Memoirs

of THE

Queensland Museum

Brisbane
June, 1976

Volume 17
Part 3

Volume 17
Part 3

Memoirs

OF THE

Queensland Museum

Published by Order of the Board

Mem. QdMus. 17(3): 341-65, pis. 43-51. [1976]

REVISION OF THE MARSUPIAL GENUS PLANIGALE
TROUGHTON (DASYURIDAE)

Michael Archer

Queensland Museum

ABSTRACT

Planigale Troughton, 1928 is revised and a new generic diagnosis provided. Five species are
recognized: P. ingrami (Thomas, 1906) (including the forms subtilissima Lonnberg, 1913, and
brunnea Troughton, 1928); P. tenuirostris Troughton, 1928; P. gilesi Aitken, 1972; P. maculata
(Gould, 1851) (formerly regarded as a species of Antechinus Macleay, 1 841 and including the form
Phascogale minutissima sinualis Thomas, 1926); and P. novaeguineae Tate and Archbold, 1941.
These species are placed in three groups: the P. maculata group, the P. ingrami group, and the P.
gilesi group. P. gilesi is regarded as the most specialized species, having completely lost the last
upper and lower premolar. Two additional forms of Planigale noted, one from northwestern
Western Australia and another described by Lundelius and Turnbull (1973), may represent
additional species.

The genus Planigale is considered to be related to Ningaui Archer, 1975 as well as to some
Antechinus-Uke dasyurids such as Phascogale Teminck, 1824. There is also some affinity with
Sminthopsis Thomas, 1888, but the common features that some species of these two genera share
may be the result of convergence in arid adaptation.

In 1928 Troughton proposed that Phascogale
subtilissima Lonnberg and P. ingrami Thomas
should be included with Planigale tenuirostris
Troughton in the genus Planigale. Subsequently,
Planigale novaeguineae Tate and Archbold and
Planigale gilesi Aitken were described. The place-
ment of maculatus Gould in Antechinus by most
modern authors (e.g. Tate 1947, Ride 1970) is
anomalous. Archer (1975) considers it to be within
Planigale.

Some non -Planigale species (in Antechinus ) exhi-
bit characters previously believed confined to
Planigale — in particular, flatheadedness (Ride
1970) — while inclusion of P. novaeguineae and P.
gilesi in Planigale indicates that not all Planigale
are minute in size. Minute dasyurids are now
known to occur in another dasyurid genus, Ningaui
Archer. These developments in classification and
taxonomy emphasize the need for revision in the
genus Planigale.

METHODS

Cranial measurements are given in Table 1 . The
method of taking measurements is shown in Figure
1. Characters of particular importance in
differentiating species of Planigale are nasal
lengths, widths, contact between premaxillae and

nasals, distance between posterior lacerate for-
amen and anterior edge of alisphenoid bulla (or
tympanic wing), transverse distance across for-
amen magnum, and skull depth in front of
alisphenoid bullae. Similarly, external measure-
ments were made of the ear (from notch to tip of
pinna), supratragus (maximum length), nose-vent
(anus), tail vent, and hind foot (less claws). Unless
otherwise noted, measurements were made on
preserved materials from which skulls had already
been removed. External measurements should not
be considered directly comparable with measure-
ments of fresh material (e.g. Lidicker and Marlow
1970).

Some small dasyurids have distinctive mor-
phological dental and cranial characteristics.
Within genera, such as Sminthopsis , these charac-
teristics are sufficiently constant to diagnose spec-
ies. In other genera, such as Planigale , these
distinctive morphological characters are less com-
mon. Only Planigale gilesi in lacking P4 is ob-
viously and consistently distinct from other species.
Recognition of other species of Planigale requires
examination of size and consideration of relative
cranial and dental proportions. A statistical sum-
mary of absolute size in species of Planigale is given
in Table 1. A summary of cranial, dental, and
external proportions, as ratios, is given in Table 2.

342

MEMOIRS OF THE QUEENSLAND MUSEUM

These ratios are as follows:

1 . Basicranial length/zygomatic width (BL/ZW):
an estimate of brachycephaly;

2. Minimum inter-orbital width/ZW (IO/ZW):
an estimate of relative mid-cranial frontal
restriction;

3. Alisphenoid bullae width (left and right
combined)/ZW (BW/ZW): an estimate of
relative width of alisphenoid bullae;

4. BW/M 1-3 : an estimate of alisphenoid bullae
width relative to M 1-3 length;

5. Bullar length (measured from posterior lac-
erate foramen to anterior edge of expanded
alisphenoid bulla/M 1 ^ 3 (BL/M 1 3 ): an es-
timate of relative length of periotic and
alisphenoid bullae (or tympanic wing)
inflations;

6. Skull depth (measured vertically immediately
anterior to alisphenoid bullae)/ZW (SD/ZW):
an estimate of relative depth of skull;

7. SD/BL: an estimate of skull depth relative to
skull length;

8. SD/(C9 - M 4 ): an estimate of skull depth
relative to length of cheek-tooth row;

9. SD/IO: an estimate of skull depth relative to
mid-cranial restriction;

10. Foramen magnum diameter (maximum
transverse)/SD (FM/SD): an estimate of re-
lative size of foramen magnum;

11. M I_3 /ZW: an estimate of width of skull
relative to M 1-3 length;

12. (C 1 M 4 ) - (M 1 M 4 )/(M^ 3 ) ( = C l

P 4 /M'- 3 in Table 2): an estimate of length of
(C 1 - P 4 ) relative to M 1-3 length which is also
an estimate of relative cheek-tooth row crowd-
ing;

13. (I 1 -M 4 )-(M 1 _ 4 )/(M 1 _ 3 )(=Ii-P 4 /M 1 3 in
Table 2): an estimate similar to 12 above, but
for lower teeth and involving lower incisors;

14. Nasal length (maximum)/BL (NL/BL): an
estimate of relative length of nasals;

15. Nasal width (maximum across both
nasals) /ZW (NW/ZW): an estimate of relative
width of nasals;

16. NW/NL: an estimate of relative length and
width of nasals;

17. Premaxillary-nasal suture/NL (PN/NL): an
estimate of relative length of premaxillary-
nasal suture;

1 8. Minimum distance between premaxillary and
maxillary vacuities/BL (VV/BL): an estimate
of relative palatal evacuation;

1 9. Dentary length (from posterior edge articular
condyle to anterior edge of Ij alveolus)/BL
(DL/BL): an estimate of relative length of
dentary;

20. Tail-vent length/head-body length (TV/
HB): an estimate of relative length of tail;

21. Hind foot length/HB (HF/HB): an estimate
of relative length of foot;

22. Length of supratragus of ear/ear height from
notch (ST/E): an estimate of supratragus
length relative to ear height.

These ratios were selected following an overall
examination of specimens which indicated that
although there were apparent differences in cranial
structure, expression of some of these was confused
by differences in relative size. As a result, ratios
were computed using absolute measurements for
each specimen. Means of these ratios were then
obtained.

Cranial, dental and external terminology is that
used by Archer (1975, 1976). Abbreviations for
specimen numbers unless otherwise indicated are
as follows: AMNH, American Museum of Natural
History; BM, British Museum (Natural History);
C, National Museum of Victoria; D, Victorian
Fisheries and Wildlife Department; JM or J,
Queensland Museum; B, Butler collection in the
Western Australian Museum; 67.8.73, example of
number in fossil collection. Western Australian
Museum; NTM, Northern Territory Museum. The
following institutions all have M as a prefix to their
mammal specimens. To distinguish them, a prefix
indicating the institution has been added to the

number as follows: AM M , Australian

Museum; SAM M , South Australian

Museum; WAM M , Western Australian

Museum.

Family DASYURIDAE
Genus Planigale Troughton

Plcmigale Troughton, 1928, p. 282.

Type Species: Planigale ingrami brunneus Troughton,
1928, by original designation. (Not P. ingrami Thomas,
1906, as cited by Tate 1947, and Laurie and Hill 1954).

Diagnosis

Dasyurids smaller than Antechinus, and differ
from these in having extremely reduced maxillary
vacuity; very small paracone on M 1-3 and small
talonids on M,^ 3 . Differ from Sminthopsis and
Antechinomys Krefft, 1867 in possessing straight
uncurled external edge on supratragus of ear;
short, broad pentadactyl hind foot; enlarged
metatarsal granules; nasals broadened posteriorly;
apparent lack of squamoso-frontal contact on
external surface of skull; posterior cingula present
M 1-3 ; reduced protocone and paracone on M 1-3 ;
reduced talonid on M^ 3 ; P 4 single-rooted or

ARCHER: REVISION OF PLANIGALE

343

absent; palatine vacuity lacking. Differ from Nin-
gaui in possessing very broad hind foot; straight
uncurled external edge of supratragus of ear;
posterior cingula on M 1 3 ; nasals markedly
broadened posteriorly; no palatine vacuities.

Description

Tail thin without brush or crest and approxi-
mately equal to, or longer than, nose-vent length.

Supratragus of ear with straight external edge.
Helix curls beneath root of supratragus. Anterior
edge of tragus bears short hairs. Up to two fold
lines for retraction of pinna. Notch on posterior
edge of pinna variably present.

Mysticial vibrissae on each side in 6-7 ill-defined
rows with 2-6 vibrissae in each row; 1-3
supra-orbital vibrissae; 6-8 genal vibrissae; 3-4
carpal vibrissae.

Hindfoot broad with 7 post -digital pads includ-
ing 3 interdigital, 1 hallucal, 1 post-hallucal, 1
anterior and 1 posterior outer metatarsal pad. All
pads with apical granules appearing transversely
striated, although striae do not normally exist as
physical ridges on surface of granules. Sole naked
except near heel. Hallux clawless.

Pelage generally lacks distinctive markings.
Abnormal variation includes spots.

Median groove pronounced and to top of
rhinarium. Groove demarcating whole of external
rim of rhinarium. Nostrils centrally situated on
each side.

Five to twelve nipples have been noted in this
study. Fleay (1967) records fifteen juveniles at-
tached to one female P. maculata (as P. ingrami ).

Nasals broadly widened posteriorly.
Premaxillary-nasal contact shorter than maxillary-
nasal contact. Lacrimal foramen on rim of orbit or
just anterior to it. Infra-orbital foramen opens onto
surface of maxilla without contact with jugal.
Postorbital process on frontal absent. Prominent
venous foramen in frontal on dorsal rim of orbit.
Contact between squamosal and frontal ap-
parently lacking on outside of skull. Variably
developed, anterior, dorso-lateral extensions
(horns) of squamosal and parietal present.
Postero-mesial edge of palatine at point of contact
with frontal in orbit extremely variable in shape.
Cranium flattening reasonably constant
intraspecifically but various interspecifically. Pre-
maxillary vacuity short, not extending posteriorly
beyond posterior edge of canine alveolus. Maxil-
lary vacuity very short not exceeding length of
premaxillary vacuity. Palatine vacuity lacking.
Postero-lateral palatal vacuity lacking or so incom-
plete as to be barely suggested. Posterior palatal

spine generally not well-developed. Pterygoid with
spinous hamular process. Alisphenoid tympanic
wing poorly developed with short periotic contact.
Periotic with low, variably enlarged tympanic
wing. Ectotympanic large with only small portion
covered by alisphenoid tympanic wing. Transverse
canal of basisphenoid very small to absent (or
indistinguishable) from anterior end of foramen
pseudovale. Variably developed entocarotid canal
adjacent to elongate narrow foramen pseudovale.
Opening of eustachian canal large. Internal jugular
canal enclosed in tube formed by basioccipital and
periotic. Tiny posterior lacerate foramina variable
in number and antero-mesial to paroccipital pro-
cess. Condylar foramina tiny to large and variable
in number (includes hypoglossal foramen). Paroc-
cipital process barely inflated as continuation of
anterior tympanic wing of periotic. Stapes col-
umnar, but generally (perhaps universally) with
very tiny stapedial depression near contact of foot
plate with columella. Depth of dentary ventral to
teeth variable. Masseteric fossa wide. Mandibular
foramen beneath M, or M 2 .

I’ largest incisor and separated from I 2 4 by
diastema. I 2-4 incisor crown length and height
variable but I 3 generally highest. C 1 tallest crown in
upper dentition. Base of C' enamel crown often far
above alveolar rim with tiny posterior cuspule and
barely distinguishable to absent buccal and lingual
cingula. P 4 , when present, largest upper premolar.
P‘ never larger than, but sometimes subequal to,
P 3 . Small posterior cuspule on P 1 4 sometimes
absent. Cingulum around P' 4 generally entire.
Paracone very reduced and close to metacone of
M 1 3 . Paracone increases in height from M 1-4 .
Basal antero-posterior length of protocone de-
creases from M 1 4 . Tiny protoconule variably
present on M 1-3 . No metaconule on M 1-4 . Prepro-
tocrista continuous with anterior cingulum, which
is complete on M 1-4 . Postprotocrista continuous
with posterior cingulum on M 1 3 which extends
buccally for approximately two-thirds posterior
length of each molar. Prefossa tilts postero-
buccally and not enclosed posteriorly except by
posterior cingulum of M 1-3 . Metacone absent on
M 4 . Paracrista of M 1 variably reduced to absent.
Paracrista of M' 4 increases in length posteriorly.
Metacrista of M 2 never shorter than metacrista of
M 1 and M 3 . Stylar cusp A not clearly distinguish-
able on any molar, possibly homologous with part
of antero-buccal cingulum of M‘. Stylar cusp B of
M 1 often completely indistinguishable and may be
totally reduced or fused with paracone. Stylar cusp
B, as defined by buccal edge of paracrista, present
on M 2 4 but largest in M 2 ~ 3 . Stylar cusp C tiny,
only variably present on posterior flank of stylar

344

MEMOIRS OF THE QUEENSLAND MUSEUM

ARCHER: REVISION OF PLANIGALE

345

cusp B or anterior flank of stylar cusp D of M 2 3 .
Stylar cusp D large on M'~ 2 , tiny on M 3 . Stylar
cusp E tiny to absent on M 1-3 . Generally, where
stylar cusps C and E tiny to absent, buccal fossae of
each molar tilt buccally and are not enclosed on
either side of stylar cusp D. Ectoflexus greatest in
M 3 and decreases anteriorly. Buccal length of M'~ 4
decreases posteriorly. Anterior width of M 1-3
increases posteriorly with M 4 anterior width never
greater than that of M 3 . 1, crown taller and longer
than I 2 . 1 2 taller and longer of subequal to I 3 . Small
posterior cingular cusp on C 1 - P 4 variably
developed or absent. Buccal and lingual cingula
developed on C, - P 3 not generally complete at
anterior edge. Development of cingula on P 4 varies
with size of tooth. C t widest antemolar tooth. P 4
varies from absent to peg-shaped to oval. When P 4
present, generally single-rooted but rarely double-
rooted. Protoconid considerably largest trigonid
cusp of M,_ 4 . Metaconid height of M^ 3 exceeds
paraconid, but smaller or subequal to paraconid of
M 4 . Difference between paraconid and metaconid
decreases posteriorly in M,_ 3 . Middle of pro-
toconid posterior to middle of buccal side of
trigonid root. Hypoconid decreases in height
posteriorly in M,_ 4 being absent or miniscule on
M 4 . Entoconids tiny to absent on M, 4 but when
present most noticeable on M 2 . Variable tiny to
absent parastylid on buccal opening of trigonid.
Variable tiny to absent metastylid on postero-
buccal edge of metaconid. Conspicuous anterior
cingulum on M,_ 4 but may be only partly de-
veloped on M,. Lingual confluence of posterior
cingulum and posterior hypocristid defines hy-
poconulid. Prominent hypoconulid notch in an-
terior cmgula of M 2 _ 4 . Well-developed carnassial
notch between protoconid-paraconid and
protoconid-metaconid but, generally, carnassial
groove very tiny to absent. Talonid wider than
trigonid of M , , narrower or subequal to trigonid of
M 2 , and progressively narrower than trigonid of

M 3 _ 4 . Paracristid M 3 longer or subequal to that of
M 2 which is longer than paracristid of M 4 which is
longer than paracristid of M a . Metacristid of M 2 _ 3
subequal and larger than paracristid of M 4 which is
larger than that of M , . Posterior hypocristid of M 2
longer than that of M, and subequal to that of M 3 .

Generic Discussion

Troughton’s (1928) concept of Planigale is based
on the relatively flat-headed forms, P. tenuirostris
and the subtilissima and brunnea forms of P.
ingrami. As a result, he considers flat-headedness
an important characteristic of the genus. Although
P. maculata and P. novaeguineae have relatively
less-flattened skulls, other dental and cranial
characters are the same as in other species of
Planigale and quite unlike any other dasyurid
genus. Thus flat-headedness is not maintained as
constant in Planigale. Flat-headedness probably
enables the individual to squeeze into narrow
crevices as suggested by Troughton (1967), Walker
(1967) and others, and permits use of the head as an
efficient wedge to raise objects, such as stones or
bark, covering insects. Many dasyurids, including
P. maculata and the subtilissima form of P. ingrami ,
studied in captivity use their heads for this purpose.
Some Antechinus are also flat-headed (Ride 1970,
Archer and Calaby in preparation), indicating that
this adaptation has developed in more than one
dasyurid lineage.

Species of Planigale generally lack an external
squamosal-frontal contact. Exceptions are pre-
sumably abnormal. For example, in a specimen of
P. gilesi (AM M7393) the left side of the skull shows
clear exclusion of a squamosal-frontal contact,
while the right squamosal may just contact the
frontal. The bones are semi-transparent rendering
identification of sutures difficult, and internal
suture relationships may be apparent on the
outside. These bones overlap rather than directly

Fig. 1: Terminology and mensuration in the skull and dentary of Planigale (based on P. maculata). a., alisphenoid;
a.c.d., articular condyle of dentary; a.ps., angular process; ar, anterior border of ascending ramus; co.f, condylar
(and/or hypoglossal) foramen; e., ectotympanic; e.f., entocarotid foramen;/., frontal; f.i.j.c., foramen for internal
jugular canal;/m., foramen magnum;//?., foramen pseudovale; /.r., foramen rotundum; /, jugal; max.v., maxillary
vacuity; m.f , mental foramen; o.c., occipital condyle; pa., palatine; p.d., posterior border of dentary; pg.f.,
postglenoid foramen; p.h., parietal horn; p.i.f , posterior lacerate foramen; p.m., mastoid part of periotic; pmx.,
premaxilla; pmx.v., premaxillary vacuity; p.p., petrosal part of periotic; p.ps., paroccipital process; ps., presphenoid;
pt., pterygoid; s.e.s., squamosal epitympanic sinus; s.f, sphenorbital fissure; sq., squamosal; t.a.r., tip of ascending
ramus; t.c., transverse canal; t.w.a., tympanic wing of alisphenoid;/!- F, positions from which cranial measurements
were made: A, basicranial length; B, maximum zygomatic width; C, outside bullar distance; D, inside bullar distance;
E, C‘-M 4 ; F, M 1 4 ; G, M'~ 3 ; H, LM 3 RM 3 ; /, interorbital width; J, maximum width of foramen magnum; K, inter-
palatal vacuity distance; L , dentary length; M, L-M 4 ; N, Mj_ 4 ; O, M,_ 3 ; P, tip of angular process to articular
condyle; Q, articular condyle to anterior border of ascending ramus; R, maximum nasal length; S, maximum nasal
width; T, minimum nasal width; U, nasal-premaxillary suture length; V , bullar length; W, line to which V is measured
and represents antero-most level of alisphenoid tympanic wing.

346

MEMOIRS OF THE QUEENSLAND MUSEUM

abut, so that while they may not contact outside the
skull, they may do so inside. A similar situation
occurs in WAM M2846, a specimen of the
subtilissima form of P. ingrami , where both sides
are in doubt.

All generic characters given as diagnostic are
separately shared with at least one other dasyurid
genus, and no single character can be considered
unequivocably diagnostic of Planigale. As with
Ningaui (Archer 1975), it is to be expected that
distribution among other dasyurids of characters
found in Planigale will in part reveal intra-familiar
relationships.

It has been suggested that Planigale is a
derivative of an Antechinus- like dasyurid (e.g.
Troughton 1967, Ride 1970). The narrow molars,
non-transversely orientated hypocristids, lack of
palatine vacuities, wide nasals, apparent lack of
squamosal-frontal contact on the outside of the
cranium, wide feet, and short ears, are characters of
Antechinus- like dasyurids such as Phascogale, but
are not characters of the genus Sminthopsis. Nin-
gaui provides a structural link between Planigale
and Sminthopsis. This link is further suggested by
S. ooldea, which exhibits mild paracone and talonid
reduction, features well-developed and characteris-
tic of Ningaui and Planigale. In general, species of
Planigale appear most similar to Ningaui , Ante-
chinus, and Phascogale, but also distantly similar to
Sminthopsis. Direct similarity to Sminthopsis is
minimal and involves characters which may be
arid-adaptations achieved independently in Smin-
thopsis and Planigale.

Development of widespread arid habitats in
Australia may have resulted in independent de-
velopment of arid-adapted characters in several
Sminthopsis and Antechinus species groups and in
other dasyurids such as Dasycercus Peters, 1875
and Dasyurides Spencer, 1896. Some presumably
arid-adapted characters are small body size, re-
latively short premolar rows, high-crowned teeth,
dolichocephaly, and well-fenestrated palates. All
except the last two are characteristic of Planigale ,
and some species such as P. tenuirostris are
dolichocephalic.

The Planigale maculata group

This group comprises two species, P. maculata
Gould and P. novaeguineae Tate and Archbold.

Planigale maculata (Gould)

(Plates 43, 44, 51C-D)

Antechinus maculatus Gould, 1851, letterpress to pi. 44.

Antechinus minutissimus Gould, 1852, letterpress to pi.

45.

Phascogale minutissima sinualis Thomas, 1926, p. 634.

Planigale maculata Archer, 1975, p. 248.

Types

Antechinus maculatus Gould, 1851

Holotype: BM53. 10.22.21, skin and skull, adult male,
collected by J. Strange. The holotype has not been
examined.

Type Locality: Gould (1851, letterpress to plate 44)
says \ . . was procured in the brushes near the river
Clarence, a little to the southward of Moreton Bay.’
Gould (1854, p. 284) says ‘Brushes of the River Clarence,
on the east coast of Australia.’ Thomas (1888, p. 293)
records the locality as ‘Clarence R., Moreton Bay . . .’.
Tate (1947, p. 131) says the type specimen came \ . . from
Clarence River, south of Moreton Bay, southern Queens-
land

Antechinus minutissimus Gould, 1852

Holotype: BM53.10.22.20, skin and BM54.10.21.5,
skull, adult male, collected by J. Strange. The holotype
has not been examined.

Type Locality: Gould (1852, letterpress to plate 45)
says ‘. . . habitat of the A. minutissimus is the districts on
the eastern coast of Australia, in the neighbourhood of
Moreton Bay.’ Gould (1854, p. 285) says ‘Hab. Brushes
of the east coasts of Australia.’ Thomas (1888, p. 293)
gives ‘Cressbrook, Moreton Bay . . .’.

Phascogale minutissima sinualis Thomas, 1926

Holotype: BM26. 3.1 1.194, skull and carcase in al-
cohol, juvenile male, obtained by Captain G. H. Wilkin’s
Expedition, 19 January 1925. The holotype has not been
examined.

Type Locality: Thomas (1926, p. 634) says ‘Hab.
Groote Eylandt, Gulf of Carpentaria.’

Material Examined

Data sheets for specimens examined available in the
library of the Queensland Museum.

Queensland: Mapoon Mission (AM M8149); Aber-
gowrie (JM833); Coen (AM M8150); Wenlock (J8170);
Mt Molloy (e.g. J 16477); Mt Garnet (J7766); Herberton
(J8244); East Funnell Creek above Sarina (AM M6983);
Sarina (AM M6840); Yeppoon Crossing, Rockhampton
(AM M8336); Rockhampton (J19668); Mt Larcom
(J7002); Calliope (J8070); Biloela (J9856); Biggenden
(J4374); Saunders Beach Rd, N. Townsville (Qd
Museum); Upper Ross Store, Townsville (JM826); Major
Creek, Woodstock (JM823); Lansdowne Stn, Woodstock
(Qd Museum); Collinsville (WAM M6203); Coorgango
Stn, nr Proserpine (J20256); Russell I, Moreton Bay
(J 10826); West Burleigh (J 13 171); Tamborine (J 16685);
Purga (e.g. J4105); Upper Brookfield, Brisbane (J 13396);
Maryborough (AM M662); Coogan Range, nr Yarraman
(J 13272); Aurukun Mission (e.g. C1483); Adel’s Grove,
Lawn Hill Ck (AM M5636).

New South Wales: 8 km W. Ballina (AM M8338);
Bunnan, nr Scone (AM M7555); Boomi Creek, Urben-
vilie (CSIRO no. CM492); Wallaby Knob, Tooloom
(CSIRO no. CM233).

ARCHER: REVISION OF PLANIGALE

347

Northern Territory: Fogg Dam, Humpty Doo (e.g.
WAM M8095); Darwin; Mataranka Homestead (AM
M7382); King River (C7820); Katherine (AM M7043);
Jim Jim Ck (W.A. Museum).

Western Australia: Drysdale River (W.A. Museum);
Barrow Island (WAM Ml 1020).

Distribution of specimens shown in Fig. 2.

Diagnosis

Large species similar in size to P. gilesi, but
differs from P. gilesi in possessing three premolars
above and below and having non-reduced stylar
cusp B. Differs from P. novaeguineae in smaller
mean size and in several cranial ratios including

higher BW/ZW and BW/M 1 3 which reflect re-
latively larger size of alisphenoid bullae. Differs
from P. ingrami in being larger with non-flattened
skull; larger M 2 stylar cusp D; relatively shorter
supratragus of ear; greatly enlarged P 4 , almost
twice P 1 crown height; non-reduced stylar cusp B;
relatively conspicuous transverse canal foramen;
paritals with relatively shorter, antero-dorsal ex-
tensions (horns); small condylar foramen; and
several cranial ratios including large mean SD/IO
and PMX-NE/NL and smaller mean BL/M 1-3 ,
FM/SD, and NW/NL. Differs from smaller
P. tenuirostris in possessing relatively conspicuous
transverse canal foramina, larger M 3 stylar cusp D,
shorter head, and in lower BL/ZW ratio.

Fig. 2: Distribution of Ptanigale maculata (solid squares represent modern specimens examined; solid triangles
represent literature records given by Marlow 1958 and Thomas 1926), and P. gilesi (solid dots represent modern
specimens examined).

348

MEMOIRS OF THE QUEENSLAND MUSEUM

Description

Tail thin. Tail length variable being shorter than
head-body length in typical form, longer in sinualis
form, and, in general, relatively shorter than in
other species. Davies (1960) lists external measure-
ments for specimens of sinualis form from Humpty
Doo, N.T. (as P. ingrami), which indicate mean
TV/HB value of 0-90. This compares closely with
series from Groote Eylandt described by Johnson
(1964) with mean TV/HB value of 0-93. Type
specimen of sinualis Thomas has TV/HB value of
090. Compared with this, typical P. maculata here
examined have mean TV/HB value of 0-82.

Supratragus of ear longer in typical form than
sinualis form, and relatively shorter than in most
other species.

Nipple number varies from 5 to 10 (to possibly
15, Fleay 1967) in typical form, and 8 to 12 in
sinualis form. Thomas (1888) records 8, Pocock
(1926) records 6, Fleay (1965) records 7-9 (as P.
ingrami , here regarded as including P. maculata, see
below), for typical form and notes (1967) specimen
with 15 young. In specimens of typical form
examined here, two had 5 nipples (J21325, J8244),
two had 6 (AM M6840, J2204), one had 7 (AM
M662), one had 8 (J3345) and three had 1 0 (J21321,
J19668, and J4374). From single locality, Mt
Molloy, NE. Qd, J21325 had 5,andJ21321 had 10.
Davies ( 1 960) records litters of 8 and 1 2 for animals
from Humpty Doo, N.T., representing sinualis
form. Aslin (1975) notes litters of 8, 10 and 11 for
wild caught Humpty Doo animals. Johnson (1964)
records 10 nipples for single specimen of sinualis
form from Groote Eylandt. Specimen from Darwin
has 10 nipples.

Pouch morphology varies in typical form,
perhaps as function of reproductive stage. Juv-
eniles held in captivity and examined live (5 August
1973) had poorly-developed, inconspicuous pouch.
Adults, including live mother of juveniles noted
above, had well-developed, deep pouch, but size
and position of opening vary. Live adults appear to
have ability to contract entrance to small, pos-
teriorly positioned, circle. One individual examined
(5 August 1973) had well-developed pouch which
opened posteriorly, with walls on sides and 10 mm
deep wall at anterior end. Some preserved adults
(e.g. J21321 from Mt Molloy), apparently lactat-
ing, have pouches widely open postero-ventrally,
presumably to accommodate larger young. Wool-
ley (1974) comments on pouch morphology in P.
maculata.

Mean IO/ZW ratio lower than in most other
species, reflecting relatively restricted interorbital
regions. Mean FM/ZW, and NW/NL ratios
lower than in most other species. Mean

(C'-M 4 )-(M' 4 )/(M 1-3 ) lower in typical than sinu-
alis form and, in general, lower than in most other
species. Mean SD/IO ratio higher in typical than
sinualis form, reflecting less-flattened condition of
typical skulls, and higher in this ratio than all other
species except P. novaeguineae.

Discussion

Although the holotypesof Antechinus maculatus,
Antechinus minutissimus and Phascogale minu-
tissimus sinualis have not been examined, topo-
typical material of maculatus and minutissimus
has been examined. Consideration of this material,
type descriptions, and descriptions given by
Thomas (1888), Tate (1947), and Johnson (1964)
leaves no doubt about the synonomy presented
here. No topotypical specimens of sinualis have
been examined. Few cranial measurements are
given by Thomas (1926) and Johnson (1964) for
sinualis but, where given, they closely correspond
to measurements for Humpty Doo Planigale
specimens.

Fleay (1965) records P. ingrami from Gin Gin,
Gunalda, Numinbah Valley and Burleigh in
SE.Qd. These records are apparently unsupported
by museum specimens of P. ingrami. Several
specimens (e.g. J 1 3526 and J 1 3 1 7 1 ) collected by Mr
Fleay from owl pellets at Burleigh, represent P.
maculata. Fleay ( 1 965) also reports P. ingrami from
Monto, SE.Qd. The only museum specimen from
Monto in the Queensland Museum (J 15783) ap-
pears to represent P. tenuirostris. It is possible that
the animals studied by Fleay may have represented
both P. maculata and P. tenuirostris, although
photographs, nipple counts and measurements
given suggest only P. maculata.

Ride (1970) reports P. maculata from Western
Australia. The specimen regarded by Ride (pers.
comm.) to be P. maculata (WAM M3432) from
Tambrey, Coolawanyah Station, may represent an
undescribed taxon, and is discussed below (p. 357).

In general, P. ingrami and P. maculata are
allopatric. The only instance of apparent sympatry
between them occurs at Major Creek, Woodstock,
NE.Qd {P. maculata JM823 and P , ingrami
JM764). Marlow (1962) suggests they are also
sympatric at Coen, NE.Qd. Examination of this
material (AM M8148 and AM M8150) suggests
both specimens represent P. maculata.

Specimens referred to Planigale by Van Deusen
(1969) from northern Australia have not been
examined. If P. ingrami occurs in areas of the
Northern Territory other than the Barkly Table-
land, it should be easy to distinguish this
extremely flat-headed, tiny species from P. mac-
ulata.

ARCHER: REVISION OF PLANIGALE

349

Forms of P. metadata: There are at least two
distinctive, allopatric forms of P. maculata. The
typical form includes samples from Mt Molloy,
Townsville, and other localities in eastern Queens-
land and northeastern New South Wales gen-
erally in and east of the Great Divide. A non-
typical form occurs in northern Northern Ter-
ritory, northwestern Queensland, Barrow Island
and the Drysdale River area of Western Australia.
No attempt has been made here to assess the
possible statistical basis for recognizing subspecies.
This non-typical form includes the type of Phas-
cogale minutissima sinualis Thomas and the name
sinualis is used here only as a convenient means of
reference. This use is not to be interpreted as formal
recognition of subspecific status. At present, too
little information is available about Planigale from
Cape Y ork to determine the affinities of specimens
from Aurukun Mission. They differ in several
respects from sinualis and may represent a third
form. Measurements of spirit carcases provide a
mean TV/HB value of 0-77, the lowest for any
maculata series measured here. Shorter tails are
recorded by other workers. Fleay (1965) reports
two inch tails and three inch head and body
measurements for male P. maculata (as P. ingrami )
which gives a TV/FIB ratio of 067. The mean ST/E
value of the Aurukun Mission animals is very low
and compares only with the mean figure for P.
tenuirostris. Absolute size of almost all characters
is smaller in Aurukun Mission specimens than in
any other maculata measured. The only Aurukun
female with a distinct pouch area appears to have
nine very small nipples. It was evidently collected in
August.

Although it is concluded here that the types of
maculatus Gould and sinualis Thomas are
conspecific, it is clear that they also represent
different forms. The only major geographic barrier
which appears to separate the allopatric ranges of
these forms is the Gulf of Carpentaria. The
Aurukun Mission population from the eastern
edge of the Gulf of Carpentaria may be unique as a
result of its isolation from the typical form by the
inland areas of the Cape York Peninsula and from
the sinualis form by the Gulf. It is probable that
during the Plesitocene, with lowered sea levels, the
Gulf of Carpentaria was not a significant barrier.

Habit and Reproduction

Typical maculata: In N.S.W. they are rare and
inhabit sclerophyll and rain-forest on the eastern
side of the Great Divide (Marlow 1958). In Qd they
are recorded from hollow logs and under sheets of
iron, in timber country, flooded marsh, and rocky

areas (Fleay 1965); most specimens brought to the
Queensland Museum from Brisbane area have
been killed by cats; one was collected from a Cane
Toad’s stomach (Covacevich and Archer 1975).

In captivity they sometimes build saucer-shaped
nests of dry grass or in deeper grass, more elaborate
nests similar to those of Blue Wrens (Flfeay 1965).
Animals held by the author seemed opportunistic,
building nests in hollow logs, between sheets of
newspaper, and in boxes. As many as five in-
dividuals may nest together. Food preferred is
insects but eggs, lean meat, chicken and honey are
accepted. Small lizards and mammals are avoided.
Movements in captivity indicate they are adept
climbers, not hesitating to jump or drop distances
of over 30 cm.

Fleay suggests from field observations and
breeding in captivity that individuals of the typical
form are summer breeders, earliest pouch develop-
ment and pregnancy taking place in October and, if
no pregnancy occurs, the pouch stops development
by mid-January. Captive animals from Mt Molloy
held by the author mated on 11 September 1973.
This resulted in birth, but actual date of birth was
not noted.

Sinualis form: At Humpty Doo, N.T., they
occupy Pandanus and Melaleuca fringe areas
bordering the Adelaide River flood plain (Davies
1960). In W.A. one individual was collected in
hummock grass beneath Acacia on sandstone
boulders in sand, on the edge of a tributary of the
Drysdale River (Dr D. Kitchener, pers. comm.).

Aslin (1975) gives breeding data for this form in
captivity, noting that it is polyoestrous with a
gestation period of 19-20 days. Litters were born in
February, March, April, July, September, October
and December, single females having two or
possibly more litters per year. Males were capable
of breeding at least to 24 months in age. These
observations are supported by the combined
observations of Thomas (1926), Davies (1960), and
Johnson (1964) which suggest this form is also
polyoestrous in the field.

Planigale novaeguineae Tate and Archbold
(Plate 45)

Planigale novaeguineae Tate and Archbold, 1941, pp.

7 8.

Type

Holotype: AMN HI 08561, skull and skin, adult male,
collected by G. H. H. Tate, 20 January 1937. The
holotype has not been examined.

Type Locality: Tate and Archbold (1941, pp. 7-8)
state: \ . . Rona Falls, near Port Moresby, Central
Division, Papua: 250 metres . . .’.

