobite Hunting in the North. Aust. Mus. Mag., V, No. 9, pp. 305-312.

Foord, A. H. 1890. Description of Fossils from the Kimberley District, Western Australia. Geol.
Mag., Lond., (dec. iii), VII, pp. 98-106.

272 MEMOIRS OF THE QUEENSLAND MUSEUM.

Grabau, A. W. 1937. Palaeozoic Formations in the Light of the Pulsation Theory. Vol. ITI,
Cambrovician Pulsation System Part II. Peking.

Gregory, J. W. 1903. The Heathcotian: A Pre-Ordovician Series and its Distribution. Proc.
R. Soc. Victoria, XV (NS), pp. 148-175.

Grénwall, K. A. 1902. Bornholms Paradoxideslag og deres Fauna. Danmarks geol. Unders., II,
No. 13.

Girich, G. 1907. Versuch einer Neueinteilung der Trilobiten. Centralbi. f. Min. Geol. u. Palaeont.,
p. 129.

Hall, J. and Whitfield, R. P. 1877. Fossils of the Potsdam Group. U.S. Geol. Explor. of the
40th Parallel, IV, Pt. 2, pp. 198-302.

Holm, G. and Westergaard, A. H. 1930. A Middle Cambrian Fauna from Bennett Island. Mem.
Acad. Sei. U.R.S.S., Ser. 8, Cl. Phys.Math., XXI, No. 8.

Howell, B. F. 1933. The Classification of the Trilobite Subfamily, Centropleurinae. Meddel. fra
Danks geol, For., VIII, pt. 3.

1935. Some New Brunswick Cambrian Agnostians. Bull. Wagner free Inst. Sci.,
X, No. 2, pp. 13-16.

Illing, V. C. 1916. The Paradoxidian Fauna of a Part of the Stockingford Shales. Quart. J. geol.
Soc., LX XI, pp. 386-448.

Jaekel, O. 1901. Uber die Organisation der Trilobiten I. Z. disch. geol. Ges., LIII, pp. 133-171.
1909. Uber die Agnostiden. Z. dtsch. geol. Ges., LXI, pp. 380-401.

;

Jensen, H, I. 1915. Report on the Geology of the Country between Pine Creek and Tanami.
Bull. Northern Territory, No. 14.

Kiaer, J. 1916. The Lower Cambrian Holmia Fauna at Tomten in Norway. Vid.—Selsk. Skr.,
Math—Nat. Kl., No. 10,

King, W. R. B. 1930. Notes on the Cambrian Fauna of Persia. Geol. Mag., Lond., LX VII, pp.
316-327,

1937. Cambrian Trilobites from Iran (Persia). Palaeont. Indica (NS) XXII, Mem.
No. 5.

Kobayashi, T. 1933. Upper Cambrian of the Wuhutsui Vasin, Liaotung, with special Reference
to the Limits of the Chaumitian (or Upper Cambrian) of Eastern Asia, and its Sub-
divisions. Jap. J, Geol. and Geogr., XI, Nos. 1 and 2, pp. 55-155.

1934. Middle Cambrian Fossils from Kashmir. Amer. J. Sci., X XVII, pp. 295-304.
— 1935. The Cambro-Ordovician Formations and Faunas of South Chosen.
Palaeontology Part III: Cambrian Faunas of South Chosen, with a special Study of

the Cambrian trilobite Genera and Families. J. Fac. Sci. Imp. Univ. Tokyo, Sect. 2,
IV, Pt. 2.

1936a. On the Parabolinella Fauna from Province Jujuy, Argentina, with a note on
the Olenidae. Jap. J. Geol. and Geogr., XIII, Nos. 1 and 2, pp. 85-102.

1936b. Three Contributions to the Cambro-Ordovician Faunas. Jap. J. Geol. and
Geogr., XIII, Nos. 1 and 2, pp. 163-184. :

1937. The Cambro-Ordovician shelly Faunas of South America. J. Fac. Sct. Imp.
Univ. Tokyo, Sect. 2, IV, Pt. 4.

Lake, P. 1906—. A Monograph of the British Cambrian Trilobites. Palaeontogr. Soc. Mon.

Lochman, C. 1938. Upper Cambrian Faunas of the Cap Mountain Formation of Texas. J.
Palaeont., XII, No. 1, pp. 72-85.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 273

Lorenz, T. 1906. Beitrige zur Geologie und Palaeontologie von Ostasien unter besonderer
Beriicksichtigung der Provinz Schantung in China, II: Palaeontologischer Teil.
Z. disch. geol. Ges., LVITI, pp. 53-108.

Mansuy, H. 1912. Etude géologique du Yun-nan oriental. Ile Partie: Paléontologie. Mem.
Serv. geol. Indochine, I, fasc. 2.

Mason, J. F. 1935. Fauna of the Cambrian Cadiz Formation, Marble Mountain, California. Bull.
S. Calif. Acad. Sci., XXXIV, pp. 97-118.

Matthew, G. F. 1886. Illustrations of the Fauna of the St. John Group, continued. No. 7i7—2
Description of new Genera and Species (including a Description of a new Species of
Solenopleura, by J. F. Whiteaves). Proc. and Trans. R. Soc. Canada, III, Sect. 4, pp.
29-84.

1888. Illustrations of the Fauna of the St. John Group. No. IV. Part I. Description
of a new Species of Paradoxides (Paradoxides regina). Part IJ. The smaller Trilobites
with Eyes (Ptychoparidae and Ellipsocephalidae). Proc. and Trans. R. Soc. Canada,
V, Sect. 4, pp. 115-166.

1891. Illustrations of the Fauna of the St. John Group. No. VI. Proc. and Trans.
&. Soc. Canada, IX, Sect. 4, pp. 33.

1896. ‘Faunas of the Paradoxides Beds in Eastern North America, No. 1. Trans.
New York Acad. Sct., XV, pp. 192-247,

1899. Studies on Cambrian Faunas, No. 3. The Cambrian Fauna of Mount Stephen,
B.C. Trans. R. Soc. Canada, V, Ser. 4, pp. 39-66.

Miquel, J. 1905. Essai sur le Cambrien de la Montagne Noire. Bull. Soc. géol. France, Ser. 4, Vv;
pp. 465-483.

Mitchell, J. 1922. Descriptions of two new Trilobites, and Notes on Griffithides convexicaudatus
Mitchell. Proc. Linn. Soc. N. S. Wales, XLVII, Pt. 4, pp. 535-540.

Moberg, J. C. 1898. En Trilobit fran Skanes Dictyograptusskiffer. Geol. Fér. i Stockholm Féorh.,
XX, No. 6, pp. 317-324.

Moberg, J. C. and Segerberg, C. O. 1906. Bidrag till Kannedomen om Ceratopygeregionen med
sirskild Hansyn till dess Utveckling i Fogelsangstrakten. Meddel. fran. Lunds. geol.
Fdltkiubb, Ser. B, No. 2.

’ Nicholas, T. C. 1916. Notes on the Trilobite Fauna of the Middle Cambrian of the St. Tudwal’s
Peninsula (Carnarvonshire). Quart. J. geol. Soc., LXXI, pp. 451-472.

Obrutschew, W. A. 1926. Geologie von Sibirien. Berlin.
Opik, A. 1937. Trilobiten aus Estland. Pub. Geol—Inst. Univ. Tartu, No. 52.

Poulsen, C. 1927. The Cambrian, Ozarkian and Canadian Faunas of northwest Greenland,
Meddel. om Groniand, LXX, pp. 237-343.

Raymond, P. E. 1913. Trilobita (in Zittel-Eastman ‘‘ Textbook of Palaeontology,” 2nd Ed.)
London. ;

1920. The Appendages, Anatomy, and Relationships of Trilobites. Mem. Connecticut
Acad, Arts and Sei., VII.

1928. Two new Cambrian Trilobites. Amer. J. Sci., Ser. 5, XV, pp. 309-313.

Reed, F. R. C. 1899. Woodwardian Museum Notes: A new Trilobite from Mount Stephen, Field,
B.C. Geol. Mag., Lond., (dec. iv), VI, pp. 358-361.

1910. The Cambrian Fossils of Spiti. Palaeont, Indica, Ser. 15, VII, Mem. No. 1.

274 MEMOIRS OF THE QUEENSLAND MUSEUM,

Resser, C. E., 1933. Preliminary generalised Cambrian Time Scale. Bull. geol. Soc. Amer., XLIV,

pp. 735-756.
— 1935. Nomenclature of some Cambrian Trilobites. Smithson. misc. Coll., XCITI,
No. 5.