350

MEMOIRS OF THE QUEENSLAND MUSEUM

Material Examined

Data sheets are available in the library of the
Queensland Museum.

New Guinea: <$, Waigani Swamp 16 km N. Port
Moresby, coll. H. Cogger, 23 December 1963 (AM
M9091); $, New Guinea, no other data (J4368).

Diagnosis

Similar to P. maculata but differs in larger size;
tendency for I 4 to exceed I 2 in crown length; several
dental and cranial ratios including lower mean
BW/ZW, BW/M 1 - 3 , FM/SD, and higher mean
SD/ZW and Q-P^M^; and in variable ten-
dency for tail to be longer than head and body.
Differs from other species of Planigale by same
features which distinguish P. maculata.

Description

Tail length variable, shorter than head and body
in holotype and specimens noted by Ziegler (1972)
but longer than head and body in J4368.

Hind foot, from description by Tate and Arch-
bold (1941, p. 8), dried in holotype with ‘Faint
traces of striations on pads (rest of foot normally
granulated as in Antechinus and other genera)’.

Nipple number unknown.

Mean BW/ZW and BW/M 1-3 values lowest in
Planigale reflecting very small alisphenoid bullae of
P. novaeguineae. Mean SD/ZW value highest
reflecting relatively deep skull of P. novaeguineae.
Mean FM/SD value lowest indicating both narrow
foramen magnum and relatively deep cranium.
Mean (I r M 4 HM 1 _ 4 )/M 1 _3 value highest.

Discussion

The type specimen was previously the only
specimen of Planigale recorded from New Guinea.
Ziegler (1972) records three additional specimens,
from Balimo, 450 km WNW. of the type locality.
Two additional specimens were collected at Waig-
ani Swamp, 16 km N. of Port Moresby, New
Guinea, by Dr H. Cogger of the Australian
Museum in 1963. Mr B. Marlow of the Australian
Museum will describe these specimens elsewhere,
but in the meantime has kindly allowed me to
examine the skull of the adult male specimen, AM
M9091. Menzies (1972) records numerous speci-
mens of this species obtained from owl pellet
material collected from the floor of a rock shelter
near Mt Eriama, about 16 km from Port Moresby
and 13 km from Rouna. A specimen in the
Queensland Museum (J4368) identified on the label
by Mr C. W. De Vis as coming from New Guinea,
represents Planigale. The specimen is represented

by a skull, dentaries and dry, shrivelled and faded,
carcase. A manuscript in De Vis’ handwriting
states that the fur is \ . . short and silky throughout
. . . mammae not apparent . . . pads smooth of the
three at the bases of the digits and the outermost
has a small backward-curved prolongation; the
plantar pad on the hallucal side elongate and semi-
divided, that on the outer side opposite and shorter
. . . above dark fawn, sides of muzzle distinctly
darker but without a definite stripe; edges of eyelids
nearly black, chin and throat nearly white passing
into pale fawn on the rest of the lower surface, feet
and tail brown above, paler brown below . . . length
of head and body 69 mm . . .’. He expresses all other
body measurements as percentages of the head-
body length. Converting these values, the tail is
704 mm, the hind foot length (possibly including
claws) 13-8 mm, hind foot breadth 3 0 mm, the ear
(it is not apparent how it was measured and the
figure seems small) 6.2 mm, the forearm and manus
1 1 -0 mm. The habitat is stated to be ‘New Guinea,
locality unrecorded’.

Differences between J4368, and AM M9091
from Waigani Swamp, include the greater skull
depth of AM M9091, 6-3, compared with 5-6 for
J4368, indicating an SD/IO value of 134 for AM
M9091 and M2 for J4368. Other differences are
minor by comparison, and it is likely that these are
sexual, AM M9091 being a male and J4368 a
female. In other series, such as those of P. maculata ,
males almost invariably have deeper skulls and
narrower interorbital values than females. Tate
and Archbold’s (1941) description of the type of P.
novaeguineae indicates that although the skull was
damaged in preparation, the braincase was, in their
opinion, very flat. This specimen is a male, and
therefore the degree of flattening may vary within,
as well as between, sexes. Ziegler (1972) notes other
possible sexual differences in P. novaeguineae from
Balimo.

J4368 is smaller in many cranial and dental
measurements than AM M9091. Measurements
given by Tate and Archbold (1941) for the holotype
are generally intermediate between these two.
J4368 indicates that there is not as great a size
difference between P. novaeguineae and other
Planigale as the type specimen alone suggests. For
example, AM M6893, male P. maculata from East
Funnell Creek above Sarina, Queensland, is in
most characters only just smaller than J4368 and in
BL, ZW, OBW, R LM\ VV, C-AP, and PLF-AB
even exceeds J4368. Further, in some specimens
(e.g. AM M6983 and AM M8336) of P. maculata ,
I 4 slightly exceeds I 2 in length, as in P. nova-
eguineae.

ARCHER: REVISION OF PLANIGALE

351

Tate and Archbold (1941 , p. 8) describe the hind
foot of the holotype as having ‘Faint traces of
striations . . .’ on the pads. Tate (1947, p. 134) says
of this specimen that it is \ . . the only one in which
the foot pads are distinctly striated’. In alcohol
specimens of P. maculata examined in the present
study, most apical granules of the interdigital
hallucal and metatarsal pads have visible striae
without surface ridges. In some (e.g. C7428) the
covering skin is thin on the apical granules and very
low ridges appear present on the surface of the pad.
In some alcholic specimens of other species (e.g.
AM M5021 P. ingrami) the apical granules are also
striate with extremely low ridges on the pads. It has
been noted in Sminthopsis (Archer, in preparation)
that dehydration may emphasize striations by
causing shrinkage of tissues over underlying sub-
surface ridges. It is possible that the difference in
opinion about the striate condition of the type
specimen of P. novcieguineae given by Tate and
Archbold (1941) and Tate (1947) may be the result
of seven years of dehydration. In any case, the fact
that other species sometimes have striated apical
granules indicates that P. novaeguineae is not
unique in this respect. It may, however, be a more
common feature of specimens of that species.

Habitat and Reproduction

The type specimen (Tate and Archbold 1941, p.
8) was caught on a ‘. . . great rock-strewn slope . . .
in a dryish place beneath an overhanging rock. The
hillside was comparatively barren of vegetation
. . .’. Ziegler (1972) notes that specimens caught at
Balimo occurred in ‘grass’. Menzies (1972, p. 404)
notes that with the possible exception of a doubt-
fully identified juvenile Pseudocheirus, all of the
species occurring with P. novaeguineae in the owl
pellet material from near Port Moresby \ . . are
savanna dwellers. Closed forest lies within a few
miles of the site and so well within the hunting
range of a medium-sized bird but it is clear that it
hunts only in the savanna.’ He suggests that the
relative abundance (6-5%) of the dasyurid material
(over 90% of which was Planigale ), may represent
the abundance of these animals in the small
mammal fauna of the Port Moresby savannas.

The Planigale ingrami group

This group comprises two species, P. ingrami
(Thomas) and P. tenuirostris Troughton.

Planigale ingrami (Thomas)

(Plates 46, 47, 51 A-B)

Phascogcile ingrami Thomas, 1960a, pp. 541-2.

Phascogale subtilissima Lonnberg, 1913, pp. 9-10.

Planigale ingrami brunneus Troughton, 1928, pp.

282-5.

Types

Phascogale ingrami Thomas, 1906a

Holotype: BM6.3.9.77, skull and skin, adult male,
collected by Mr W. Stalker, 30 April 1905. The holotype
has not been examined.

Type Locality: Thomas (1906b, p. 541 2) states
‘Buchanan, Alexandria, 600' . . . central part of Northern
South Australia’.

Phascogale subtilissima Lonnberg, 1913.

Holotype: Stockholm Museum no. 2482, skull and
mounted skin, juvenile male, collected by the Swedish
Scientific Expeditions, 2 February 1911. Photographs of
skull have been examined.

Type Locality: Lonnberg (1913, p. 9) says 'caught in
crack of the earth on a plain near Noonkambah . . .’.

Planigale ingrami brunneus Troughton, 1928

Holotype: AM M2 174, skull and carcase in alcohol,
adult female, donated by Mr F. L. Berney. The holotype
has been examined.

Type Locality: Troughton (1928, p. 285) gives
'Wyangarie, on the Flinders River, Richmond district,
northern Queensland.’

Material Examined

Data sheets for specimens examined available in the
library of the Queensland Museum.

Queensland: Richmond (e.g. J7655); Leslew Downs, nr
Richmond (JM824); Alex Ck, approx. 8 km from Leslew
Downs (JM763); Wyangerie, nr Richmond (AM M21 74);
Major Ck, nr Townsville (JM764); Red Falls, nr Lol-
worth Ck, 88 km NW. Charters Towers (Qd Museum);
Charters Towers (Department Primary Industries, Town-
sville); Karumba, nr Normanton (e.g. AM M8468); Old
Normanton (e.g. C3260).

Northern Territory: 200 km W. Burketown (e.g. AM
M5022); owl pellets, Brunette Downs (N.T. Museum and
noted in Parker 1973).

Western Australia: Ord River area (WAM M2846);
Argyle Downs Stn (W.A. Museum); Wotjalum Mission,
nr Derby (WAM M3191); cave surface, approx. 16 km
SE. Fitzroy Crossing (e.g. 71.12.30); cave, Windjana
Gorge (72.9.64); cave, between Kununarra and Ninbing
Stn (JM827).

Distribution of specimens shown in Fig. 3.

Diagnosis

Smallest species, also differing from other species
in having tail commonly longer than head and
body and in certain external, cranial, and dental
ratios including highest mean BL/M 1 3 and
NW/NL. Also differs from P. maculata and P.
novaeguineae in lower mean SD/IO, PN/NL and
higher mean FM/SD. Also differs from P. tenu-
irostris in lower mean SD/IO. Also differs from P.
gilesi in having P4, and lower mean PN/NL.

352

MEMOIRS OF THE QUEENSLAND MUSEUM

Fig. 3: Distribution of Planigale ingrami (solid diamonds represent modern specimens examined; hollow diamonds
represent cave specimens examined; inverted solid triangle represents record given by Van Deusen 1969), P.
tenuirostris (solid dots represent modern specimens examined; solid triangle represents record given by Aitken 1971),
and Planigale sp. (solid squares represent specimens examined from Ooldea and Tambrey; hollow squares represent
cave specimens from Ayers Rock and Madura).

Description

Tail invariably thin. Tail-vent length generally
longer than head-body length, and longer in
subtilissima than brunneus and typical forms.
Thomas’s (1906b) description of body measure-
ments of lectotype and paratype of typical form,
give mean TV/HB value of 0-75. Not clear from
Thomas’s description if all specimens (Thomas
indicates five) collected by Stalker had shorter tails
than head-body measurements. He states (p. 541)
‘Tail of medium length . . .’.

Supratragus of ear relatively long, compared
with other species and relatively longer in sub-
tilissima than typical form.

Six to 10 nipples have been recorded. Heinsohn
(1970) notes some individuals with 12 young. Three
individuals of subtilissima form have 10 nipples,
fourth appears to have only 9 and is presumably
abnormal. Holotype P. i. brunneus has 6 nipples.

Pouch morphology may distinguish subtilissima
from other forms (as suggested by Woolley 1974),
although subtilissima form not unique in poss-

ARCHER: REVISION OF PLANIGALE

353

ession of accessory anterior pockets. Less well-
developed pockets occur in P. ingrami from
Richmond, and may occur in all forms but
possibility cannot at present be checked. More
detailed examination is required of possible chan-
ges in pouch morphology as a function of repro-
ductive condition.

Thomas (1906b, p. 541) says of typical form
‘General colour above not unlike that of paler wild-
living forms of Mus musculus , something between
Ridgway’s “wood-brown"' and “broccoli-brown”,
the hairs slaty grey, with pale tips . . . Under surface
paler, with a yellowish tinge ... the hairs slaty at
base except on the chin. Crown like back. Cheeks
and chin whitish. A whitish-buffy line just over
each eye." Two females of subtilissima form exam-
ined live are not identically coloured: one much
darker than other with black nose; other noticeably
lighter with caramel-coloured nose. This variation
not clearly attributable to different habitats and
may represent normal intra-specific variation.
Heinsohn (1970) says of P. ingrami from near
Townsville that individuals from areas of black
basalt rock tend to be black, whereas those from
non-basalt areas are grey or grey-brown. In-
dividuals of brunnea form may be distinguished
from typical form by brown basal fur.

Absolute size of many dental and cranial charac-
ters comparable with P. tenuirostris, smaller than
P. maculata, P. gilesi and P. novaeguineae. Mean
M, 3 length P. ingrami 3T compared with 3 0 for
P. tenuirostris , 3-6 for P. gilesi , 3-6 for P. maculata ,
and 41 for P. novaeguineae . Various mean cranial
ratio values distinctive as follows: mean SD/IO
value comparable with that of P. gilesi, lower than
that of P. tenuirostris and much lower than that of
P. maculata and P. novaeguineae , indicating very
flat head and broad interorbital region of P.
ingrami: ; mean BL/M 1 3 value P. ingrami higher
than all other species, reflecting proportionately
long periotic and alisphenoid bullae; mean FM/SD
value comparable with P. gilesi and P. tenuirostris
but considerably larger than P. maculata and P.
novaeguineae ; mean NW/NL value higher than
that of any other species, indicating very short and
wide (posteriorly) nasals of P. ingrami; mean
PN/NL value comparable with that of P. tenu-
irostris but smaller than that of P. gilesi, P.
maculata, and P. novaeguineae. Intraspecifically,
subtilissima form generally distinguishable from
typical form by relatively smaller P 4 . Difference in
crown height between P 4 and P 1 noticeable, but less
than two times height of P 1 . One individual (AM
M5021) of typical form from 200 km W. Bur-
ketown similar, with only slightly enlarged P 4 . Both
forms have well-developed posterior P 4 talon.

Specimens of brunnea form have highest mean
FM/SD value of any Planigale population,
reflecting relatively very flat heads and wide
foramen magnum. This form also distinguished
from subtilissima form by higher mean
PMX-NAS/NL value and relatively longer lower
premolar row. P 4 also relatively larger.

Discussion

Cranial measurements given by Thomas (1906b)
for P. ingrami are similar to those of specimens
referred in the present study to P. ingrami. The very
flat head and wide interorbital width values are
very similar to those of specimens from Richmond
and 200 km W. of Burketown. The SD/IO value of
the lectotype, 0-87, is the same as that of the mean
value for specimens from 200 km W. of Burketown
and comparable with that of specimens of the
brunnea form from Richmond. It is also compara-
ble with the mean value for individuals of the
subtilissima form. This feature, in conjunction with
many other similar cranial and dental ratios and
absolute measurements, indicates the general sim-
ilarity of the type specimen to those of other
samples regarded in this study as p. ingrami.

Heinsohn (1970) records several individuals of
P. ingrami from the Townsville area. Examination
of these specimens (including JM823 and JM764)
indicates that P. maculata and P. ingrami are
sympatric at Major Creek, Woodstock. Marlow
(1962) records P. ingrami (AM M8148) from Coen,
Queensland. This specimen has been examined and
appears to be a juvenile P. maculata. Specimens
referred by Fleay (1965, 1967) to P. ingrami
probably represent P. maculata and possibly
include some P. tenuirostris. Ride (1970) refers
specimens from Laverton, Western Australia to
Planigale cf. P. ingrami (plate 35). These specimens
have been described as Ningaui ridei (Archer 1975).
Ride (1970, p. 1 20) also refers to P. ingrami from ‘. .

. Kimberley and central W.A. . . This material
has been examined and, like all modern specimens
of Planigale ingrami examined from the Kimberley
region, appears to represent the subtilissima form
of P. ingrami. One fossil sample from the southeas-
tern edge of the Kimberley region is unusual (see
below).

Forms of P. ingrami: There are at least two
allopatric forms of P. ingrami. The typical form
includes samples from the Barkly Tableland, in an
area 200 km west of Burketown; and other areas in
northeastern Queensland as far east as the Towns-
ville area. The subtilissima form includes samples
from the Kimberley Region. The brunnea form

354

MEMOIRS OF THE QUEENSLAND MUSEUM

includes samples from the Richmond area. Speci-
mens from Old Normanton and Karumba,
Queensland, are not clearly referable to any of
these three and either represent a fourth form of P.
ingrami or perhaps a mixed sample of more than
one species. No attempt has been made here to
assess the possible statistical basis for recognizing
subspecies. Use of the formal names for these
allopatric forms here is a matter of convenience
and must not be interpreted as recognition of
their subspecific status.

Lonnberg (1913) describes Phascogale sub-
tilissima on the basis of a specimen that was shown
by Tate (1947) to be juvenile. Tate considers the
subtilissima form to be a race of ingrami and in this
view is followed by Marlow (1968). Troughton
(1928, 1967) and Ride (1970) regard it as a full
species. Close similarity has been noted here
between specimens of the subtilissima and typical
forms. Although modern and fossil Kimberley
specimens demonstrate morphological extremes,
they are in most respects just one step beyond
specimens from 200 km west of Burketown, which
are also geographically closest to the Kimberley
population. A dine may exist which links animals
from the area west of Burketown, via the Barkly
Tableland, to the Kimberley region. More speci-
mens are required to test this possibility.

Fossil specimens (including 72.9.65 and
71.12.29-31) collected from surficial cave deposits
in the southern Kimberley region, associated with a
small mammal fauna which will be described
elsewhere, represent at least nine individuals. They
differ in some absolute measurements from speci-
mens of the subtilissima form including mean
I,-M 4 length of the fossil specimens which is 5-9
mm as opposed to 6 7 mm for the subtilissima form.
This results in a (I 1 -M 4 )-(M 1 _ 4 )/(M,_ 3 ) value of
0-83 for the fossil sample as opposed to 0-90 for the
modern sample. This is also the lowest figure for
any Planigale population. Further, mean length of
the dentary of the fossil sample, 10-6 mm, is less
than that of all other Planigale .

Troughton (1928) considers the brunnea form to
differ from the typical form in possessing brown
basal fur, longer tail, premaxillary vacuity which
extends posteriorly to the middle of the C 1 alveolus,
clear maxillary vacuity, P 1 barely two thirds of the
height of P 3 rather than subequal, and broader
nasals. Topotypical specimens J7655-6, give an
idea of variation unavailable to Troughton. Both
Queensland Museum specimens have premaxillary
vacuities which extend beyond the level of the
anterior edge of the C 1 . This is an almost universal
condition in P. ingrami examined in the present
study. It seems likely that the type of the species is

unusual in this respect (Thomas 1906b). Both
Richmond specimens and all Planigale specimens
examined have small, distinct maxillary fenestra,
and this is not a useful diagnostic character for any
subgeneric taxonomic rank. If these vacuities are
not apparent, palatal skin has not been removed
from the bony palate. Regarding premolar grad-
ient, in J2655 P 3 is only slightly larger than P 1 and
in J7656 P 3 is almost equal to P 1 in height.
Regarding relative nasal breadth, the NW/NL
value of the holotype of brunneus , using
Troughton’s (1928) measurements, is 0-49. This
value for J7655 is also 0-49, while the measure for
J7656 is 0-43. Regarding tail length, although
measurements of J7655-6 are unknown, other
samples serve to show that there is considerable
variation in this character (see Table 2). This leaves
basal fur colour as a possibly useful diagnostic
character. This condition is indeterminable in
J7655 6. Flowever, Thomas (1906b) notes vari-
ation in fur colour. It therefore seems unlikely that
the brunnea form is differentiable from the typical
form using characters given by Troughton (1928),
with the possible exception of basal fur colour.

Specimens from Old Normanton (C3259 60)
and Karumba (AM M8467-9, M9144) are so
variable that they may represent more than one
species. Two specimens (both males) AM M8468
and AM M8469 from Karumba are more robust
and broader-skulled than the third. AM M8467
(also male), from this locality. Specimens from Old
Normanton (one male, one female) resemble in
most respects AM M8467.

Habitat and Reproduction

Typical ingrami: In N.T. they inhabit area
around Alexandria draining inwards to Poly-
gonum swamp (Thomas 1906b); at locality 200
km W. Burketown, they occur in tussocky grass,
dry swamps and along perennial streams flowing
out of the coastal ranges westward from Burke-
town (Troughton 1967); at New Castle Waters,
one was found under bark near a small stream (Van
Deusen 1969); generally not uncommon on black-
soil plains, dry swamps and perennial watercourses
of the Gulf drainage (Parker 1973). In Qd, 137 km
SW. Townsville, one was found drinking from a
rock pool (Heinsohn 1970).

In N.T., 200 km W. Burketown, they are said
(Albert De Lestang, letter in Queensland Museum
dated 8 October 1930) to breed in February to
April, producing litters of 4 to 6 young. In NE.Qd
they have litters of 4 to 12, reproduction occurring
around December to March (Heinsohn 1970).

ARCHER: REVISION OF PLANIGALE

355

In captivity, an individual of the typical form
constructed a series of covered runways and a
hollow nest chamber in dry grass (Heinsohn 1970).

Subtilissima form: In W.A. they inhabit tussocks
of grass near the Kimberley Research Stn and were
also collected from piles of wet decomposing grass
(Rudeforth 1950 and pers. comm.); on isolated hills
on Argyle Downs Stn, while the waters of Lake
Argyle were rising, they were collected from clumps
of spinifex (Dr D. Kitchener and Mr W. H. Butler,
pers. comm.).

One or two females from Argyle Downs Stn
collected live in December-January 1971-2 had an
unspecified number of young about 0-5 cm long in
her pouch. A single juvenile was preserved before
the mother was transported to Perth. She evidently
consumed the remaining young in transit. During
January, the other female’s pouch underwent
enlargement, elongation of pouch hairs and, to-
wards the end of February, regression. No juveniles
were born. Woolley (1974) notes changes in pouch
morphology of these same individuals and con-
cludes that this form breeds in summer months,
unlike the majority of dasyurids.

Brunnea form: Habitat unknown. Holotype has
pouch young but month of collection is not
recorded.

Old Normanton: Habitat unknown. One speci-
men (AM M9144) with 8 (or possibly 9) nipples
died June 1965, The pouch was developing but
clearly not in breeding condition.

Planigale tenuirostris Troughton
(Plate 48)

Planigale tenuirostris Troughton, 1928, pp. 285 7.
Type

Holotype: AM M3933, skull and carcase in alcohol,
adult female, collected by R. Helms in May or June 1 890.
The holotype has been examined.

Type Locality: Troughton (1928, p. 287) says ‘Col-
lected at Bourke or Wilcannia, New South Wales, during
the Darling River floods.’

Material Examined

Data sheets for specimens examined available in the
library of the Queensland Museum.

Queensland: Cunnamulla (AM M6957); Warwick
(J7559); Pittsworth (J3096); Roma (J3824); 16 km NE.
Longreach (J17549); Glenmorgan (J 10109); Belmont, via
Rockhampton (J 14089).

New South Wales: Bourke or Wilcannia, Darling River
(AM M3933); Bellata (AM M6879); Cullubri, 43 km SSE.
Nyngan (AM M8151); Fowlers Gap (e.g. JM831).

Distribution of specimens shown in Fig. 3.

Diagnosis

Small, very similar to P. ingrami but differs in
relatively shorter supratragus of ear; tail being
generally shorter than head-body length; and in
several cranial ratios including lower mean

BL/M 1 - 3 and NW/NL, BW/M 1 - 3 , BL/M 1 3 and
higher mean SD/IO, SD/ZW, and BL/ZW. Differs
from P. novaeguineae and P. maculata in being
smaller; in having relatively reduced transverse
canal foramina; smaller stylar cusp D on M 3 ;
longer head; wider interorbital distance; and in
several cranial ratios including lower mean

BL/ZW, PN/NL, SD/IO and higher mean
NW/ZW, BL/ZW, BL/M 1 3 , FM/SD, and
FM/ZW. Differs from P. gilesi in having P4.

Description

Mean TV/HV value, 0-87, indicates relatively
short-tailed condition.

Mean ST/E value, 0-28, lowest of any Planigale
except P. maculata from Aurukun Mission. Mean
absolute supratragus length, 2-6 mm, shortest.

Troughton (1928) notes holotype has 11 nipples
but suggests normal number is 10. J7559 from
Warwick, Queensland, has 7. J3096 from

Pittsworth, Queensland, has 8.

Mean BL/ZW value highest indicating relative
dolichocephaly. Mean NW/'NL value lower than
any P. ingrami , demonstrating relatively narrow
nasals. Other differences indicated in Tables 1-2.

Discussion

Troughton (1928) describes characters which he
believes are useful in diagnosing P. tenuirostris.
Having examined larger series of specimens of P.
ingrami than were available to Troughton, it is
clear that some of these characters are also variably
present in P. ingrami. For example, Troughton
(1928) says P. tenuirostris has 10 or 12 nipples in
contrast to P. ingrami which has 6. As noted above,
the subtilissima form may have 10 nipples, the
typical form of P. ingrami from 200 km west of
Burketown may have 10, the typical form of P.
ingrami from near Townsville (Heinsohn 1970)
may have 6 to 12, P. gilesi has 12(Aitken 1972), and
some P. maculata may have 8 to 15 (Fleay 1965,
1967, Davies 1960). Troughton (1928) considers
that P. tenuirostris has more elongate premaxillae
than P. ingrami. In some cases, this appears to be
true. However, mean premaxilla-nasal contact
length, a measure of this character, is 2-4 mm in
specimens of P. tenuirostris , and 2-3 mm in
specimens of P. ingrami from near Richmond,
indicating little, if any, difference. Troughton
(1928) describes P 1 as being two-thirds the size of

356

MEMOIRS OF THE QUEENSLAND MUSEUM

P 3 in P. tenuirostris and subequal in typical P.
ingrami (Thomas 1906). This character appears
variable in P. ingrami (see above) and P. tenu-
irostris. Troughton (1928) considers P. tenuirostris
to differ from the brunnea form of P. ingrami in its
narrower hind foot, a condition which may relate
to some circumstance of death. Among specimens
of P. ingrami , AM M5022 has a hind foot
noticeably wider distally than AM M4744 from the
same locality. The widely spread foot of AM
M5022 suggests the animal may have died and been
fixed in muscular spasm. Finally, Troughton ( 1 928)
considers P. tenuirostris to differ from the brunnea
form of P. ingrami in possessing a posterior notch
on the lower part of the pinna of the ear. On the
basis of larger series, this appears variable in both
P. tenuirostris (e.g. slight or absent in AM M7313
and pronounced in holotype) and P. ingrami
(small notch present in AM M5022 but not present
in holotype of P.i. brunneus).

Van Deusen (in Fleay 1965) suggests P. tenu-
irostris may be a race of P. ingrami. The evidence
seems inadequate to decide one way or the other.
The present study only suggests that specimens
referred here to P. tenuirostris exhibit some ex-
tremes in size and proportion, and most appear to
inhabit more central or arid areas.

Marlow (1958) refers some specimens , (AM
M7033, M7393, M7819 and M7820) to this species
from New South Wales, Aitken (1971) refers these
to P. gilesi.

Habitat and Reproduction

Aitken (1972) notes P. tenuirostris (AM M6879
and AM M7313) is sympatric with P. gilesi (AM
M7033) from Bellata, New South Wales, and that
(1971) a dehydrated specimen of P. tenuirostris
(SAM M8405) was found at the bottom of a
disused stone tank at Mulga Creek Well, near the
northeastern tip of the Flinders Range, South
Australia. In the same tank was a specimen of
Sminthopsis crassicaudata.

Troughton (1928) notes the holotype, collected
during the Darling River floods in May or June,
1890, has a well-developed pouch with enlarged
nipples.

Holotype

The holotype has been examined. Troughton
(1928, fig. 2) illustrates the ear, nose, hind foot, and
skull of the holotype. Specimen numbers given by
Troughton appear to be confused. The holotype is
stated to be AM M3856, but AM M3933 in the
Australian Museum is labelled as the holotype.
AM M3856 is (pers. comm. Mr B. Marlow, 1975)

listed in the Australian Museum catalogue as
Pteropus poliocephalus from Tamborine, south
Queensland.

The Planigaie gilesi group
This group contains only P. gilesi Aitken.

Planigaie gilesi Aitken
(Plate 49)

Planigaie gilesi Aitken, 1972, pp. 1-14.

Type

Holotype: SAM M8046, dry skin and skull, torso in
spirit, adult male, collected by messrs P. Aitken, A.
Kowanko, J. Forrest and J. Howard, 29 June 1969. The
holotype has been examined.

Type Locality: Aitken (1972, p. 1) gives ‘No. 3 Bore,
Pastoral Property of Anna Creek, South Australia (lat.
28° 18'S, long. 136° 29' 40"E).’

Material Examined

Data sheets for specimens examined available in the
library of the Queensland Museum.

Queensland: Durrie Stn, 97 km E. Birdsville (J21973).
South Australia: Anna Creek (e.g. SAM M8406).

New South Wales: Bellata (e.g. AM M7033); Brewar-
rina (e.g. AM M7819); Fowlers Gap; Mt King, 27 km N.
Tibooburra (AM M9829).

Distribution of specimens shown in Fig. 2.

Diagnosis

Large species of Planigaie differing from all
others by having only two upper and lower
premolars on each side as adult condition.

Description

Tail thin to slightly incrassated. Shorter than
head and body length. Mean TV/HB value 0-96.

Supratragus of ear relatively long. Mean ST/E
value 0-36.

Apical granules of interdigital pads of hindfoot
have striae which do not reflect incident light and
hence do not occur as ridges on surface of granule.
Aitken (1972) records twelve nipples.

Cranium wide and very flat. Parietal horns
extend anteriorly to level near anterior end of
cerebral hemispheres. Mean SD/IO value, SD/C 1
M 4 and several other cranial and dental ratios (see
Table 2) very similar to those of otherwise smaller
P. ingrami.

Discussion

P. gilesi is the most distinctive species of the
genus because of the reduced premolar number.
When premolar reduction or loss occurs in other

ARCHER: REVISION OF PLANIGALE

357

dasyurids (e.g. some Antechinus, Neophascogale
Stein, 1933, Phascolosorex Matschie, 1916, Myoi-
ctis Gray, 1858, Dasycercus , Dasyuroides, Dasy-
urus Matschie, 1916) invariably the posterior
premolar is reduced. This predisposition to reduce
or lose P4 in dasyurids has previously been noted
(e.g. Thomas 1887, Bensley 1903, Tate 1947). In
genera in which only some species have lost
premolars, others show reduction of P4. For
example, all Antechinus rosamondae Ride, 1964
have two premolars above and below. Some
individuals of A. macdonnellensis (Spencer, 1896)
show the same condition. Others show a small P4
above and below. Still others lack P4 altogether.
There is therefore a structural gradient of loss
within the genus. In Planigale, two trends in
premolar size arfe evident. In all species P, and P 1
are reduced, possibly the result of the very large
canines, and (except in P. gilesi ) P 4 is markedly
reduced while P 4 is grossly enlarged into a tall
shearing blade. Only the suhtilissima form of P.
ingrami does not show gross enlargement of P 4 , but
even here, P 4 is larger than P 3 . If these trends in
Planigale are used to interpret tooth loss in P. gilesi,
the most logical conclusion is that P 1 and P 4 have
been lost. If this were so, the anterior upper
premolar (P 3 ) should shear behind the posterior
lower premolar (P 3 ), a situation which does not
exist. If P 1 and P 4 were lost, there must have been
an intermediate stage in which premolars did not
occlude in order that the anterior upper premolar
and posterior lower premolar could bypass one
another. Such an intermediate stage of non-
occlusion is improbable and it is concluded that P.
gilesi, like all other dasyurids which exhibit pre-
molar reduction, has lost P 4 and P 4 .

Loss of P 4 and P 4 in P. gilesi, in spite of the trend
for reduction of P 1 rather than P 4 in other
Planigale, indicates the magnitude of the structural
gap between P. gilesi and other Planigale.

Habitat and Reproduction

Aitken (1972) describes the habitat as bullrush
and sedge plant associations developed around a
bore drain in an area where average annual rainfall
is less than 125 mm per year. Other sympatric
species include Canis familiaris, Sminthopsis frog-
gatti, Pseudomys desertor, Rattus villosissimus,
Mus musculus, and introduced mammals such as
rabbits, foxes, cats, camels, horses and cows.
Aitken (1972) concludes P. gilesi is at least partially
insectivorous.

None of the specimens examined by Aitken
(1972) indicate the breeding season. Of two females
with undeveloped pouches, AM M7033 (according

to the label) was collected in June (Aitken says 27
Feb.) and SAM M841 1 was collected in August.

Holotype: The holotype has been examined.
The skull and dentary are figured by Aitken (1972,
plate 3).

Planigale, incertae sedis

Two specimens of Planigale examined in this
study are not clearly referable to any described
species. Both are damaged, and until better mat-
erial comes to hand, they should not be named. A
third form is described by Lundelius and Turnbull
(1973).

J 16732

A juvenile collected by A. S. Le Souef from
Ooldea, Transcontinental Railway, South Aus-
tralia (Fig. 3). Because it is juvenile (P 4 has not
completely erupted), most cranial and dental
measures cannot be meaningfully compared with
those of adults of other species. It is most similar to
P. tenuirostris. The bullae appear to be very much
smaller than those of P. tenuirostris. P 4 (excavated)
is 0-45 mm long; P 3 is 0-70 mm long; P! is 0-60 mm
long.

WAM M3432 (Plate 50)

This specimen, the basis of Ride’s (1970, p. 120)
recognition of Planigale maculata in Western
Australia, consists of a broken skull and somewhat
damaged skin of an old adult male, collected by Mr
W. H. Butler on 3 August 1958, from Tambrey,
Coolawanyah Station, Hammersley District of
Western Australia (Fig. 3). P 4 is double-rooted, a
very unusual condition in Planigale . M 1-3 /ZW
value, 0-32, is lower than the mean of any other
Planigcde population. (C 1 - M 4 ) - (M^^/M 1-3
value, 0-82, is larger than the mean for any other
Planigale population. This specimen may represent
an unnamed species of Planigale but the specimen
is too incomplete and isolated to permit adequate
comparisons. P 4 is 0-56 mm long; P 3 0-65 mm long;
and Pj 060 mm long.

The Madura Form

Lundelius and Turnbull (1973) describe a
Quaternary Planigale from Madura Cave, on the
Roe Plain of the Western Australian Nullarbor
(Fig. 3). After comparing it with several specimens
of P. ingrami and one specimen of P. maculata, they
conclude (p. 27) \ , . the Madura Cave material
cannot be referred to any of the described species of
pygmy antechinuses’, and consider (p. 18) it is
similar to P. maculata in having P 1 crowded out of

358

MEMOIRS OF THE QUEENSLAND MUSEUM

alignment in the tooth row, a feature which
contrasts with P. ingrami. An examination oflarger
series of specimens than those available to Lun-
delius and Turnbull suggest this character is
variable in P. ingrami and P. maculata. Two out of
5 modern P. ingrami have P 1 crowded out of
alignment (e.g. J7656) and 1 (J 17549) has LP 1
crowded out of line while RP 1 is straight. In 2 other
specimens, P 1 is straight on both sides. In 10 P.
maculata (a single sample from Mt Molloy), 3
specimens (e.g. J16730) have P 1 crown straight, 5
(e.g. J 16481) have the crown slightly crowded out
of alignment, and 2 (e.g. J 16729) have the crown
markedly crowded out of alignment. In P. tenu-
irostris all 4 specimens examined have the P 1 crown
straight, as does the single specimen of P. gilesi
checked for this character. Lundelius and Turnbull
also suggest (p. 20) that the Madura Planigale
resembles P. maculata in having a less-indented
ectoloph immediately anterior to stylar cusp D
(their mesostyle) than is present in P. ingrami . An
examination of this character in 10 P. ingrami and
10 P. maculata similarly suggests the character is
not constant in either of the two modern species. In
J 15891, the ectoloph is as deeply indented as in
most P. ingrami and more deeply indented than
others. It reveals that stylar cusp D is in general
larger in P. maculata than P. ingrami and in this
respect the Madura Planigale more closely re-
sembles the latter. Lundelius and Turnbull also
note (p. 22) that in the Madura Planigale, P 4 is
generally straight while P, is crowded out of
alignment, and in this respect it is unlike P. ingrami.
In a series of modem P. ingrami examined in this
study, orientation of P 4 appears to be variable. In
J 17549, P 4 is straight while P 3 and P, are slightly
crowded out of alignment. In J7656 and J7655, P 4 is
crowded out of alignment. In P. maculata, some
specimens have P 4 straight (e.g. J 16477, J 16730),
slightly turned (e.g. J16729, J 1 648 1 ), or markedly
turned (e.g. J 16482, J 15891). Similar variation
occurs in P. tenuirostris. The Madura Planigale is
very distinctive in regard to the large size of P 4 , as
noted by Lundelius and Turnbull. No P 4 of any
modern Planigale examined in the present study is
of comparable length, not even the two-rooted P 4
ofWAM M3432.

Summary of Resemblance Within Planigale

Typical P. maculata and P. novaeguineae are very
similar, and may prove conspecific. A general trend
of increasing size exists in coastal populations
north from New South Wales into New Guinea.
Other trends such as reduction in size of alis-
phenoid bullae are also demonstrated. P. nova-

eguineae appears, in most respects, to represent no
more than one end of this cline. Northern Territory
and northwestern Queensland P. maculata (poss-
ibly all referable to P. m. sinualis ) are distinct from
the typical form.

P. ingrami and P. tenuirostris are similar and it is
not clear that they are separate species. These two
forms are similar to P. gilesi, which because of
premolar loss involving a reversal of a trend
developed in the remaining species of the genus, is
otherwise very distinct from all other Planigale.

The subtilissima form of P. ingrami is distinctive,
while the brunnea form may be identical with the
typical form of P. ingrami.

There may be additional undescribed forms of
Planigale represented by single specimens or small
samples, at present too inadequate to assess
taxonomically.

ACKNOWLEDGMENTS

During the course of this study, the author was
supported by a Fulbright Scholarship, a grant in
aid from The American Explorers Club, and a
Research Assistantship to Dr W. D. L. Ride
(Director, Western Australian Museum) who was
in receipt of a Research Grant from the Australian
Research Grants Committee.

Miss J. Covacevich (Queensland Museum), Mr
B. J. Marlow (Australian Museum), Miss J. Dixon
(National Museum of Victoria), Mr S. Parker
(formerly of the Arid Zone Research Centre, Alice
Springs), Mr J. Calaby (C.S.I.R.O. Wildlife, Can-
berra), Mr P. Aitken (South Australian Museum),
Dr G. E. Heinsohn (James Cook University), Mr J.
Bannister and Dr D. Kitchener (Western Aus-
tralian Museum) kindly allowed the author to
examine specimens from collections in their charge.
Specimens and/or information about their col-
lection were provided by Miss J. Covacevich, Mr F.
Little, Mr G. Ingram, Mrs L. Huxley, Mr L. G.
Marshall, Dr D. Kitchener, Mr W. H. Butler, Mr

A. Borsboom, Mr S. Morton, Mr R. G. Hobson,
Mr B. J. Marlow, Dr H. Cogger, Mr S. Parker, Mr

B. F. Rudeforth, Dr S. J. J. F. Davies, Dr G.
Gordon, MrP. Aitken, and Dr G. E. Heinsohn. Dr
W. D. L. Ride (formerly of the Western
Australian Museum), Dr A. Bartholomai and Mr
B. Campbell (Queensland Museum), con-
structively criticised drafts of this paper. Dr Ride
allowed access to his photographs of the type
specimen of Phascogale subtilissima Lonnberg. Mr
A. Easton and Mr A. Elliot (Queensland Museum)
helped with aspects of photography. Miss P.
Rainbird and Mrs C. Farlow (Queensland
Museum) typed drafts of this paper. My wife.

ARCHER: REVISION OF PLANIGALE

359

Elizabeth, has helped to maintain numerous live
specimens of Planigale over a period of four years.

LITERATURE CITED

Aitken, P. F., 1971. Planigale tenuirostris Troughton.
The Narrow-nosed Planigale, an addition to the
mammal fauna of South Australia. S. Aust. Nat. 46:
18.

1972. Planigale gilesi (Marsupialia, Dasyuridae); a new
species from the interior of south eastern Australia.
Rec. S. Aust. Mus. 16: 114.

Archer, M., 1975. Ningaui, a new genus of tiny
dasyurids (Marsupialia) and two new species, N.
timealeyi and N. ridei, from arid Western Australia.
Mem. Qd Mus. 17 : 237-49.

1976. The basicranial region of marsupicarnivores
(Marsupialia), inter-relationships of carnivorous
marsupials, and affinities of the insectivorous per-
amelids. J. Linn. Soc. Lond ., in press.

Asun, H., 1975. Reproduction in Antechinus maculatus
Gould (Dasyuridae). Aust. Wildl. Res. 2: 77-80.

Bensley, B. A., 1903. On the evolution of the Australian
Marsupialia: with remarks on the relationships of
the marsupials in general. Trans. Linn. Soc.
Lond.(ZooL) (2) 9: 83-217.

Covacevich, J. and Archer, M., 1975. The distribution
of the Cane Toad, Bufo marinus, in Australia and its
effects on indigenous vertebrates. Mem. Qd Mus. 17 :
305-10.

Davies, S. J. J. F., 1960. A note on two small mammals of
the Darwin area. J. Roy. Soc. W. Aust. 43 : 63-6.

Fleay, D., 1965. Australia’s ‘needle-in-a-haystack’ mar-
supial. Viet. Nat., Melb. 82: 195-204.

1967. Planigale holds record family number. Viet. Nat.,
Melb. 84 : 202.

Gould, J., 1851. The mammals of Australia.’ Letterpress
to plate 44.

1852. The mammals of Australia.’ Letterpress to plate
45.

1854, Description of two new species of mammalia of
the genus Antechinus. Proc. zool. Soc. Lond. 1851 :
284-5.

Hftnsohn, G. E., 1970. World’s smallest marsupial the
flat-headed marsupial mouse. Animals 13 : 220-2.

Johnson, D. H., 1964. Mammals of the Arnhem Land
Expedition. Pp. 427 51 in R. L. Specht (Ed.)
’Records of the American Australian Scientific
Expedition to Arnhem Land, 4.’ (Melb. Univ. Press:
Melbourne).

Laurie, E. M. O. and Hill, J. E., 1954. ‘List of land
mammals of New Guinea, Celebes and adjacent
islands 1758 1952.’ 175 pp. (British Museum (Nat-
ural History): London).

Lidicker, W. Z. Jr, and Marlow, B, J., 1970. A review of
the Dasyurid marsupial genus Antechinomys Krefft.
Mammalia 34: 212 27.

Lonnberg, E., 1913. Results of Dr E. Mjobergs Swedish
Scientific Expeditions to Australia 1910-13. Mam-
mals. K. svenska VetenskAkad. Handl. 52 (1): 1-10,

Lundelius, E. L. Jr. and Turnbull, W. D., 1973. The
mammalian fauna of Madura Cave, Western Aus-
tralia Part 1. Fieldiana, Geol. 31: 1 35.

Marlow, B. J., 1958. A survey of the marsupials of New
South Wales. C.S.I.R.O. Wildl. Res. 3: 71 1 14.

1962. On the occurrence of Antechinus maculatus
Gould and Planigale ingrami (Thomas) (Mar-
supialia, Dasyuridae) in Cape York Peninsula,
Queensland. J. Mammal. 43: 433-4.

1968. ‘Marsupials of Australia.’ (2nd Ed.) 141 pp.
(Jacaranda Press: Brisbane).

Menzies, J. I., 1972. The relative abundance of Planigale
novaeguineae and other small mammals in the south
Papuan savannas. Mammalia. 36: 400-5.

Parker, S. A., 1973. An annotated checklist of the native
land mammals of the Northern Territory. Rec. S.
Aust. Mus. 16 : 1-57.

Pocock, R. I., 1926. The external characters of Thyla-
cinus, Sarcophilus, and some related marsupials.
Proc. zool. Soc. Lond. 68: 1-48.

Ride, W. D. L., 1964. Antechinus rosamondae, a new
species of dasyurid marsupial from the Pilbara
District of Western Australia; with remarks on the
classification of Antechinus. W. Aust. Nat. 9 : 58-65.

1970. ‘A guide to the native mammals of Australia.’ xiv
and 249 pp. (Oxford Univ. Press: Melbourne).

Rudeforth, B. F., 1950. Some notes on an interesting
marsupial. Scope (University of Western Australia)
1 : 10 - 11 .

Tate, G. H. H., 1947. On the anatomy and classification
of the Dasyuridae (Marsupialia). Bull. Amer. Mus.
nat. Hist. 88: 97-156.

1951. Notes on Australian marsupials rare or little-
known in the United States. Amer. Mus. Novit. 1528 :
1 - 6 .

Tate, G. H. H. and Archbold, R., 1941. New rodents
and marsupials from New Guinea. Amer. Mus.
Novit. 1101 : 1-9.

Thomas, O., 1887. On the homologies and succession of
the teeth in the Dasyuridae with an attempt to trace
the history of the evolution of mammalian teeth in
general. Phil. Trans. 178 : 443-62.

1 888. ‘Catalogue of the Marsupialia and Monotremata
in the Collection of the British Museum (Natural
History).’ xiii and 401 pp. (British Museum (Natural
History): London).

1906a. A collection of mammals made by Mr W.
Stalker in the Northern Territory of South Aus-
tralia. . . . Ab. Proc. zool. Soc. Lond. 32: 6.

1906b. On mammals from northern Australia pre-
sented to the National Museum by Sir Wm. Ingram,
Bt., and the Hon. John Forrest. Proc. zool. Soc.
Lond. 1906 : 53C43.

1926. On various mammals obtained during Capt.
Wilkin’s Expedition to Australia. Ann. Mag. nat.
Hist, (9) 9 : 625-35.

Troughton, E., 1928. A new genus, species, and
subspecies of marsupial mice (Family Dasyuridae).
Rec. Aust. Mus. 16 : 281-8.

1967. ‘Furred animals of Australia,’ (9th Ed.) xxxii
and 384 pp. (Angus and Robertson: Sydney).

360

MEMOIRS OF THE QUEENSLAND MUSEUM

Van Deusen, H. M., 1969. Feeding habits of Planigale. J.
Mammal. 50: 616-8.

Walker, E. P., 1964. ‘Mammals of the world. Volume 1.’
Xvii and 644 pp. (Johns Hopkins Press: Baltimore).
Woolley, P., 1974. The pouch of Planigale subtilissima

APPENDIX

Since this paper went to press I have had an
opportunity to examine specimens of Planigale in
the United States and England, and can make the
following comments.

Specimens in the collections of the American
Museum of Natural History include:

P. maculata (AMNH 160075-6, near Townsville, Qd.
AMNH 193959, Gunalda, N. of Gympie, Qd. AMNH
160374, Monto, Qd. AMNH 18380, the sinualis form.
Red Bank Mine, 18 miles (29 km) W. Wollogarang,
N.T.). AMNH 193959 and 160374 were donated by Mr
D. Fleay and support the suggestion made above that at
least some of the specimens reported by Fleay (1965)
represent P. maculata rather than P. ingrami.

P. novaeguineae (AMNH 108561, Holotype, Rona
Falls, nr Port Moresby, Papua: measurements: C 1 M 4 ,
8-2; M l - 4 , 4-9; M 1 - 3 , 4-3; I,-M 4 , 9-5; M^, 5-3; M t _ 3 , 4-0;
nasal length, 9-2; maximum width of nasals, 4-0;
minimum width of nasals, 1-7; pmx-nasal suture, 4-1; the
skull is badly smashed and lacks RIi_ 3 , RPj, RP 4 , LIi~P 4 ,
and RM 2-4 ; foot pads are faintly striated, clearly the
result of dehydration).

P. ingrami (AMNH 160313, Karumba, Qd, appears to
represent this species but I have not examined the skull so
reference here is tentative).

Specimens in the collections of the British
Museum (Natural History) include:

P. maculata (BM53. 10.22.21, Holotype, Clarence R.,
Moreton Bay, Qd: measurements: C ! -M 4 , 6-7; M 1-4 , 4-0;
M 1 - 3 , 3-7; R-LM 3 , 6-2; IPVD, 4-5; I r M 4 , 7-3; M,_ 4 ,
4-6; Mj_ 3 , 3-4; nasal length, 6-8; minimum width of
nasals, 0-8; pmx-nasal suture, 2-9; the skull is only repre-
sented by a rostrum with damaged lower jaws, and lacks
RP 1 and LPp Skin with number 53.10.22.21. is type and
has locality as Clarence R., Moreton Bay; vague white
spots are apparent on ventral side and flanks; it has
damaged feet and scrotal area. BM54.10.21.5, Holotype

and other dasyurid marsupials. J. Roy. Soc. West.
Aust. 57: 11-15.

Ziegler, A. C., 1972. Additional specimens of Planigale
novaeguineae (Dasyuridae: Marsupialia) from Ter-
ritory of Papua. Australian Mammalogy 1: 43-5.

of minutissima, Cressbrook, N.S.W.: measurements: ZW,
10-5; C 1 M 4 , 6-8; M 1 - 4 , 4-2; M 1 - 3 , 3-8; R-LM 3 , 6-4; 10,
4- 1; IPVD, 4-4; DL, 13-5; I t -M 4 , 7-7; M t 4 , 4-4; M^ 3 , 3-4;
C-AP, 4-0; C-AR, 4-4; SD, 5-3; nasal length, 7 1;
maximum nasal width, 3-2; minimum nasal width, 1-2;
pmx-nasal suture, 2-5; the rear of skull is badly damaged.
Skin with number 53.10.22.20, is type and has locality as
Cressbrook, Moreton Bay; it has damaged feet and
ventral surface; BM26.3.1 1.194, Holotype of sinualis ,
Groote Eylandt: measurements of this juvenile with M 3
erupting were not made except for Mj_ 3 which is 3-7; skin
is in good condition. BM76.3.29.2, Peak Downs, Qd.
BM91. 6.28.1, N.S.W. BM25.8.1.133, locality?.
BM75.14. 1.23.5-6, Gin Gin, Qd).

P. novaeguineae (BM 73.145, Mt Eriama, about 10
miles (16 km) from Port Moresby and 8 miles (13 km)
from Rouna, Papua).

P. ingrami (BM6.3.9.77, Holotype, Buchanan,
Alexandria, N.T.: measurements: BL, 1 7 0; ZW, 9-4;
OBW, 7-2; IBW, 2-3; C 1 M 4 , 6-0; M'“ 4 , 3-5; M 1 3 , 3-1;
R-LM 3 , 5 7; IO, 3-9; IPVD, 3-9; DL, 12-9; I,-M 4 , 6-8;
Mj 4 , 3-9; Mi_ 3 , 3-0; C-AP, 3-7; C-AR, 3 7; SD, 3 6;
bullar length, 4-9; nasal length, 6-7; maximum nasal
width, 2-8; minimum nasal width, 1-0; pmx-nasal suture,
2-5; FM, 3-7; skull represents very old individual with
worn teeth and lacks RM 2 although loss occurred during
life; skin is in good condition. BM6.3.9.76, BM6.3.9.78,
Buchanan, Alexandria, N.T.. BM6.3.9.79, Bluff Hole,
Alexandria, N.T.. BM6.3.9.80, Alexandria, N.T..

BM25.4.9.8, the subtilissima form, Derby, W.A.).

I am grateful to the C.S.I.R.O. Endowment fund
for making the trip possible and to Drs K.
Koopman, S. Anderson, and Mr H. Van Deusen
(American Museum Natural History) and Dr I.
Bishop and Mr J. Hill (British Museum, Natural
History) for allowing me to study specimens in
their respective institutions.

Abbreviations: BL, basicranial length; ZW, zygomatic width; OBW, outside bullar width; IBW, inside bullar width;
FM, maximum width foramen magnum; IO, minimum interorbital width; VV, inter-palatal vacuity distance; DL,
dentary length; C-AP, articular condyle to anterior border of ascending ramus; C AR, articular condyle to tip of
angular process; NL, nasal length; NW, maximum nasal width; NWMN, minimum nasal width; PN, nasal-
premaxillary suture length; PLF-AB, bullar length from posterior lacerate foramen to anterior end alisphenoid
tympanic wing; SD, skull depth; N, number of specimens in sample; x ±t, sample mean ± one standard error; O.R.,
observed range; s, standard deviation; CV, coefficient of variation.

ARCHER: REVISION OF PLANIGALE

361

TABLE 1: Absolute Measurements in Species of Planigale

N

F

+1

IX

OR s

CV

N

+1

IX

OR

s

CV

Planigale novaguineae

P.

maculata (total)

BL

2

21.4+.98 20.0-22.7 1.38

6.46

38

18.3+.14

16.4-

20.2

0.86

4.67

ZW

2

12.3+.75 :

LI . 5-13.1 1.06

8.64

40

10.6+.10

9.3-

12.0

0.66

6.24

OBW

2

8.1+.57

7.6- 8.5 0.80

9.88

40

7.3+.07

6.7-

8.3

0.41

5.68

1 BW

2

3 . 2+ . 4 6

2.9- 3.5 0.65

20.35

38

2.6+.04

2.1-

3.1

0.25

9.60

tf-M 4

2

8.3+.50

7.9- 8.6 0.70

8.52

43

6.9+.05

6.4-

7.6

0.31

4.49

M 1 ' 4

2

4 . 8+ . 33

4.6- 4.9 0.47

9.84

44

4.3+.02

4.0-

4.6

0.15

3.47

M 1 " 3

2

4 . 3+ . 27

4.2- 4.4 0.38

8.73

48

3.9+.02

3.7-

4.2

0.13

3.29

R-LM 3

2

7 . 2+ . 4 6

6.9- 7.5 0.65

9.04

41

6.4+.04

5.8-

7.1

0.29

4.46

FM

2

4 . 3+ . 38

4.1- 4.5 0.53

12.35

37

4.0+.04

3.5-

4.2

0.25

6.02

10

2

4 . 9+ . 33

4.7- 5.0 0.47

9.64

42

4. 2+. 04

3.8-

4 . 8

0.24

5.72

VV

2

S .6+.75

4.8- 6.4 1.06

18.98

42

4.5+.06

3.6-

5.3

0.38

8.46

DL

2

16.6+.78 :

15.7-17.4 1.10

6.61

45

14.0+.13

12.6-

15.7

0.84

5.99

ll-M 4

2

9.6+. S 3

9.2-10.0 0.75

7.83

44

8.1+.06

7.5-

8.9

0.37

4.54

M w

2

S .4+.33

5.2- 5.5 0.47

8.74

44

4.7+.02

4.4-

5.0

0.16

3.33

M 1-3

2

4.1+.33

3.9- 4.2 0.47

11.52

46

3.6+.02

3.3-

3.9

0.16

4.57

C-AP

2

4.7+. 33

4.5- 4.8 0.47

10.05

42

4.1+.04

3.7-

4.6

0.23

5.62

C-AR

2

5.0+.53

4.6- 5.4 0.75

15.03

42

4.2+.04

3.5-

4.7

0.29

6.83

NL

2

9.0+.65

8.4- 9.6 0.92

10.32

40

7.4+.09

6.4-

8.4

0.54

7.29

NW

2

3 . 8+ . 27

3.7- 3.9 0.38

9.88

41

2.9+.04

2.3-

3.5

0.28

9.49

NWMN

2

1.6+.22

1.5- 1.6 0.32

19.76

40

1.3+.02

1.1-

1.6

0.13

10.38

PN

2

3.5+. 35

2.9- 4.0 0.50

14.17

39

3.0+.06

2.3-

3.6

0.37

12.30

PLF-AB

I 2

4 . 7+. 22

4.6- 4.7 0.32

6.73

37

4 . 4+. 04

3.8-

4.8

0.22

4.99

SD

2

6.0+.50

5.6- 6.3 0.71

11.79

41

4.8+.04

4.2-

5.4

0.29

5.96

P .

m.(Qld)

P .

m.(Mt Molloy)

BL

23

18.3+ .19

16.9-20.2 0.91

4.99

7

18.9+.35

18.0-

20.0

0.94

4.97

ZW

24

10.6+1.34

9.3-12.0 0.70

6.57

7

11.1+.22

10.2-

11.9

0.57

5.16

OBW

24

7.2+1.15

6.7- 8.3 0.41

5.65

6

7.4+.14

7.0-

7.8

0.33

4.51

1 BW

22

2.6+1.75

2.2- 3.0 0.21

8.22

6

2.6+.10

2.3-

3.0

0.25

9.69

CLm 4

26

6.9+1.04

6.5- 7.4 0.30

4.28

7

7 . 2+ . 0 6

7.0-

7.4

0.16

2.24

M 1-4

27

4.3+ .68

4.0- 4.6 0.15

3.60

7

4.4+.04

4.3-

4.6

0.11

2.45

M 1 ' 3

31

3.9+ .61

3.7- 4.2 0.13

3.28

9

4.0+.03

3.9-

4.1

0.08

1.98

R-LM 3

25

6.4+ .95

5.9- 7.1 0.30

4.75

7

6.6+.07

6.4-

6.9

0.19

2.83

FM

22

3.9+ .93

3.5- 4.2 0.17

4.36

7

3.9+.08

3.6-

4.2

0.20

5.13

10

25

4.3+1.14

4.0- 4.8 0.24

5.70

7

4.5+.06

4.3-

4.8

0.16

3.58

VV

26

4.5+2.04

3.6- 5.3 0.38

8.37

7

4.6+.13

4.3-

5.3

0.35

7.52

DL

28

13.9+1.49

12.6-15.7 0.85

6.10

7

14.5+.25

13.5-

15.5

0.66

4.53

I 1 -M 4

27

8.1+ .89

7.5- 8.8 0.38

4.64

7

8.4+.05

8.3-

8.6

0.14

1.68

M,-4

28

4.7+ .59

4.4- 4.9 0.15

3.14

7

4.8+.02

4.8-

4.9

0.04

8.33

M 1-3

31

3.6+ .81

3.4- 3.9 0.16

4.50

9

3.7+.03

3.6-

3.8

0.09

2.34

C-AP

26

4.1+1.57

3.7- 4.6 0.26

6.43

6

4.4+.06

4.2-

4.5

0.15

3.36

C-AR

26

4.1+1.79

3.5- 4.7 0.30

7.35

6

4.2+.12

2.9-

4.7

0.31

7.12

NL

25

7.4+1.48

6.5- 8.3 0.55

7.38

7

7.9+.12

7.4-

8.3

0.31

3.97

NW

25

2.9+1.72

2.5- 3.4 0.25

8.58

7

2.9+.06

2.6-

3.0

0.15

5.21

NWMN 25

1.3+2.22

1.1- 1.6 0.14

11.09

7

1.2+.05

1.1-

1.5

0.14

11.75

PN

25

3.0+2.34

2.3- 3.6 0.35

11.72

7

3.3+.13

2.8-

3.6

0.35

10.73

PLF-AB 22

4.3+1.00

4.0- 4.8 0.20

4.67

7

4.4+.03

4.4-

4.6

0.09

2.07

SD

24

4.9+1.05

4.4- 3.4 0.25

5.16

7

5.0+.07

5.9-

5.4

0.18

3.62

P .

m. (Townsville )

P. m.(sinualis form.N.T.)

BL

2

18.7+.57

18.2-19.1 0.80

4.28

10

18.5+.06

17.6-

19.9

0.20

1.10

ZW

2

10.9+.27

10.8-11.0 0.38

3.45

11

1 Q .8+.12

10.1-

11.4

0.39

3.57

OBW

2

7 . 4+ . 22

7.0- 7.8 0.33

4.51

11

7.6+.09

7.1-

8.0

0.30

3.95

1 BW

2

2 . 8+ . 22

2.7- 2.8 0.32

11.29

11

2 .8+. 06

2.3-

3.1

0.21

7.46

CLm 4

2

6 . 8+ , 37

6.6- 7.0 0.53

7.81

11

7.1+. 09

6.6-

7.6

0.29

4.06

M 1 ' 4

2

4 . 4+ . 33

4.2- 4.5 0.47

10.73

11

4. 3+ .'04

4,1-

4.5

0.14

5.35

M 1-3

3

4.0+.12

3.8- 4.2 0.21

5.30

ii

3.9+.04

3.7-

4.1

0.12

3.13

R-LM 3

2

6 . 5+ . 27

6.4- 6.6 0.38

5.78

10

6.4+.04

6.3-

6.6

0.12

1.84

FM

2

3 . 9+ . 22

3.8- 3.9 0.32

8.11

10

3.9+.08

3.5-

4.2

0.24

6.15

10

2

4 . 4+ . 22

4.3- 4.4 0.32

7.19

11

4.3+.08

3.9-

4.7

0.26

5.95

VV

2

4.2+.50

3.8- 4.5 0.71

16.84

11

4.5+.09

4.1-

5.0

0.31

6.96

DL

2

14.4+.33 :

14.2-14.5 0.47

3.28

11

14.2+.16

13.7-

15.5

0.53

3.73

I1-M4

2

8.1+.43

7.8- 8.3 0.60

7.41

11

8.3+.11

7.7-

8.9

0.36

4.33

M1-4

2

4 . 6+ . 27

4.5- 4.7 0.38

8.17

10

4.7+.07

4.4-

5.0

0.21

4.40

M 1-3

3

3.5+.14

3.4- 3.8 0.23

6.69

10

3.7+.06

3.3-

3.9

0.20

6.51

C-AP

2-

4 . 3+ . 33

4.1- 4.4 0.47

10.98

11

4.4+.06

3.9-

4.6

0.20

4.93

C-AR

2

4 . 6+. 33

4.4- 4.7 0.47

10.27

11

4.2+.07

3.3-

4.6

0.23

5.48

NL

2

7 . 8+. 43

7.5- 8.0 0.60

7.69

10

7.6+.13

7.0-

8.4

0.41

5.43

NW

2

3.1+.33

2.9- 3.2 0.47

15.23

10

3.1+.07

2.8-

3.5

0.22

7.06

NWMN

2

1.5+.27

1.4- 1.6 0.38

25.07

10

1.3+.04

1.2-

1.5

0.12

9.38

PN

2

3 . 2+ . 27

3.1- 3.3 0.38

11.75

10

2.8+.06

2.4-

3.6

0.18

6.54

PLF-AB 2

4 . 4+ . 27

4.3- 4.5 0.37

8.55

10

4.6+.14

4.4-

4.8

0.43

9.41

SD

2

5.1+.22

5.0- 5.1 0.32

6.20

11

4.6+.08

4.2-

5.0

0.26

5.52

362

MEMOIRS OF THE QUEENSLAND MUSEUM

Table I (cont’d)

N

XI

1+

OR

s

CV

N

XI

1+

OR

s

CV

P.

maculatafArakun Mission)

P. ingrami (total )

BL

3

17.2+.42

16.4-

17.7

0.72

4.21

11

15.9+.16

14.8-

17.3

0.52

3.25

ZW

3

9 . 9+ . 26

9.4-

10.3

0.45

4.57

14

9.2+.20

8.4-

10.9

0.76

8.26

OBW

3

7.1+.14

6.8-

7.2

0.23

3.30

14

7.0+.10

6.5-

7.9

0.39

5.59

1 BW

3

2.4+.21

2.1-

2.8

0.36

15.00

12

2.4+.09

1.8-

2.9

0.32

13.53

C'-M*

3

6.7+.12

6.5-

6.9

0.21

3.16

15

5. 9+.08

5.4-

6.5

0.31

5.30

M 1 " 4

3

4.2+.04

4.1-

4.2

0.07

1.67

15

3.6+.04

3.2-

3.8

0.16

4.45

M 1 ' 3

3

3.8+0

-

-

-

16

3.3+.04

2.9-

3.5

0.16

4.94

R-LM 3

3

6.0+.12

5.8-

6.2

0.20

3.33

15

5.4+.07

4.8-

6.2

0.29

5.33

FM

3

3.8+.04

3.7-

3.8

0.07

1.84

12

3.6+.04

3.4-

3.8

0.14

3.84

10

3

3.9+.04

3.8-

3.9

0.07

1.79

15

3.9+.06

3.4-

4.2

0.24

6.05

VV

3

4.1+.22

3.8-

4.5

0.38

9.29

14

3.7+.09

3.3-

4.6

0.35

9.57

DL

3

13.2+.25

12.7-

13.5

0.44

3.30

14

11.8+.19

10.4-

13.7

0.71

6.00

ll-M 4

3

7.8+.15

7.6-

8.1

0.25

3.27

15

6.8+.07

6.3-

7.3

0.27

3.95

Mi-4

3

4.6+.04

4.5-

4.6

0.07

1.52

15

4.0+.04

3.7-

4.2

0.15

3.77

Mi- 3

3

3.5+.04

3.5-

3.6

0.07

2.00

15

3.1+.04

2.7-

3.3

0.15

4.78

C-AP

3

4.0+.12

3.8-

4.2

0.20

5.00

13

3.6+.08

3.2-

4.0

0.29

7.93

C-AR

3

4.2+.12

4.0-

4.4

0.21

5.05

14

3.4+.10

2.8-

4.0

0.36

10.72

NL

2

6 . 7+ . 26

6.4-

6.9

0.36

5.31

14

6.3+.14

5.4-

7.5

0.53

8.49

NW

3

2.8+.15

2.6-

3.1

0.25

9.10

14

2.9+.09

2.3-

3.5

0.32

11.07

NWMN

2

1.4+.07

1.3-

1.4

0.10

7.14

14

1.3+.04

1.0-

1.5

0.15

11.87

PN

2

2.9+.07

2.8-

2.9

0.10

3.45

14

2.2+.10

1.7-

2.8

0.36

16.39

PLF-AB 3

4.1+.15

3.8-

4.3

0.25

6.22

13

4.5+.05

4.1-

4.7

0.19

4.16

SD

3

4.5+.10

4.3-

4.6

0.17

3.84

13

3.5+.08

3.2-

4.2

0.30

8.61

P.

i. (Richmond )

R i.(Normanton )

BL

4

15.9+.20

15 . 4 -:

17.1

0.40

2.49

3

16.1+.67

15.0-

17.3

1.15

9.40

ZW

5

9.0+.21

8.5-

9.8

0.47

5.27

4

9 . 8+ . 57

8.4-

10.9

1.14

11.65

OBW

5

7.1+.12

6.8-

7.4

0.28

3.92

4

7 . 2+ . 34

6.5-

7.9

0.68

9.45

1 BW

5

2.2+.14

1.8-

2.6

0.31

14.00

3

2.7+.17

2.4-

2.9

0.29

10.80

cL-m*

5

5.9+.06

5.7-

6.0

0.14

2.39

4

6.2+.13

5.9-

6.5

0.25

4.06

M 1 " 4

5

3.6+.17

3.4-

3.7

0.39

10.75

4

3.8+.04

3.7-

3.8

0.08

2.14

M 1 ' 3

5

3.3+.05

3.1-

3.4

0.11

3.39

4

3.4+.06

3.3-

3.5

0.10

2.94

R-LM 3

5

5.4+.09

3.1-

5.6

0.21

3.93

4

5.7+.16

3.2-

6.2

0.31

5.48

FM

4

3.8+.03

3.7-

3.8

0.06

1.50

3

3.6+.04

3.5-

3.6

0.07

1.94

JO

5

3.8+.06

3.7-

4.0

0.13

3.48

4

3.9+.20

3.4-

4.2

0.40

10.51

VV

4

3.7+.12

3.4-

3.9

0.24

6.59

4

4.0+.24

3.6-

4.6

0.49

12.23

DL

4

11.9+.30

11 . 4 -:

12.8

0.61

5.11

4

12.4+.68

10.9-

13.7

1.36

11.01

I 1 -M 4

5

6.9+.09

6.6-

7.1

0.20

2.90

4

7.1+.20

6.7-

7.3

0.41

5.74

M,-4

5

4.0+.03

3.9-

4.0

0.07

1.75

4

4.2+.04

4.1-

4.2

0.08

1.93

M 1-3

5

3.0+.03

3.0-

3.1

0.07

2.33

4

3.2+.04

3.1-

3.3

0.08

2.54

C-AP

3

3.6+.06

3.5-

3.7

0.10

2.78

4

3.8+.13

3.4-

4.0

0.26

6.96

C-AR

4

3 . 4+ . 22

3.1-

4.0

0.44

12.93

4

3.7+.16

3.3-

4.0

0.31

8.40

NL

4

6.3+.18

5.9-

6.6

0.36

5.70

4

6.8+.12

6.3-

7.5

0.24

3.59

NW

4

2 . 9+ .13

2.6-

3.2

0.25

8.67

4

3.0+.28

2.3-

3.5

0.57

18.83

NWMN 4

1.2+.05

1.1-

1.3

0.10

8.33

4

1.2+.10

1.0-

1.5

0.21

17.33

PN

4

2.3+.11

2.0-

2.5

0.22

9.72

4

2.5+.17

2.0-

2.8

0.34

13.66

PLF-AB 4

4 . 6+. 05

4.5-

4.7

0.10

2.17

4

4.4+.09

4.2-

4.6

0.18

4.15

SD

4

3.4+.10

3.2-

3.7

0.21

6.12

4

3.8+.19

3.3-

4.2

0.37

9.84

R

i. (201km W. of Burketown)

BL

2

15.4+.45

14.8-

15.9

0.78

5.07

ZW

3

8.8+.21

8.5-

9.2

0.36

4.10

OBW

3

6.9+.10

6.8-

7.1

0.17

2.51

1 BW

2

2.1+0

2.2-

2.2

P.

i.(owl pellet, Barkly Tbld)

C’-M 4

3

5.5+.14

5.4-

5.8

0.23

4 . 26

8

5.4+.08

5.3-

5.6

0.11

2.09

M 1 ' 4

3

3.4+.12

3.2-

3.6

0.21

6.24

8

3.4+.04

3.3-

3.6

0.11

3.32

M 1 ' 3

3

3.1+.12

2.9-

3.3

0.20

6.45

8

3.1+.05

2.9-

3.3

0.13

4.22

R-LM 3

3

5.2+.09

3.1-

5.4

0.16

3.04

FM

3

3.6+.09

3.4-

3.7

0.16

4.39

10

3

3.8+.10

3.7-

4.0

0.17

4.55

VV

3

3.4+.07

3.3-

3.5

0.12

3.60

5

3.5+.09

3.3-

3.8

0.20

5.71

DL

3

11.2+.26

10.7-

11.6

0.46

4.09

15

10.9+.08

10.5-

11.5

0.31

2.89

I 1 -M 4

3

6.5+.12

6.3-

6.7

0.20

3.08

15

6.2+.05

6.0-

6.6

0.18

2.83

Mi-4

3

3.9+.09

3.7-

4.0

0.16

4.05

16

3.7+.04

3.6-

4.1

0.15

4.00

M 1-3

3

2.9+.12

2.7-

3.1

0.20

6.90

16

2.9+.02

2.8-

3.1

0.10

3.32

C-AP

3

3.5+.04

3.5-

3.6

0.07

2.00

13

3.3+.04

3.0-

3.5

0.16

4.87

C-AR

3

3.3+.04

3.2-

3.3

0.07

2.12

17

3.1+.04

2.8-

3.4

0.18

4.42

NL

3

5.8+.07

5.7-

5.9

0.12

2.11

NW

3

2.9+.04

2.9-

3.0

0.07

2.41

NWMN

3

1.2+.04

1.2-

1.3

0.07

3.83

PN

3

1.9+.07

1.8-

2.0

0.12

6.42

PLF-AB 3

4.4+.17

4.1-

4 . 7

0.30

6.81

SD

3

3.3+.07

3.2-

3.4

0.12

3.70

ARCHER: REVISION OF PLAN 1 GALE

363

Table 1 (cont’d)

N

+1

IX

OR

s

cv

P.

tenuirostris

BL

8

16.3+.27

15.2-

17.4

0.76

4.67

ZW

8

8.7+.13

8.3-

9.3

0.36

4.10

OBW

9

6.5+.09

6.2-

6.9

0.27

4.11

1 BW

9

2.3+.05

2.1-

2.5

0.15

6.70

C’-M 4

12

6.1+.06

5.8-

6.3

0.20

3.24

M 1 ' 4

12

3.7+.04

3.4-

3.8

0.15

4.00

M 1 ' 3

12

3.3+.03

3.1-

3.5

0.12

3.65

R-LM 3

11

5.5+.06

5.2-

5.8

0.19

3.40

FM

7

3.6+.05

3.5-

3.8

0.13

3.58

10

11

3.7+.04

3.5-

4.0

0.14

3.70

VV

9

3.9+.08

3.5-

4.2

0.23

5.87

DL

12

12.2+.13

11.4-

13.0

0.45

3.71

I 1 -M 4

12

7.1+.08

6.6-

7.4

0.28

4.01

M,-4

13

4.1+.07

3.8-

4.8

0.25

6.06

M 1-3

13

3.1+.04

2.9-

3.8

0.15

4.75

C-AP

11

3.5+.03

3.3-

3.7

0.11

3.11

C-AR

12

3.4+.06

3.1-

3.7

0.21

6.08

NL

11

6.7+.16

5.8-

7.6

0.54

8.06

NW

7

2.7+.09

2.4-

3.2

0.23

8.60

NWMN 10

1.0+.05

0.8-

1.2

0.15

14.53

PN

10

2.4+.06

2.1-

2.7

0.19

7.97

PLF-AB 10

4.2+.08

3.7-

4.5

0.24

5.66

SD

10

3.9+.07

3.6-

4.1

0.21

5.27

P. i

.(subtilissima form)

BL 15.5

ZW 8.8

OBW 6 . 8

IBW 2.2

C’-M 4 5.8

M 1 ' 4 2.5

M 1 ' 3 3.2

R-LM 3 5.2

FM 3.5

10 4.1

VV 3.7

DL 11 . 6

l,-M 4 6.8

M,- 4 4.0

M 3 3.1

C-AP 3.5

C-AR 3.3

NL 6.2

NW 2.9

NWMN 1I3

PN 1.7

PLF-AB 4 . 4

SO 3.5

N

P.

x ± T
gilesi

OR

s

CV

4

18.4+.65

16.6-

19.4

1,31

7.11

3

10.6+.64

9.5-

11.7

1.10

10.40

4

7 . 6+. 23

6.9-

8.0

0.47

6.17

4

2.8+.05

2.7-

2.9

0.10

3.57

5

7.0+.24

6.2-

7.6

0.53

7.56

5

4.4+.14

4.0-

4.8

0.30

6.91

5

3.9+.13

3.5-

4.3

0.29

7.36

5

6.4+.18

5.7-

6.7

0.40

6.20

4

3.8+.07

3.7-

4.0

0.15

3.97

5

4.3+.07

4.1-

4.5

0.16

3.67

5

4.4+. 18

4.0-

4.9

0.39

8.95

5

14.1+.40

12.7-

15.0

0.89

6.33

5

8.2+.14

7.7-

8.9

0.31

3.81

5

4.7+.16

4.3-

5.1

0.36

7.74

5

3.6+.11

3.3-

3.9

0.24

5.95

4

4.1+.18

3.6-

4.3

0.35

8.55

5

4.0+.13

3.7-

4.4

0.30

7.50

5

7 . 7+ . 23

6.9-

8.3

0.50

6.56

5

3.2+.06

3.0-

3.4

0.14

4.41

4

1.4+.12

1.0-

1.5

0.24

17.49

3

2.9+.14

2.8-

3.2

0.23

8.07

4

4.8+.10

4.5-

4.9

0.20

4.17

4

3.9+.04

3.8-

4.0

0.09

2.21

N

x ± T

OR

s

CV

P. i. (fossil Kimberleys)

DL

7

10.6+.12

10.2-11.1

0.32

3.06

li-M<

8

5.9+.10

5.4- 6.3

0.28

4.78

M,-4

9

3.7+.05

3.4- 3.8

0.15

4.05

M 1-3

4

2.8+.05

2.7- 2.9

0.10

3.57

C-AP

6

3.4+.05

3.2- 3.5

0.13

3.94

C-AR

6

3.1+.07

2.9- 3.3

0.17

5.39

P. i.brunneus Holotype P.t.Holotype

ZW

8.9

OBW

6.8

OBW

6.8

1 BW

1.8

1 BW

2.5

C’-M 4

5.7

C’-M 4

6.0

M 1-4

3.4

M 1 " 4

3.5

M w

3.1

M 1 ' 3

3.2

R-LM 3

5.1

R-LM 3

5.5

FM

FM

10

3.7

10

VV

VV

3.5

DL

11.7

DL

12.3

I 1 -M 4

6.6

ti-M 4

6.6

M 1-4

3.9

Ml*

3.9

Mm

3.0

M 1-3

3.0

C-AP

3.5

C-AP

3.5

C-AR

3.1

C-AR

3.5

TABLE 2: Ratios of Cranial, Dental and External Characters in Planigale

364

MEMOIRS OF THE QUEENSLAND MUSEUM

O Tf* -X TP -V <

Z «j

^ QJ

i-T

S i

0)

<u

& °

I a

X |

c/5 cx

a

°

g 6

5 c

G —

S-H r

u H
.G V3

o '5

cfl tJO

G c

■ G JU

<D ^
> O
G o

8 =
■C Hh

G X

rA i

O «

JU o

Oh tt

3 o

cn G

G

$

O

H .