1936. Second Contribution to Nomenclature of Cambrian Trilobites. Smithson.
misc. Coll., XCV, No. 4.

1937. Third Contribution to Nomenclature of Cambrian Trilobites. Smithson.
misc. Coll., XCV, No. 22.

1938a. Middle Cambrian Fossils from Pend Oreille Lake, Idaho. Smithson. misc.
Coll., XCVII, No. 3.

1938b. Cambrian System (restricted) of the southern Appalachians. Geol. Soc.
Amer. spec. Paper No. 15.

Richter, R. 1932. Crustacea (In ‘“‘ Handwérterbuch der Naturwissenschaften,” 2nd Ed., Jena,
pp. 840-863). ;

Rominger, €. 1888. Description of Primordial Fossils from Mt. Stephen, N.W. Territory of Canada.
Proc. Acad. nat. Sci. Philad., 1887, pp. 12-19.

Saito, K. 1934. Older Cambrian Trilobita and Conchostracha from north-western Korea. Jap.
J. Geol. and Geogr., XI, Nos. 3-4, pp. 211-237. :

1936. Older Cambrian Gastropoda ete. from north-western Korea. J. Fac. Sev.
Imp. Univ. Tokyo, Sec. 2, IV, Pt. 3, pp. 345-367.

Saito, K. and Sakakura, K. 1936. Description de deux nouvelles especes de Trilobite. J. geol.
Soc. Japan, XLITI, pp. 112-117.

Schuchert, C. 1936. [Review of : Whitehouse, F. W., ‘‘ The Cambrian Faunas of Northeastern
Australia, Parts 1 and 2]. Amer. J. Sci., Ser. 5, XXXIT, pp. 397-398.

Stérmer, L. 1933. Are the Trilobites related to the Arachnids? Amer. J. Sci., (5) X XVI, pp.
147-157.

Stubblefield, C. J. 1936. Cephalic Sutures and their Bearing on current Classifications of
Trilobites. Biol. Revs., XI, pp. 407-440.

Sun, Y. C. 1924. Contributions to the Cambrian Faunas of north China. Palaeont. Sinica, Ser.
B, Vil, Fase. 2.

Swinnerton, H. H. 1915. Suggestion for a revised Classification of Trilobites. Geol. Mag., Lond.,
Dec. 6, IT, pp. 487-496, 538-545,

1919. The facial Suture of Trilobites. Geol. Mag., Lond., (dec. vi), VI, pp. 103-110.

Thomas, D. E. 1935. Cambrian. Handbook for Victoria, Aust. and N. Z’land Assoc. Adv. Sci.,
pp. 91-95.

Tillyard, R. J. 1936. Tracing the Dawn of Life further Backwards. Cawthron Inst. Pub.

Toll, E. von 1899. Beitrage zur Kenntniss des sibirischen Cambrium. Mem. Acad. Imp. Sci.
St. Petersb., Ser. B, Cl. Math.—Phys., VIII, No. 10.

Troedsson, G. T. 1937. On the Cambro-Ordovician Faunas of western Quruq Tagh, eastern
T’ien-Shan. Palaeont. Sinica, (NS) B, No. 2.

Tullberg, S. A. 1880. Om Agnostus-Arterna i de Kambriska Aflagringarne vid Andrarum. Sver.
geol. Unders., Ser. C, No. 42, pp. 1-37.

Vogdes, A. W. 1925. Palaeozoic Crustacea. Trans. San Diego Soc. Nat. Hist., IV, pp. 1-154.

Wade, A. 1924. Petroleum Prospects: Kimberley District of Western Australia and Northern
Territory. Special Parl. Paper, Commonwealth of Australia,

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 275

Walcott, C. D. 1886. Second. Contribution to the Studies on the Cambrian Faunas of North
America. U.S. geol. Surv., Bull. No. 30.

= 1887. Fauna of the ‘“ Upper Taconic”? of Emmons in Washington County, New
York. Amer. J. Sci., Ser. 3, XXXIV, pp. 187-199.

1889. Descriptions of new Genera and Species of Fossils from the Middle Cambrian.
Proc. U.S. nat. Mus., XI, pp. 441-446.

1890. The Fauna of the Lower Cambrian or Olenellus Zone. U.S. geol. Surv., 10th
A.R., pp. 511-774.

1905. Cambrian Faunas of China. Proc. U.S. nat. Mus., XXIX, pp. 1-106.

1910. Cambrian Geology and Palaeontology No. 6—Olenellus and other genera of
the Mesonacidae. Smithson. misc. Coll., LIII, No. 6.

1912, Cambro-Ordovician Boundary of British Columbia with Description of Fossils.
Smithson. misc. Coll., LX VII, No. 7.

1913. The Cambrian Faunas of China. Carnegie Inst. (Washington). Pub. No. 54,
pp. 1-276.

1914. Cambrian Geology and Palaeontology III—No. 1. The Cambrian Faunas of
Eastern Asia. Smithson. misc. Coll., LXIV, No. 1.

1916. Cambrian Geology and Palaeontology III, No. 5: Cambrian Trilobites.,
Smithson. misc. Coll., LXIV, No. 5.

1918. Cambrian Geology and Palaeontology IV, No. 4—Appendages of Trilobites.
Smithson. misc. Coll., LXVII, No. 4.

1921. Notes on the Structure of Neolenus. Smithson. misc. Coll., LX VII, No. 7.

1924. Cambrian Geology and Palaeontology V, No 2—Cambrian and Lower Ozarkian

Trilobites. Smithson. misc. Coll. LXXV, No. 2.

1925. Cambrian Geology and Palaeontology V, No. 3—Cambrian and Ozarkian
Trilobites. Smithson. misc. Coll., LXV, No 3.

Walcott, C. D. and Resser, C. E. 1924. Trilobites from the Ozarkian Sandstones of the Island
of Novaya Zemlya. Rep. Sci. Results Norwegian Exped. to Nov. Zemlya, 1921, No. 24.

Wallerius, I. D. 1895. Undersékningar 6fver Zonen med Agnostus laevigatus i Vestergotland.
Akad. Afh., Lund.

Warburg, E. 1925. The Trilobites of the Leptaena Limestone in Dalarne. Bull. geol. Inst. Univ.
Upsala, XVII.

Westergaard, A. H. 1922. Sveriges Olenidskiffer. Sver., geol. Unders., Ser. Ca. No. 18.
1936. Paradoxides oelandicus Beds of Oland. Sver. geol. Unders., Ser. C, No. 394.

Whitehouse, F. W. 1927. [Notes accompanying an Exhibit of Fossils] Proc. R. Soc. Q’land.,
XXXIX, pp. vil-viil.
1930. The Geology of Queensland. @’land Handbook for. Aust. Assoc. Adv. Sci.,
pp. 23-39.
1931. Report of the Palaeontologist. A. R. Dept. Mines Q’land. for 1930, pp. 141-142.

1936. The Cambrian Faunas of North-Eastern Australia. Part 1: Stratigraphical
Outline. Part 2: Trilobita (Miomera). Mem. Q’land. Mus., XI, pp. 59-112.

Woodward, H. 1884. Note on the Remains of Trilobites from South Australia. Geol. Mag.,
Lond., (dec. iii) I, pp. 342-344.

276 MEMOIRS OF THE QUEENSLAND MUSEUM.

7. EXPLANATION OF PLATES.

PLATE XIX,
(All figures natural size except figs. 2b, 5b and 6.)

Fig. 1. Mesodema venulosa sp. nov. Holotype (page 190).
From Yelvertoft Homestead (Redlichia Stage).

a is the broken edge of the palpebral lobe as it arises from the frontal lobe of the glabella.
b is the post-palpebral surface, arising from the second glabellar lobe.

Figs. 2-11. Redlichia idonea sp. nov. (page 191).
From Yelvertoft Homestead (Redlichia Stage).

2a, b. A small individual. The depression on the glabella probably corresponds with
the position of the hypostome. Fig. 2b is enlarged three diameters.

3. Another specimen similar to 2a.

4. The holotype.

5a, b, An internal mould with curious segmental structures. 5a shows the entire-
specimen, with eight thoracic and two cephalic processes, the anterior of the two
cephalic structures being considerably larger than the others. 5b, enlarged three
diameters, shows, in detail, the three posterior processes of 5a. In 5a, on the
cephalon, there is the cranidium of a second, smaller, individual.

6. The most complete specimen found—a young individual (magnified three diameters).

7, 8 and 9. Cranidia of three individuals of average size.

10. The posterior portion of the test, showing the spine arising from the axis on the
eleventh thoracic segment.

11. A larger specimen showing the thoracic spine.

The originals of figures 1, 3, 4, 5, 6, 7, 8 and 11 are in the University of Queensland Collection.

The original of figures 2, 9 and 10 are in the Schmidt Collection, on loan to the University of Queens-
land.