"G

a<

o

25

2

-5 g>

JSi

.j-5*

MZ/-

O

r* tt
r> n

o o

w © w o

i i

vd in vo cn

mm rn n

o o o o

Tp CM
CO • rH *

w O O

■ — TP

o

m »

w o

r*' in

o o

co

o o

o o o o

co co
o o

CO CO

o o

— . Tf « PO — '

f

— » — r

Mz/wa

as/wa

' o s -' O O O —'O w O — O

ro m

o o

oi/as

r- r-

o o

Ge/as

mz/qs

Ur "'" 10

-W/MG

0

1

o »

-Si MZ/MG
X)

C3

H

w O

I

O CO

H CO
£ ®
c/T *5

IS

Q c

w <L> G

in

n j 1 , cn

G -=S cd ™ ♦

g ~ 2 MZ/OI ?
PQ

Hi’S

Z “>

^ co
o o

£ T3

£ g
o C

b o

HZ/GS

Pd 2

o §

h o

5. w
ix o

a |
5 §

o

(N £X
T3 |

a s

ID CM
CO (O

o o

— . r-
CM *

o

o o o o

CO

CO CO

o o

O — CO — o

■*0 w O — '-'rH — r

o o o o o o oo

00 00

o o

O O rH O

IN CM
O O

o m
tp co

o o

Z

i’x O

O rH r-v rH

CO VD
rH CTi

rH O

rH O

CM CM U) U0

HO rH rH HO

os 00
o o

— ^ SO — f

— ' Ol r-» in

o w <

w o v -' O s -' O o

r- so r-

o o o o

VO VO

o o

vo in

o o

— * IN

— CM

— (N — CN r

o o o o

■ m — «

CM CM

o o

o o o o

<N <N

o o

Tf

^ o

o o

o c

^ o
I

CO VO

r-nrH — r^CM CM r^P

vo — -tr

<N

CM *

o

— - Tp

VO •
*— o

r-> -^p
CM *

— * o

—w m
00 *
o

r*. VO
rH in
rH »

'-'O

^P TP

o o

oo oo oo oo

*T r*.

O *^P
rH *
"■'©

CO

o o

oo oo oo oo

o o o o

r-* CO «CO r-^r*. — * os «c

r; & ztz z« zc: zai g«

1X0 IX O IXO 1X0 1X0 1X0 1X0 1X0 1X0 1X0

rH >1

d Uj
o a

^p

w o

1

rH rv

oo oo oo oo

— ^o — , <

f- rH o m

<r» os co as

O O O o

— . VO . — V3 «

O O o o

' o ^ O v — O <

CM CM CM CM

O O O o

rH TP TP —.TP — , f

- O o

1 I

• vo r- r-

o o o o

fO in

’ 1 -' pH — ' r

■. in — . tp — .

| . fO . CM

' o ^ o ^

I I

> O vO in n> r

o o o o

r-* r— * r*'

2 ^
i^< o

VO VO

o o

a o

VO VO

o o

CM CM

o o

r*

Tp TP

o o

z qj z a; z pi
i X d I X d 1X0

rn m

o o

os os
o o

TT rH

cn os
o o

!n in
o o

<N CM

o o

2 «

ix o

Oif

0*1

ARCHER: REVISION OF PLANIGALE

365

H/1S

eii/an

0 o o

1 } I

(N cm p* *p m

m cm m cm run

o o o o o o

*-»• «-t — *

--o v ~' o o s - o

w ' O o

o o o o

VO

*-*.•-4 —

CN • r4
' O — •

VO l/> in to

— ' o

I

vo *r

— <n

« o
I

in rH

m m
o o

w O o

I I

m cn aj r*

^ m — * m

m * m •

— - o w o

i i

in m m r-j

o o o o

« o w o

t i

m cm vo m

. rn

IN •
o

05 VO

— m
n* •
^ o

o o oooo o o o o o o oo o o o o

eii/Ai,

‘-'rH w O

— 05

m *

— o

— — '-'O

o o o o

r-

o o

p* co

CO P-

o o

^ i— I — pH *H

co * in * <n • n *
*—' iH *■' (H ' iH ■*“ ' H

HO rH *H

iQ/ia

— P*
IN ♦

— O

r- r-
o o

— r-

05 *
— 'O

— p* — p*

oo oo oo oo oo

— p*
05 •

^ o

in m
r- r-

o o

—

05 *

—• o

m <n
p* r-

o o

— p*
in •
o

— p-
in *
^ o
I

in m

r- p*
o o

p* r-
o o

'IQ/AA

’—o w o

I I

vo tp m iH

IN IN CM (N

o o o o

IK/Mil

TN/flN

MZ/MN

w o '-'o

*p *p tp m

o o o o

— » m
IN •
O

rH O

m m
o o

— . m — <

— * cm —

o o o o

IN CM
o o

> tp ■— n- —

o o o o o o
. in

o o O o

05 CO

-‘M 1 — **

oo • n* ♦

w o —O
I I

H cm <n m

0 ~ o

1 I I

on* vo in o vo

mo mm Mfi

o o o o o o

IN
<n •
w o

O w O '-'O

o o o o

w O — o
I I

oo in vo in

— O o

o o o o o o

IN CM
O O

0

1

O CO

IN CM
o o

IN CM
O O

. IN '■'IN —CM —

— o w o w o

m m
o o

— vn

rH TP
rH •

o

w o

I I

m Tp h vo

Tp TP MCI

o o o o

o o o o

— CM «r

<n cm
o o

CN CM
o o

' O — O O o

mm mm mm mm

oo oo oo oo

Tp ^

o o

« in i

o w o *— o

rH co mi

oo oo oo oo

CN IN

o o

r- p>

TP *P

o o

0

1

in av

TO/1K

— ' o

o o o o

in rp ^ tp — n*

cn • r* • cm *

w O ‘-'O ^ o

1 I I

HCl IN iH CM 05

MCI TT tp t r m

O O O O O o

— TP

in •
*“* o

tp m

o o

mo *h co

o o o o

— o w o

I I

m p- o oo

mm tp m

o o o o

o o

n* tp

o o

o — o

— o

— O — O

r _ T cn • m * cm • r- • w 1 ■ u« - ' ■ ^

=— *W v «H — rH ‘-'iH '-'rH — 1 ‘-'rH w rH rH r

m m

O O

m m
o o

O O O o

C' o

CM •
r rH

— o

oo oo oo oo

— m
m *
v o

c-i

— co
cn •
^ o

VO vo
o o

— r>

•

w o

o o o o

o ~ o

in r-

rH •

*- o

— r-
•— o

— P*
05 *

v O

oo oo oo oo

o o o o

So? 2 o2 S 05 ZP5 S« So5 305 SpI S05 20* 2«

ixo ixo (xo ixo ixo ixd ixo ixo ixo ixo ixo

ix d

5°:

o

2 05 2 05

1X0 1X0

2 05

ix o

2 05

ix d

WAM3432

Tambrey, W.A. 0.82 0.74 0.40 0.32 0.47 0.34 0.22 0.97

366

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 43

A D, PlanigaJe maculata (typical form). J16477, adult, Mt
Molloy, Qd. A, x 51. B-D, x 3 7.

ARCHER: REVISION OF PLAN IG ALE

Plate 43

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 44

A-D, Planigale maculata ( sinualis form). WAM M8095,
adult, Humpty Doo, N.T. A, x 5-6. B-D, x 3-8.

ARCHER: REVISION OF PLANIGALE

Plate 44

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 45

A-D, Planigale novaeguineae. J4368, adult, New Guinea. A,
4-6. B-D, x 3 6.

ARCHER: REVISION OF PLANIGALE

Plate 45

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 46

A-D, Planigale ingrami {typical form), JM824, adult,
Richmond, Qd. A, x 6-8. B-D, x 4-7.

ARCHER: REVISION OF PLANIGALE

Plate 46

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 47

A-D, Planigale ingrami {subtilissima form). WAM M2846,
Ord River area, W.A. A, x 6-8. B-D, x 4-7.

ARCHER: REVISION OF PLANIGALE

Plate 47

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 48

A-D, Planigale tenuirostris. J3096, adult, Pittsworth, Qd. A,
x 6-2. B D, x 41.

ARCHER: REVISION OF PLANIGALE

Plate 48

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 49

A-D, Planigale gilesi. J21973, adult, Durrie Stn, nr Birdsville,
Qd. A, x 5-9. B-D, x 4-4.

ARCHER: REVISION OF PLANIGALE

Plate 49

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 50

A- D, Planigale sp. WAM M3432, adult, Tambrey, W.A. A,
5-6, B-D, x 4 0.

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 51

A B, Planigale ingrami (the subtilissima form), Ord River
area, W.A. C-D, Planigale maculata (the typical form), Qd.

ARCHER: REVISION OF PLAN 1G ALE

Plate 51

Mem. QdMus. 17(3): 367-71, pi. 52. [1976]

PHASCOLARCTID ORIGINS AND THE POTENTIAL OF THE SELENODONT
MOLAR IN THE EVOLUTION OF DIPROTODONT MARSUPIALS

Michael Archer

Queensland Museum

ABSTRACT

Perameloids are regarded as ancestral to diprotodonts. Of known diprotodonts, the selenodont
forms are structurally the best ancestors for the group. Dental and some cranial similarities
between perameloids and selenodont diprotodonts are marked and indicate that bunodont
diprotodonts such as burramyids are specialized derivatives of selenodont forms.

The majority of diprotodonts may be allocated into one of four groups based on dental
morphology. Selenodont diprotodonts are probably monophyletic although two lineages can be
recognized. Bunodont diprotodonts are almost certainly polyphyletic and contain forms with
secondarily simplified molars. Ektopodont diprotodonts are monophyletic. Lophodont dipro-
todonts may be either polyphyletic or monophyletic.

Ektopodont diprotodonts have developed a type of lophodonty that is also partly developed in
some phalangerids. This is achieved by a marshalling into rows of crenulations and conules. These

transverse rows function as lophs and indicate
lophodonty could have been achieved.

It is commonly believed that bunodont bur-
ramyids such as Cercartetus are structurally the
most primitive living diprotodonts (e.g. Tyndale-
Biscoe 1973). Phascolarctos and other selenodont
marsupials are regarded as specialized forms which
probably evolved from ancestral bunodont dipro-
todonts. This view has been adopted in part
because of the well-known secondary development
of selenodonty in many eutherian groups (such as
the origin of selenodont perissodactyls from bun-
odont condylarths) and in part because Phascol-
arctos and all other selenodont marsupials are
assumed to be highly specialized leafeaters whereas
bunodont diprotodonts are omnivores.

Recent basicranial (Archer 1976a) and dental
investigations (Archer 1976b) of marsupicar-
nivores and perameloids have led to an alternative
hypothesis presented here that selenodont dipro-
todonts evolved directly from perameloids and that
they, not the bunodont diprotodonts, are struc-
turally the most primitive.

The origin of marsupial lophodonty also re-
quires new consideration in view of recently
discovered Miocene diprotodonts. Ride (1971) has
suggested an ingenious hypothesis for the origin of
marsupial lophodonty which differs from the
traditional view of Bensley (1903). The idea

one of perhaps three ways in which marsupial

presented below is yet a third way in which the
evolution of lophodonty in marsupials may have
occurred.

Terminology of teeth is shown in Plate 52 and
follows in part that used by Archer (1975a, 1975b).
Basicranial terminology is that given by Archer
(1976a). Family names follow the usage of Kirsch
(1968).

Perameloids and their Relationship to
Diprotodonts

Ride (1964) divides Australian marsupials into
three orders: Marsupicarnivora, including dasy-
urids and thylacinids; Peramelina, including only
the superfamily Perameloidea which contains per-
amelids and thylacomyids; Diprotodonta, includ-
ing all ten families of Australian diprotodonts.

Molars of perameloids differ from those of
marsupicarnivores mainly in having a very large
stylar cusp C which is subequal in size to stylar cusp
D, and in lacking a crest which directly links the
paracone and metacone. Lacking this crest, the
posterior protocrista links with stylar cusp C and
the anterior metacrista links with stylar cusp D,
thereby providing a transverse valley through
which the hypoconid passes from the protoconal

368

MEMOIRS OF THE QUEENSLAND MUSEUM

basin to the buccal side of the tooth. The lower
molars of perameloids differ from those of most
marsupicarnivores in having a relatively high
talonid and, in all groups (but particularly so in
thylacomyids), a reduced paraconid.

The ways in which perameloid molars differ from
those of marsupicarnivores are also the ways in
which they are similar to molars of selenodont
diprotodonts. Winge (1941) believes that the
phascolarctid molar is structurally primitive
among diprotodonts. The phascolarctid upper
molar may easily be seen as a slightly modified
perameloid molar. The modifications required to
transform the upper molars of Perameles (a
structurally ancestral peramelid) into upper molars
of phascolarctids would be a reduction in size of the
stylar cusps with approach of stylar cusp C to stylar
cusp D, enlargement of the paracone and hypocone
(the modified metaconule), greater development of
the anterior and posterior cingula, and an increase
in size of the metacone of M 4 . All of the principal
shearing crests are comparable in the two groups.
To similarly transform the lower molars, the
perameloid paraconid must be reduced, the crista
obliqua must intersect and connect to the para-
cristid, and the paracristid must not contact the tip
of the metaconid.

Perameloids, like diprotodonts but unlike mar-
supicarnivores, are syndactylous. For this reason
perameloids (although polyprotodont) are re-
garded by most authors (e.g. Osgood 1921, Ride
1964) as the group most likely to have been
ancestral to diprotodonts. Opponents of this view
must hold that syndactyly has developed at least
twice, once in perameloids and at least once in
diprotodonts (Thomas 1 888, Kirsch 1 968). There is
no evidence for this (Jones 1924) and the only
recent examination of syndactyly (Marshall 1972)
has failed to provide reasons for regarding syn-
dactyly to have evolved more than once. It is to be
expected that if perameloids are ancestral to
diprotodonts, traces of this ancestry might be
evident in the teeth and basicrania of structurally
ancestral diprotodonts.

Structurally Ancestral Diprotodonts

Burramyids include living forms which are
generally regarded (e.g. Thomas 1888, Troughton
1967, Tyndale-Biscoe 1973) as most closely re-
sembling hypothetical ancestral diprotodonts.
These authors refer to similarity in molar mor-
phology to some marsupicarnivores such as dasy-
urids and also to their low chromosome number.
This similarity consists of the subtriangular shape
of the burramyid upper molar which lacks or has

only a poorly-defined hypocone. Bensley (1903)
regards these forms as indicative of an intermediate
condition between ancestral tribosphenic mar-
supials which lack the hypocone and more advan-
ced phalangerids which have well-developed hy-
pocones. If burramyids are ancestral to phascolarc-
tids, the latter must have redeveloped a complete
stylar shelf as well as a majority of the shearing
crests which, although absent in burramyids, are
present in perameloids and marsupicarnivores.
Alternatively, the burramyid condition could be a
simplification of a more complex morphology such
as that of phascolarctids and some petaurids, a
view held by Winge (1941) and accepted here.

Of all ‘possum’ groups (phalangerids in the sense
of Ride 1 964; there is no corresponding taxon in the
more recent classification of Kirsch 1968) the
closely related phascolarctids and vombatids are
also the only ones to have a variably developed
alisphenoid-frontal contact on the side of the
braincase, a feature found in all perameloids
(Archer 1975b summarizes the distribution of this
character in marsupials). Further, the 2N chromo-
some number of perameloids (not including thyla-
comyids) and vombatids is 14, and of phascolarc-
tids 16, 14 being regarded as structurally primitive
among marsupials (Sharman 1974). Serologically
(Kirsch 1967, 1968), phascolarctids and vombatids
are closely related, but perameloids group with
dasyurids. However, the serological distance bet-
ween phascolarctids and perameloids may be the
result of relatively rapid protein evolution in
phascolarctids (and vombatids).

At present, available evidence suggests phascol-
arctids (and possibly other selenodont forms) are
the group best regarded as structurally ancestral to
other diprotodonts.

Diversity of Molar Patterns among
Diprotodonts

With the exception of Tar sipes whose dental
morphology is deceptively simple, presumably the
result of degeneration from a more complex
ancestral pattern, all diprotodont molars may be
categorized as being either selenodont, bunodont,
lophodont, or what may be referred to as ekto-
podont. With the exception of the ektopodont
pattern (Plate 52), these types are figured by
Bensley (1903).

Selenodont Diprotodonts: Selenodont forms
include all phascolarctids ( Phascolarctos , Per-
ikoala , Pseudokoala , Litokoala ) some petaurids
( Pseudocheirus , Hemibelideus , and Schoinobates )
and possibly vombatids ( Vombatus , Lasiorhinus,
Phascolonus, Rhizophascolonus ) to judge from

ARCHER: PHASCOLARCTID ORIGINS

369

unworn teeth. It has been suggested (e.g. by Kirsch
1968) that phascolarctids and selenodont petaurids
represent separate lineages, the development of
selenodonty in the two lineages possibly being the
result of convergence. Bensley (1903), and Turn-
bull and Lundelius (1970) point out significant
differences in molar form in the two groups.
However, it is also possible that the two groups had
a common selenodont ancestor, differences noted
in the modern representatives being nothing more
than specialization developed later by each group.
Bensley (1903) regards the Phascolarctos molar
pattern as a derivative of the Pseudocheirus con-
dition, while Winge (1941) interprets a structural
trend which goes from Phascolarctos to Pseu-
docheirus. Kirsch (1967, 1968) has shown that
serologically Phascolarctos groups with vombatids
rather than the other selenodont forms which
group with the remaining diprotodonts. If phascol-
arctids (and vombatids) have undergone rapid
protein evolution relative to other diprotodonts,
their serological uniqueness could obscure re-
lationships that may exist with other selenodont
diprotodonts. Sharman (1974) notes that because
Phascolarctos has 2N=16 chromosomes, it is
closer to the assumed primitive number of 14, and
differs from other selenodont diprotodonts which
range from 20 to 22. The significance of this is
difficult to interpret in view of the fact that within
one family (the macropodids), the range is 10 to 32.

In view of the generally held notion (Troughton
1967, Ride 1970) that selenodont diprotodonts are
strictly herbivorous, it seems appropriate to point
out here that Common Ringtails {Pseudocheirus
peregrinus ) held in captivity by the author in-
variably show a decided preference for insects if
given a choice between these and any type of leaf or
fruit.

Bunodont Diprotodonts: Bunodont forms
include some petaurids ( Petaurus , Gymnobelideus,
and Dactylopsila ), burramyids ( Acrohates , Dis-
toechurus , Cercartetus, and Burramys), phalan-
gerids ( Trichosurus , Wyulda, and Phalanger ), pot-
oroine macropodids ( Hypsiprimnodon , Bettongia,
Caloprymnus, Aepyprymnus, Potorous, and Pro-
pie opus), and thylacoleonids ( Thylacoleo and Wak-
eleo ). Bensley (1903) regards the more tri tubercular
forms such as Distoechurus to be structurally
ancestral to other bunodont diprotodonts, con-
sidering the absence of a hypocone to be struc-
turally primitive. He also suggests that bunodont
and selenodont forms may have been independent-
ly derived from tritubercular (Bensley’s hypothet-
ical properamelid) ancestors. This seems doubtful
considering that selenodont and bunodont dipro-

todonts have many characters in common such as
diprotodonty, reduced upper incisor number,
wrinkled enamel, fasciculus aberrans, serological
characters, and highly modified basicranium in-
volving fusion of the ectotympanic (although
fusion does not occur in some Phascolarctos) which
are not present in known tritubercular groups. It
seems more reasonable to regard selenodont and
bunodont diprotodonts as having been derived
either from one another or from other diprotodont
ancestors, rather than independently derived from
tritubercular ancestors. Winge’s (1941) view, that
bunodont forms were derived from selenodont
forms, is accepted here because of the presence of
traces of selenodonty and the common occurrence
of wrinkled enamel in bunodont diprotodonts.
This interpretation implies that the more tri-
tubercular and less selenodont forms such as
Distoechurus are in fact highly specialized forms,
and not, as Bensley (1903) believes, structurally
primitive.

The unity of the bunodont diprotodonts is very
doubtful and several independent origins, possibly
from selenodont forms of different sorts, are
probable. Bunodont non-macropodids have a 2N
chromosome number of 14 to 20 (Sharman 1974).
Phalanger , regarded here as a structurally primitive
bunodont form, has 14 but so do burramyines
which are regarded here as structurally advanced.
Kirsch (1967, 1968) also regards the bunodont
forms to represent several distinct serological
groups.

Ektopodont Diprotodonts: Ektopodont

forms are represented by the late Miocene species
Ektopodon serratus (Stirton, Tedford and Wood-
burne 1967). They are characterized in part by
transverse serrate ridges formed by numerous small
upside down V-shaped longitudinal crests. Ekto-
podon was originally described as a possible
monotreme, but Woodburne (pers. comm.) sug-
gests it is a diprotodont following the discovery of
an older and simpler species.

Lophodont Diprotodonts: Lophodont forms
include macropodine macropodids and diprotod-
ontids. Traditional views of the origin of lophod-
onty from bunodonty (such as proposed by Bensley
1903) involve evolution of crests or lophs which
link the protocone to the paracone and the
hypocone to the metacone. In lower molars it is
generally assumed that the paracristid and para-
conid become reduced, the metacristid develops as
the anterior lophid, and the hypocristid develops as
the posterior lophid. The crista obliqua becomes
the midlink of macropodines.

370

MEMOIRS OF THE QUEENSLAND MUSEUM

The importance of Ektopodon in the present
context is that it demonstrates that diprotodont
lophs may not be homologues of the crests of other
marsupials. Thomas (1888, p. 193) notes that in
some species of Phalanger the molars have distinct
transverse ridges. These transverse ridges could be
regarded as incipient lophs. Close inspection of
unworn molars reveals a striking similarity to
molars of Ektopodon. The buccal half of the
transverse ridges of upper molars and the lingual
half of the ridges of lower molars appear to consist
of numerous upside-down V-shaped longitudinal
crests. The lingual half of the transverse ridges of
upper, and the buccal end of the transverse ridges
of lower molars consist of short steep-sided ridges
which appear to be the homologues of ridges in
these positions of molars of Phascolarctos. The
remainder of the crown surface of Phalanger
molars are covered in small wrinkles and crenu-
lations, as are the teeth of selenodont and many
bunodont diprotodonts. This suggests the possi-
bility that lophs may have evolved through a
marshalling together of wrinkles, conules and small
ridges already present in ancestral selenodont

vombat ids
phascolarctids

AT-.

selenodont

diprotodontids

potoroi nes

macropodines

► ^

\ potoroines

phalangerids

ektopodontids
tarsiped ids

burr^imyids

R i d e’s
1971 '

peramel ids jr
hypothesis ♦ , . ,

• perameloids

r t hy lacoleonids

thylacomyids

thylacinids

dasy u r ids

I

didelphids

Fig. 1: Some traditional concepts of origins for Aus-
tralian marsupial families. The two macropodid sub-
families are treated separately. These concepts are
based largely on Bensley (1903) with modifications
suggested by later authors. Wynyardiids are not shown
because their teeth are unknown. Dashed lines indicate
alternative origins, such as Ride’s (1971) hypothesis for
the origin of macropodids.

forms. Overriding this organization in Phalanger ,
as in most bunodont forms, are the remnants of
selenodont crests, now modified to form triangular
buttresses at the ends of transverse lophs. This
transformation is as readily performed on lower as
it is on upper molars.

In completely lophodont forms, the lophs are not
clearly modified ancestral ridges. It is possible that
marsupial lophodonty evolved more than once and
in very different ways. Ride’s (1971) interesting
hypothesis for origin of macropodine molars
assumes that a tritubercular pattern was ancestral
to the lophodont pattern, that stylar cusps became
the buccal ends of the upper lophs, and the
paracone and metacone were incorporated along
the length of the lophs. An alternative is that the
lophodont molar has been derived from a sel-
enodont molar in the manner outlined above.
Lateral selenes became triangular buttresses (hom-
ologues of which occur in many macropodids), and
transverse lophs were formed by a marshalling of
conules, ridges and wrinkles. The already well-
developed hypocones, reduced paraconids, and
enlarged M4 of selenodont forms could have been
characters directly utilized by ancestral lophodont
forms. If this latter hypothesis for the origin of
lophodonty is accepted, bunodont potoroine mac-
ropodids could be regarded as derivatives of
lophodont macropodines, a conclusion also accep-
ted by Ride (1971).

diprotodontids

Fig. 2: Concepts of origin suggested in the present work.
Alternatives are indicated by dashed lines.

ARCHER: PHASCOLARCTID ORIGINS

371

CONCLUSIONS

Various views of descent noted above are
contrasted in Figs. 1-2. It is suggested here (as
shown in Fig. 2) that selenodont diprotodonts
arose directly from peramelids and were the base
stock for all other diprotodont radiations.

The relationship between ektopodontids and
other diprotodonts is unclear. They exhibit a
transverse lophodont molar pattern which, al-
though possibly convergent on other lophodont
forms and derived from phalangerids, indicates a
unique way in which lophodonty could be de-
veloped from selenodonty.

It is not clear how lophodonty was achieved in
macropodids or diprotodontids. It could have
developed from either the bunodont, ektopodont,
or selenodont pattern. Ride (1971) suggests the
additional possibility that lophodonty developed
as a modification of a more or less tribosphenic
pattern.

ACKNOWLEDGMENTS

The author has benefited by discussion on this
subject with Drs J. A. W. Kirsch, M. O. Wood-
burne, and A. Bartholomai. Dr A. Bartholomai
and Mr B. Campbell kindly read and criticised a
draft of this manuscript.

LITERATURE CITED

Archer, M., 1975. The development of premolar and
molar crowns of Antechinus flavipes (Marsupialia,
Dasyuridae) and the significance of cusp ontogeny in
mammalian teeth. J. Roy. Soc. West. Aust. 57 :
118-25.

1975b. Ningaui, a new genus of tiny dasyurids (Mar-
supialia) and two new species, N. timealeyi and N.
ridei , from arid Western Australia. Mem. Qd Mus.
17 : 237-49.

1976a. The basicranial region of marsupicarnivores
(Marsupialia), inter-relationships of carnivorous
marsupials, and affinities of the insectivorous per-
amelids. J. Linn. Soc. Lond ., in press.

1976b. The dasyurid dentition and its relationships to
that of didelphids, thylacinids, borhyaenids (Marsu-
picarnivora) and peramelids (Peramelina, Mar-
supialia). Aust. J. Zool. Suppl. Series 39 , in press.

Bensley, B. A., 1903. On the evolution of the Australian
Marsupialia; with remarks on the relationships of
the marsupials in general. Trans. Linn. Soc. Lond.,
Zool. (2) 9 : 83-217.

Jones, F. W., 1924. The mammals of South Australia.
Part II. The bandicoots and the herbivorous mar-
supials.’ Pp. 133-270 (Govt. Print.: Adelaide).

Kirsch, J. A. W., 1967. ‘Comparative serology of
marsupials.’ (Ph.D. thesis, University of Western
Australia).

1968. Prodromus of the comparative serology of
Marsupialia. Nature, Lond. 217 : 418-20,

Marshall, L. G., 1972. Evolution of the peramelid
tarsus. Proc. Roy. Soc. Viet. 85 : 51-60.

Osgood, W. H., 1921. A mongraphic study of the
American marsupial, Caenolestes. Publ. Field Mus.
(Zool. Ser.) 14 : 1-156.

Ride, W. D. L., 1964. A review of Australian fossil
marsupials. J. Roy. Soc. West. Aust. 47 : 97-131.

1970. ‘A guide to the native mammals of Australia.’ Pp.
xiv and 249. (Oxford University Press: London).

1971. On the fossil evidence of the evolution of the
Macropodidae. Aust. Zool. 16 : 6-16.

Sharman, G. B.. 1974. Marsupial taxonomy and phylo-
geny. Aust. Mammal. 1 : 137-54.

Simpson, G. G., 1945. The principles of classification and
a classification of mammals. Bull. Amer. Mus. Nat.
Hist. 85 : xvi and 350.

Stirton, R. A., Tedford, R. H., and Woodburne, M.
O., 1967. A new Tertiary formation and fauna from
the Tirari Desert, South Australia. Rec. S. Aust.
Mus. 15 : 427-62.

Thomas, O., 1888. ‘Catalogue of the Marsupialia and
Monotremata in the collection of the British
Museum (Natural History).’ Pp. xiii and 401.
(British Museum (Natural History): London).

Troughton, E., 1967. ‘Furred animals of Australia.’ Pp.
xxxii and 384. (Angus and Robertson: Sydney).

Turnbull, W. D. and Lundelius, E. L., 1970, The
Hamilton fauna, a late Pliocene mammalian fauna
from the Grange Burn, Victoria, Australia. Fiel-
diana : Geol. 19 : 1-163.

Winge, H., 1941. ‘The interrelationships of the mam-
malian genera.’ Vol. I. Pp. xii and 4 18. (C. A. Reitzels
Forlag: Kobenhavn).

372

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 52

A. Dendrolagus lumholtzi, a lophodont macropodid. RM 2 (x

4-5) and RM 2 (x 4-5) scanning electron microscope

stereophotographs.

B. Ectopodon serratus , an ectopodontid. Shown as ?RM
(reversed and modified from Fig. 7 of Stirton, Tedford and
Woodburne 1967).

C. Phalanger nudicaudatus, a bunodont phalangerid. RM 2 (x

41) and RM 2 (x 4-2) scanning electron microscope

stereophotographs.

D. Phascolarctos cinereus , a selenodont phascolarctid. RM 2 ( x
3*1) and RM 2 (x 3*2) stereophotographs.

E. Perameles bougainville, a tribosphenic peramelid. RM 2 (x

7*9) and RM 2 (x 8*1), scanning electron microscope

stereophotographs.

F. Didelphis virginiana, a tribosphenic didelphid. RM 2 (x 4*1)
and RM 2 (x 3-8), stereophotographs.

Abbreviations: acd, anterior cingulid; aecd, anterior entocristid; ahyc, anterior hypocrista; amec, anterior metacrista;
aprc, anterior protocrista; co, crista obliqua; end, entoconid; hid, hypolophid; hy, hypocone; hycd, hypocristid; hyd,
hypoconid; hyld, hypoconulid; mcl, metaconule; me, metacone; mec, metacrista; meed, metacristid; med, metaconid;
mel, metaloph; mst, metastylar corner of tooth; pa, paracone; pac, paracrista; pacd, paracristid; pamed, para-
metacristid; peed, posterior entocristid; phyc, posterior hypocrista; pmed, posterior metacristid; ppre , posterior
protocrista; pr , protocone, prd, protoconid; prl, protoloph; prld, protolophid.

ARCHER: PHASCOLARCTID ORIGINS

Plate 52

Mem. QdMus. 17(3): 373-7, pi. 53. [1976]

THE GENUS W ALLABIA TROUESSART (MARSUPIALIA: MACROPODIDAE) IN
THE UPPER CAINOZOIC DEPOSITS OF QUEENSLAND

Alan Bartholomai

Queensland Museum

ABSTRACT

The previously monotypic genus Wallabia Trouessart, 1905, is re-defined to include a fossil
species, W. indra (De Vis, 1895) from the Upper Cainozoic fiuviatile deposits of the Darling
Downs area, southeastern Queensland. ‘ Halmaturus ’ vishnu De Vis, 1895, is shown to be
synonymous with W. indra.

The present study comprises part of an overall
revision of the fossil macropodids from the Pleis-
tocene fiuviatile deposits in the eastern Darling
Downs and the Chinchilla Sand of Late Pliocene
age in the west of this area. The majority of the
larger grazing macropodines were referred by De
Vis ( 1 895) to Halmaturus Illiger, a junior secondary
hymonym of Macropus Shaw. Bartholomai (1967,
1973a, 1973b, 1975) has shown that the material
included representatives of the genera Troposodon
Bartholomai, 1967, Protemnodon Owen, 1874,
Fissuridon Bartholomai, 1973 and Macropus Shaw,
1790, in addition to material which is here referred
to Wallabia. This material is relatively uncommon
compared with other large grazing elements in the
faunas represented, and the available sample is too
small to permit statistical evaluation of the pop-
ulation from which it was drawn.

All measurements throughout are in millimetres.

Genus Wallabia Trouessart, 1905

Wallabia Trouessart, 1905, p. 834 (type species Kan-
gurus bicolor Desmarest, 1804 = Kangurus ualabatus
Lesson, 1827, by subsequent designation of Iredale
and Troughton, 1934).

In addition to designating Wallabia bicolor
(Desmarest) as the type species for the genus
Wallabia , Iredale and Troughton (1934) followed
Trouessart (1905) in listing all of the extant,
medium-sized brush wallabies within this genus.
These species previously were regarded mainly
within the genus Macropus. No attempt was made
to take into account any related fossil species, but
later, Troughton (1937, 1957) indicated the generic

distinctness of species now referred to Pro-
temnodon by Bartholomai (1973a). Troughton
presented no convincing arguments to support this
separation. Tate (1948) used Protemnodon widely
for both living and fossil wallabies, while Stirton
(1963) separated living and fossil representatives,
using Wallabia for living species and Protemnodon
for extinct forms.

Bartholomai (1973a, 1975) has introduced cran-
ial morphological differences, results of compara-
tive reproductive physiological studies by Sharman
et al. (1966) and chromosome studies by Sharman
(1961) to indicate the distinctness of Wallabia
bicolor , the only living species referred to the genus,
from all other living wallabies. Research by Kirsch
(1968) on marsupial haemoglobin suggests that the
species of wallabies and kangaroos, including W.
bicolor , Megaleia and Lagorchestes are closely
associated, but Bartholomai (1975) indicates that
this cannot be verified from the fossil record
because of general deficiencies in the known fossil
samples available.

Calaby (1966) concludes, on the basis of be-
haviour and distinctive dental characters, that
Wallabia should be recognized as a monotypic
genus. This conclusion is supported here for living
forms. Only a single species is added from the
known fossil record.

Generic Diagnosis: Medium sized macro-
podids; cranium with rostrum little deflected;
diastema short; premaxillae relatively narrow in
occlusal view; infraorbital canal very short; for-
amen ovale open, less anterolaterally directed than
in Macropus ; additional foramina usually present
in alisphenoid bulla, lateral to, connected but well

374

MEMOIRS OF THE QUEENSLAND MUSEUM

separated from foramen ovale, presumably for
mandibular branch of trigeminal nerve; basioc-
cipital very broad between petrosals; anterior
margin of basioccipital with reduced elevation;
basisphenoid slope low; dorsal margin of supraoc-
cipital narrowly U-shaped. Palate shallowly ex-
cavate and slightly downturned posteriorly.
Masseteric foramen almost below anterior cheek
teeth. I 3 groove at posterior one-third of lateral
surface; DP 3 and anterior upper molars with strong
ridges from paracone and metacone uniting across
median valley; forelink absent. Lower molars with
relatively low, rectilinear lophids and links; lateral
margins of lophids bulbous; posterior surface of
hypolophid not ornamented.

Wallabia indra (De Vis, 1895)

(Plate 53, Figs. 1-6)

Halmaturus indra De Vis, 1895, pp. 1 12 3, pi. 17, figs.

18, 20.

Halmaturus vishnu De Vis, 1895, pp. 114-6, pi. 17, figs.

3-4.

‘ Halmaturus ’ vishnu De Vis: Stirton, 1959, p. 124;

Bartholomai, 1966, pp. 121-2, pi. 16, figs. 1-3.

‘ Halmaturus ’ indra De Vis: Bartholomai, 1966, pp.

116-7, pi. 15, figs. 4-6.

Material: Holotype, F3595, partial left mandibular
ramus with P 2 M t (unerupted P 3 removed by fenes-
tration, and no longer in the Queensland Museum
collections), Darling Downs, SE.Q., (figd in part, De Vis,
1895, pi. 17, figs. 18, 20; figd Bartholomai, 1966, pi. 15,
figs. 4-6). Preservation suggests derivation from the
Chinchilla Sand, of Late Pliocene age.

Also referred is the lectotype of ‘ Halmaturus ’ vishnu,
F3860, a partial left mandibular ramus with P 3 -M 4 ,
adult. Darling Downs, SE.Q., (figd in part, De Vis, 1895,
pi. 17, figs. 3-4; figd Bartholomai, 1966, pi. 16, figs. 13).
Preservation suggests derivation from the Pleistocene
fluviatile deposits of the eastern Darling Downs.

In addition, 7 juvenile mandibular rami, 8 adult
mandibular rami and one maxillary fragment are referred
from the following localities in the Darling Downs:
Chinchilla; Middle Gully System, Chinchilla Rifle Range
(Rifle Range No. 78, Par. Chinchilla); Dalby at 34-35 feet
(c. 11 m) in a pump well; and from the eastern and
western Darling Downs (particular localities unspecified).

Specific Diagnosis: A relatively large species,
with mandibular symphysis very slightly elevated,
and with diastema relatively short. Lower cheek
teeth low; P 2 with only one set of vertical labial and
lingual ridges transecting crest; P 3 approximately
as long as M 4 , the longest molar; longitudinal crest
nearly straight; anteriorly, base markedly tumes-
cent; crest transected by 3-4 sets of ridges. DP 3
protolophid very narrow at crest, rectilinear, with
forelink descending directly from protoconid;
labial moiety of trigomd basin much reduced.

Lower molars relatively broad with lophid margins
broadly curved from crown base to crests in
anterior view; lophid crests somewhat rectilinear;
midlink moderately poorly developed; posterior
cingulum absent. Upper molars relatively low, with
strong anterior ridge from paracone, broad anter-
ior cingulum, and posterior ridges from metacone
and hypocone of similar strength, uniting well
above crown base, delimiting slight posterior
fossette.

Description: Mandible narrow, rather shallow.
Symphysis not ankylosed, set at very low angle to
base of mandible. Diastema relatively short, geni-
ohyal pit very shallow, below anterior margin of P 3 .
Ventral margin of ramus rounded. Mental foramen
moderately large, oval, well anterior to P 3 , and just
below diastemal crest. Ramus with shallow labial
groove from below P 3 extending posteriorly to
below M 2 and occasionally to below centre of M 3 .
Lingually, broad depression leads posteriorly to
pterygoid fossa. Post-alveolar shelf short, leading
to mesial wall of coronoid process. Masseteric crest
raised to below level of alveolar margin. Angle of
mandible, condyle and bulk of coronoid process
not preserved.

I] unknown.

P 2 relatively short, robust, subovate in basal
outline; longitudinal crest secant, curving slightly
lingually in its posterior extension; transected
mesially by a single set of vertical labial and lingual
ridges, with production of well defined cuspule at
crest; crown basally with labial and lingual tumes-
cences, continuous around anterior margin, with
production of small anterior basal cuspule.

DP 3 molariform, subtriangular in basal outline;
lophids moderately low, with hypolophid crest
much broader than protolophid; protolophid rec-
tilinear but with hypolophid somewhat convex
posteriorly; protoconid positioned above crown
axis. Trigonid basin narrow, extremely poorly
developed labially, short, its length being much less
than distance between lophids. Forelink high,
strong, descending without curvature anteriorly to
point labiad to mid-point of high anterior cin-
gulum, occasionally ornamented labially and lin-
gually by a set of weak accessory ridges; antero-
lingual fossette developed in trigonid basin in
conjunction with slight, variable anterior ridge
from metaconid. Posterior ridge from protoconid
moderately strong, curving labially to unite with
moderately strong midlink, curving antero-
lingually from hypoconid; posterior ridge from
metaconid weak, descending into lingual extremity
of rounded talonid basin; labial moiety of talonid
poorly developed descending at high angle from

BARTHOLOMAI: WALLABIA IN THE UPPER CAINOZOIC

375

TABLE 1: Measurements for Wallabia indra (De Vis), Mandible

Specimen

P 2

dp 3

P 3

M,

m 2

m 3

m 4

F3595*

6-7 x 4-3

7-1 x 4-2

8-5 x 5-6

—

—

—

F3860

— ■

10-1 x 41

7-8 x —

9-4 x 6-5

10-4 x 7-5

10 7 x 7-3

F4743

—

—

- x 4.7

7-6 x 5-7

9-0 x

10-6 x 7-6

112 x 7-5

F4741

—

—

10 3 x 41

7-5 x —

8-5 x —

10-0 x -

10 8 x 7-0

F4746

—

—

— —

—

8-7 x 5-7

9-8 x 6-8

10-0 x 6-8

F4749

—

6-7 x 3-4

—

- x 4-9

—

—

—

F4742

—

—

—

—

—

9-8 x 7-3

10-4 x 7-2

F4747

—

—

i — t.

—

8-5 x 6 0

10 0 x 7-2

10-8 x 7-5

F3597

—

—

—

—

8-8 x 6-1

10 7 x 7-4

10-8 x 7-5

F2496

—

—

—

7-3 x 4-8

8-3 x 5-6

9-8 x 6-2

—

F4753

—

—

8-0 x 5-7

8-9 x 6-5

—

F4751

—

—

7-5 x 51

8-7 x 6-1

—

—

F4752

—

—

—

8-1 x 4-9

91 x 5-9

—

—

F4744

—

—

—

—

9-0 x 6-2

9-6 x 6-9

—

F3601

—

—

—

8-1 x 5-1

8-5 x 6-1

8-5 x 6-1

—

*Holotype

midlink. Anterior ridge from entoconid weak.
Posterior of hypolophid rounded, unornamented,
occasionally with slight posterolabial basal swell-
ing.

P 3 elongate, subovate in basal outline, only very
slightly shorter than M 4 . Longitudinal crest secant,
extremely slightly concave labially, or straight,
transected by three or four sets of vertical labial
and lingual ridges, with production of cuspules at
crest; strength of ridges decreases posteriorly; base
of crown markedly tumescent, produced to form
noticeable cingulum anteriorly.

M, < M 2 < M 3 < M 4 ; molars subrectangular
in basal outline, slightly constricted across talonid
basin in anterior molars, more strongly constricted
in posterior molars; lophids relatively low, almost
rectilinear, with hypolophid somewhat more
convex posteriorly; protolophid narrower than
hypolophid in M, and M 2 , but broader in M 3 and
much broader in M 4 ; lateral surfaces of lophids
markedly convex. Trigonid basin usually very
broad, its length almost equalling distance between
lophids. Forelink low, moderately strong, un-
ornamented, descending anterolingually from pro-
toconid, across labial moiety of trigonid basin,
usually uniting with low anterior cingulum, labiad
to axis of crown; very weak accessory ridge
descends anteriorly from metaconid towards tri-
gonid basin; lingual position of trigonid near
horizontal, labial portion reduced and sloping.
Slight ridge descends posteriorly from metaconid,
occasionally uniting with similar ridge from en-
toconid across lingual margin of talonid basin.
Midlink from hypoconid low, crossing labial

moiety of talonid basin to unite with slight ridge
from protolophid, labiad to axis of crown. Pos-
terior of hypolophid unornamented, occasionally
with swollen base delimited as slight posterior
cingulum.

TABLE 2: Measurements for Wallabia indra (De Vis),
Maxilla

Specimen

M 2

M 3

M 4

F4740

8-6 x 7-4

9-6 x 7-7

9-7 x 7-2

Upper incisors, P 2 , DP 3 , P 3 and M 1 unknown.

M 2 <M 3 <M 4 ; molars subrectangular in basal
outline, somewhat constricted across median val-
ley. Lophs low, with protoloph slightly narrower
than metaloph in M 2 and somewhat broader in M 3
and M 4 , slightly anteriorly curved. Anterior cin-
gulum relatively low, labially subhorizontal, but
lingually ascending slightly, short; forelink not well
defined, but anterior ridge from paracone strong,
ascending slightly lingually to labial margin of
cingulum. Posterior ridge from paracone weak,
ascending into median valley. Midlink low, weak,
curving posterolingually across median valley to
unite with slight ridge from near mid-point of
metaloph; lingual moiety of median valley V-
shaped, labial moiety sharply U-shaped; low ridge
occasionally present across lingual extremity of
valley. Posterior ridges from metacone and hy-
pocone moderately strong, approximately equally
developed, uniting to delimit slight posterior
fossette below axis of crown. Base of crown
somewhat swollen posteriorly and posterolin-
gually.

376

MEMOIRS OF THE QUEENSLAND MUSEUM

Discussion: Bartholomai (1966) redescribed and
refigured the holotype specimen of Wallabia indra
(De Vis), the only specimen referred by De Vis
(1895) to the species. In the same publication,
Bartholomai (1966) selected F3860 as the lectotype
of ‘ Halmaturus’ vishnu De Vis. The holotype
specimen of W. indra has preservation suggesting
its derivation from the Chinchilla Sand of Late
Pliocene age, while the lectotype of ‘FT vishnu
appears to have been collected from Pleistocene
fluviatile deposits of the eastern Darling Downs.
The specimens are not accompanied by accurate
locality data, but material here referred to W. indra
has been derived from both deposits.

Neither the holotype of W. indra nor the
lectotype of ‘FT vishnu is figured herein, as adequate
illustrations are presented in Bartholomai (1966
pi. 15, figs. 4-6, and pi. 16, figs. 1-3, respectively).

Although the samples from these areas are small,
sufficient evidence exists to indicate almost com-
plete overlap in size, and mandibular measure-
ments are presented in Table 1. Morphologically,
no differences are apparent in the material at
present known, and relegation of ‘FT vishnu to
synonymy is considered justifiable. It should be
noted, however, that the complete dentition of the
species is as yet unknown. The single referred
maxilla, measurements for which are povided in
Table 2, is from the Chinchilla Sand, at Chinchilla.

It is usual for material from the eastern Darling
Downs to be specifically distinct from that in the

Chinchilla Sand, but as indicated for the genus
Troposodon by Bartholomai (1967), this is not
without exception. Geologically, the time
difference between the Chinchilla Sand and the
Pleistocene fluviatile deposits does not appear to be
excessive and the occasional occurrence of the same
species in both stratigraphic units is not surprising.

Association of upper and lower jaw fragments
has been effected by considerations of both size and
morphology, and while fragments have not, as yet,
been located together, their present association is
considered correct. No post-cranial remains have
been referred to W. indra although considerable
disarticulated material, derived from small macro-
podine species, is present in the collections of the
Queensland Museum.

Remarkable morphological similarity exists be-
tween W. indra and the extant W. bicolor (Des-
marest), the type species of the genus. Indeed, the
only distinction which can be drawn at this stage is
in regard to the sizes of the species involved. While
it has been shown by Bartholomai (1975), with
respect to M. agilis siva (De Vis), that size
differences may be generally insufficient for even
subspecific distinction, the W. indra sample is too
small to enable statistical evaluation to be under-
taken at this stage. The log difference diagram
shown in Fig. 1, is based on single specimens, but
nevertheless illustrates the similarity in proportions
existing in the cheek teeth of W. indra and W.
bicolor. The similarly elongate nature of P 3 com-
pared with the molars is particularly evident.

LENGTH P 3
BREADTH P 3
LENGTH M,
BREADTH M,
LENGTH
BREADTH
LENGTH M 3

breadth m 3

LENGTH M 4
BREADTH M 4

LOG DIFFERENCE SCALE

OIO 000 0 0 6 0 04 0 02 0 0 02 0 04 0 0 6 0 00 010 012 014 016 018 0-20 022 0 24 026 0 28 0 30

Fig. 1: Log difference diagram illustrating proportional relationships of Wallabia indra (F3860, 7) and W. bicolor
(J4890, T), using Macropus parryi (J10756) as standard.

BARTHOLOMAI: WALLABIA IN THE UPPER CAINOZOIC

377

LITERATURE CITED

Bartholomai, A., 1966. The type specimens of some of
De Vis’ species of fossil Macropodidae. Mem. Qd
Mus. 14 : 115-25.

1967. Troposodon, a new genus of fossil Macropodinae
(Marsupialia). Mem. Qd Mus. 15 : 21-33.

1973a. The genus Protemnodon Owen (Marsupialia:
Macropodidae) in the Upper Cainozoic deposits of
Queensland. Mem. Qd Mus. 16 : 309-63.

1973b. Fissuridon pearsoni, a new fossil macropodid
(Marsupialia) from Queensland. Mem. Qd Mus. 16 :
365-8.

1975. The genus Macropus Shaw (Marsupialia: Macro-
podidae) in the Upper Cainozoic deposits of Queens-
land. Mem. Qd Mus. 17 : 195-235.

Calaby, J. H., 1966. Mammals of the upper Richmond
and Clarence Rivers, New South Wales. CSIRO Div.
Wildl. Res. Pap. 10 : 3-55.

De Vis, C. W,, 1895. A review of the fossil jaws of the
Macropodidae in the Queensland Museum. Proc.
Linn. Soc. N.S.W. 10 : 75-133.

Iredale, T., and Troughton, E. Le G., 1934. A check-
list of mammals recorded from Australia. Mem.
Aust. Mus. 6: xi 4- 122 pp.

Kirsch, J. A. W., 1968. Prodromus of the comparative
serology of Marsupialia. Nature 217 : 418-20.

Owen, R. , 1 874. On the fossil mammals of Australia. Part

VIII. Family Macropodidae: Genera: Macropus ,
Osphranter, Phascolagus, Sthenurus and Pro-
temnodon. Phil. Trans. 164 : 245-87.

Sharman, G. B., 1961. The mitotic chromosomes of
marsupials and their bearing on taxonomy and
phylogeny. Aust. J. Zool. 9 : 38-60.

Sharman, G. B., Calaby, J. H., and Poole, W. E., 1966.
Patterns of reproduction in female diprotodont
marsupials. In Rowlands, I. W., Editor, Compara-
tive Biology of Reproduction in Mammals. Symp.
Zool. Soc. Lond. No. 15, pp. 205-32 (Acamedic
Press: London).

Stirton, R. A., 1963. A review of the macropodid genus
Protemnodon. Bull. Dept. Geol. Uni. Calif. 44 :
97-162.

Tate, G. H. H., 1948. Studies on the anatomy and
phylogeny of the Macropodidae (Marsupialia).
Results of the Archbold Expeditions, No. 59. Bull.
Amer. Mus. Nat. Hist. 91 : 237-351.

Trouessart, E. L., 1905. Catalogus Mammalium tarn
viventium quam fossilium. Quinquennale sup-
plementum (1899-1904), Fasc. IV. Cetacea, Eden-
tata, Marsupialia, Allotheria, Monotremata; pp.
753-873 (R. Friedlander and Son: Berlin).

Troughton, E. Le G., 1937. Descriptions of some
New Guinea mammals. Rec. Aust. Mus. 20 : 1 17-27.

1957. The kangaroo family: brush wallabies. Aust.
Mus. Mag. 12:186 92.

378

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 53

Wallabia indra (De Vis, 1895)

Fig. 1: Lateral view of adult maxilla, F4740, Chinchilla Sand,
Chinchilla, Darling Downs, x 1.

Fig. 2: Stereopair of occlusal view of F4740, x 1 .

Fig. 3: Lateral view of juvenile mandible, F4749, western
Darling Downs, x 1.

Fig. 4: Stereopair of occlusal view of F4749, x 1 .

Fig. 5: Lateral view of adult mandible, F4741, Dalby, at 34-35
feet in pump well, Darling Downs, x 1.

Fig. 6: Stereopair of occlusal view of F4741, x 1.

BARTHOLOMAI: WALLABIA IN THE UPPER CAINOZOIC

Plate 53

Mem. QdMus . 17(3): 379-97, pis. 54-8. [1976]

RESULTS OF THE RAY E. LEMLEY EXPEDITIONS, PART 1. THE ALLINGHAM
FORMATION AND A NEW PLIOCENE VERTEBRATE FAUNA FROM NORTHERN

QUEENSLAND

Michael Archer and Mary Wade
Queensland Museum

ABSTRACT

The Allingham Formation, a lake and stream deposit with a rich fauna of fossil vertebrates,
contains some boulders of basalt. It is paraconformably overlain by part of the earliest known
phase of the Nulla Basalt, the Allensleigh ‘flow’ which has given dates varying from 4' 5 to 4 Myr.
The fossiliferous deposit may result from damming by early flows and fit stratigraphically
between the earliest and main flows of the Allensleigh Phase. The age is thus Lower Pliocene, and
the fauna is significant for correlation as well as containing the first known occurrence of
Perameles allinghamensis n. sp., Phascolonus lemley i n. sp., and Koobor jimbarratti n. gen and sp.
The fauna, referred to as the Bluff Downs local fauna, is similar to but apparently slightly older
than the Chinchilla local fauna. This assessment is based on a consideration of the structurally
more ancestral character of some of the Bluff Downs species. This supports the suggestion of
Bartholomai (1972) that the Chinchilla local fauna is late Pliocene in age. Faunal differences from
comparable Kalimnan faunas such as the Hamilton and Kanunka local faunas are attributed to

slight age differences, different geographical
accumulation.

Fossil bones were discovered, collected, and
reported to the Queensland Museum in 1973 by
Messrs J. Barratt and W. Snewin of Ayr. Following
information from Mr Barratt that the outcrop
underlay basalt, Museum collecting trips were
made to the area in 1973 and 1974; both parties
were ably assisted by Mr Barratt. Dr Ray E.
Lemley participated in and helped to finance the
second trip. Specimens described below are de-
posited in the Queensland Museum.

The geology is the joint responsibility of both
authors. The taxonomic work has been carried out
by Archer.

THE ALLINGHAM FORMATION

The name Allingham Formation is proposed
here for the sequence of terrigenous clays, silts,
sands, calcareous sands, and Chara limestones
underlying the Allensleigh ‘flow’ of the Nulla
Basalt, overlying laterite, and outcropping on Bluff
Downs Stn, along the banks of Allingham Creek
(at Lat 19° 43'S, Long. 145° 36E) a tributary of
the Burdekin River, north Queensland. The type
section (position indicated in Fig. 1 A, B, and strati-

ecological settings, or different modes of

graphic columns detailed in Fig. 2) is the western
end of the most extensive outcrop, on the north
bank of Allingham Creek, 5 6 km west of Emu
Valley Homestead. The name Allingham For-
mation is in reference to Allingham Creek. The
thickness at the type section, which is also the most
extensive vertical outcrop observed, is 18 metres.

As indicated in Fig. 2, lateritic detritus is present
in whatever portion of the Allingham Formation
overlies the laterite. At and adjacent to the type
section this largely detrital bed has an independent
character (Fig. IB, a; Fig. 2, a). Here it is a fine
ochraceous sediment with laterite clasts and light-
coloured component which is now decalcified
apart from one thin lens of Chara limestone (Fig. 2,
section 2); its upper part is a claystone of titurated
laterite which grades upward into a grey sandy silt,
which is the base of a series of sands and silts (Fig.
IB, b; Fig. 2, b). These beds change upward again
gradationally, into sandy Chara limestone (Fig. IB,
c; Fig. 2, c) with rare patches of secondary
slicification, which in turn grades up into sands and
clays (Fig. IB, d; Fig. 2, d). In creeks at the north
end of the amphitheatre (Fig. 1 A), 0-2 km north of
Allingham Creek, a brownish-grey clay is the

380

MEMOIRS OF THE QUEENSLAND MUSEUM

+ . . f, , + + + + + + + + + + +

k'T ' U ‘ t h\\" u t'A + + + + + + +

\xxxxxxxx^ + + + + + + +

W * x X "V,+ + i + I V' 1 ^

Af-* XXX l/ft, ji / / | ) 1 U x x^y v

^// /mi it

ALLINGHAM CREEK

BUJfF

BASALT + RtMfNAblf

SPRING

CREEK

LEGEND

+ BA + $aLt V+% + 4 rt buiV *

+ + + + + +y+ + + +^< x
+ + + + + ,> , * -

x-S 4

*• ih'x :
,rtl"x' X X ’•

300m
I

5 , X x

^ + t + 4 + + v'tjnntt

, + ?A§Al f T +

T . T ■ -

+ I + I 4 + +^f

+ ; • 4 Dqwj\is, \ ^ + [ 1 1 j

14-44-4-14- 1/4-4-r

4 1-44 4 4 4 /4 4 4 4

: -4-4/41 +

> i t - -r A * ♦ * H

444- 4/ 4-44

4 4 i 4 I /+ 1 + +

444 + 44-4-4-

I- + • / 4 I + +

, , + 4 + i I- - I +

4 4 4 4 1-4 H + * + 1 .

+ + + + i + h +

44+4 + 4 4/44

4 + 4444/4 4

++++++ if I +,

^J + ‘<V+VJLewti+ V+*
ISte^ .%* , VZ+t/Attfr-V
:|^C V + V+n&T fc- + +V

» J t I + + T + i + 1 +
> 1 ** + -I 4 4 1 + + I + +

4 4 I 4

+ t + I

® x x.O + + ' + + + +

■i S ,V.’WW, , ,V.',

>\'\ + . 1 + + + + + + I 4 +

+ , + T • + 4 + + 4 + -| +

Quaternary alluvium and soil
Basalt Scree

+ ±_i Allensleigh 'Flow'

Allingham Formation
ES53 Lower Tertiary laterite

Fence line

/m/t/f/t Basalt plateau edge

Fig. 1A: Locality plan of Allingham Creek showing type section (D-D) of Allingham Formation. F-H = exposed
contacts of Allensleigh flow with: F, Allingham Formation; G, H, laterite mottled zone. Line D-D indicates position
of largest cliff exposure of Allingham Formation with 1-3, positions of sections. Details of section stratigraphy
shown in Fig. 2. Line 4—4 indicates non-vertical transect which is presented as vertical section 4—4 in Fig. 2.

Fig. IB: Face of cliff exposure (D-D of Fig. 1A) along Allingham Creek, type section. Positions of sections (13) are
also shown. The laterite basement is deeply eroded and the sediments filling the hollows (bed a) contains much more
laterite detritus than overlying beds (b-d). Stratigraphic positions of a-d are also indidated in Fig. 2.

ARCHER AND WADE: THE ALLINGHAM FORMATION

381

4-4

Water level Allingham Creek, April, 1974

Fig. 2: Stratigraphic sections (1-3) of Allingham Formation along face of cliff exposure (D-D of Fig. 1 A) or interpreted
(4 4) from exposures along transect (4-4, Fig. 1 A) north and east of cliff exposure. Vertical sections 1-3 and lower
part of 4-4 were measured; the upper gently sloping part of section 4-4 was estimated. Beds a-d are informal
designations of similar lithology. The top of bed d in section 1 and section 2 has not been observed because it is
covered by basalt tors. The base of bed a is partly below the water level of Allingham Creek.

382

MEMOIRS OF THE QUEENSLAND MUSEUM

uppermost of the horizontal beds of the Allingham
Formation. This is not likely to represent volcanic
ash as it does not occur in the type section along
Allingham Creek (Fig. IB).

The general area has been described by Wyatt
and Webb (1970) in the course of mapping and
dating the flows of the Nulla Basalt (see also Wyatt
1968, 1969; Wyatt etal. 1965, geological map). The
fossil occurrence on the eastern extremity of Bluff
Downs station was straddled by traverses, and its
area mapped by photo interpretation (Wyatt and
Webb 1970, fig. 1) as covered by the Allensleigh
‘flow’. Here Allingham Creek has cut through the
basalt and the fossiliferous deposit into the wide-
spread and deep laterite that Wyatt and Webb
describe as basement in Basalt River just to the
north. They correlate this laterite with the wide-
spread and deep laterite that Exon, Langsford-
Smith and McDougall (1970) date as older than
basalts of 23 ± 1 million years, in an area running
north from Amby to the level of Injune. In the area
exposed by Allingham Creek there is a trend for the
more westerly exposures of laterite to be stripped to
mottled zone, while more easterly exposures have a
varying thickness of iron-rich laterite crust pre-
served. Surface irregularity was several metres and
at the northwesternmost exposure the fossiliferous
beds wedge out between the laterite mottled zone
and the basalt. To the north and east a ledge-
forming Chara limestone forms most of the fossil
beds and has been traced continuously for 3 km
east of the main fossil localities. The top of the bed
approximates the 1100' contour throughout its
known extent (contours from Hillgrove 1:100,000
military map series R631, sheet 8058). It extends
further than it was followed, and has been observed
as an outlier on the road 1 km NE. of Emu Valley
Homestead, another 4 km to the east, lying directly
on laterite weathering to buckshot gravel. In this
spot the Allensleigh ‘flow’ is reduced to loose
boulders. This is probably the outcrop of limestone
Wyatt (1969, p. 302) describes as ‘. . . on the track to
Emu Valley Homestead . . .’ and likens to \ . . that
which occurs as old lake or swamp deposits west of
Eumara Springs Homestead . . without giving
distances or stratigraphic data. The latter locality is
not specifically mentioned by Wyatt and Webb
(1970) and must be relocated to find its relationship
to the Allensleigh ‘flow’.

In the Allingham Creek area on Bluff Downs, the
Allensleigh ‘flow’ is usually seen as a jumble of
small tors overlying the fossiliferous sediments or
the lateritic basement. Original contacts have been
noted in three places (Fig. 1A, F-H). In all three,
deep tor weathering has occurred but, without the
physical removal of weathered material from

between the tors, ground water still continues its
attack. The continuous basalt is now much more
deeply weathered than the free tors. In the absence
of erosion surfaces or other divisions between
them, this physico-chemical explanation for the
differential weathering seems more likely than
differing ages, especially as two of these occur-
rences occupy lows in the lateritic basement (south
and west of the fossiliferous deposits) and one
overlies the fossiliferous deposit. Wyatt (1969)
reached a similar conclusion studying exposures to
the north, notably in Basalt River. At the back of
the shallow amphitheatre surrounding the out-
crop shown in Fig. 1 A, F; Fig. 2, 4 — 4, the strongly
weathered basalt lies conformably on a brownish-
grey clay overlying a metre of limey sand which in
turn grades into 3 metres of hard Chara limestone.
This limestone unconformably overlies iron-
enriched laterite crust and has a basal con-
glomerate of buckshot gravel and finer laterite
detritus. Traced westwards, along Allingham
Creek, the limestone overlies laterite mottled zone
and much of it changes laterally to sands and clays.
Some of these are channel deposits which contain a
minor amount of basalt as boulders and pebbles.
These must indicate a basalt earlier than the
Allensleigh ‘flow’ or the early extrusion of part of
the ‘flow’. Wyatt and Webb have already stated
that the datings obtained on this ‘flow’ range from
4-5 Myr, 10 km WNW. near Bluff Downs Homes-
tead, to 4 Myr further north (both datings ± 3%).
They suggest the ‘flow’ may more properly be
regarded as a flow series with a span of at least half
a million years. The presence of basalt boulders in
sediment underlying the main flow lends weight to
their suggestion but does not rule out an earlier
phase of the Nulla Basalt, possibly of minor extent.
Neither interpretation is at variance with the visual
evidence of conformity between the top of the
fossiliferous sediment and the local base of the
Allensleigh Phase. The fossiliferous sediment is
thus dated as no less than 4 ±12 Myr and not much
older, i.e. Lower Pliocene (Harland, Gilbert Smith,
and Wilcock 1964; Riedel 1973).

Observation of the lateral extent of the fossili-
ferous beds is hampered by the basalt cover and by
Pleistocene channel deposits related to a roughly
parallel course of Allingham Creek, which cut the
deposit in two. The fossiliferous beds are probably
an age-equivalent to the unconsolidated sands
below the Allensleigh Phase at Eumara Springs
(Wyatt and Webb 1970, pp. 40, 47). These are
roughly 30 m lower than the fossiliferous beds, and
24 km to the east of the outlier on Emu Valley road.
Both also could be age-equivalents of the Cam-
paspe Beds, which are poorly dated, if the older of

ARCHER AND WADE: THE ALLINGHAM FORMATION

383

the possible dates Wyatt and Webb suggest applies
to the Campaspe Beds; even if this should be so, the
deposits were not in lateral continuity and the
strong limestone component provided by the
growth of Char a gives the fossiliferous deposits
described here a distinctive lithology over most of
their outcrop.

Apart from the cliff containing the type section,
outcrops of the terrigenous lower members of the
Allingham Formation (a and b) are confined to the
southern to eastern side of Allingham Creek and to
its bed. The easternmost terrigenous outcrop yet
found is 1 km west of the Emu Valley-Bluff Downs
boundary fence and lies on lateritic mottled zone. It
may have been covered directly by the basalt as its
top is indurated, but erosion has removed an
upper contact. A major outcrop, capped by Chara
limestone, runs south from opposite the entry of
Spring Creek, a western tributary, for about 300 m.
This outcrop contains a very strongly calcareous
development of bed c, Chara limestone in close-
packed lenses. Here the limestone laterally replaces
sand, both lateral equivalents overlying 0-5 m of
derived laterite lying on laterite mottled zone in
situ. The remaining southern outcrop is from 1 to
1 -4 km south of the type section, and is fossiliferous
sands and clays overlain by basalt. Outcrops in the
creek bed are of a temporary nature due to shifting
alluvium. They occupy pockets in the laterite.

The conditions of deposition indicated by the
sediments show that the laterite surface was partly
stripped and deeply gouged, before its flooding by
water sufficiently permanent both to accumulate
sediment and allow the growth of the widespread
Chara flora. This flora, together with waterlaid
sands, clays and channel deposits, suggests that an
existent stream widened into a shallow lake.
Animal fossils are relatively rare in the widespread
limestone and are mainly scattered tortoise plates.
In contrast, the terrigenous sediments contain
many broken bones, some complete bones and
skulls, and rarely articulated bones. By far the most
common fossils throughout are scattered tortoise
plates and crocodile teeth. Fish remains are
relatively rare, which may indicate a seasonal
constriction of the water body. Then, as now, the
area sloped gently eastward (Wyatt and Webb
1970), and evidence of diastrophism, other than
that provided by partial stripping of the laterite, is
lacking. The cause of this erosion is scarcely likely
to have been the damming mechanism which
started deposition. The presence of a limited
amount of basalt in some of the channel deposits
suggests another mechanism, damming by a more
easterly flow or portion of a flow, than supplied the
basalt boulders.

BLUFF DOWNS LOCAL FAUNA
Michael Archer

The Bluff Downs local fauna is described below
and may be summarized as follows. Dental termi-
nology follows that used by Archer (1974, 1975a).
Local fauna refers to a faunal assemblage from a
particular area in the sense used by Tedford (1970);
it is an informal term. Prefixes to specimen
numbers include AM, Australian Museum; WAM,
Western Australian Museum; J or JM, Queensland
Museum modern specimens; F, Queensland
Museum fossil specimens.

Arthropoda

Crustacea

Unidentified gastrolith
OSTEICHTHYS
Teleostei

Unidentified spines and vertebrae
Reptilia
?Cheliidae
IChelodina sp.

Crocodilidae
Palimnarchus sp.