PLATE XX.
(All figures natural size except fig. 6.)

Figs. 1-5. Redlichia idonea sp. nov. (page 191). Five specimens of large size from Yelvertoft.
Homestead (Redlichia Stage).

. The largest cranidium.

. Portion of a thoracic axis.

A free cheek.

A hypostome attached to the rostrum.
A pygidium.

wee iS

Fig. 6. Paradoxides peregrinus sp. nov. Holotype (x. 10) (page 194).
From the Templeton River area, north west of Mount Isa (Dinesus Stage).

The originals of figs. 1 and 2 are in the Schmidt Collection, on loan to the University of
Queensland ; figs. 3, 4 and 5 in the University of Queensland Collection ; fig. .6 in the collection of
the Geological Survey of Queensland.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 277

PLATE XXI.
(All figures natural size except figs. 4 and 16.)

Fig. 1. Xystridura browni (Etheridge fil.), holotype (page 198).
From the spoil heap of a well, 5 miles north of Alexandria Homestead (Dinesus
Stage).
Figs. 2-16. Xystridura saint-smithi (Chapman). (page 199).
2. Holotype of species. From Beetle Creek, 12 miles north west of Mount Isa (Dinesus
Stage).
3a and 4. Holotype of “ Bathyuriscus nitidus*’ Chapman. Fig 4 is enlarged two
diameters. Same locality and horizon.
3b. Holotype of ** Bathyuriscus olenelloides *’ Chapman. Saime locality and horizon.
5. Holotype of “* Marjumia milest”’ Chapman. Same locality and horizon.
6. Holotype of “‘ Marjumia conspicabilis’? Chapman, From Shepherd Creek, near
Miles Creek, north branch of the Templeton River (Dinesus Stage).
7. Paratype of “ Marjumia conspicabilis ’ Chapman. Same locality and horizon.
8. Holotype of “‘ Marjumia elegans”? Chapman. From Beetle Creek, 12 miles north
west of Mount Isa (Dinesus Stage).

9. Holotype of ‘‘ Dikelocephalus dunstani”? Chapman. From the Thornton River
(Dinesus Stage). ;
10. Holotype of ‘“ Milesia templetonensis”? Chapman. From Beetle Creek, 12 miles
north west of Mount Isa (Dinesus Stage). :
11. Complete specimen, rather narrower than usual. Same locality and horizon.
12. A typical, complete specimen. Same locality and horizon.
13. A cranidium and part of the thorax. In the upper part, a hypostome. Same locality
and horizon. °
14, A hypostome attached to part of the rostrum. Same locality and horizon.
15. The pygidium of a large specimen, with the last thoracic segment attached. Same
locality and horizon.
-16. The two smallest specimens examined (x. 3). Same locality and horizon. Note the
long genal spines and the meta-cranidial spines.

The original of fig. 1 is in the Australian Museum Collection ; fig. 2 in the collection of the
Commonwealth Geological Survey, Canberra ; figs 3-10 in the collection of the Geological Survey
ofjQueensland ; fig. 11 in the private collection of Mr. J. Wadley ; figs. 12-16 in the University of
Queensland Collection.

PLATE XXII.

(Figures natural size, unless otherwise stated.)

Figs. 1, 2. Xystridura saint-smitht (Chapman). (page 199).

1. An internal mould on which the following features may be noted : The wide doublure
of the thorax; the doublure of the cephalon extending around the genal angle ;
the rostrum extending into the angle formed di the doublure near the genal angle ;
the hypostome in place.

From the Dinesus Stage of Beetle Creek, 12 miles N.W. of Mount Isa,

2. A worn specimen from the No. 1 bore, Alroy: Downs (Dinesus Stage).

278 MEMOIRS OF THE QUEENSLAND MUSEUM.

Figs. 3-5. Lyriaspis sigillum sp. nov. (page 204).
3. The holotype (x. 2).
4, A specimen minus the free cheeks (x. 2).
5. An incomplete specimen.
- From the Dinesus Stage of Beetle Creek, 12 miles N.W. of Mount Isa.
Fig. 6. Lyriaspis alroiensis (Etheridge fil.). Holotype (page 203).
Specimen preserved on an aboriginal artefact picked up at 8 miles east of Alroy
Downs. The species is a member of the Dinesus Stage fauna.
Fig. 7. Chancia vicenalis sp. nov. Holotype (page 205).
From the Dinesus Stage of the Templeton River area, probably from Beetle
Creek. :
Figs. 8, 9. Hlathriella plebeia sp. nov. (page 207).
8. Holotype, a cranidium (x. 2).
9. An immature individual (x. 4).
From the base of a hill immediately west of Tyson’s Bore, Glenormiston
(Elathriella Stage).
Fig. 10. Hlathriella sp. (x. 2) (page 208).

Locality and horizon as for figs. 8, 9.

Figs. lla, b. Hurostina trigona sp. nov. Holotype (page 209).
lla. The cranidium magnified 2 diameters.
llb. The same (x. 4).
From the Hurostina Stage of No. 4 Bore, Alroy Downs.
Figs. 12a, b and 13. Protemnites elegans sp. nov. (page 210). >
12. Two cranidia. The larger is the holotype. Fig. 12a is x.2; fig. 12b is x. 4.
13. A pygidium (x. 4).
Locality and horizon as for figs. 8, 9.

Fig. 14. Anomocarella (?) sp. (page 227).
A cranidium from the same locality and horizon as figs lla, b.
Figs. l5a, b and 16. Nepea narinosa sp. nov. (page 212).
. 15a The holotype (a cranidium).
15b. The same (x. 2).
16. Another cranidium.
From the Amphoton Stage of the road crossing of Waroona Creek (“* Split Rock ’’)
between Camooweal and Mount Isa.
Figs. 17-20. Asthenopsis levior sp. nov. (page 214).
17. Holotype, a complete individual.
18. The cranidium of a larger specimen.
19. A pyyidium.
20. Thorax and pygidium. ‘
From limestones of the Papyriaspis Stage at the road crossing of V Creek, between.
Undilla and Thorntonia.

The original of figure 6 is in the coHection of Mr. A. L. Merrotsy. The original of figure 7
is in the collection of the Geological Survey of Queensland. All other specimens are in the
collections of the University of Queensland.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 279

PLATE XXTII.

(All figures natural size unless otherwise stated.)

Figs. 1-8. Papyriaspis lanceola sp. nov. (page 218). :
1. Holotype (x. 2).
2, 3. Other moderately complete specimens (x. 2).
4. A young individual (x. 2).
5, 6. Cranidia.
7. Thorax and pygidium.
8. Pygidium of a large specimen. :
From the Papyriaspis Stage at the road crossing of V Creek, between Undilla
and Thorntonia.
Fig. 9. Rhodonaspis longula sp. nov. (x. 2). Holotype (page 220).
About two miles south of Tyson’s Bore, Glenormiston (Rhodonaspis Stage).
Figs. 10-12. Rhodonaspis prosecta sp. nov. (page 220).
10. A cranidium, the holotype, (x. 2).
11, 12. Pygidia (x. 2).
From the Rhodonaspis Stage at the base of Polly’s Lookout, Glenormiston.
Fig. 13. Rhodonaspis sp. (x. 2) (page 221).
A fragmentary cranidium from the Hugonocare Stage, ant miles north of the
20 Mile Bore, Glenormiston.
Fig. 14. Olenus (?) sp. (page 222).
A cranidium from the Elathriella Stage at the base of a hill immediately west of
Tyson’s Bore, Glenormiston.
Figs. 15-18. Eugonocare tessellatum sp. nov. (page 226).
15. Holotype (a cranidium), (x. 3).
16 and 17. Two pygidia (16 is magnified two diameters).
18. A free cheek.

From the Hugonocare Stage about four and a half miles north of the 20 Mile Bore
Glenormiston.
Figs. 19, 20. Hugonocare propinquum, sp. nov. (page 227).
19. A pygidium (the holotype), (x. 2).
20. A cranidium.
From the Hugonocare Stage, five miles north of the 20 Mile Bore, Glenormiston.
Fig. 21. Anomocare (7?) angustum sp. nov. Holotype (x. 2) (page 224).
From the Papyriaspis Stage at the road crossing of V Creek, between Undilla
and Thorntonia.
Figs. 22-28. Anomocare confertum sp. nov. (All figures x. 2) (page 222).

22-24. Cranidia. Fig. 22 is the holotype. On 23 and 24 portions of the pre-glabellar
field are preserved.

25-27. Portions of pygidia.

28. A free cheek.

From the Anomocare Stage five miles east of the road crossing of Harris Creek,
between Camooweal and Old Thorntonia,

All specimens figured on this plate are in the University of Queensland Collection.