Agamidae

Small unidentified agamid similar to Amphi-
bolurus spp.

Varanidae
Varanus sp.

Boidae
? Morelia sp.

?Elapidae

Small vertebrae
Aves

Ciconiidae

Xenorhynchus asiaticus (Lathan, 1790)

Mammalia

Peramelidae

Perameles allinghamensis n. sp.

Vombatidae

Phascolonus lemleyi n. sp.

Phascolarctidae

Koobor jimbarratti n. gen. and sp.
Thylacoleonidae
Thylacoleo sp.

Macropodidae
Protemnodon sp.

Macropus sp. cf. M. dryas (De Vis, 1895)

M. ( Osphranter ) sp. cf. M. woodsi Bartholomai,
1975

Macropodid similar to Thylogale
Small macropodid, genus indet.

Diprotodontidae
Zygomaturus sp.

Euryzygoma sp.

Nototheriine, genus indet.

Unidentified families
One tooth fragment
Coprolites

384

MEMOIRS OF THE QUEENSLAND MUSEUM

SYSTEMATICS

Arthropod a

Crustacea
(Plate 540

F7829 represents a crustacean gastrolith, the
only specimen recovered,

OSTEICHTHYS

Teleostei

(Plate 54a-b)

Fish spines (e.g. F7771) and vertebrae (e.g.
F7772) are small and relatively uncommon. The
largest vertebra is only 7 mm in diameter and the
largest spine is 25 mm long.

Reptilia

?Cheliidae

Fragments of tortoises were the most common
fossils. These cannot at present be referred with any
certainty to a particular species or even genus. A
comparison of various Allingham fragments with
materials described by De Vis (1894, 1897) as
Chelymys uberrima, C. arata and Chelodina in-
sculpta show some similarities. A fragment (F7796)
possibly referable to Chelodina exhibits curious
pock-markings, presumably the result of disease or
invertebrate predation.

Crocodilidae

Crocodilian teeth are the next most common
vertebrate remains. Variation in form and size is
considerable and it is possible that more than one
species is represented. As well as teeth there are
large crocodilian vertebrae, limb bones, scutes and
skull fragments.

Palimnarchus sp.

(Plate 54 c-e)

Large compressed teeth (e.g. F7763, Plate 54d)
with serrated edges probably represent a species of
this genus. Almost identical teeth occur in the
Chinchilla Sand. Some Allingham teeth (e.g.
F7764, Plate 54e), show occlusal wear, a not
uncommon characteristic of Palimnarchus (M.
Flecht, pers. comm.).

Several small crocodile teeth (e.g. F7767, Plate
54c) exhibit extensive vertical fluting. In Crocodilus
johnstoni, this feature is common in most teeth. In
C. porosus it sometimes occurs on smaller but not
larger teeth. It may similarly have occurred on
some teeth of Palimnarchus, and cannot be used to

distinguish a second taxon unless associated mat-
erial proves distinctive. Longman (1924) describes
C. nathani from fossil material found at Tara
Creek, Maryvale Station, north Queensland. This
species is dubiously distinct from Palimnarchus and
no attempt has been made to distinguish it among
the Allingham crocodilian remains.

A thorough revision of late Cainozoic croco-
dilians must be made before all the Allingham
crocodile remains can be positively identified.

Agamidae

(Plate 54i)

A small right dentary fragment (F7812) with
seven teeth represents an agamid lizard similar to
some species of Amphibolurus.

Varanidae

Varanus sp.

(Plate 54h, j)

At least two vertebrae (F7774, and F7777)
represent a species of this lizard genus. A recurved
tooth crown (F7813, Plate 54h) may also be
referable to Varanus. The tooth is 1 5 mm long and 6
mm wide at the base, somewhat polished by
stream-abrasion, without serrations on posterior
or anterior cutting edges, and lacks vertical fluting
around the crown base.

In modern comparative material of Varanus the
teeth have very fine serrations on anterior and
posterior cutting edges as well as vertically fluted
crown bases. In Megalania prisca, Varanus dirus
and Notiosaurus dentatus these same characters
occur, except that the anterior cutting edge has
fewer serrations than the posterior edge.

Fejervary (1918) refers Notiosaurus to Megalania
and suggests that material previously referred to V.
dirus may represent two forms: the holotype, which
comes from the Darling Downs, possibly repre-
senting Megalania prisca', the referred specimen,
which comes from Chinchilla, representing some-
thing else. Varanus emeritus is not represented by
teeth. Of all these fossil varanids and megalanids
the Allingham tooth is most similar to the referred
specimen of V. dirus from Chinchilla but differs in
being less recurved and in lacking the serrations
and vertical fluting. Its degree of recurving is
matched by some modern Varanus (e.g. V. varius)
and it is possible that if fluting and serrations were
fine enough on the Allingham tooth, they could
have been removed by abrasion.

The single Allingham tooth is twice the size of
teeth of a seven foot specimen of Varanus salva-
dorii, the largest specimen of a species of Varanus in

ARCHER AND WADE: THE ALLINGHAM FORMATION

385

the Queensland Museum. The fossil tooth differs
morphologically in having a wider and less slender
crown, and a more rounded anterior cutting edge.

One of the Allingham varanid vertebrae (F7774,
Plate 54j), a posterior rib-bearing lumbar vertebra
(around the 21st or 22nd position), is similarly
unique. It compares favourably with vertebrae of
Varanus species but not M. prisca. Vertebrae of
megalanids from Chinchilla have not been de-
scribed. The Allingham vertebra differs from
comparably sized vertebrae of V. salvadorii in being
markedly taller, having an antero-posteriorly
shorter neural spine, a pronounced vertical crest on
the posterior edge of the neural arch, and sub-
rounded, more vertically inclined prezygapo-
physes.

The other varanid vertebra (F7777), an anterior
caudal vertebra of uncertain position, is about
equally distinct from modern species of Varanus.
The ventral pedicels for the haemal arch, the
relatively anteriorly situated cotyle, and the
relatively reduced neural spine indicate that this
vertebra represents a varanid, but the pronounced
ventral crest, relatively high neural arch and
reduced transverse processes are unmatched by
vertebrae of any modern species with which
comparison has been made ( V. salvadorii , V. varius
and V. gouldii).

Boidae
(Plate 54k)

Three vertebrae (including F 7775 ) represent a
very large boid, morphologically very similar to
modern species of Morelia. No comparative
material has been available in the size range of this
Allingham snake, so minor differences in mor-
phology may be attributed to allometry.

It is difficult and often misleading to estimate
sizes of animals based on fragmentary remains, but
compared with a 5-3 m specimen of Morelia
spilotes, the Allingham boid may have been over
6-2 m long. Worrell (1970) notes a record of a
modern Australian boid (Liasis amethistinus) of
8-7 m. Considering this, the Allingham snake was
probably no larger than some modern boids.

?Elapidae
(Plate 54g)

Two small vertebrae (including F7826) may
represent elapids. They compare favourably with
species of Pseudechis but not enough colubrid
material has been available to be certain of even the
familial identity of these vertebrae.

Aves

ClCONIIDAE

Xenorhynchus asiaticus (Lathan, 1790)

A fragment of a tarsometatarsus (F7036) repre-
sents this modern stork (pers. comm. J. van Tets,
2 1 .i. 1 974) known as the Blackheaded Stork.

Other bird remains are presently under study by
P. Rich.

Mammalia

Peramelidae

Perameles allinghamensis n. sp.

(Fig. 3; Plate 55c)

Holotype: F7821, isolated RM 2 ; Allingham For-
mation, Lower Pliocene, site 5, Allingham Creek, Bluff
Downs Stn, north Queensland.

Diagnosis: Very large peramelid differing from
other species of Perameles in having better-
developed antero-buccal cingulum; more closely
approximated protocone and paracone; paracrista
which buccally contacts stB which is posterior to
parastylar corner of tooth.

Description: Measurements shown in Fig. 3.

Crown showing slight wear on tips of all cusps.
Metacone broken and metastylar corner missing.
Tip of hypocone damaged. Anterior cingulum
short and complete but just barely so beneath
paracrista. Anterior cingulum formed by
confluence of antero-buccal cingulum and prepro-
tocrista. Postprotocrista descends to meet on
hypocone (or metaconule). Crest leaves hypocone
postero-buccally and descends to base of metacone
where it forms very short, reduced posterior

Fig. 3: Measurements (mm) ofF7821, RM 2 , holotype of
Perameles allinghamensis n. sp. Hatched area indicates
damage.

386

MEMOIRS OF THE QUEENSLAND MUSEUM

cingulum before terminating along flank of met-
acone. Hypocone small and separated from pro-
tocone by deep vertical fissure on postero-lingual
face of crown. Metacone higher than subequal
paracone and protocone which are higher than
hypocone. StB shorter than large subequal stC and
D. StA indistinguishable from parastylar corner of
tooth but short buccal crest links stB with antero-
buccal cingulum. This crest may be homologue of
stA. Anterior paracrista (homologue of dasyurid
paracrista) forms prominent crest linking paracone
and stB. Posterior paracrista similarly links para-
cone and stC. Prominent anterior metacrista links
metacone and stD. Posterior metacrista (homo-
logue of dasyurid metacrista) damaged but pre-
sumably extended to metastylar corner of tooth.
Presence or absence of stE unknown. Prefossa
between bases of protocone, paracone and met-
acone extends to buccal surface of crown. No crest
links stC with D. Similarly, no crest links
stB with C. StC gently recurved anteriorly.
StD gently recurved posteriorly. Ectoloph virtually
non-existent as result of failure of stylar cusps to be
united by crests. Ectoflexus greatest between stC
and D.

This tooth is considered to be an M 2 because of
the size of the stylar cusps, hypocone and relative
sizes of the paracone and metacone.

Discussion: Species of Perameles available for
examination have been P. nasuta (e.g. J 108 16), P.
bougainvillei (e.g. WAM Ml 0576), P. eremiana
(WAM Ml 575), and P. gunnii (e.g. AM M2640).
This includes all modern species recognized by
Ride (1970). Most species of the genera Micro-
peroryctes, Peroryctes, Echymipera, Isoodon, Chae-
ropus, Macrotis and Ischnodon have been
examined. Photographs of the only known
material of Rhynchomeles have been made avail-
able by courtesy of the British Museum. This
includes all modern and fossil perameloid genera
recognized by Tate (1948) and Stirton (1955).

P. allinghamensis occupies a somewhat in-
termediate structural position between Perameles
and those species of Echymipera which have been
examined (E. rufescens, both subspecies, and E.
kalubu). As in other species of Perameles , stC of P.
allinghamensis is relatively discrete and conical on
M 2 . This is true but to a lesser extent in species of
Echymipera where stC is sometimes linked to the
paracone by a small crest. This latter condition is
common and better-developed in other peramelids
such as Peroryctes and Microperoryctes. The
anterior cingulum is better-developed in Echy-
mipera than it is in Perameles and in this respect P.
allinghamensis resembles Echymipera. P. alling-

hamensis may be an ancestor of Perameles ,
Echymipera , or both. It is referred here to Pera-
meles because of the crest and stylar cusp
morphology. When more material is discovered, it
is probable that it will warrant generic separation
from all modern peramelids.

Other fossil perameloids include Ischnodon aus-
tralis (referred elsewhere to the Thylacomyidae,
Archer and Kirsch in preparation); an unnamed
peramelid from the Hamilton local fauna (Turn-
bull and Lundelius 1970) represented by fragments
of lower molars; Perameles tenuirostris Owen
which is regarded by Lydekker (1887) as syn-
onymous with P. nasuta; P. wombeyensis which has
been synonomized with Isoodon macrourus by
Wakefield (1972); and an unnamed peramelid from
the Fisherman’s Cliff local fauna (Marshall 1973)
represented by a fragmentary upper molar. There is
some doubt about the provenance of this last
specimen (Mr P. Crabb, pers. comm.). The possible
peramelid noted by Woodburne (1967) from the
Miocene Alcoota local fauna now appears to
represent a thylacoleonid (Dr W. A. Clemens, pers.
comm.).

Origin of Specific Name: The specific name is in
reference to the Allingham Formation and
Allingham Creek.

VOMBATIDAE

Phascolonus lemley n. sp.

(Fig. 4; Plate 56)

Holotype: F7819, left dentary with M, 4 ; Allingham
Formation, lower Pliocene, site 5, Allingham Creek, Bluff
Downs Stn, north Queensland.

Referred Material: F7818, LI a ; F7768-70, isolated
molars: same locality as holotype.

Diagnosis: Very large vombatid, differing from
Phascolonus gigas in markedly longer dorso-
ventral section of I t ; shorter cheek-tooth row;
shallower masseteric fossa; smaller and less pro-
truding ventro-lateral rim of masseteric fossa;
broader posterior border of dentary below arti-
cular condyle; shallower symphysis; and edge of
ectocrotaphyte plate and articular condyle which
extend relatively farther postero-dorsally. Differs
from Phascolomys magnus and P. medius in being
larger; having markedly longer dorso-ventral sec-
tion of I,; proportionately much longer premolar;
and deeper masseteric fossa. Differs from all other
vombatids in its much larger size as well as
morphological characters.

Description: Measurements shown in Fig. 4.

Dentary broken at point anterior to mental
foramen and posterior to I, alveolus. Coronoid

Fig. 4: Measurements (mm) ofF7819, holotype of Phascolonus lemleyiw. sp., and F7819. A-C, F7819, left dentary with
D, F7818, LI,. Hatched areas indicate damage.

—~LZl

388

MEMOIRS OF THE QUEENSLAND MUSEUM

and angular processes broken at bases. Mesial and
distal tips of articular condyle broken. I : and P 4
missing, although proximity of (F7818) to
dentary in quarry (less than 20 cm), unabraded
condition of open root of I : and alveolus of I]
indicate that isolated \ x probably drifted out of
dentary shortly before burial. Associated with
dentary in same quarry 10 cm away was an
articulated macropodid hindlimb, suggesting that
adjacent objects in quarry may be parts of one
individual.

I, with flat horizontal occlusal surface. Wear
striae on occlusal surface extend antero-posteriorly
from anterior tip for distance of 39 cm, presumably
distance through which dentary dislocates during
thegosis. Prominent ventral keel. Dorsal surface
forms narrow shelf which inclines lingually.
Prominent depression runs length of I, on dorsal-
buccal surface. Less pronounced longitudinal
striae run length of I,.

P 4 alveolus indicates tooth had only very shallow
medial lingual groove, although there were two
poorly differentiated (in comparison with molars)
columns. Alveolus length suggests P 4 slightly
shorter than M,.

M , 4 markedly divided into two columns, buccal
grooves being sharper and more deeply incised
than lingual grooves. Talonid (posterior column)
wider than trigonid (anterior column) in M l5
subequal in M 2 _ 3 and narrower than trigonid in
M 4 . Worn trigonid height subequal to worn talonid
height M|_ 2 , but taller than worn talonid M 3 _ 4 .
Talonid length longer than trigonid length M, but
shorter than trigonid length M 2 _ 4 .

Dentary massive (but markedly less so than in P.
gigas ). Symphysis extends posteriorly to level of
middle of M 2 . Base of ascending ramus leaves body
of dentary at level of anterior end M 4 . Masseteric
fossa deep for a vombatid but shallow compared
with P. gigas. Postalveolar ridge sharply curved
postero-laterally and forms dorsal rim of dental
canal. Pre-alveolar ridge forms sharp crest which
extends as far anteriorly as broken anterior edge of
dentary. Ectocrotaphyte plate relatively narrow
(compared with P. gigas). Mylo-hyoid groove
relatively large. Width of superangular cavity
(pterygoid fossa) exceeds that of ectocrotaphyte
plate. Inferior dental canal has large oriface
(mandibular foramen) but rapidly tapers down to
small canal. Masseteric foramen very large, almost
as large as mandibular foramen.

Ectocrotaphyte ridge and articular condyle ex-
tend postero-dorsally farther than they do in
material described by Stirling (1913). Posterior
neck of dentary below condyle broad and flat,

unlike condition illustrated by Owen (1872, plate
138, fig. 1) but somewhat similar to specimen
described as ‘Mandible “C” ’ by Stirling (1913).
Compared with Owen’s (1872, plate 36, fig. 1)
illustration of P. gigas, ascending ramus of P.
lemleyi exhibits longer masseteric fossa, and longer
and less-curved border anterior to articular con-
dyle.

Discussion: Wombats of the genus Phascolonus
(P. gigas and Sceparnodon ramsayi) have all been
placed in the synonomy of P. gigas Owen. The
taxonomic positions of medius Owen and magnas
Owen are unclear (Tate 1951). In some respects
such as size and premolar morphology, they
resemble P. gigas and make generic boundaries of
Phascolonus difficult to recognize. Synonomy of
ramsayi with P. gigas enables the generic diagnosis
of Phascolonus to include widely spatulate upper
incisors. Upper incisors of P. lemleyi are not yet
known, but overall similarity of lower incisors of P.
lemleyi to those of P. gigas suggest the upper
incisors are spatulate. Differences in lower incisors
of P. lemleyi and P. gigas include much longer
cross-sectional length of the former which results in
a relatively longer occlusal wear surface. Possible
significance of this is not clear although increase in
cross-sectional length of I, in other wombats
appears to correlate with increase in width of I 1
such as may be observed in a structural sequence
from Vombatus ursinus , through Phascolomys
magnus, to Phascolonus gigas. If this relationship is
maintained in P. lemleyi , its upper incisors were not
only relatively but absolutely wider than those of P.
gigas.

A specimen (F834) collected at Freestone Creek,
Darling Downs, Queensland, resembles P. lemleyi
in cross-sectional length of I,, and length of M 1-4 .
It differs in having a shorter P 4 comparable in size
with magnus, and raises a question about the
number of Pleistocene species of Phascolonus.
Stephenson (1963) describes Diarcodon parvus as a
diprotodontid similar to but smaller than Scepar-
nodon ramsayi, which he also regarded as a
diprotodontid. There can be no doubt that Scepar-
nodon is a vombatid (Ride 1967) and the diprotod-
ontid affinity of all of the material referred to D.
parvus is doubtful. Some of the upper incisors may
represent a species of Phascolonus.

Origin of Specific: Name: The specific name is in
honour of Dr Ray E. Lemley, Queensland Museum
Associate, who very kindly helped us on several
occasions by financing and accompanying expe-
ditions.

ARCHER AND WADE: THE ALLINGHAM FORMATION

389

Phascolarctidae

Koobor n. gen.

Type Species: Koobor notabilis (De Vis) [ =
Pseudocheirus notabilis De Vis 1889],

Diagnosis: Small phascolarctids, similar to
Phascolarctos but differ in being smaller; lacking
extensive fine crenulations on molar crown surface;
lacking well-developed pockets or basins between
bases of protocone and hypocone; having distinctly
shorter and bicuspid P 4 without significant longitu-
dinal crest development; having well-developed
basin buccal to paracone between ectoloph, pre-
paracrista and postparacrista; having stylar crests
well-developed adjacent to stB; having well-
developed gap between stC and stD; and overall
crown outline which is relatively longer than wide.
Differs from Litokoala in having narrower molars;
better-developed and enclosed basin buccal to
paracone; less crenulations; and no well-developed
metaconule. Differs from Perikoala (as judged by
comparison with fragmentary material described
by Stirton, Tedford and Woodburne 1967) by
having narrower and less crenulated molars.
Differs from Pseudokoala and all pseudocheirines
(. Pseudocheirus , Pseudochirops, Schoinobates, Pet-
ropseudes, Hemibelideus ) in having lingually dis-
placed paracone and metacone; lacking complex or
well-developed metaconule on M 1 2 ; lacking well-
developed protoconule on M 2 3 ; having better-
developed anterior cusp on P 4 ; and having larger
gap between stC and stD.

Origin of Generic Name: Koobor is an Abor-
iginal mythological Koala-boy who was always so
thirsty that he stole his companions water con-
tainers and hid with them up a tree. When
discovered, he was punished by his enraged
companions and turned into the Koala, a creature
who thereafter never drank water (Roberts and
Mountford 1970).

Koobor jimbarratti n. sp.

(Fig. 5; Plate 55b)

Holotype: F7822, isolated RM lor2 , Allingham For-
mation, Lower Pliocene, site 5, Allingham Creek, Bluff
Downs Stn, north Queensland.

Diagnosis: Differs from Koobor notabilis (only
other species) in having well-developed parastyle;
poorly-developed anterior cingulum; preproto-
crista which contacts preparactista; more obtuse
angle enclosed by pre- and postmetacristae; and
poorly-developed buccal crests at ends of pre- and
postmetacristae.

Description: Measurements shown in Fig. 5.

Transverse fracture occurs through tooth. Sur-
face of enamel lightly pitted by chemical erosion.
All primary cusps subequal in height. Metacone
and hypocone just closer together than paracone
and protocone. Preprotocrista contacts base of
preparacrista midway along length of prepara-
crista. Postprotocrista passes postero-buccally to
midline of tooth then joins prehypocrista. Posthyp-
ocrista passes postero-buccally to form small
posterior cingulum before contacting buccal end of
postmetacrista. Preparacrista short, markedly cur-
ved, and contacts stB. Postparacrista longer, less
curved, and contacts stC. Pre- and postparacrista
indirectly connected buccally by ectoloph crest
connecting stB and stC. Premetacrista straight and
runs to position of stD. Postmetacrista straight and
runs to metastylar corner of tooth where it
connects with upturned posterior cingulum. Low
crest on ectoloph from buccal end of premetacrista
does not extend posteriorly as far as metastylar
corner. No crest connects stC and position of stD.
Parastylar crest connects anterior end of prepara-
crista to parastylar corner of tooth. Below point of
contact between preprotocrista and preparacrista,
two small vertical crests connect short antero-
buccal cingulum to premetacrista and prepro-
tocrista, enclosing very small pocket between them.
Antero-buccal cingulum poorly-defined or absent
along lingual half of tooth, although basal crown
swelling occurs. Shallow basin occurs between
bases of protocone and hypocone. Side of tooth
between pre- and postmetacristae not enclosed
buccally. A lingual crest occurs on metacone
extending antero-lingually to midcrown basin.

Fig. 5: Measurements (mm) of F7822, RM 1 or RM 2 ,
holotype of Koobor jimbarratti n. gen. and sp.

390

MEMOIRS OF THE QUEENSLAND MUSEUM

Smaller lingual crest extends postero-lingually
from metacone tip to posterior cingular basin.
Poorly-developed protocone rib extends buccally
from protocone tip to base of metacone. Rib and
crest development on paracone and protocone not
clear.

Discussion: The species of Koobor are clearly
not pseudocheirines but they are similar to phascol-
arctids, and of these, perhaps closest to the middle
Miocene Perikoala. They may be late Tertiary
representatives of the same phascolarctid group.
Litokoala and Phascolarctos are not repre-
sentatives of this group and may have been
independently derived from other Miocene phas-
colarctids.

The Chinchilla species is better known than the
Allingham species and, in some respects such as the
better-developed buccal basins, is structurally
more advanced.

Origin of Specific Name: Koobor jimbarratti is
named in honour of Mr Jim Barratt who, with Mr
Wally Snewin, originally discovered the fossil sites
along Allingham Creek.

Thylacoleonidae

Thylacoleo sp.

(Plate 55a)

Material: F 7762 , right dentary fragment with half of
P 4 and M,, roots of I ls and alveoli for M,; F7807,
posterior fragment of LP 4 ; F7808, posterior fragment of
LP 4 .

Discussion: This is a very small species of
Thylacoleo comparable in size to a form recorded
by Merrilees (1968, p. 14) as ‘ Thylacoleo , probably
not carnifex ’ from a paraconglomerate interpreted
by Merrilees to be Pleistocene in age, at Wonberna,
near Balladonia, Western Australia. It differs from
this form in having a relatively much larger M x . It is
also similar to T. crassidentatus from the Pliocene
Chinchilla Sand, but differs in having a pro-
portionately much shorter P 4 .

Thylacoleonids are separable into at least two
distinct types: Thylacoleo species, so far known
only from Pliocene and younger sediments; and
Wakeleo species, so far known only from middle to
late Miocene sediments. A third type may be
represented by an undescribed form from the
middle Miocene Etadunna formation. The Alling-
ham form, although clearly referable to Thyla-
coleo, is also similar to Wakeleo in so far as it
exhibits a relatively short P 4 and large M,. It is
tempting to assume orthogenesis and see the
Allingham thylacoleonid as a link in a chain

leading directly from Wakeleo to Thylacoleo .
However this type of reasoning is almost always
found to be fallacious when the fossil history of a
group becomes better known. Indeed, the occur-
rence of a thylacoleonid with a short P 4 in
supposedly Pleistocene deposits at Wonberna is
sufficient reason to doubt that premolar length can
be directly correlated with time. As in most groups,
the late Tertiary radiation of thylacoleonids prob-
ably resulted in a complex of forms, each the
product of different selective pressures, rather than
all being subject to one single pressure, namely an
increase in carnassial length at the expense of other
cheek-teeth.

Macropodidae

Comments about the forms represented here are
only preliminary. Formal descriptions will be given
by Archer and Bartholomai (in preparation). There
are at least five macropodid species represented, of
which four can be assigned to genera: Protemnodon
sp.; Macropus cf. M. dryas; Macropus ( Osphran-
ter) sp.; and IThylogale sp. Concepts of Chinchilla
species of Macropus employed here are those of
Bartholomai (1975).

Protemnodon sp.

(Plate 57a-b)

A maxilla (F7810), fragmentary upper premolar
(F7814), isolated molars (e.g. F7809), and a
dentary with dP 3 and M, (F7812) represent a small
species of this genus. It is most similar in mor-
phology to P. chinchillaensis and P. devisi from the
Pliocene Chinchilla Sand and P. otibandus from the
Pliocene Otibanda Formation. The upper molars
of the maxillary fragment (F7810) differ from those
of P. devisi in being smaller; lacking a lingual
cingulum between the bases of the protocone and
hypocone; having a less well-developed midlink,
and a better-developed posterior pocket. They
differ from P. chinchillaensis in having a wider
metaloph on M 4 and a less well-developed midlink.
Overall, they most closely resemble P. chinchilla-
ensis. An isolated RM 4 (F7811) is even more
Minilar lo those of P. chinchillaensis than is the M 4
of F7810. It has a comparably developed midlink
and a very narrow metaloph. The isolated RP 4
fragment (F7814) is relatively narrow and shows
other differences that appear to distinguish it from
P. chinchillaensis. Considering variation in pre-
molar morphology shown by Bartholomai (1973),
these differences may not be significant. DP 3
(F7812) however, appears to be markedly different
from that tooth in P. chinchillaensis as well as P.
devisi. The protoconid is either absent or in-

ARCHER AND WADE: THE ALLINGHAM FORMATION

391

distinguishable from the metaconid, and the buccal
side of the trigonid is incised by a prominent
vertical fissure. There is also a buccal cingulum
between the bases of the trigonid and talonid. It is
possible that this single tooth is abnormal. Com-
parably unique though different morphologies are
reported as abnormalities by Archer (1975b).

Detailed comparisons with P. otibandus will
have to await formal description of the Bluff
Downs material. Plane (1972) and Bartholomai
(1973) indicate broad similarities between P.
otibandus , P. devisi and P. chincbillaensis , and the
apparently long geological history of at least P.
otibandus, from late Miocene to late Pliocene time.

Macropus sp. cf. M. dryas (De Vis, 1895)
(Plate 57c~d)

This maeropodid is well-represented by frag-
mentary dentaries (e.g. F7823), and maxillary
fragments (e.g. F7780). F7823 closely resembles
F2508 from Chinchilla which is referable to M.
dryas. Differences include a slightly more massive
P 3 which also lacks the prominent crest extending
postero-lingually from the posterior end of the
longitudinal crest; less antero-posteriorly oriented
paracristid; narrower anterior cingulum on dP 4 ;
and longer P 4 with fewer and less well-defined
vertical ribs along the longitudinal crest. These
differences appear reasonably constant in all speci-
mens examined. Upper teeth also resemble M.
dryas, and no consistent differences have been
noted in premolar or molar morphology.

Macropus (Osphranter) sp., cf. M. woodsi Bartho-
lomai, 1975
(Plate 57e)

F7785, an isolated RM X represents a species of
Macropus morphologically similar to Macropus
(Osphranter) pan and woodsi from the Chinchilla
Sand. It is approximately the size of M. woodsi and
smaller than M. pan. Characters that suggest
relationship with these Chinchilla species include a
prominent isolated enamel crest or cusp buccal to
the midlink in the buccal side of the mid-valley;
relatively narrow anterior cingulum with forelink;
midlink without accessory crests; and well-
developed posterior pocket. This Allingham form,
at present known from one tooth, could well prove
to be an ancestor of either Macropus woodsi, M.
pan or both.

Three isolated premolars (F7789 91) resemble
those of modern M. (Osphranter ) but may repre-
sent either M. ( Osphranter ) cf. M. woodsi, or yet
another unknown, even unrelated, maeropodid.
An unworn LP 4 (F7791) shows a basic pattern

shared by many macropodine genera such as
Macropus, Petrogale, Prionotemnus, and Wallabia,
with a principal longitudinal ribbed crest sup-
ported at each end by a large cusp; well-developed
lingual cingulum and cingular basin; low postero-
lingual cusp connected to the longitudinal crest by
a small transverse crest; and small posterior pocket
formed between the transverse crest, the postero-
lingual cusp, the large posterior cusp of the
longitudinal crest, and a small posterior cingulum.
A slightly worn LP 4 is more similar to that tooth in
Osphranter than other forms examined. It is a
simple tooth, indistinctly bilobed with a very
reduced postero-lingual crest.

Cf. Thylogale

Two isolated lower molars (F7794-5) are
difficult to distinguish from teeth of modern
Thylogale but show too few distinctive structures to
enable reference to any particular modern or fossil
genus.

An isolated upper right molariform tooth
(F7785) may represent a dP 4 and is similarly
difficult to distinguish from corresponding teeth of
Thylogale (e.g. T. stigmatica).

Small Macropodine
(Plate 57Q

F7784, an isolated lower molar, is unlike other
Allingham macropodids noted above in possessing
a relatively horizontal posterior cingulum, such as
occurs in some Protemnodon. This feature com-
bined with its Macropus- like crown morphology
and small size makes it unlike any Pliocene or
Quaternary macropodids examined.

Diprotodontidae

Zygomaturus sp.

(Plate 58d)

An isolated RP 4 (F7776) represents a species of
Zygomaturus. Three species of Zygomaturus are
currently recognized: Z. trilobus (many Pleistocene
deposits); Z. gitli (Beaumaris); and Z. keanei
(Alcoota). The Allingham Zygomaturus differs
from all of these in having a much larger hypocone
and smaller protocone so that transverse tooth
width is markedly greatest along a line through the
hypocone and metacone, and a much better
developed buccal cingulum and cingular pocket.

Closer comparison may be made with an as yet
undescribed specimen (F3829) of Zygomaturus
from the Chinchilla Sand. The Chinchilla specimen
is similar in having a large hypocone and a well-
developed buccal cingulum. It differs from the

392

MEMOIRS OF THE QUEENSLAND MUSEUM

Allingham tooth in being much larger in all
dimensions; in having an even longer buccal
cingulum; and in having the protocone and
Lypocone farther apart. Other differences are
obscured by wear on the Allingham specimen.

This zygomaturine is important first in suggest-
ing a closer relationship with a Chinchilla species
than any other zygomaturine, and secondly in
demonstrating differences which may be in-
terpreted as indicating it could be ancestral to the
Chinchilla species.

The importance of zygomaturines in correlation
has been suggested by Stirton, Tedford and
Woodburne (1968), and they may be useful in
interpreting the age of the Chinchilla Sand relative
to other late Tertiary mammal bearing deposits.

Euryzygoma sp.

(Plate 58a)

The most common diprotodontid from the
Allingham Formation is referable to the noto-
theriine genus Euryzygoma. Only one species is
named, E. dunense, from the Chinchilla Sand.

The Allingham Euryzygoma is represented by
one complete (F7891) and two partial skulls as well
as several dentaries and isolated teeth. If it were not
for the enormous variation apparently exhibited by
E. dunense from Chinchilla, it would be tempting to
believe that more than one species was represented
by the Allingham remains. This may yet prove to be
the case. The possibility that there is more than one
Chinchilla species is also under examination. Until
this problem is resolved, the specific identity of the
Allingham Euryzygoma must remain uncertain.

Although premolar morphology can be matched
in the two samples (e.g. F7765 from Allingham,
and some of the teeth included in F5812 from
Chinchilla), several cranial differences include
morphology of the zygomatic arch which in the
Allingham Euryzygoma more closely resembles less
specialized nototheriines than does E. dunense.

Nototheriine, genus indet.

(Plate 58b-c)

A small nototheriine is represented by several
dentary fragments including F7766, an isolated
worn RP 4 ; F7830, a maxillary fragment of an as yet
unprepared skull containing M 3 4 . This animal
differs from the Allingham Euryzygoma in being
much smaller, having a very reduced M 4 with
markedly narrow metaloph, and in several charac-
ters of the dentary.

It resembles in molar and dentary morphology
several specimens from Chinchilla regarded pre-
viously as Euowenia grata , a form whose generic

status is in doubt. M 4 of F519 (holotype of
Euowenia grata) from Chinchilla is similar but has
a wider metaloph than the Allingham specimen.

Taxonomic assessment of this small Allingham
nototheriine will have to await preparation of the
skull.

Incertae Sedis
(Plates 55d-f, 57g)

A single tooth fragment (F7792) may represent
an otherwise unknown family. It represents a
medium to large-sized animal, presumably mar-
supial, that has a well-developed cingulum, and at
least two small twinned cusps. Twinned cusps are
rare in marsupials. They occur in some perameloids
(e.g. Macrotis) and phascolarctids.

Coprolites (e.g. F7761) are common in the
deposit. Size and shape suggest they were produced
by a medium to large-sized animal, possibly a
corcodilian, large snake, or diprotodontid, and
that they were deposited in water. Some of the
largest are too massive to have maintained their
shape had they been deposited on hard ground or
transported. Further, they bear no impressions
such as might be expected if they were deposited on
an irregular terrestrial surface.

DISCUSSION

At this stage in our knowledge of the Bluff
Downs local fauna, twenty-two taxa including
thirteen mammals have been recognized. There is
no representation of monotremes, dasyurids, thyla-
cinids, thylacomyids, phalangerids, petaurids, bur-
ramyids, myrmecobiids, wynyardiids, notoryctids,
or tarsipedids. Except for the last four, all are
represented in older as well as younger deposits and
their absence from the Bluff Downs local fauna
cannot be the result of absence from the Australian
continent at that time. In some families, repre-
sentative modern forms occur in most broad
ecological habitats so that absence of monotremes,
dasyurids, petaurids, and burramyids may be the
result of chance sampling. Notoryctids, myrme-
cobiids and thylacomyids are represented in the
modern fauna by arid-adapted forms, most of
which are rare, and their absence from the Bluff
Downs local fauna may reflect a similar Pliocene
rarity or an unsuitable environment.

Aspects of the Palaeoenvironment

Accuracy of interpretation depends on both
extent to which the sample represents the con-
temporaneous fauna in diversity and species abun-
dance, and extent that ecological requirements of

ARCHER AND WADE: THE ALLINGHAM FORMATION

393

fossil forms may be interpreted from those of their
nearest living relatives. Major uncertainties re-
main.

Certainties are that Chara, crustaceans, and fish
indicate persistent fresh water, and tortoises,
crocodiles and Black-headed Storks are supporting
evidence for at least seasonal bodies of water.
Modern tortoises and crocodiles can migrate
considerable distances to find suitable water and
therefore are not evidence for permanent water.
When the modern Allingham Creek is running,
small fish are abundant and tortoises common.
When the creek is not running tortoises form a
relatively much larger part of the biomass in
waterholes (pers. comm. J. Barratt), presumably a
reflection of their ability to migrate.

Among carnivores known from the fossil fauna,
the dog-sized Thylacoleo is the largest mammalian
carnivore. The crocodile Palimnarchus was prob-
ably capable of killing any of the mammals
represented in the fauna. Although tortoises could
have formed at least part of the food supply of
Palimnarchus , not one of the hundreds of plates
preserved show evidence of tooth marks which
would suggest such predation. Further, there are
very few fish remains in the deposit, suggesting
they were not an abundant source of crocodile
food. It is possible that Palimnarchus hunted
mainly mammals (and/or birds), either waiting for
them to come to water, or possibly even pursuing
them near shore.

Mammals represented in the deposit suggest
arboreal ( Koohor jimharratti ) as well as terrestrial
habitats. The much greater abundance of ter-
restrial forms suggests that the surrounding area
was savannah woodland. Terrestrial forms include
numerous grazing kangaroos and diprotodontids,
including a species of Protemnodon which may
have been a browser. There is evidence from
macropodid post-cranial remains of a Tree
Kangaroo-like form which may have been the same
species of Protemnodon. Considering the ap-
parently rapid evolution and radiation of kan-
garoos in the late Tertiary (Bartholomai 1972), it is
perhaps particularly unwise to extrapolate to these
lower Pliocene forms, habitat requirements of
supposedly similar modern forms. Thus, Macropus
( Osphranter) cf M. woodsi may not resemble most
modern members of the subgenus Osphranter in
showing any preference for rocky hills or slopes.

In summary, there is evidence to suggest that
bodies of water were present for at least months at a
time. It is also probable that these lakes, rivers, or
swamps were not permanent. It is possible that the
Bluff Downs local fauna represents a riparian
assemblage.

Comparison with other Kalimnan and Post
Kalimnan Local Faunas

Comparisons of the Bluff Downs local fauna
should be made with the Chinchilla, Hamilton,
Awe, Fisherman's Cliff, Kanunka, Palankarinna,
and Beaumaris local faunas. This comparison is
summarized in Table 1 where only species occur-
ring in the Bluff Downs local fauna are considered.

Chinchilla: Except for the peramelid, all Bluff
Downs mammal species are closely comparable
with forms from the Chinchilla local fauna of
southeastern Queensland. Frequently the Bluff
Downs member of each species pair is structurally
ancestral in terms of size and morphology. For this
reason, differences between the two local faunas are
not regarded here as evidence of merely ecological
or geographical differences, but rather as evidence
for a difference in age, with Chinchilla appearing to
be the younger of the two. We support the
suggestion of Bartholomai (1973) of a late Pliocene
age for Chinchilla. It is not likely to be Pleistocene
because of the closer similarities of Chinchilla
species to those of Bluff Downs than to those from
eastern Darling Downs Pleistocene deposits. For
example, there is no undoubted record of Dipro-
todon, Nototherium, Procoptodon, Fissuridon , Mac-
ropus (Macropus), or Sarcophilus from Chinchilla
or Bluff Downs, nor is there any record of
Euryzygoma, Macropus cf. M. dryas or Pro-
temnodon cf. P. chinchillaensis from eastern Darl-
ing Downs deposits.

Hamilton: The Pliocene Hamilton local fauna
of Victoria (Turnbull and Lundelius 1970) is not
broadly comparable with either the Bluff Downs or
the Chinchilla local faunas since it contains mostly
small mammals represented by isolated teeth. Only
Palorchestes near P. painei appears to suggest that
Hamilton is structurally older than Chinchilla
which contains P. parvus. Teeth from the Hamilton
deposit came from a soil overlain by a basalt dated
at 4-35 ± 01 Myr (Turnbull and Lundelius 1970).
Therefore the Hamilton local fauna may be closely
comparable in age to the Bluff Downs local fauna.

Awe: The Pliocene Awe local fauna of Papua
(Plane 1967) contains two structurally simple
species of Protemnodon and in this respect they
resemble Bluff Downs and Chinchilla species. Awe
zygomaturines ( Kolopsis rotundus and Kolopsoides
cultridens) are structurally more primitive than the
Bluff Downs Zygomaturus , and clearly resemble
late Miocene Alcoota zygomaturines. However,
the Awe nototheriine ( Nototherium watutense ) is
not very different from the Bluff Downs noto-
theriines ( Euryzygoma and ‘ Euowenia ’). Lack of a

394

MEMOIRS OF THE QUEENSLAND MUSEUM

TABLE 1: The Distribution of Elements of the Bluff Downs Local Fauna in Other Late Tertiary and Early

Quaternary Local Faunas

Bluff Downs
local fauna

E. Dari. Downs

Kanunka

Fisherman’s Cliff

Chinchilla

Hamilton

Awe

Palankarinna

Beaumaris

crustacean

s?

s?

fish (teleost)

s?

s?

s?

s?

s?

tChelodina sp.

s?

s?

f?

s?

Palimnarchus sp.

f

s?

f

f

cf. Amphibolurus sp.

Varanus sp.

s?

f

s?

? elapid

boid

s?

Xenorhynchus asiaticus

c

c

c

c

c

c

Perameles allinghamensis

g‘ ?

sf

Phascolonus lemleyi

g

s?

f

g

Koobor jimbarratti

g

Thylacoleo sp.

g

g

g

Protemnodon sp.

g

s?

g

g

s?

g

M. ( Prionotemnus) cf. M. dryas

Sg

Sg

Sg

s?

Sg

M. (Osphranter) cf. M. woodsi

Sg

Sg

s?

small macropodid, cf. Thylogale

s?

small macropodid, gen. indet.

Zygomaturus sp.

g

g?

g

s?

sf

g

g

Euryzygoma sp.

g

nototheriine, gen. indet.

g?

g

s?

sf

sf

Maximum number of mammal

species in common

0

2?

0

4?

2?

0

0

0

Abbreviations: s, either same species or else not yet demonstrated to be a different species; sg, same
subgenus; g, same genus but different species; sf, same subfamily; f, same family; c, same class.

premolar referable to N. watutense prohibits close
comparison, but it is clear that all three forms are
broadly similar. Anderson (1937) compares the
holotype with known nototheriines and concludes
it is closest in size to Euowenia grata. Considered
as a whole, the Awe zygomaturines suggest that the
Awe local fauna is structurally older than the Bluff
Downs local fauna although Page and McDougall
(1972) suggest that a date of 31 Myr may be the
oldest reliable date associated with the Awe local
fauna. Plane (1967) reports dates between 5-7 and
7-6 Myr for intercalcated pyroclastics.

Fisherman’s Cliff: Marshall (1973) regards the
Fisherman’s Cliff local fauna to be probably late
Pliocene or early Pleistocene in age. The macro-
podids include a species of Protemnodon which
Marshall considers is most similar to P. devisi (as P.
cf. P. otibandus ) from Chinchilla. The specimen is
not figured. The diprotodontids include a possible
species of Diprotodon which suggests this local
fauna is Pleistocene in age.

Kanunka: The Kanunka local fauna from the
Lake Eyre Basin of South Australia is briefly

described by Stirton, Tedford and Miller (1961)
who consider it to be ?Pleistocene in age. Their
preliminary comments suggest similarity to the
Bluff Downs local fauna, but do not exclude
comparison with Pleistocene deposits such as those
from the eastern Darling Downs. The apparent
absence of Diprotodon favours a late Pliocene age,
although it is probably younger than the Bluff
Downs or Chinchilla local faunas.

Palankarinna: Stirton, Tedford and Wood-
burne (1968) regard the Palankarinna local fauna
from the Lake Eyre Basin of South Australia to be
late Pliocene in age. Faunal diversity is low. The
perameloid ( Ischnodon australis ) is a thylacomyid
and hence not comparable with Perameles alling-
hamensis. Diprotodontids include the noto-
theriine Meniscolophus mawsoni and the zygomat-
urine Zygomaturus keanei. It is apparent that
generic boundaries of Meniscolophus , Euryzygoma
and ‘ Euowenia ’ need re-examination. However, M.
mawsoni appears to differ specifically from Chin-
chilla and Bluff' Downs nototheriines. Z. keanei
and the Zygomaturus from Bluff Downs differ at

ARCHER AND WADE: THE ALLINGHAM FORMATION

395

what is probably a specific level, but it is not yet
possible to assess which form is structurally
ancestral to the other. Bartholomai (1975) has
recently recognized the Palankarinna macropodid
Prionotemnus palankarinnicus from Chinchilla.

Beaumaris: Stirton, Tedford and Woodburne
(1968) review the Beaumaris local fauna from
southern Victoria, considering it to be early
Pliocene in age. Zygomaturus gilli is, like Z. keanei
and Z. trilobus, unlike the Bluff Downs Zygo-
maturus which resembles, but dilfers from, the
Chinchilla Zygomaturus. Zygomaturine taxonomy
and biostratigraphy is in need of careful re-
examination and at this stage it does not enable us
to assess the relative ages of the two local faunas.

In summary, the Bluff Downs local fauna
compares most closely with the Chinchilla local
fauna but may prove to be also similar in
composition to the Kanunka and Hamilton local
faunas. Apparent differences between these and
other Kalimnan local faunas may be the result of
differing ages, latitudes, conditions of accumu-
lation, or ecological setting.

ACKNOWLEDGMENTS

Mr Jim Barratt deserves special thanks for giving
so much of his time, with his wife Tup and son
Lionel. Thanks also are extended to Mr Wally
Snewin who, with Jim, first discovered the Ailing-
ham fossils. Mr Allistair McDougal at Bluff
Downs Station kindly allowed us access to the site
and has been continuously helpful during the two
years of field work. Mr Stan Blacklock and Mr
John Stone of Emu Valley Station have similarly
provided generous assistance. Dr Ray E. Lemley,
Associate of the Queensland Museum, was most
helpful during the 1973 field season by financing
the expedition and sorting concentrate in the field.
Mr Andrew Elliot of the Queensland Museum
provided valuable field assistance. Dr George
Heinsohn, Ms Sally Cooper, Messrs Stewart
Frusher, James Oliver, Tony Paladin, David Flett,
Peter Channells, and Michael Guinea, all of James
Cook University, helped in aspects of excavation in
1974. We are also grateful to Dr Alan Bartholomai,
Queensland Museum, Dr Richard Tedford, Amer-
ican Museum of Natural History, Dr Max Hecht,
Queens College and Mr Peter Crabb, Monash
University for comments and advice. Mr Jerry Van
Tets, C.S.I.R.O., Canberra, kindly provided the
bird identification. Mr John Hardy, Queensland
University, produced the scanning electron micro-
scope photographs (Plate 55b-c). Mr Allan
Easton, Queensland Museum, produced the other

photographs used in the plates. Mrs E. Archer
prepared the holotype of Phascolonus lemleyi , and
Mr A. Elliot prepared the skull of Euryzgoma sp.
(Plate 58a) and the dentary of the small
nototheriine (Plate 58c).

LITERATURE CITED

Anderson, C., 1937. Palaeontological notes No. IV:
Fossil marsupials from New Guinea. Rec. Aust.
Mus. 20: 73-6.

Archer, M., 1974. The development of the cheek-teeth
in Antechinus flavipes (Marsupialia, Dasyuridae). J.
Roy. Soc. West. Aust. 57 : 54—63.

1975a. The development of premolar and molar crowns
of Antechinus flavipes (Marsupialia, Dasyuridae)
and the significance of cusp ontogeny in mammalian
teeth. J. Roy. Soc. West. Aust. 57 : 118-25.

1975b. Abnormal dental development and its
significance in dasyurids and other marsupials.
Mem. Qd Mus. 17 : 251-65.

Bartholomai, A., 1972. Aspects of the evolution of the
Australian marsupials. Presidential address. Proc. R.
Soc. Qd 83: v-xviii.

1973. The genus Protemnodon Owen (Marsupialia:
Macropodidae) in the upper Cainozoic deposits of
Queensland. Mem. Qd Mus. 16 : 9-23.

1975. The genus Macropus Shaw (Marsupialia:
Macropodidae) in the upper Cainozoic deposits of
Queensland. Mem. Qd Mus. 17 : 195 235.

De Vis, C. W., 1894. The lesser chelonians of the
Nototherian drift. Proc. R. Soc. Qd 10 : 123 7.

1895. A review of the fossil jaws of the Macropodidae
in the Queensland Museum. Proc. Linn. Soc. N.S. W.
10 : 75-133.

1897. The extinct fresh-water turtles of Queensland.
Ann. Qd Mus. 3: 3-7.

1 889. On the phalangistidae of the post-Tertiary period
in Queensland. Proc. R. Soc. Qd 6: 105-14.

Exon, N. F., Langford-Smith, T., and McDougal, I.,
1970. The age and geomorphic correlations of deep-
weathering profiles, silcrete, and basalt in the Roma
region, Queensland. J. Geol. Soc. Aust. 17 : 21 30.

Fejervary, G. J., 1918. Contributions to a monography
on fossil Varanidae and on Megalanidae. Ann. hist.-
nat. Mus. Hungar. 16 : 341-467.

Harland, W. B., Gilbert Smith, A., and Wilcock, B.,
1964. The Phanerozoic time scale. Supplement to Q.
Jl Geol. Soc. Lond. DOS: i-viii, 1-458.

Longman, H. A., 1924. Some Queensland fossil verteb-
rates. Mem. Qd Mus. 8: 16-28.

Lydekker, R., 1887. ‘Catalogue of the fossil mammalia
in the British Museum (Natural History).’ Part V
(Marsupials, 146-289). (British Museum (Natural
History): London).

Marshall, L. G., 1973. Fossil vertebrate faunas from the
Lake Victoria region, S.W. New South Wales,
Australia. Mem. Nat. Mus. Viet. 34 : 151-72.

Merrilees, D., 1968. Man the destroyer: late Quaternary
changes in the Australian marsupial fauna. J. Roy.
Soc. West. Aust. 51 : 1-24.

396

MEMOIRS OF THE QUEENSLAND MUSEUM

Owen, R., 1872. On the fossil mammals of Australia (Part
vi): Genus Phascolomys, Geolfr. Phil. Trans. 1872 :
173-96.

Page, R. W., and McDougall, I., 1972. Ages of mineral-
ization of gold and porphyry copper deposits in the
New Guinea highlands. Econ. Geol. 67 : 1034-48.

Plane, M. D., 1967. Stratigraphy and vertebrate fauna of
the Otibanda Formation, New Guinea. Bur. Min.
Resour. Aust. Bull. 86 : i vi, 1-64.

1972. A New Guinea fossil macropodid (Marsupialia)
from the marine Pliocene of Victoria, Australia.
Mem. Nat. Mus. Viet. 33 : 33-6.

Ride, W. D. L., 1967. On Sceparnodon ramsayi Owen,
1884: The selection of a lectotype, the clarification of
its type locality, and on its identity with Phascolonus
gigas (Owen, 1859). Rec. S. Aust. Mus. 15 : 419-25.

1 970. ‘A guide to the native mammals of Australia.’ Pp.
xiv + 249 (Oxford Univ. Press: Melbourne).

Riedel, W. R., 1973. Cenozoic planktonic micropaleon-
tology and biostratigraphy. Ann. Rev. Earth and
Plan. Sci. 1 : 241-68.

Roberts, A., and Mountford, C. P., 1970, The
Dreamtime’. 79 pp. (Rigby Ltd.: Adelaide).

Stephenson, N. G., 1964. On fossil giant wombats and
the identity of Sceparnodon ramsayi. Proc. Zool. Soc.
Lond. 142 : 537^16.

Stirling, E. C., 1913. On the identity of Phascolomys
( Phascolonus) gigas, Owen, and Sceparnodon ram-
sayi, Owen; a description of some of its remains.
Mem. Roy. Soc. S. Aust. 1 : 127-78.

Stirton, R. A., 1955. Late Tertiary marsupials from
South Australia. Rec. S. Aust. Mus. 11 : 247-68.

1967. A new species of Zygomaturus and additional
observations on Meniscolophus, Pliocene Palankar-
inna fauna, South Australia. Bur. Min. Resour. Aust.
Bull. 85: 129-47.

Stirton, R. A., Tedford, R. H., and Miller, A. H.,
1961. Cenozoic stratigraphy and vertebrate paleon-
tology of the Tirari Desert, South Australia. Rec. S.
Aust. Mus. 14 : 19 61.

Stirton, R. A., Tedford, R. H., and Woodburne, M.
O., 1968. Australian Tertiary deposits containing
terrestrial mammals. Univ. Calif. Publ. Geol. Sci. 77:
1-30.

Tate, G. H. H., 1948. Studies in the Peramelidae
(Marsupialia). Bull. Am. Mus. nat. Hist. 92 : 313 46.
1951. The wombats (Marsupialia, Phascolomyidae).
Amer. Mus. Novit. 1525 : 1-18.

Tedford, R. H., 1970. Principles and practices of
mammalian geochronology in North America. Proc.
N. Amer. Paleon. Com., Sept., 1969: 666-703.
Turnbull, W. D. and Lundelius, E. L., 1970. The
Hamilton fauna, a late Pliocene mammalian fauna
from the Grange Burn, Victoria, Australia. Fiel-
diana: Geol. 19 : 1-163.

Wakefield, N. A., 1972. Palaeoecology of fossil mammal
assemblages from some Australian caves. Proc. R.
Soc. Vic. 85: 1-26.

Worrell, E., 1970. ‘Reptiles of Australia’. Pp. xv + 169.

(Angus and Robertson: Sydney).

Wyatt, D. H,, 1968. Townsville, Qd. 1:250,000 Geologi-
cal Ser. Explan. Notes Bur. Miner. Res. Geol.
Geoph. Aust., SE/55-14.

1969. A note on the geology of the Bluff
Downs-Allensleigh area. Qd. Govt. Min. J. 17 :
296 303.

Wyatt, D. H., and Webb, A. W., 1970. Potassium-argon
ages of some northern Queensland basalts and an
interpretation oflate Cainozoic history. J. Geol. Soc.
Aust. 17 : 39-51.

398

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 54

Bluff Downs invertebrates and lower vertebrates

a, ¥1112 teleost vertebra

b, F7771, teleost spine

c, F 7767 , vertically fluted crocodile tooth

d, ¥1163, tooth of Palimnarchus sp.

e, F7764, tooth of Palimnarchus showing occlusal wear of tip

f, F7829, crustacean gastrolith

g, F7826, small snake vertebra, possibly elapid

h, F7813, tooth possibly referable to Varanus sp.

i, F7812, fragment of an agamid dentary

j, ¥1114, vertebra of Varanus sp.

k, F7775, vertebra of a large boid

Unless otherwise indicated, line represents one cm.

ARCHER AND WADE: THE ALLINGHAM FORMATION

Plate 54

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 55

Bluff Downs mammals

a, F7762, right dentary of Thylacoleo sp. with broken P 4 and M ,

b, F7822, stereopair scanning electron microscope photo-
graphs, RM 1 or RM 2 , holotype Koobor jimbarratti n. gen.
and sp.

c, F782 1 , stereopair scanning electron microscope photographs,
RM 2 , holotype Perameles allinghamensis n. sp.

d-f, F7792, stereopair, tooth fragment of unknown type of
mammal

Line represents one cm.

ARCHER AND WADE: THE ALLINGHAM FORMATION

Plate 55

I-

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 56

Phascolonus lemleyi n. sp.
a, F7818, LI], lingual view

b-d, F7819, left dentary, holotype, Phascolonus lemleyi n. sp.
Line represents one cm.

ARCHER AND WADE: THE ALLINGHAM FORMATION

Plate 56

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 57

Bluff Downs macropodids and coprolite

a, F7810, right maxillary fragment with parts of RM 1 ' 2 and
M 3-4 , Protemnodon sp.

b, F7812, right dentary fragment with RdP 4 - M 1? Protemnodon
sp.

c-d, F7823, right dentary with RP 3 , dP 4 , M x , and P 4 (excavated,
d), Macropus sp. cf. M. dryas

e, F7785, isolated upper molar, Macropus ( Osphranter) sp., cf.
M. woodsi

f, F7784, isolated lower molar, small macropodine of uncertain
affinities

g, F7761, coprolite

Line represents one cm.

ARCHER AND WADE: THE ALLINGHAM FORMATION

Plate 57

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 58

Bluff Downs diprotodontids

a, F7891, skull, Euryzygoma sp.

b, F7830, left maxillary fragment, small nototheriine

c, left dentary fragment with damaged LM 2 _ 4 , small nototh-
eriine

d, F7776, isolated RP 4 , Zygomaturus sp.

Line represents one cm.

Plate 58

ARCHER AND WADE: THE ALLINGHAM FORMATION

Mem. QdMus. 17(3): 399-411, pis. 59-60. [1976]

DARDURUS, A NEW GENUS OF AMAUROBIID SPIDER FROM EASTERN
AUSTRALIA, WITH DESCRIPTIONS OF SIX NEW SPECIES

Valerie Todd Davies

Queensland Museum

ABSTRACT

A new genus Dardurus (fam. Amaurobiidae) is described, with six new species D. spinipes , D.
silvaticus, D. tamborinensis (ecribeWate), D. ncmorctlis. D. saltuosus and D. agrestis. Thecribellum
looks like a flat semi-circular colulus and a magnification of 200x was needed to see the spinning
areas. The species show a reduction in the number of spinning tubes with a corresponding
reduction in the length and number of hairs in the calamistrum until in D. tamborinensis neither
spinning tubes nor calamistrum are present. The spiders live a sedentary life in small, decorated
silk tubes with two openings. Scanning electron micrographs of the cribella of Ixeuticus
longinquus (Koch, L., 1867) and Stiphidion face turn Simon, 1902 are included for comparison.

In the latter half of the nineteenth century
collections of spiders were made around the sea
ports of Queensland and New South Wales for the
Godeffroy Museum, Hamburg. About 600 spiders
were described in Die Arachniden Australiens and
its Supplement by Koch (1871-81) and Keyserling
(1881-89). At about the same time Thorell de-
scribed 46 species from Cape York Peninsula in
Studi sui Ragni Malesi e Papuani, vol. 3. Between
1893 and 1920 Rainbow contributed much to the
knowledge of Australian spiders. He was en-
tomologist at the Australian Museum, Sydney for
many years and not only described about 200
spiders but also wrote much on their webs and
behaviour. In 1911 he published a Census of
Australian Araneidae where he listed approxi-
mately 1200 species. Included among these were
100 spiders described by Hogg from specimens in
the British Museum. From 1926 Hickman,
Australia’s foremost araneologist, described about
70 spiders most of which were from Tasmania and
some of which required the erection of new
families. Recently Main has revised and re-defined
some of the mygalomorph genera. It will be seen
from this brief review that the araneomorphs of E.
Australia have been rather neglected since 1920, so
it is not surprising that a new genus, like Dardurus ,
can be found in a suburban garden.

The term ‘amaurobiid’ is used in the broad sense
of Lehtinen (1967) not in the restricted sense of
Forster and Wilton (1973). All specimens have

been deposited in the Queensland Museum (QM).
Measurements of the cephalothorax length (CL),
cephalothorax width (CW), abdomen length (AL),
abdomen width (AW), and measurements of leg
segments were made with an ocular micrometer
and converted to millimetres.

Dardurus nov. gen.

Small size. Both rows of eyes straight if viewed
from above, procurved if viewed from front;
anterior median eyes smallest. Cheliceral boss
present. Labium wider than long. Maxillae narrow
at base, wide and truncated at apex with well
marked serrula and scopula. Sternum widely
truncate, produced posteriorly between fourth
coxae. Trochanteral notch absent. Anterior tibiae
with more than 3 pairs of conspicuous long ventro-
lateral spines. Superior tarsal claws with 7 pecti-
nations; inferior claw smooth. Palpal tarsus longer
and wider than other tarsi. Plumose hairs absent.
Trichobothria in a single row of increasing length
distally on metatarsi and tarsi, irregularly placed
on tibiae. Trichobothria on cymbium. Six spin-
nerets, anterior laterals largest. Undivided cribel-
lum or flat semicircular colulus. Cribellum absent
in <$. Calamistrum proximal or absent. Epigynum
with undivided fossa. Apophyses on patella and
tibia of S palp. Embolus curved, spiniform;
conductor membraneous. Four unbranched
tracheal tubes arising from a median posterior

400

MEMOIRS OF THE QUEENSLAND MUSEUM

spiracle. The spider lives on the underside of logs or
in soil in a camouflaged silken tube with two
openings.

The generic name is derived from the Aboriginal
work ‘dardur’ — a bark hut.

Type Species: Dardurus spinipes n. sp.

Dardurus spinipes n. sp.

(Figs. 1-10, 19a; Plates 59A, 60B)

Material Examined

Holotype: Open sclerophyll forest on Brisbane River,
Roedean St, Fig Tree Pocket, Brisbane, SE.Q., V. Davies,
22.V.74, . , QM W4877.

Paratypes: Open sclerophyll forest on Brisbane R..
Roedean St, Fig Tree Pocket, Brisbane, SE.Q,, V. Davies,
15.vi.74, Id, QM W4878; 4.iii.74, L. 1 juv., QM W4879;
21.i.73, 1 , QM W4880; 12.viii.73, 1$, Id, QM W4881;
9.ix.73, 3 , QM W4882; 2.viii.73, 2$, Id, 1 penult, d, 1
juv., QM W4883; 22.V.74, 3,', 4d, 1 juv., QM W4884;
15.vi.74, 2$, Id, QM W4885; 15.vi.74, 12, QM W4886;
l.viii.74, 3:, QM W4887. Under log in bank near Little
Yabba Creek, Conondale Ra., SE.Q., R. Raven,
3 l.viii.74, 6 , Id, QM W4888; 31,viii. 74, 4 , 2 juv., QM
W4889.

Description of Female

CL 1 50; CW 098; AL 1 78; AW MO.

Cephalothorax and legs are orange brown,
abdomen is grey-black with lighter grey pattern
(Figs. 1 , 2). Small white patches (‘thoracic patches’)
one on each side of the thoracic fovea similar to
those noticed by Forster and Wilton (p. 1 66, 1973)
in some of the female cribellate Amphinectidae
from New Zealand. The long spines on tibiae and
metatarsi 1 and II are reddish brown in colour.

Both rows of eyes are procurved if viewed from
the front, straight if viewed from above (Figs. 3a,
b). The ratio of eyes AME:ALE:PME:PLE is
6:10:9:9. There are 4 teeth on retromargin of
chelicerae; 2 large and 3 small teeth on promargin
(Fig. 4). The maxillae are narrow at the base, wide
and truncated at the apex with a well marked
serrula and scopula. The labium is wider than long
in ratio 1:0-63. The sternum is longer than wide,
1:0-87.

Notation of spines: Palp: tibia, p. 1-2.2; tarsus,
numerous spines (Fig. 5). First leg (Fig. 6): femur,
d.l.p.l. distal; tibia, v.2.p.0. 1.1. 1 .l.(l).r.0.1 . 1.1.1;
metatarsus, v.l.p.O.l.l.l.r.O.l.l.l; tarsus, 0.
Second leg: femur, d. 1.1.0; tibia, v.2.

p.0.1. (l).l.l.r.O.l.O.l.l.; metatarsus, d.l. 1.0.0.

p.1.1. l.l.r.0. 1.1.1. Third leg: tibia, d.O.l.O.v. 1.0.1.
p.l.l.O.r. 1.1.0; metatarsus, numerous small spines.
Fourth leg: tibia, v.l.0.1.0.2.p.0.1.0.!.0.r.0.1.0.1.0;
metatarsus, several spines. Calamistrum proximal,
consisting of 7 rather sparse curved hairs (Fig. 7).
Superior tarsal claws with 7 pectinations (Fig. 8);
the inferior claw smooth.

Six spinnerets; anterior laterals (AS) largest, in
ventral view tending to obscure the rest.
ASxribellum, 1:1-10. The cribellum has the ap-
pearance of a large flat colulus. The electron
scanning microscope shows spinning tubes are
present in 6 transverse alternating rows with about
15-16 tubes in each row (Plate 60, B). The
epigynum (Figs. 9a, b, 19a) occupies a large part of
the ventral abdominal surface and has well marked
lateral ridges. The fossa is undivided, i.e. there is no
median ridge or guide. Width of external epigynum
2} x length.

Variation: Cephalothorax lengths were from
1-14-1-50. Sometimes there are 2 large and only 2
small teeth on the promargin of the chelicerae.
There was some variation in the arrangement of
spines on the corresponding legs from left and right
sides of the spiders. This is indicated by the use of ( )
in the notation of spines. An extra prolateral spine
is not uncommon on the first and second tibia. The
spiders from Conondale Range were darker in
colour and there was a dorsal spine on the fourth
femur.

Description of Male

QM W4878: CL 1*20; CW 0-80; AL 1*32; AW
0 - 88 .

Cephalothorax lengths varied from 1-02 to 1-46
mm. Males are similar to females in colour, general
structure and spination of the legs except that the

TABLE 1: Leg Measurements of D . spinipes ■. and (<3)

Leg

Femur

Patella

Tibia

Metatarsus

Tarsus

Total

1-06(0-98)

0-46(0-40)

1-00(0-96)

0-82(0-80)

0-28(0-34)

3-62(3-48)

ii

0-84(0-74)

0-42(0-36)

0-66(0-60)

0-66(0-64)

0-26(0-26)

2-84(2-60)

iii

0-68(0-62)

0-36(0-30)

0-38(0-40)

0-56(0-52)

0-24(0-24)

2-22(2-08)

iv

0-94(0-90)

0-42(0-34)

0-72(0-74)

0-84(0-80)

0-30(0-30)

3-22(3-08)

palp

0-40(0-42)

0-20(0-14)

0-28(0-20)

0-42(0-66)

1-30(1-42)

0.2mm

DAVIES: DARDVRVS , A NEW GENUS OF AMAUROBIID

401

Figs. 1-8: $ D. spinipes. L, lateral view; 2, ventral view; 3a, eyes from top; 3b, eyes from front; 4, chelicera; 5, tarsus of
palp; 6, first leg; 7, fourth leg; 8, superior tarsal claw.

1 0.1 mml

0.2mm

402

MEMOIRS OF THE QUEENSLAND MUSEUM

Fig. 9: 9 D. spinipes. 9a, external epigynum, cleared; 9b, internal epigynum.

Fig. 10: £ D. spinipes. 10a, palp, dorsal; 10b, palp, ventral; 10c, palp, retrolateral; lOd, palp, prolateral.

DAVIES: DARDbRUS , A NEW GENUS OF AMAUROBIID

403

prolateral distal spine on femur I is absent and
there are no ‘thoracic patches’ near the fovea.
There is a colulus in the penultimate and mature
It is smaller than the $ cribellum and has no
spinning tubes. Neither the penultimate nor the
mature male has a calamistrum. Trichobothria are
present on the cymbium. The palp has a finger-
shaped anterior retrolateral apophysis on the
patella and a very complex one with several
protuberances on the retrolateral surface of the
tibia (Fig. 10a, b, c). An apophysis arises from the
tegulum and curves over the embolus which is
retrolateral to the apophysis. A small ventral flange
on the latter appears to keep the embolus in place.
The spiniform embolus rests on a membraneous
conductor. Whether the apophysis is a median
apophysis or whether this is absent is uncertain.

Habits and Life History

All the spiders were collected from open sclero-
phyll forest. They live in small (10-13 mm) tubes
with 2 openings (Plate 59A) and are found on
rotting wood or sometimes under stones or in soil.
The tube which is decorated with bark debris
(hence the generic name) or soil particles has no
web outside it. Females were found throughout the
year; penultimate males from February until May
and males till the end of August. Males were often
found with one palp broken off and it was not
unusual to find them occupying the same tube as
females. Egg sacs containing 3-6 eggs were found
from August to March.

Dardurus silvaticus n. sp.

(Figs. 11-13, 19c; Plates 59B, 60A)

Material Examined

Holotype: Rain forest, Mt Glorious, 32 km NW.
Brisbane, SE.Q., V. Davies, 19.vii.74, $, QM W4890.

Paratypes: Rain forest, Mt Glorious, SE.Q., V.
Davies, 19.vii.74, lrf, QM W4891; 20.vi.74, 2$, IJ, QM
W4892.

Description of Female

CL 1*18; CW 0-88; AL 1 28; AW 110.

The spider is similar in colour and pattern to D.

spinipes however the abdominal pattern shows an
extension of the pale patches over the posterior half
of the abdomen giving it a lighter colour (Fig. 11).
Small ‘thoracic patches’ are present. The eyes are
similar to D. spinipes. There are 4 teeth on the
retromargin of the chelicerae. The labium is wider
than long 1 :0-66. Sternum longer than wide 1:0-91.

Notation of Spines: Palp: tibia, p.1.2; tarsus,
numerous small spines, First leg femur, d.l.(l).0.
p.l. distal; tibia, v.2.p.0.1.1.1.1.r.0. 1.1.1. 1.(1);
metatarsus, v.l.p.O.l.l.l.r.0. 1.1.1; tarsus, 0.
Second leg: femur, d. 1.1.0; tibia, v.2.p. 0.1. 1.1.1.
r.0. 1 .0. 1. 1; metatarsus, d. 1 . 1 .0.O.p, 1.1.1. 1 .r.0. 1.1.1.
Third leg: tibia, d.O.l.O.v.l.O.l.p. 1.1.0. r. 1.1.0;
metatarsus, numerous small spines. Fourth leg:
tibia, v. 1 . 1 . 1.0. 1 .p.0.(l).0. 1 .0.r.0. 1 .0. 1 .0; meta-

tarsus, several spines. Calamistrum, proximal and
well developed consisting of 9 curved hairs. Ratio
of AS:cribellum 1:1-18.

The cribellum has about nine rows of spinning
tubes with at least 25 tubes in a row (Plates 59B,
60A) which is many more than D. spinipes. In both
species the tubes are unsegmented. The epigynum
(Figs. 12a, b, 19c) is less chitinized than D. spinipes
with weak lateral ridges and a large open fossa
almost the width of the epigynum.

Variation: Cephalothorax of $ QM W4892 is
1-28 mm long. As in D. spinipes there may be extra
spines on the first and second tibiae.

Description of Male

QM W4891: CL 1-10; CW 0-78; AL M2; AW
0-74.

The colour and pattern is similar to $. There are
no ‘thoracic patches’ on the cephalothorax.

Notation of Spines: First leg: femur, d.l.O.p.l.
distal; tibia, v.2.p.0. 1 . 1 . 1 . 1 .r.0. 1 . 1 . 1 . 1 .; metatarsus,
v.l. p.0.1.1. 1. r.0. 1.1.1; Second leg: femur, d.l; tibia,
v. 2. p.0.1 . 1.1.1. r.0. 1.0. 1.1; metatarsus, d.l. 1.0.0.
p.l. 1.1.1. r.0. 1.1.1. Third leg as in female. Fourth
leg: tibia, v. 1.0. 1.0.1. p.0.1. 0.0.0. r. 0.(1). 0.1.0. There
is no calamistrum.

Male palp: The tibial apophysis on the palp (Fig.
1 3a, b) is less complex than D. spinipes with fewer

TABLE 2: Leg Measurements of D. silvaticus $ and (^)

Leg

Femur

Patella

Tibia

Metatarsus

Tarsus

Total

i

0-84(0-88)

0-40(0-40)

0-72(0-86)

0-64(0-76)

0-26(0-32)

2-86(3-22)

ii

0-66(0-70)

0-34(0-34)

0-52(0-54)

0-52(0-60)

0-22(0-28)

2-26(2-46)

iii

0-52(0-58)

0-28(0-28)

0-30(0-32)

0-40(0-48)

0-20(0-22)

1-70(1-88)

iv

0-80(0-84)

0-32(0-30)

0-60(0-64)

0-68(0-78)

0-28(0-30)

2-68(2-86)

palp

0-38(0-38)

0-20(0-18)

0-22(0-20)

—

0-32(0-54)

1-12(1-30)

404

MEMOIRS OF THE QUEENSLAND MUSEUM

protuberances. The embolus lies on a mem- MtTamborine, SE.Q. V. Davies, C. L. Wilton, R. Raven,

braneous conductor retrolateral to the apophysis. 10.vii.74, 2J, 1$, QM W4895.

Dardurus tamborinensis n. sp.

(Figs. 14-15, 19b)

Material Examined

Holotype: Rain forest, Curtis Falls track, Mt Tam-
borine 50 km S. Brisbane, SE.Q. V. Davies, 22.vi.75, I,
QM W4893.