280 MEMOIRS OF THE QUEENSLAND MUSEUM.

PLATE XXIV.
(All figures natural size unless otherwise stated.)

Figs. 1-3. Dinesus ida Etheridge fil. (page 229).
1. A complete individual, slightly flattened.

2. A specimen complete except for the posterior portion.
3a. An exterior (a wax squeeze of an external mould) and 3b an internal mould of the

‘same specimen. ;
From the Dinesus Stage of Beetle Creek, 12 miles north west of Mount Isa.

Figs. 4-6. Idamea venusta sp. nov. (page 232).
4 and 5. Cranidia (x. 4). Fig. 4 is the holotype.
6. A pygidium (x. 4).
From the Rhodonaspis Stage fourteen miles south of Glenormiston homestead.

Fig. 7. Idamea superstes sp. nov. Holotype (x. 4) (page 233).
A cranidium from the Elathriella Stage at the base of a mill, immediately west of
Tyson’s Bore, Glenormiston.

Figs. 8-10. Corynexochus plumula sp. nov. (page 234).
8. Holotype, a cranidium (x. 8).
9. A pygidium (x. 10).
10. A pygidium (x. 6).
From the Hugonocare Stage, four and a half miles north of the 20 Mile Bore,
Glenormiston.

Figs. 11, 12. Ellipsocephalus (?) sp. (page 231).
Two very small cranidia. Fig. 11 x. 10; fig. 12 x. 20.
From the Lugonocare Stage five miles north of the 20 Mile Bore, Glenormiston.

Figs. 13-20. Amphoton spinigerum sp. nov. (page 236).
13-15. Cranidia. Figs. 14a and 14b are two views of the holotype.
16. A pygidium.

17, 18. Free cheeks.
19. Portion of the combined cranidium and free cheek.

20. A thoracic segment, showing the long, axial spine.
From the Amphoton Stage of ‘* Split Rock”? on Waroona Creek, at the main
road crossing between Camooweal and Mount Isa.

Figs. 21-23. Amphoton serotinum sp. nov. (page 238).
2la, b. Two figures of the holotype, a cranidium. 21b is natural size (a photograph of
a Wax squeeze of the external mould). 21a is slightly enlarged and is a photograph
of the internal mould.
22. A pygidium (x. 2).
23. Two cranidia (x. 2).
From the Papyriaspis Stage, a mile and a half west of the road crossing of V
Creek, between Undilla and Thorntonia,

The original of figures 4, 5 and 6 are in the collections of the Geological Survey of Queensland.
All other specimens are in the University of Queensland Collection.

THE CAMBRIAN FAUNAS OF NORTH-EASTERN AUSTRALIA. 281

PLATE XXV.

(All figures natural size unless otherwise stated.)

Figs. 1-3. Notasaphus modicus sp. nov. (page 243).

1. Holotype.

2. A cranidium.

3. A pygidium.

From the Dinesus Stage of Beetle Creek, 12 miles north west of Mount Isa.
Figs. 4, 5. Notasaphus fergusoni Gregory, the genotype of Notasaphus, shown for comparison.
(page 241).
4. A cranidium.
5. The lectotype, a pygidium.

From the Dinesus Stage of Knowsley in Victoria.

Figs. 6 and 7a. Charchaqia erugata sp. nov. Holotype (page 239).

Fig. 6 is magnified by four diameters and fig. 7a by three. On the left of 7a a free cheek
and a pygidium can be seen.

From the Fugonocare Stage, four and a half miles north of the 20 Mile Bore
Glenormiston.
Fig. 7b. Eugonocare tessellatum sp. nov. Holotype (x. 3), (page 226).

From the Hugonocare Stage about four and a half miles north of the 20 Mile Bore,
Glenormiston.

Figs. 8a, b. Proceratopyge nectans sp. nov. (page 249).
8a. Holotype, a cranidium (x. 2).
8b. A pygidium (x. 2).

From the Glyptagnostus Stage, 16 miles south of Glenormiston homestead.

Figs. 9-11. Proceratopyge rutellum sp. nov. (page 250).

9. A cranidium, the holotype (x. 2).
10. Another cranidium.
ll. A pygidium (x. 2).

From the Hugonocare Stage. Figs. 10 and\11 are from 44 miles north of the 20
Mile Bore, Glenormiston. Fig. 9 is from half a mile further north.

Fig. 12. Proceratopyge lata sp. nov. Holotype (x. 2). (page 248).

From the Eugonocare Stage five miles north of the 20 Mile Bore, Glenormiston.

Fig. 13. Pygidium belonging to either Proceratopyge lata or P. rutellum (x. 2).

From the same locality and horizon as figs. 9 and 12.

Fig..14. Proceratopyge polita sp. nov. Holotype (x. 3), (page 251).

From the Rhodonaspis Stage, at the western base of Polly’s Lookout, Glenor-
miston.

Fig. 15. Oryctocephalus discus sp. nov. Holotype, a cranidium (x. 2), (page 252).

From the Dinesus Stage of Beetle Creek, 12 miles north west of Mount Isa.

282 MEMOIRS OF THE QUEENSLAND MUSEUM.

Fig. 16. Elathriella plebeta sp. nov. <A cranidium (x. 2). (page 207).

From the Hlathriella Stage at the base of a hill immediately west of Tyson’s Bore,
, Glenormiston.

Figs. 17, 18. Homagnostus cf. obesus (Belt). (page 261).

17. A cranidium doubtfully associated with the species. (x. 3).
18. A pygidium of the typical Homagnostus form (x. 3).
From the Rhodonaspis Stage at the western base of Polly’s Lookout, Glenor-
miston. ;

Fig. 19. Goniagnostus scarabeus sp. nov. Holotype (x. 3), (page 260).

From the Papyriaspis Stage at the road crossing of V Creek, between Undilla
and Thorntonia

Fig. 20. Goniagnostus cf. nathorsti (Brégger) (x. 3), (page 259).

From the same stage as fig. 19, but from a mile and a half further west.

Figs. 21-23. Goniagnostus purus sp. nov. (page 258).

21. A cephalon (x. 6) showing traces of pleural furrows near the margin.
22. A cranidium with smooth pleura, the usual type (x. 4).
23. A pygidium (x. 4), the holotype.

From the Agnostus seminula Stage, eleven and a half miles north of Yelvertoft
Dip, on the road to the Paradice Goldfield.
Fig. 24. Agnostus seminula sp. nov. Holotype (x. 3), (page 254).
From the same locality and stage as figs 21-23.

Figs. 25-26. Hypagnostus clipeus sp. nov. (page 263).
25. Holotype (x. 4).
26. Paratype (x. 3).

From the Papyriaspis Stage at the road crossing of V Creek, between Undilla
and Thorntonia.
Fig, 27. Ceratagnostus magister sp. nov. Holotype. (x. 3), (page 256)
From the same stage and locality as figs. 25-26.

Figs. 28, 29. Phalacroma cf. nudum (Beyrich).

(x. 4), (page 262).
28. A pygidium.

29. A crushed and imperfect specimen with head, thorax and tail preserved.

From the Papyriaspis Stage a mile and a half west of the road crossing of V
Creek, between Undilla and Thorntonia.

Specimens 4 and 5 in the collection of the Geological Survey of Victoria. All other specimens
in the University of Queensland Collection.

MEMOIRS OF THE QU

INSLAND MUSEUM, Vou. XI, Prare XIX,

F, IF, Whitehouse

MEMOIR Vou. XI, Poatve XX.

FEF. OW, Whitehouse.

MEMOIRS OF THE QUEENSLAND MUSEUM, Vow. XI, Prarnr XXII.

fF, IF, Whitehouse.

MEMOIRS OF THE GUELNSLAND MU.

SEUM, Vou. XI, Prare XXIL.

Ir, i, Whitehouse,

Pirate XXIII.

<I
< 7)

VOL.

‘UM,

vi

MUSE

THE QUEENSLAND

OF

}MOTRS

ME

whan dae

hesstee
He

Whitehouse,

I

iP

MEMOIRS OF THE QUEENSLAND MUSEUM, Vou. XI, PLrate XXIV.

Ir, W. Whitehouse,

MEMOIRS OF THE QUEENSLAND MUSEUM, Vou. XI, PLrare XXV,

t

F, W. Whitehouse.

A CENTRAL QUEENSLAND WOMBAT.

‘By H. A. Lonoman, DIREcTOR.

(Plate X XVI.)

In July, 1937, a single specimen of a large male wombat was collected for the
Queensland Museum by Messrs. Charles and H. Greensill Barnard at Epping Forest
Station, 75 miles west of Clermont. Special collecting trips were undertaken by
these naturalists, who had heard of the existence of large wombat-like animals in
that district.

Although these marsupials were widely distributed in Queensland in the
Pleistocene and two present-day species were known to occur sparingly in southern —
parts of the State, it was somewhat surprising to have definite evidence of living
wombats in a locality in central Queensland. This extends their range by over 400

miles.
This: Epping Forest wombat is obviously of the Jlatifrons-gillespiei type,

generically designated as Lasiorhinus. It is here described as a new subspecies—
Lasiorhinus latifrons barnardt.

Description.—The pelt, which was received with a complete skull and limb
bones, is slightly over four feet in maximum length (Reg. No. 6239). The texture
of the fur is somewhat silky as in latifrons and gillespiev. The rhinarium is completely
clothed with short, close hairs. The ears are of the elongated type (incomplete in
this pelt), being over 60 mm. in length.

General dorsal colour brown, but mottled with grey, interspersed here and
there with entirely black hairs: Basal portion of dorsal hairs dark brown. Rhinarium
hairs brown. Tufts of white hairs around the outside of the ears. Outside of ears
well haired, but inside sparsely clothed. Ventral surface dirty grey, gradually
merging into dorsal colour. Basal portion of ventral hairs dark brown. Feet dark
brown. Tail scantily clothed with short dark hairs. Long black hairs are prominent
on either side of the sacral region, in the median area of which the hairs are very
thick and tufted. Average length of body hairs 25 mm.

Crania.—The skull associated with the pelt has a maximum length of 191
mm..and a basal length of 172 mm. The breadth across the zygomatic arches is
156. The nasals have a length of 77 and a combined breadth at maximum of 67. The
length of the palate is 113. The upper molar series, which is well worn, has a length
of 57-5 mm. The skull is evidently that of a fully mature wombat. The post-orbital
processes, typical of Lasiorhinus, are very rugose and this characteristic also applies

284 MEMOIRS OF THE QUEENSLAND MUSEUM.

to the region over the orbits, the lachrymal, and the maxillary area adjoining the
nasals. There is a slight anterior projection of the frontals in the median line between
the nasals. A prominent projecting knob or nasal spine (bifid) is present on the
premaxillaries. The lachrymal has a prominent tubercle. The parietal platform
is broad. The anterior nares are widely open at their lateral borders, as in typical
latifrons and the gillespiet crania. The anterior edge of each premaxillary is more
or less vertical in lateral aspect where it joins the nasals, and there is no anterior
projection of the superior border as in mitchelli. The relatively wide nasals form
practically the whole of the narial roof.

The lower contours of the mandible are markedly convex, as in typical
latifrons. The length of the symphysis with incisors (84 mm.), exceeds one half of
the maximum length of the mandible (154 mm.). The length of the mandibular
molar series is 55 mm.

In comparison with the records given by Spencer and Kershaw (1910), the
actual dimensions of the skull and the length of the molar series are exceptionally
large for an existing wombat.

A second incomplete large skull (No. 6240) was also found near the burrows.

In October, 1937, two additional crania were obtained in the same locality by
Messrs. Barnard (Reg. No. 6283-6284). One of these exhibits such a condition of
asymmetry that opportunity is taken to illustrate it (Plate XXVI). Evidently
development of the left side of the skull was stunted in early life, probably due to
hemiplegia. This is markedly shown by the dimensions of the molar series ; that of
the right side is 55 mm., whereas the left series only attains 48. The molar series is
well worn and the individual evidently lived to maturity. - Pathological conditions
such as these are apparently very rare among marsupials.

A ffinities.—In his comprehensive description of “A New Species of Hairy-
nosed Wombat,”’ the late C. W. De Vis! recognised the affinities of his Phascolomys
gillespiei from the Moonie River, south-western Queensland, with the South Australian
latifrons, but considered that certain cranial characteristics entitled it to specific
distinction. Baldwin Spencer and Kershaw, however, in their review of “‘ The
Existing Species of the Genus Phascolomys”’ (p. 65) refer to the wide range of
latifrons and express considerable doubt as to whether De Vis’ species is specifically
distinct from P. latifrons. Iredale and Troughton in their Check List of Australian
Mammals (1934) give a generic name “* Wombatula ”’ for De Vis’ species, but this
was done without autoptic examination and no diagnosis was recorded.*

1 De Vis, C. wW. Annals of the Queensland Museum, No. 5, 1900.
2 Baldwin Spencer & J. A. Kershaw. Mem, Nat. Mus., Melb. No. 3, 1910.
3 Tredale & Troughton, Memoir VI, Austr. Mus. Sydney, 1934.

A CENTRAL QUEENSLAND WOMBAT. 285

In his excellent descriptions of the Common and of the Hairy-nosed Wombats,
Wood Jones adopts Gray’s genus Lasiorhinus (1863) for the characteristic South
Australian latifrons and notes its wide range. The significant structural differences
between latifrons and mitchelli warrant generic distinction. ’

It is of interest to note that whilst the common naked-nosed wombat was
first recorded by Owen from fossil remains and subsequently found as an existing
species, the living latifrons type has now, conversely, a wider range through records
of fossils. Mandibular fragments from the Wellington Valley, New South Wales,
were recorded by Owen as P. latifrons.

A fairly complete fossil cranium of the latifrons type was received in 1917
from Clermont through Mr. Power, Warden of the Mines Department, who thought
that it came from Cotherstone, some 35 miles east of Clermont. In this cranium the
length of the molar series is 53 mm. and the width of the hind lobe of the second
molar is 10-5. This cranium, which is very definitely fossilised, is registered as
F. 1279 in our series.

Another fossil skull with mandible attached and with several associated limb-
bones, vertebrae and pelvic fragments was found during the excavation of a well at
Glengallen Plains, Darling Downs, in 1936. This skull is registered as L. latifrons
(F. 2526).

Obviously the correct status of existing species of wombats should be con-
sidered in association with some of the fossil species, such as P. krefft1 Owen (1872),
P. medius Owen (1872) and thomsoni Owen (1872). In the paper in which he
described P. angustidens, De Vis considered that P. mitchelli was really distinct from
the existing species.°®

L. Glauert’s P. hacketti from Western Australia® is allied to P. mitchell.

In his Catalogue of Fossil Mammals (1887)? R. Lydekker treated Owen’s
mitchelli and platyrhinus as distinct species, but in his Handbook to the Marsupialia
(1896)® he placed ‘platyrhinus as a synonym (p. 125).

Owing to the paucity of significant material, the writer has diffidence in
attempting a comprehensive revision of the several names that have been given to
fossil fragments of wombats. It is evident that these marsupials had a far wider
range in the past and that. they were decidedly variable. In records of the total
length of the molar series in adult crania variations of several millimetres were given
by Spencer and Kershaw for both mitchelli and latifrons.

4 Wood Jones, The Mammals of South Australia, Pt. IT, 1924.

5 De Vis, C. W. Pr. Linn. Soc. N.S.W., VI, 1891.

6 Glauert, L. Rec. West. Aus. Mus. Vol. I, 1910.

? Lydekker, R. Catal. Foss. Mamm. Brit. Mus. Pt. V, 1887.

8 Lydekker, R. Handbook to the Marsupialia, Lloyd’s Nat. Hist, 1896.

286 MEMOIRS OF THE QUEENSLAND MUSEUM.

In the total length of their molar series the new Epping Forest crania are
distinctly larger than those of latifrons and even of mitchelli, but they are much
smaller than P. medius. Their general facies, however, is so obviously with L.
latifrons that until additional material is forthcoming it seems best to regard these
central Queensland wombats only as a new sub-species.

LASIORHINUS LATIFRONS BARNARDI.

This new sub-species as described above is based mainly on its larger size.
As in typical latifrons the frontals project slightly between the nasals, whereas in
gillespiet the nasals project posteriorly into the frontal region. As there is considerable
variation in this region it is doubtful whether much significance should be attached to
minor deviations of sutures.

The infraorbital foramen is circular or sub-oval, as in gillespiet and typical
latifrons, and not laterally compressed as in mitchelli.

Messrs. Barnard record that the total length of the animal as measured when
shot, was 3 feet 4 inches, with a girth of 2 feet 8 inches. The length of the diminutive
tail was 24 inches. The eyes were brown in colour.

In view of the variation in colour recorded for species of wombats the
differences between gillespie: and latifrons are not very significant.

The two skins, male and female, received by De Vis from Bullamon Station,
on the Moonie River, show that the gillespiei form was very variable. The cranial
distinctions noted by De Vis such as the development of the lachrymal protuberance
and of the nasal spine are not very distinctive characters in this group. Although
Owen laid some stress on the direction of the sutures between the frontals and nasals,
there is evidently much individual variation in this character. De Vis’ species
is more reasonably designated to-day as a sub-species—Lasiorhinus latifrons gillespici.

General notes—Mr. Charles Barnard reports that there wére many burrows
in the district, but very few tracks of the animals were seen. The Station hands said
that the wombats had never been seen far from the burrows, which are in sandy
ridges in fairly-well timbered country, but carrying very poor grass. They said
that the wombats came out to lie on top of the burrows on very hot days and that
they also came out on rainy days. Mr. Barnard notes: ‘We could see no trace of
their feeding out anywhere. They made light pads running from one burrow to
another. The burrows were generally not more than one hundred yards apart and
were mostly in groups of three or four.”

Only three animals were seen, one of which was shot. As suggested by Messrs.
Barnard, it is probable that these wombats were much more numerous in earlier

MEMOIRS OF THE QUEENSLAND MUSEUM, Vou. XI, Plate XXYI.

H, A. Longman,

A CENTRAL QUEENSLAND WOMBAT. 287

years, but successive periods of drought have brought them to the verge of extinction.
It was considered that these wombats obtained most of the food underground.
Although they may be partly rhizophagous, the examination of stomach contents
of the specimen shot (thoughtfully preserved by Messrs. Barnard) shows that this
particular animal had been feeding on ‘“‘ the above ground parts of plants.” We
are indebted to Dr. D. A. Herbert for examination of this material, which consists
largely of fragments of stems and leaves, including awns of the Comet River Grass,
Perotis rara.

It is of interest here to record that Mr. Griffiths of the Monkland School,
Gympie, informed me that about twenty years ago (from August, 1937) he saw
wombats “‘ distinctly ’’ in the Tambo district.

The Common Wombat, P. mitchelli, extends into the southern portion of
Queensland in the Stanthorpe district, from whence several specimens have been
obtained in recent years. Iredale and Troughton (1934, p. 34), following Mathews
and Iredale (1912, p. 14),° designate this species as Perry’s hirsutum, but it seems
more probable that the wombat illustrated by Perry and named Opossum hirsutum
came from a Bass Strait island and should be kept in the synonymy of uwrsinus.

® Mathews and Iredale, Vict. Naturalist, Vol. X XIX, 1912.

A BICEPHALOUS SNAKE.

By H. A. Loncman, DIRECTOR.

(Plate XXVIL.)

In September, 1938, we received a specimen of a bicephalous snake from Mr.
J. P. Forbes, of Mullumbimby, New South Wales, with the information that it was
obtained about sixteen years ago. .

As this is the first local record of a bicephalous snake known to the writer,
it has been thought worthy of brief notice.

The snake is obviously a young specimen of the Common Black Snake
Pseudechis porphyriacus. As it is only ten inches (250 mm.) in maximum length,
it is doubtful whether it lived to feed, as it is approximately the length and size of
Black Snakes at birth. Almost immediately behind the head the two anterior regions
are united to a single body, and the scales are numerous and irregular in this region.
A short distance posteriorly, however, the normal series of seventeen body scales may
be counted. There are no special peculiarities to be noted in either of the two heads
and the lepidosis of the posterior part of the body is normal. As the specimen has
been preserved for so many years, it seems undesirable to make dissections.

In 1925 J. Strohl gave records of sixty-eight instances of this abnormality.1
J. Miguel Ladeiro in 1935 illustrated with a radiograph a specimen of Tropidonotus
natrix in the Coimbra Museum.?

11925. J. Strohl, Ann. Sci. Nat. Zool., 8, pp. 105-132.
21935, J. Miguel Ladeiro, Mem. Estudos Mus. Zool. Univ. Coimbra, Ser. IV, No. 2.

MEMOIRS OF THE QUEENSLAND MUSEUM , Vou. XI, Purate XXVIII,

A BICEPHALOUS SNAKE,

‘

ABORIGINAL MIDDENS OF POINT CARTWRIGHT
| DISTRICT.

By G. K. Jackson.

(Figures 1-7.)

On many parts of the Queensland coast, old aboriginal camping grounds
with the usual supply of shells and stone chips occur at intervals at places suitable
for such a purpose, and regulated, of course, by the supply of food and water in the
locality.

Between Point Cartwright at the mouth of the Mooloolah River and Lake
Garrimundi just to the north of Caloundra there occurs a stretch of uninhabited
coast, consisting of a line of sand dunes backed by a large plain or boronia swamp,
stretching west to the Mooloolah River. In the folds of these sandhills, a number of
fine kitchen middens are located. During the years 1932 to 1934, several of these
were in excellent condition, appearing as though they had only recently been
abandoned. However, at the beginning of the year 1938, they were found to be
suffering from the effects of heavy seas which, during the winter months, had swept
far up into the sandhills.

These old camping grounds lie within the rich territory formerly inhabited by
the Kabi tribe, and formed bases periodically visited during their nomadic wanderings.

The camps occur only among those sandhills which are devoid of trees, one
reason for this being that they were less likely to be disturbed there by a hostile party,
while sleeping off the effects of a feast of rich coastal food. The enemy naturally
would find it hard to creep upon a camp situated in the centre of an extensive white
sand dune. Another point in favour of these positions is the absence of mosquitoes
and sandflies in the summer months, which would not be the case in amongst the low
scrubby vegetation of the timbered sandhills.

Fresh water is abundant even when the swamps are suffering from a spell of
drought, and the blacks had only to dig a foot or more into the sand at the base of
the depressions on the western side of the dunes to obtain a plentiful supply.

In an examination of one of these middens the most striking feature is the
quantity of “‘ eugarie”’ shells (Donax deltoides), Q.E. 3302, spread about the sand.

290 MEMOIRS OF THE QUEENSLAND MUSEUM.

This shell-fish formed a large item on the menu of the coastal blacks, and was
obtained in the surf at low water by digging with the foot. A curious point to observe
is the fact that the shells are more or less of a uniform size; neither the very small
nor the very large ones appear to be present. Possibly the larger ones lacked the -
flavour of those of medium size, or else overfishing had prevented full development.
This is rather an interesting point in view of the fact that many of the large middens
on. Stradbroke Island are filled with shells of maximum size.

Being a bivalve the “‘ eugarie ’ was opened by being subjected to heat alongside
a small fire. This accounts for the fact that theshells are nearly all in perfect
condition, only showing the results of weathering during the time they have been
lying around the sandhills.

- Apart from utilising the flesh for food, the Aborigines used the shells of these
animals as planes and scrapers for smoothing boomerangs and spears, and for reducing
the latter to a sharp point, just as the more inland blacks used the fresh water mussel
shell. |

Passing from the shell deposits, we next turn to the stones and rocks which have
been carried to this area for the purpose of manufacturing various implements. The
largest are the grinding and pounding stones which are perhaps the most local in
origin. They are made from the same rocks which form most of Point Cartwright
and Moffats Head to the North and South of the area respectively, that is various
grades of sandstone ranging from a very fine-grained variety up to a fairly coarse

grit. They were apparently carried along the miles of beach to their present location
‘by the unfortunate women.

The variety of stones other than these is very extensive. The commonest
is quartzite of various kinds, this rock being excellent for the making of knives,
scrapers, etc. We find also trachyte, granite, hornblende-porphyrite, flint, quartz,
opal, agate and petrified wood.

Amongst other implements used extensively in this district by the Aborigines
were those associated with the preparation of fern root for food. The fern, Blechnum
serrulatum, known to the blacks as “‘ bungwall,” is very common in all coastal swamps.
It possesses a thick rhizome growing perpendicularly in the soil. This was dug up
by the women with the aid of yam sticks, or by hand, after which it was cleaned and
roasted in hot ashes. Then came the chopping-up process, with the aid of a stone
implement, specially suited for this purpose. The root was placed on a log or piece
of wood, and the chopper held in one hand. After having been reduced to small
pieces, it was either eaten thus or else further ground and pounded on grinding stones,

then mixed with a little water and made into small cakes. It was apparently a very
nourishing food. |

ABORIGINAL MIDDENS OF POINT CARTWRIGHT DISTRICT. 291

The chopper used for this purpose is an implement of quartzite or some other
suitable rock, flaked away on one margin after the style of a chisel, and possessing
secondary chipping along its edge. Figs. 1, 2 and 3.

Mr. N. B. Tindale, of the South Australian Museum, informs me that these
implements are similar to a characteristic type found in Southern Australia, and
belong to the same type as the “ Sumatra implements ” of the Malay Peninsula.
‘They have not previously been recorded from Queensland.

39

A similar implement collected by myself at the Piccabeen Lakes, Northern
New South Wales, in 1930, gives an idea of their coastal distribution both north
and south of the Moreton Bay Area.

This specimen is of red quartzite and the edge has been formed by the removal
of two large flakes, and shows distinct polishing along either face. The back in this
case has been flaked to form the usual rounded hand grip.

Several miles to the westward of these sandhills lies the Mooloolah River,
and I am indebted to Mr. W. Potts, of Mooloolaba, an old pioneer of the district,
for showing me some further native middens along portions of its bank. They were
composed of piles of oyster shells, sometimes rising in the bank to a height of seven
feet above the river.

Mr. Potts informed me that the blacks used to dive for oysters, either from the
bank or from their canoes: native name “ Kumbur” (‘‘ The Australian Race,’’
E. M. Curr, Vol. III, p. 143), made from the bark of swamp mahogany trees. The
oysters, of course, grew on the bed of the river. One can easily pick the parts where
they are most abundant, as nearby are numerous shells. In barren portions of the
river, the banks are likewise barren of shells. |

After a spell of coastal food the Aborigines, probably at the suggestion of
some old man, would move camp and journey up into the rain-forests of Buderim
Mountain or some other locality, and so they travelled from place to place, as the
instinct or the call for change of food prompted them.

_ An interesting feature of this area, which has disappeared through denudation
within the past few ‘years, was the presence of rock carvings or “ peckings ”’ on the
shore-platforms of Pt. Cartwright and Moffats and Wickham Heads.

In the year 1932 quite a number were still in existence but the alternate wetting
by sea water and sun drying caused the sandstone to flake away, eliminating all
trace of these relics. They must at one time have been very extensive. These rock
carvings showed emu and kangaroo footprints, and also those of a variety of smaller
birds.’ Dingo footprints were a fairly-common feature. I was lucky enough to see
one quite good “ pecking ” of a human footprint. Other markings of an indefinite
character occurred, but some of these were very vague.

292 MEMOIRS OF THE QUEENSLAND MUSEUM.

ip ee
OF 3342

CKT

Figs. 1-7.—Implements from Point Cartwright Middens. r

ABORIGINAL MIDDENS OF POINT CARTWRIGAT DISTRICT. 293

LIST OF SPECIMENS.

Fig. (1) Q.E. 3318.—A chopper for “‘ bungwall” fern. In size 91 mm. x 73 mm. x 41mm. Com.
posed of a water-worn quartzite pebble, which has been reduced to a chisel edge on one

side by flaking. It shows secondary chipping and flaking and possesses a good chopping
edge.

Fig. (2) Q.E. 3342.—Another “ bungwall”’ chopper, showing considerable use, the cutting edge
being rounded off by continuous pounding. In size 120 mm. x 67 mm. x 55mm. Com-
posed of quartzite it is very weathered on its original surface. It shows good flaking
and secondary chipping and has the same chisel edge as the former specimen.

Fig. (3) Q.E. 3297.—Chopper of quartz, which is very interesting in view of the fact that it possesses
two cutting edges on opposite margins. Both are well worn and show flaking and
secondary chipping. It is 120 mm. x 52 mm. x 60 mm. in size.

‘Q.E. 3307—Unfigured Specimen.—Another chopper made from a quartzite pebble which has been
reduced to a good edge by the striking off of one large flake. It too possesses a good
chopping edge. Size 70 mm. x 70 mm. x 30 mm.

‘Fig. (4) Q.E. 3317.—A stone knife of quartzite. Native name “ dhak-ke.’’ Size 101 mm. x 63

mm, x 24mm, This specimen is well shaped and possessses two good cutting edges, one
of which shows secondary chipping.

Fig. (5) Q.E. 3306.—This specimen is rather a fine example of a chipped back scraper 51 mm. x
29 mm. It is made of flint and shows secondary chipping. The single cutting edge has
apparently been subjected to much use.

Fig. (6) Q.E. 3306.—A scraper or knife 41 mm. x 46 mm. Made of fine-grained grey quartzite.
Shows secondary flaking and possesses two fine cutting edges.

Fig. (7) Q.E. 3306.—A small scarifier of mottled pink quartzite 43 mm. x 22mm. This specimen

shows secondary chipping and flaking, and possesses two good cutting edges. The blade
is distinctly curved to one side,

UNFIGURED SPECIMENS,

‘Q.E. 3306.—A small scraper 25 mm. x 23 mm. made of common opal. Shows secondary flaking
and possesses one cutting edge.

‘Q.E. 3306.—A side scraper showing secondary flaking 41 mm. x 33mm. Composed of fine-grained
grey quartzite, weathered on two original faces. Possesses two cutting edges.

‘Q.E. °3306.—A small spokeshave or chisel of flmt 48 mm. x 33 mm. Possesses grooved cutting

edge at broad end, formed by a large conchoidal fracture in the stone. Secondary flaking
is noticeable.

Q.E. 3306.—A small flake scraper made of fine-grained grey quartzite 30 mm. x 25 mm. Possesses
one cutting edge.

‘Q.E. 3306.—A small high-backed scraper 35 mm. x 27 mm., made of grey quartzite. Shows good
flaking and possesses one cutting edge.

Q.E. 3306.—A rather fine scraper or knife, made of quartzite 61 mm. x 45 mm. Slightly dis-

coloured by iron staining. Shows secondary chipping and flaking. It possesses three
cutting edges.

294 MEMOIRS OF THE QUEENSLAND MUSEUM.

Q.E. 3306.—A high-backed scraper 38 mm. x 32 mm. Composed of grey quartzite slightly
weathered on its outer extremities. Possesses three good cutting edges and shows |
secondary flaking.

Q.E. 3306.—Another quartzite scraper, 43 mm. x 42 mm. Shows secondary flaking and possesses.
three cutting edges.

Q.E. 3306.—A fairly well-shaped scraper of fine-grained grey quartzite, 40 mm. x 43 mm. It,
however, shows very little secondary working. Possesses one cutting edge.

Q.E. 3306.—A knife of quartzite, discoloured by oxidation, 45 mm. x 33 mm. It is rather well
shaped and possesses two cutting edges showing fair use.

Q.E. 3306.—A knife or scraper of silicious rock 58 mm. x 38 mm, Shows secondary flaking, and
possesses one good cutting edge.

Q.E. 3306.—A high backed side scraper of grey quartzite, 41 mm. x 31 mm. Shows secondary
flaking and chipping and possesses a good cutting edge.

Q.E. 3306.—A rather more irregular type of scraper 36 mm, x 41 mm. Made of grey quartzite,
slightly discoloured by oxidation. It possesses three cutting edges, two of them showing
much wear.

Q.E. 3306.—Apparently a knife which has been broken during process of manufacture, 41 mm.
x 35 mm. Composed of a grey rather coarse quartzite. Shows secondary chipping and
flaking and possesses two cutting edges.

Q.E. 3306.—A small scraper 36 mm. x 30 mm. Made of fine-grained grey quartzite. Shows rough
flaking, but has a rather unfinished appearance. Possesses three cutting edges.

Q.E. 3306.—A small scraper 35 mm. x 20 mm. with a chisel edge at one end. It is composed of a
fine-grained grey quartzite and shows secondary flaking. It possesses two cutting edges.

Q.E. 3306.—A large implement of quartzite stained by iron, 75 mm. x 57 mm. Possesses only
primary flaking, but one cutting edge shows considerable use. It was probably used
as a chopper for cutting up “‘ bungwall ” fern root.

Q.E. 3306.—A simple knife of quartzite 75 mm. x 42 mm. Made from a single flake struck from
a water worn rock. Its single cutting edge shows much wear, and it too was probably
used for cutting “ bungwall ” root.

Q.E. 3306.—A small scraper of common opal 32 mm. x 17 mm. It is rather a crude implement,
and has one poor cutting edge.

Q.E. 3306.—A scraper of quartzite 56 mm. x 32mm. Has one rather rough cutting edge. Only a
little rough flaking is visible.

Q.E. 3306.—A knife formed of quartzite, struck from a water-worn stone, 61 mm. x 24mm. Has
two cutting edges, and shows a little secondary chipping.

Q.E. 3306.—A scraper or knife of quartzite also stained brown by oxidation, 63 mm. x 44 mm.
Shows secondary flaking and possesses two good cutting edges.

Q.E 3299,—An upper grinder of trachyte. In size it is 104 mm. x 104 mm. x 61 mm. and possesses
much secondary flaking. It was used in conjunction with the large sandstone grinders
for milling ‘ bungwall ”’ root.

ABORIGINAL MIDDENS OF POINT CARTWRIGHT DISTRICT. 295

Q.E. 3300.—This implement is composed of trachyte with a badly-weathered surface, indicating
extreme age. It is apparently a stone axe in the making and shows good flaking and
secondary chipping. Size 105 mm. x 87 mm, x 46 mm. It was evidently discarded for
some reason or other before completion.

Q.E. 3284.—A lower grinding stone, composed of a fairly fine-grained grey grit. Size 94” x 8)” x
3}”. It possesses fairly deep grinding hollows on both sides.

Q.E. 3282.—A lower grinding stone composed of a fine uniform-grained grey sandstone. Size
8}” x 8” x 32”. This also possesses fairly deep ground-out hollows on both sides.

Q.E. 3286.—A triangular-shaped iower grinding stone composed of a fairly fine-grained grey
sandstone. Size 9” x 84” x 24”. It has had very little use, and the hollow on one side
is quite shallow.

Q.E. 3285.—Portion of a grinding stone composed of a fine uniform-grained grey sandstone. Size
8” x 9” x 3}”. This one has apparently been used for the purpose of sharpening axes
and other tools, as the groove is wider and larger in extent than that observed in other
specimens. It is grooved on both sides.

Q.E. 3287.—A lower grinding stone composed of a very fine-grained grey sandstone. Size
8” x 52” x 3}”. The groove in this one is fairly shallow and appears on one side of the
stone only.

Q.E. 3294.—Fragment of a lower grinding stone composed of a coarse uneven grained sandstone.
Size 6” x 5}” x 2}”. It has presumably been broken early in its use as there is little sign
of wear on the grinding surface.

Q.E. 3292.—Half of a lower grinding stone composed of a coarse sandstone. Size 62” x 6}” x 24”.
It possesses a good worn groove on one side only. This specimen is badly weathered,
and has probably been lying there for a considerable period.

Q.E. 3292.—A small lower grinding stone, composed of a fairly coarse-grained sandstone, con-
taining limonite, which has at some time been deposited into hollows in the rock. Size
64” x 53” x 3)”. It is ground out on one side only.

A NEW FISH, OF THE GENUS PRIONOBUTIS,
FROM NORTHERN AUSTRALIA.

By GILBERT P. WHITLEY, F.R.Z.S.,
ICHTHYOLOGIST OF THE AUSTRALIAN Musgeum, SYDNEY.

(By permission of the Trustees of The Australian Museum.)

Family ELEOTRIDAE.
Genus PRIONOBUTIS Bleeker, 1874.

Prionobutis Bleeker, Arch. Neerl. Sci. Nat. ix, 1874, pp. 295 & 305. Orthotype,
Eleotris dasyrhynchus Gunther, Ann. Mag. Nat. Hist. (4) i, April 1, 1868, p. 265, pl. xii,
fig. B, from Sarawak. Jd. Bleeker, Arch. Neerl. Sci. Nat. x, 1875, p. 103 and Versl.
Akad. Amsterd. (2) xi, 1877, p. 73 (fide Weber & de Beaufort, 1911). Jd. Herre, Gobies .
Philip. & China Sea, 1927, p. 52. Jd. Koumans, Zool. Mededeelingen xx, 1937, p. 21. Id.
Fowler, Proc. Acad. Nat. Sci. Philad. Ixxxix, 1937, p. 248.

A genus of small gudgeons found in marine, brackish, and fresh water in
the Indo-Australian and southern Asiatic regions, whence a few species have
been described. A characteristic feature is the presence of rows of spines on
the top of the head. Two subgenera may be distinguished :

A. Eye small, about 8 in head. Mouth reach- ;
ing backward to behind level of eyes .. Prionobutis

(type, dasyrhynchus).
AA. Eye large, less than 4 in head. Mouth

reaching to below anterior half of eye .. Themislocles, subg. nov.
(type, wardi, nov.).

PRIONOBUTIS (THEMISTOCLES) WARDI, subg. et sp. nov.
(Figure 1.)
D. vifi, 8; A.8; P.17; V.5; C. 12 et lat. brev. Sc. 26 to hypural. Tr. 9,
below first dorsal, to 5 on caudal peduncle. About 9 predorsal scales.
Head (13 mm.) 3-15, depth of body (11) 3-7, length of pectoral (12-5)

nearly 3-3, depth of caudal peduncle (5) 8-2 in standard length (41 mm.). Eye
(4 mm.) 3-25, interorbital (3) and snout (3) 4-3 in head.

A NEW FISH FROM NORTHERN AUSTRALIA. 297

*

Head blunt, longer than broad or deep. Lower jaw slightly longer than
upper; mouth extending to below anterior half of the large eye. Fine teeth
in jaws, outer ones slightly enlarged, but no canines. Tongue rounded. Eyes
dorso-lateral, armed above and behind by a row of about two dozen small
spines. Some similar spines above the nostrils and on each side of the
premaxillary processes which reach back almost to level of eyes. A few more
spines scattered on nape and along the sides of the distinct nuchal groove.
Preoperculum entire, unarmed, its limb with a few open pores. Operculum
entire, ending in a flap with a blunt: spine. Cheeks and opercles scaly. Top
of head scaly from a little behind the eyes. Interorbital flat and sunken. A

Fig. 1—Prionobulis (Themistocles) wardi. (G. P. Whitley.)

row of fleshy papillae along each side of lower surface of head. No cuticular
papillae elsewhere (but a paratype has some short rows of minute papillae on
cheeks). A small mental flap at the symphysis of the lower jaw. Gill-
membranes united across the tapering isthmus. Body compressed, deep below
the first dorsal fin and tapering to a long caudal peduncle, covered with large
thin ctenoid scales which do not extend on to fins. No auxiliary scales. Vent
large. A large genital papilla.

First dorsal fin with six weak spines. Second dorsal short-based but with
long rays. Anal fin similar to second dorsal. Pectorals long, extensive, ninth
ray longest, reaching anal fin; lower rays fringe-like. Ventral fins well
separated, inserted well forward; rays weak, fringed. Caudal rounded, upper
half the longer.

Colour in alcohol, rich brown; ground-colour of fins somewhat yellower.

Head and body crossed by several diffuse bands of darker-edged scales. The

first dark area crosses the cheek, two bands extend from first dorsal to near
¥

298 MEMOIRS OF THE QUEENSLAND MUSEUM.

the belly, another band below soft dorsal breaks into two on lower half of fish
over the anal fin, two or three more dark transverse areas across caudal
peduncle. Eye bluish. Top of head yellow. Fins speckled and mottled with
dark brownish. Caudal largely dusky (often light or only partly infuscated in
paratvpes).

Described and figured from the holotype, a specimen 41 mm. in standard
length or two inches overall, the largest of a series of ten specimens.

Localities. —Off Point Charles, near Darwin, Northern Territory of
Australia; caught in tangled rope in 12 fathoms, June 1938. Holotype (Austr.
Mus. regd. no. IA. 7849).

Same locality, July 4, 1938. Paratype, TA. 7850.

Darwin Harbour, near the jetty, 8 to 10 faths., July 30, 1938: six paratypes
(IA. 7852-3), and one (IA. 7610) in burrows on mud-flat, Darwin, July 31.

‘Off the south coast of Melville Island, Northern Territory, 8 to 13 faths.
July 1938: one paratype ([A. 7854). Specimens caught in rope tangles from
H.M.A.S. “Moresby ”’ by Mr. Melbourne Ward, naturalist to the naval surveying
expedition, after whom I gratefully name the species.

Range.—Northern Australia ; Dampierian marine region, shallow water.

This new species differs from the genotype in having much larger orbits,
and maxilla reaching to below, not behind, the eyes. From the other species
(koilomatodon Blecker, 1849: caperatus Cantor, 1850, serrifrons Rutter, 1897)
it is distinguished by the form of the head and in having papillae on chin.
Pogoneleotris microps Weber, 1908 is nearer’ the small-eved Prionobutis
dasyrhynchus (Gunther, 1868) but has auxiliary scales on the body.

gS

END OF Vou. XI, MEMOIRS OF THE QUEENSLAND MUSEUM.

4
34

Davip WuHytE, Government Printer, Brisbane.

CONTENTS.

The Skull of an Australian Aboriginal found at Stradbroke Island,
Queensland—Text-figurea 1-7 amd Plates XV-AVOI ae ia

The Cambrian Faunas of North-Eastern Australia, Part 3: The
Polymerid Trilobites—with Supplement No. 1—Plates XIK-AXV
A Central Queensland Wombat—~—Flate XXVI.. * ae =n
A Bicephalous Snake—Plate AX VIE av *e ee
Aboriginal Middens of Point Cartwright District—Figures 1-7 ..

Ce

A New Fish of the Genus Prionotwis, irom Northern Australia—
Figure 1 ee ee Ld se aes ov oe een

Joseph L, Shellahear, M.D., D.8.0.

F. W. Whitehouse, Ph.D.,
H, A. Longman ..
H. A. Longman

G. K. Jackson - an

G.P. Whitley .. ..

(Thia is the final Part of Vol. XI.)

M.Sc.

PAGH.

169-178

179.282

283-287
288

289-205

296-298