Paratypes: Curtis Falls track, Mt Tamborine, SE.Q.,
V. Davies, 22.vi.75, Id, QM W4894; Curtis Falls track.

Description of Female

CL 1-22; CW 0-84; AL 1-34; AW 0-90.

Cephalothorax and legs are pale yellow brown.
Two small white Thoracic patches’ on each side of
fovea present. Abdomen grey-black with lighter
pattern. Eyes like D. spinipes. Five teeth on
retromargin of chelicerae; 2 large and 2 small teeth
of promargin. Labium wider than long 1:062. The
sternum a little longer than broad 1:0-95.

13 b

\

j

/

Figs. 11-12: , D. silvaticus. 11, abdomen, dorsal; 12a, external epigynum, cleared; 12b, internal epigynum.
Fig. 13: d D. silvaticus. 13a, palp, dorsal; 13b, palp, ventral.

Fig. 14; y D. tamborinensis. External epigynum, cleared.

DAVIES: DARDURUS, A NEW GENUS OF AMAUROBIID
TABLE 3: Leg Measurements of D. tamborinensis $ and (<J)

405

Leg

Femur

Patella

Tibia

Metatarsus

Tarsus

Total

i

0-92(1-10)

0-38(0-40)

0-82(1-02)

0-68(0-92)

0-30(0-38)

3-10(3-82)

ii

0-72(0-88)

0-36(0-38)

0-58(0-72)

0-56(0-66)

0-26(0-32)

2-48(2-96)

iii

0-54(0-72)

0-30(0-30)

0-32(0-46)

0-44(0-62)

0-24(0-28)

1-84(2-38)

iv

0-88(1-04)

0-34(0-34)

0-66(0-84)

0-76(0-94)

0-32(0-38)

2-96(3-54)

palp

0-38(0-46)

0-20(0-14)

0-24(0-24)

—

0-36(0-56)

1-18(1-40)

Notation of Spines: Palp: tibia, p.2.1-2; tarsus,
numerous small spines. First leg: femur, d.2.p.l.
distal; tibia, v.2.p.0.1.1.1.1.(l).r.0.1.1.1.1; meta-
tarsus, v. 1 .p.0.(l). 1.1.1 .r.0. 1.1.1; tarsus, 0. Second
leg: femur, d.2.v.l. distal; tibia, v.(l).2. p.0.1. 1.1.1.
r. 0.1. 0.1.1. metatarsus, v.l. p.0.1. 1.1. 1.1.

r.0. 1.0. 1.0.1. Third leg: femur, d.l; patella, d. 1.(1);
tibia, d.0.(l).0.v.0.(l).2.p. 0.1.1. r.0. 1.(1); meta-
tarsus, numerous small spines. Fourth leg: patella,
d.l; tibia, d.l.0.0.(l).0.v.l.0.1.0.2.p.0.1.0.1.0.
r.0. 1.0. 1.0; metatarsus, several spines. No calamis-
trum.

This species is ecribellate. The colulus was
scanned and showed no spinning tubes. Anterior
spinnerets are wider than the colulus. AS:colulus
1:0-70, unlike the other species described. The
epigynal fossa is chitinised anteriorly (Figs. 14,
1 9b) and is about a third the width of the epigynum.

Variation: Cephalothorax length of $ QM
W4895 is 1-30.

Description of Male

QM W4894: CL 1 34; CW 0-88; AL 1-24; AW
0-82.

Cephalothorax lengths varied from 1-16 to 1-40.

Colour and pattern are similar to ‘Thoracic
patches’ absent. There are 5 teeth on retromargin
of chelicerae.

Notation of spines: Prolateral distal spine on
femur 1 is present as in $. This spine is also present
in D. silvaticus but is absent in other $ Dardurus
spp. examined. Apart from the presence of a dorsal
spine on Femur IV the spination is like ?.

Male palp: The embolus (Fig. 15b) arises
anterior and prolateral to the apophysis in contrast
to its position in the other species where it is

retrolateral. The apophysis here looks very like a
‘median’ apophysis. In view of the fact that D.
tamborinensis is ecribellate it is assumed that the
position of the embolus is secondary.

Dardurus nemoralis n. sp.

(Figs. 16-18, 19d)

Material Examined

Holotype: Wet sclerophyll forest, Cameron Falls
track, Mt Tamborine, 50 km S. Brisbane, SE.Q., V.
Davies, 22.vi.75, $, QM W4896.

Paratypes: Cameron Falls track, Mt Tamborine,
SE.Q. V. Davies, E. Howell, 22.vi.75, 2?, 2$, QM W4897.

Description of Female

CL 1-16; CW 0-74; AL 1-56; AW LOO.

Cephalothorax and legs are a pale grey-brown.
Small white ‘thoracic patches’ on each side of fovea
are present. The abdomen is a dark grey-black with
pale pattern posteriorly (Fig. 16). There are 4 teeth
on retromargin of chelicera. The labium is wider
than long 1 :0-62. The sternum is longer than broad
1:0-89.

Notation of Spines: Palp: tibia, p.1.2; tarsus,
numerous small spines. First leg: femur, d.l.p.l.
distal; tibia, v. 2. p.0.1. 1.1.1. r.0. 1.1. 1.1; metatarsus,
v. l.p.0. 1 . 1 . l.r.0. 1 . 1 . 1 ; tarsus, 0. Second leg: femur,
d.l. 1.0; tibia, v.L2.p.O.O.Ll.Ll.r.O.O.L0.1.1; met-
atarsus, d. 1,1. 0,0. p. 1.1. 1.1. r.0. 1.1.1. Third leg:
tibia, d. 0. 1. 0.v. 1.0. l.p.l.l.O.r. 1.1.0; metatarsus
with numerous small spines. Fourth leg: tibia,
v.LO.l.l.p.O.l.O.O.r.0. 1.0.0; metatarsus with sev-
eral spines. Calamistrum 7 hairs proximal.
AS:cribellum 1:1-10.

The epigynum (Fig. 17a, b) is very similar to D.
spinipes but has weak lateral ridges. It is to be noted
that in mounting the epigynum for drawing the sac-

TABLE 4: Leg Measurements of D. nemoralis 7 and (J)

Leg

Femur

Patella

Tibia

Metatarsus

Tarsus

Total

i

0-80(1-00)

0-36(0-44)

0-74(1-00)

0-62(0-84)

0-26(0-34)

2-78(3-62)

ii

0-70(0-78)

0-32(0-36)

0-50(0-64)

0-48(0-62)

0-24(0-28)

2-24(2-68)

iii

0-52(0-60)

0-28(0-30)

0-30(0-38)

0-38(0-54)

0-20(0-24)

1-68(2-06)

iv

0-78(0-90)

0-30(0-34)

0-54(0-74)

0-62(0-82)

0-26(0-32)

2-50(3-12)

palp

0-34(0-40)

0-18(0-18)

0-22(0-22)

—

0-30(0-58)

1-04(1-38)

406

MEMOIRS OF THE QUEENSLAND MUSEUM

Fig. 15: $ D. tamborinensis. 15a, palp, dorsal; 15b, palp, ventral.

Figs. 16-17: $ D. nemoralis. 16, abdomen, dorsal; 17a, external epigynum, cleared; 17b, internal epigynum.
Fig. 18: D. nemoralis. 18a, palp, dorsal; 18b, palp, ventral; 18c, palp, retrolateral.

feU

DAVIES: DARDURUS, A NEW GENUS OF AMAUROBIID

407

like entrance ducts have been displaced (cf. Fig.
19d) so that the ducts to the spermathecae appear
to rise anteriorly.

Variation: Cephalothorax lengths were

1-1 4—1-26.

Description of Male

QM W4897: CL 1 26; CW 0-88; AL 1 36; AW
0 - 88 .

Cephalothorax length of the other $ 1-18.

Colour and pattern are similar to

Notation of spines in c $ similar to $ except in
following: First leg: femur, prolateral distal spine
absent. Second leg: tibia, v. 1.2. p.0.0.1. 1.1.1.
r.0. 0.1.0. 1.(1). Third leg: tibia, d.0.1.0.v.(l).0.1.
p.l.l.O.r. 1.0.0. Fourth leg: tibia, v. 1.0. 1.1-2.
p.O.l.O.O.r.0. 1.0.0.

Male palp: (Fig. 18a, b, c). The tibial apophysis
lacks the median protuberances of D. spinipes. The
embolus is retrolateral to the apophysis and rests
on a membraneous conductor.

Dardurus saltuosus n. sp.

(Figs. 19e, 20)

Material Examined

Holotype: Yabbra State Forest, Richmond Range, N.
N.S.W., R. Raven, 28.ix.74, ?, QM W4898.

Paratypes: Yabbra State Forest, Richmond Range, N.
N.S.W., R. Raven, 28.ix.74, 1^, 1 juv., QM W4899.

Description of Female

CL 1-36; CW 0-90; AL 1-60; AW LOO.

Cephalothorax and legs light brown, abdomen
grey-black with lighter pattern. ‘Thoracic patches’
at each side of fovea. Four teeth on retromargin of
chelicerae. Labium wider than long 1:0-60. Ster-
num longer than wide 1:0-86.

Notation of spines: Palp: tibia, p.1-2.2; tarsus,
numerous spines. First leg: femur, d.l.l.O.p.l
distal; tibia, v.2.p.0.1.1.Ll.(l).r.0. 1. 1.1.1; meta-
tarsus, v.Lp.O.l.l.l.r.0. 1.1.1; tarsus, 0. Second
leg: femur, d. 1.1.0; tibia, v.2.p.0. 1.1.1. l.r.0.1.0.1.1;
metatarsus, d.l. l.O.O.p.l. 1.1. l.r.0. 1.1.1; Third leg:
tibia, d.O.v.l.O.l.p.O.l.l.r.O.O.l; metatarsus, num-
erous small spines. Fourth leg: femur, d. 1.0.0; tibia,

v.l.l.l.l.p.0.1.0.(l).r.0. 1.0.0; metatarsus, several
spines. Calamistrum is well developed with 12
hairs. AS:cribellum 1:1-18.

The epigynum (Fig. 19e) has well developed
lateral ridges and is twice as wide as long.

Description of Male

QM W4899: CL 1-36; CW 0-92; AL 1-33; AW
0-92.

Colour and pattern similar to ?. ‘Thoracic
patches’ absent.

Notation of spines: S differs from 9 in following
spination. First leg: femur, d. 1.0.0. Fourth leg:
tibia, v. 1.0. 1 .0. 1 .p.0.1.0. 1 .0.r.0. 1 .0. 1 .0.

Male palp: (Figs. 20a, b, c). The tibial apophysis
has large anterior and posterior protuberances but
no median ones. The embolus is retrolateral to the
apophysis as in D. spinipes.

Dardurus agrestis n. sp.

(Figs. 19f, 21)

Material Examined

Holotype: Open grazed sclerophyll area, Black Duck
Creek, near Junction View 90 km. W. Brisbane, SE.Q., V.
Davies, 24.i.73, $, QM W4900.

Paratypes: Black Duck Creek, SE.Q., V. Davies,
24.i.73, 1 penult. <$, 2 juv., QM W4901.

Description of Female

CL 1-30; CW 0-88; AL 1-48; AW 1-00.
Cephalothorax and legs light brown, abdomen
grey-black with lighter pattern. ‘Thoracic patches’
not visible. Four teeth on retromargin of chelicerae.
Labium almost twice as wide as long 1:0-56.
Sternum longer than wide 1:0-88.

Notation of spines: Palp: tibia, p. 1.2; tarsus,
numerous small spines. First leg: femur, p. 1. distal;
tibia, v. 2. p. 0. 1. 1.1. l.(l).r.0. 1.1.1. 1.(1); metatarsus,
v.Lp.O.l.l.l.r.0. 1.1. 1.(1); tarsus, 0. Second leg:
femur, d.1.0; tibia, v.2.p.0.1.LLLl.r.0.1. 0.1.0. 1;
metatarsus, d.l. l.O.O.p.l. 1.1. l.r.0. 1.1.1. Third leg:
femur, d.l; tibia, d.O.l.O.v.LO.l.p.l.LO.r.O.1.1;
metatarsus, numerous small spines. Fourth leg:
tibia, d.O.p.O.O.O.l.O.r.O.O.O.l.O.v.l. 1.1. 0.2; meta-
tarsus, several spines. Calamistrum absent.
ASxnbellum 1:1-22.

TABLE 5: Leg Measurements of D. saltuosus $ and (J)

Leg

Femur

Patella

Tibia

Metatarsus

Tarsus

Total

i

0-96(1-12)

0-42(0-46)

0-88(1-06)

0-76(0-94)

0-30(0-34)

3-32(3-92)

ii

0-74(0-86)

0-36(0-40)

0-60(0-70)

0-56(0-68)

0-26(0-26)

2-52(2-90)

iii

0-62(0-68)

0-32(0-26)

0-36(0-42)

0-50(0-60)

0-22(0-24)

2-02(2-20)

iv

0-90(1-00)

0-38(0-38)

0-68(0-80)

0-76(0-96)

0-32(0-36)

3-04(3-50)

palp

0-38(0-40)

0-24(0-20)

0-30(0-42)

—

0-40(0-70)

1-32(1-72)

408

MEMOIRS OF THE QUEENSLAND MUSEUM

Fig. 19: External epigyna. 19a, D. spinipes; 19b, D. tamborinensis; 19c, D. silvaticus; 19d, D. nemoralis ; 19e, D.
saltuosus; 19f, D. agrestis.

Fig. 20: T D. saltuosus. 20a, palp, dorsal; 20b, palp, ventral; 20c, tibia, with tarsal scar.

Fig. 21: $ D. agrestis. Internal epigynum.

DAVIES: DARDURUS, A NEW GENUS OF AMAUROBIID
TABLE 6: Leg Measurements of D. agrestis l

409

Leg

Femur

Patella

Tibia

Metatarsus

Tarsus

Total

0-98

042

0-92

0-80

0-26

3-38

ii

0-80

0-36

0-64

0-64

0-24

2-68

iii

0-60

0-32

0-38

048

0-22

200

iv

0-90

0-34

0-72

0-78

0-32

3-06

palp

042

0-22

0-30

—

0-40

1-34

The epigynum (Fig. 19f, 21) has well marked
lateral ridges.

Variation: A calamistrum of 7 hairs is present in
the juveniles.

Description of Male

No adult 3 has been collected. Measurements of
a penultimate 3 are: CL 1-20; CW 0-74; AL 1-42;
AW 0-88.

Key to Identification of Dardurus spp.

1. Four retromarginal teeth on chelicera; em-
bolus retrolateral to apophysis; calamis-
trum usually present 2

Five retromarginal teeth on chelicera; em-
bolus prolateral to apophysis; no calam-
istrum . . . . . . D. tamborinensis

2(1) Fossa of epigynum almost as wide as

epigynum 3

Fossa of epigynum not exceeding \ width of

epigynum . . 4

3(2) Epigynum with marked lateral ridges

D. agrestis

Epigynum with weak lateral ridges . .

D. silvaticus

4(2) Epigynum with marked lateral ridges 5

Epigynum with weak lateral ridges . .

D. nemoralis

5(4) External epigynum 2\ x wide as long; 3
tibial apophysis with anterior, median and
posterior protuberances . . D. spinipes
External epigynum 2 x wide as long; q tibial
apophysis with anterior and posterior
protuberances. . . . . . D. saltuosus

The decorated tubes of the different species are
indistinguishable. The only similar tube seen is that
of an undescribed sedentary clubionid which was
found while looking for Dardurus in M.E. Queen-
sland and NE. Queensland. It has a small flat silk
disc with two openings and is found under logs in
similar positions to Dardurus. At present the latter
has a distribution limited to SE. Queensland and N.
New South Wales.

Small pale ‘thoracic patches’ are present on the
mature females of D. spinipes , D. silvaticus, D.
tamborinensis (ecribellate), D. nemoralis and D.
saltuosus. The $ palpal tarsus seems unusually large
in all the species. These characters, for which no
explanation is offered, may be useful in placing
Dardurus in a higher taxon.

Cribellum and Colulus

The cribellum of Dardurus spp. provides a
further example of the reduction and eventual loss
of this spinning organ in the amaurobiids. D.
spinipes, D. silvaticus and D. tamborinensis were
examined with the scanning electron microscope
for the presence of spinning tubes on the cribellum.
D. silvaticus was found to have many more tubes
than D. spinipes and there was none on the colulus
of D. tamborinensis. There is further correlation in
the reduction of the calamistrum from 9 hairs in D.
silvaticus to 7 thin hairs in D. spinipes to the absence
of a calamistrum in D. tamborinensis. Unlike the
cribella of the other species, the colulus width of D.
tamborinensis is narrower than that of the anterior
lateral spinneret but it is still flat and semi-circular.
The unsegmented structure of the spinning tubes is
in contrast to the segmented tubes of cribella of
other spiders examined. These included Jxeuticus
longinquus ( = I. martins) (Plate 60C, D) Stiphidion
facetum (Plate 60E, F) Paramatachia decorata and
several undescribed cribellates all of which are
known to spin webs. The scanning micrographs
show interesting differences in density of the tubes
in I. longinquus and S. facetum which may be useful
in separating the two groups. The reduction in
number of spinning tubes in Dardurus spp. and
their unsegmented nature as well as the absence of
any web outside the tube suggest that the spinning
function of the cribellum is reduced or even lost. It
is certainly lost in D. tamborinensis. If the absence
of web outside the tube of the sedentary Dardurus is
associated with a decrease in dependence on the
snare for capturing food, then perhaps this is
compensated for by the development of the very
long spines on the anterior legs. They are common
features of many hunting spiders and are used in
the capture of food.

410

MEMOIRS OF THE QUEENSLAND MUSEUM

Position in the Amaurobiidae

Dardurus does not fit easily into any of the
recognised subfamilies though it has affinities with
Forster and Wilton’s (1973) Amphinectidae not-
ably in relation to the large chitinised epigynum
with lateral ridges, the undivided fossa, and the
presence of ‘thoracic patches’. The undivided
cribellum, the presence of a patellal apophysis, and
perhaps the habit suggest Lehtinen’s (1967) Ma-
tachiinae.

There are so many amaurobiid spiders yet to be
described from Australia that it seem preferable to
leave the decision of placement in the family until
more are described.

ACKNOWLEDGMENTS

Special thanks are due to John Hardy of the
Electron Microscope Centre, University of
Queensland for his skill and enthusiasm. Thanks
also to Sue Hiley for tracing and shading Figs. 3, 4,
7, 8, 10, 13 and 15.

LITERATURE CITED

Forster, R. R. and Wilton, C. L., 1973. The spiders of
New Zealand. Part iv. Otago Museum Bulletin No.
4. Dunedin.

Lehtinen, P. T., 1967. Classification of the Cribellate
Spiders and some allied families, with notes on the
evolution of the sub-order Araneomorpha. Am.
Zool. Fenn. 4: 199-468.

412

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 59

A: D. spinipes. Camouflaged tube on under side of fallen log x
3-6.

B: D. silvaticus. Scanning electron micrograph of cribelium and
anterior lateral spinnerets x 230.

DAVIES: DARDURUS , A NEW GENUS OF AMAUROBIID

Plate 59

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 60

Scanning electron micrographs A, $ D. silvaticus , spinning
tubes on cribellum x 1450; B, $ D. spinipes, cribellum x 1000,
C, ? Ixeuticus longinquus, half of divided cribellum x 450; D, I.
longinquus , spinning tubes of cribellum x 6600; E, $ Stiphidion
facetum, cribellum x 150; F, S. facetum, spinning tubes of
cribellum x 1800.

DAVIES: DARDURUS, A NEW GENUS OF AMAUROBIID

Plate 60

Mem. QdMus . 17(3): 413-6. [1976]

THE WOLF SPIDERS OF AUSTRALIA (ARANEAE: LYCOSIDAE): 7.
TWO NEW SPECIES FROM VICTORIA

R. J. McKay

Queensland Museum

ABSTRACT

Two new species of Wolf Spider, Lycosa mayama and Lycosa neboissi, from near Melbourne,
Victoria, are described.

An examination of the Wolf Spiders in the
collection of the National Museum, Victoria
(NM), kindly forwarded on loan by Mr A. Neboiss,
disclosed the presence of two undescribed species in
the suburban area of Melbourne. Because little
systematic collecting of lycosids has been done in
Victoria, the occurrence of new species near
Melbourne is not unexpected.

Thirteen species of the subfamily Lycosinae have
been recorded from Victoria (McKay 1973); New
South Wales has over 40 species recorded, and
Queensland has 32 plus a number of undescribed
species. A considerable amount of collecting will be
necessary to fully document the Australian Wolf
Spiders and at this early stage, although some most
interesting distribution patterns and speciation
problems are emerging, such questions are prema-
ture with so much of the lycosid fauna unknown.

Lycosa mayama sp. nov.

(Fig. la-d)

Material Examined

Holotype: National Museum, Victoria (NM), \ M (CL
8-3 mm), Werribee, Victoria, collector unknown, 22.ix.68.
In spirit.

Paratypes: Broadmeadows, Melbourne, Victoria, F.
P. Spry, 27.ix.07, $M (CL 6 5) 2M (CL 6-8), NM.

Description (Based on the holotype.)

Carapace medium brown with a very faint
longitudinal stripe commencing behind the PL eyes
and continuing to the posterior margin; no lateral
band visible; some darker radiating marks are
barely visible; paturon and fang dark brown;
labium, maxillae, sternum and coxae brown. Ab-
domen dark brown above and below; on the dorsal
surface are numerous scattered light and dark

spots; ventral surface with a dark brown (possibly
black in life) area from the epigastric furrow to the
base of the spinnerets, surrounded by scattered
light spots within a dark rounded field (see Fig. 1 a);
sides streaked with light and dark brown flecks.
Legs brown with diffuse dark brown to blackish
blotches.

Anterior row of eyes very slightly procurved,
AM larger than AL. Ratio of eyes AM:AL:PM:PL
= 23:17:43:32; distance AM:AM 8, AM.AL 5,
AM:PM 12, AL:PM 17, PM:PM 23. Clypeus to
AM 24. Length of first eye row 100, length of
second eye row 103 micrometer units. The distance
between the AM and AL eyes to the PM eye was
measured between the lens of the eyes.

Chelicerae with three promarginal teeth, the
middle one largest; three retromarginal teeth, the
two internal ones of about equal size, the exterior
one much smaller and almost obsolete. Labium
longer than wide.

Epigynum with a short median guide; the
transverse guide narrow at the base of the median
guide, broadening distally to almost truncate ends
(Fig. lb).

Femur of fourth leg with 1 + 1 + 1 dorsal
spines.

Variation: The colour pattern of the holotype is
partly faded due to preservation. The smaller
paratypes lack the pattern on the ventral surface of
the abdomen but in both specimens a faint
longitudinal stripe is present on the carapace. One
paratype has a brown longitudinal wedge-shaped
median stripe on the anterior dorsal surface of the
abdomen; this stripe is outlined in dark brown with
a white spot on either side midway and a pair of
more widely spaced white spots near the posterior
tip.

414

a

MEMOIRS OF THE QUEENSLAND MUSEUM

C

Fig. 1: a-d, Lycosa mayama : a, ventral surface of abdomen of holotype; b, epigynum of holotype; c-d epigvna of
paratypes.

e ~i’ L v co s a neboissi. e, carapace of holotype; f, epigynum of holotype; g, palpal organ of paratype; h, palpal organ
rotated slightly; i, palpal organ lateral 'dew.

McKAY: WOLF SPIDERS OF AUSTRALIA: 7

415

TABLE 1: Measurements of Leg Segments of
Holotype of L. mayama in mm

Leg

Femur

Patella

Tibia

Metatarsus

Tarsus

1

54

2-9

4-1

3-6

2-3

2

4-8

2-8

3-4

34

24

3

46

2-3

2-9

3-6

2-3

4

64

2-8

4-7

5-7

2-7

Palp

3-0

1-5

1-6

—

2-3

The two paratypes have the ratio of eyes
(expressed as per cent of the total width of the first
row of eyes) AM:AL:PM:PL = 21:22:17:16;
41:45:35:37; distance AM:AM 9, 11; AM:AL 6, 8;
AM:PM 17, 11; AL:PM 20, 16 (from lens of eye);
PM:PM 26, 23. The clypeus equals the diameter of
the AM. Length of the first eye row in micrometer
units 85, 88; length of the second eye row in
micrometer units 90, 93.

The epigyna of the holotype and the two
paratypes are illustrated (Fig. lb-d). Mature males
were not collected.

Diagnosis: A dark brown species; epigynum
with a short narrow median guide, and a transverse
guide that is narrow in the mid-section and has
widely expanded truncate ends.

Life History
Unknown.

Habitat

The two paratypes were collected from beneath
stones.

Distribution

Within 35 km of Melbourne, Victoria.
Derivation

From the aboriginal ‘mayama’ meaning stone.

Lycosa neboissi sp. nov.

(Fig. le i)

Material Examined

Holotype: National Museum, Victoria (NM), ,M
(C.L. 3-6 mm), Rosanna, northeast of Melbourne,
Victoria, collected by A. Neboiss, 18 July 1954. In spirit.

Paratype: Rosanna, A. Neboiss, 5 June 1954, c?M
(C.L. 3-5 mm), NM.

Description

Holotype: Carapace light brown with a faint
light median longitudinal stripe including a broad
paler area behind the PL eyes, sides darker brown
with radiating black-brown marks, and a pale
lateral band (Fig. le); area around eyes black;
paturon medium brown; labium, maxillae, sternum

and coxae uniform pale brown. Abdomen above
pale brown to greyish with some scattered dark
brown spots and blotches arranged in pairs along
the median line, but somewhat distorted due to the
wrinkled state of the abdomen; ventral surface and
sides uniform grey brown. Legs pale brown; femora
with dark blotches on the retrolateral surface and
to a lesser extent the prolateral, dorsal and ventral
surfaces; patellae with a conspicuous dark brown
spot distally on either side; tibiae with three or four
lateral blotches; remaining leg segments uniform
pale brown or very faintly blotched.

Anterior row of eyes strongly procurved, the AL
situated on a level below the AM; AM larger than
AL; PM large, the distance between slightly greater
than their diameter. Ratio of eyes AM:AL:PM:PL
= 9:6:24:17; distance AM:AM 9, AM:AL 6,
AM:PM 6, AL:PM 6, PM:PM 25. Clypeus to AM
14. Length of first eye row 46, length of second eye
row 70 micrometer units.

Chelicerae with three promarginal teeth, the
middle one largest; three retromarginal teeth of
equal size. Labium longer than wide.

Epigynum rounded, the guide occupying all
the epigynal depression (Fig. 10-

Femur of fourth leg with 1 + 1 + 1 spines.

Paratype: Carapace similar to the female.
Below carapace on the membrane above each coxa,
are a pair of dark brown to black blotches, the
posterior one of each pair being most conspicuous;
the female has similar blotches above the coxae but
these are not as well defined as in the male. Legs
marked as in the female. Abdomen with a broad
pale longitudinal median stripe, rounded anteriorly
and gently tapering to touch the base of the
spinnerets; sides darker with fine pale spots and
numerous streaks and flecks of black-brown form-
ing small black-brown blotches along the pos-
terior margin of the pale longitudinal stripe; ventral
surface with the dark pigmentation extending from
the posterior sides to surround the spinnerets, and
two widely spaced dusky bars commencing at the
base of the lung-books and converging towards the
middle of the ventral surface where they continue
parallel to the base of the spinnerets.

TABLE 2: Measurements of Leg Segments of
Holotype of L. neboissi in mm

Leg

Femur

Patella

Tibia

Metatarsus

Tarsus

1

2-5

1-3

20

1-8

1-0

2

2-3

F3

1-7

1-9

11

3

2-0

12

1-5

20

0-9

4

2-8

1-2

24

3-1

0-9

Palp

1-2

0-6

0-8

10

416

MEMOIRS OF THE QUEENSLAND MUSEUM

Anterior row of eyes strongly procurved, the
AM larger than the AL; PM larger than the
distance between. Ratio of eyes AM:AL:PM:PL =
8:6:26:20; distance AM:AM 8, AM:AL 7, AM;PM
5, AL:PM 5, PM:PM 23. Clypeus to AM 9. Length
of first eye row 44, length of second eye row 71
micrometer units.

Palpal organ with a well developed median
apophysis which is narrow at the base but flaring
out into a broad structure with a sharp blade-like
cusp on the inner tip and a curved hook on the
outer tip (Fig. lg, h). The complete structure of the
palpal organ was not studied.

Diagnosis: A small species with a rounded
epigynum having an expanded guide that fills the
depression; male palpal organ complex (Fig. lg, h,
i); patellae with a conspicuous dark brown spot on
each side, and dark marks on the membrane above
the coxae.

Life History

Unknown.

Habitat

Collected from garden in a recently developed
suburb.

Distribution

Suburban area of Melbourne, Victoria.
Derivation

Named after Mr A. Neboiss, National Museum,
Victoria in appreciation of assistance given, and for
collecting the type material of this new species.

ACKNOWLEDGMENTS

I record my appreciation of the assistance given
to me by Mr A. Neboiss during my visit to the
National Museum, Victoria, and his efforts to
collect further specimens of Lycosa neboissi for
dissection.

LITERATURE CITED

McKay, R. J., 1973. The wolf spiders of Australia
(Araneae: Lycosidae): 1. The bicolor group. Mem.
QdMus. 16 (3): 375-98.

Mem. QdMus. 17(3): 417-23. [1976]

THE WOLF SPIDERS OF AUSTRALIA (ARANEAE: LYCOS1DAE): 8.
TWO NEW SPECIES INHABITING SALT LAKES OF WESTERN AUSTRALIA

R. J. McKay
Queensland Museum

ABSTRACT

Two new species of Wolf Spider, Lycosa alteripa and Lycosa salifodina, are described from salt
lake habitats in Western Australia.

Australia, due largely to position, size and
topography is, with the exception of Antarctica,
the driest of the continents; about 93 per cent of the
land mass receives less than 510 mm, and 37 per
cent receives less than 250 mm of rainfall per year.
The driest area is located in South Australia,
around Lake Eyre, where on the average less than
125 mm of rain falls annually. Evaporation in
inland Australia is exceedingly high, and therefore
permanent bodies of standing water — lakes,
lagoons, and ponds — are rare. Much of the land
surface is of low topography and is characterised
by ephemeral rivers draining into salt lakes, and
vast areas of uncoordinated drainage.

This area of internal drainage has innumerable
playas, some saline, others gypsiferous or clayey.
The small claypans and salinas are the sumps for
local drainage and periodically fill with the runoff
from surrounding areas which may be very exten-
sive during cyclonic rains; the water evaporates
leaving brine swamps or expanses of salt-encrusted
clay, devoid of vegetation or surrounded by
samphire ( Arthrocnemum ) communities with salt-
bush and spinifex zones more distant. Some
claypans have a continuous samphire margin, the
inner part of which breaks up into clumps, and then
on to vast clay flats that progressively become
encrusted with evaporites until a layer of salt is
formed. The largest salt lakes are frequently near or
below sea level, and those of Lake Eyre and Lake
Frome form a giant internal drainage basin some
1,250,000 km 2 in area. The evaporites are pre-
dominantly sodium chloride (90-95%) with lesser
amounts of magnesium sulphate (5-7%) mag-
nesium chloride (up to 4%) and calcium sulphate
(up to 2.5%) (Johns and Ludbrook 1963), and
originate from ‘rocks which crop out in the

catchment area and, carried dominantly by shallow
groundwater to a hot arid region of interior
drainage, has then been concentrated at various
times during the Quaternary’ (Wopfner and Twid-
ale 1971, p. 138). The depth of the salt deposit
varies from a thin encrustment to about 44 cm or
more. Some playas or salinas are arranged in chains
of white saltpans which may fill and have con-
nected flow in periods of heavy rainfall, only to
evaporate and redeposit the salt along the drainage
system. The salt layer is therefore dissolved and
reformed at irregular intervals, but may remain
stable for many years during drought periods.

Figure 1 shows the extent of the uncoordinated
drainage system in mainland Australia, and the
major salt lake systems. The system is undoubtedly
an ancient one and sufficiently stable to provide a
suitable habitat for lycosid spiders. Hickman
(1944) described Pardosa eyrei from the ‘surface of
North Lake Eyre two and a half miles from the
shore’ but no details were given concerning the
burrow or life history. Immature Lycosa specimens
have also been collected from lakes Callabonna,
Ngapakaldia, and Palankarinna, in South Aus-
tralia.

Two salt encrusted habitats were investigated in
Western Australia; Lake Lefroy near Widgie-
mooltha, by Mr Athol Douglas of the Western
Australian Museum, and a small claypan near the
mouth of the Fitzgerald River by the author. The
Wolf Spiders are readily captured using a head
torch at night; the light is carried on the forehead,
and by looking along the beam, the reflection of the
light, appearing as pin points of pale emerald
green, may be seen from spiders’ eyes. On the
surface of the salt lake the spiders are quite easily
detected by this means, and individuals were

418

MEMOIRS OF THE QUEENSLAND MUSEUM

Fig. 1: Mainland Australia showing area of internal or uncoordinated drainage and distribution of salt lakes and
claypans.

spotted a hundred yards out on the salt. Two quite
unrelated species were discovered and are de-
scribed below.

Lycosa alteripa sp. nov.

(Fig. 2a-e)

Material Examined

Holotype: Western Australian Museum WAM 71-40,
c?M (C.L. 9-6 mm), claypan near mouth of Fitzgerald
River, W.A., collected by R. J. McKay and R. Prince, 1 1
July 1970. In spirit.

Paratypes: Fitzgerald River (data as holotype), I P
2<JM 2rjP IJ WAM 71-41-6. Claypan on road to
Fitzgerald River — Susetta River junction, 13 July 1970,
RJM, 1J WAM 71-47. Salt lake near Israelite Bay, W.A.,
19 April 1974, A. V. Thomas, ! M WAM 74-501.

Description (Based on the holotype.)

Carapace pale grey to smoke white; radiating
narrow wedges of charcoal grey originate near the
white foveal area, almost reach the lateral margin,
and have a pearl grey to white spot immediately
behind or at their extremity; face silver grey;
paturon smoke grey, fang black; labium and
maxillae pale brown, sternum dark brown covered
with grey hair; coxae pale grey-brown. Abdomen
smoke grey above and below, an ornate pattern of
charcoal grey is present on the dorsal surface (Fig.
2a), sides and venter uniform pale smoke grey. Legs
pale grey-brown to pinkish grey with white hair
and conspicuous dark brown spines; femora with
two slightly darker brown blotches above, sides

McKAY; WOLF SPIDERS OF AUSTRALIA: 8

419

and ventral surface pale; patellae with the proximal
half mid-brown on the dorsal surface, pale below;
tibiae of the first pair of legs dark charcoal brown,
remaining leg segments dark brown; tibiae of the
third and fourth legs with a basal dark half-band, a
mid half-band of brown, and the extremity dark
tipped; remaining leg segments pale brown; palp
pale smoke grey, cymbium darker grey.

Anterior row of eyes procurved, AM larger than
AL. Ratio of eyes AM:AL:PM:PL = 16:14:35:27;
distance AM:AM 7, AM:AL 5, AM:PM 6, AL:PM
7, PM:PM 23. Clypeus to AM 10. Length of first
eye row 75, length of second eye row 89 micrometer
units.

Chelicerae with 3 promarginal teeth on the right
side, the middle one largest; 4 promarginal teeth on
the left side, the second one largest; three retro-
marginal teeth of about equal size on both sides.
Labium longer than wide.

The palpal organ differs from all the known
Australian Lycosa species in having the secondary
conductor sclerotised (Fig. 2b, d). The median
apophysis has a recurved hook-like tip. A dissec-
tion of the male palp was not made, and therefore
the structure remains to be studied in full when
additional specimens are collected.

Variation: Males, females and juveniles are
similarly patterned, the very small examples being
somewhat more dark grey. Some specimens have
the light background colour almost pure white with
a grey-green pattern dorsally, occasionally tinged
with mauve or pink. The undersurface of the
abdomen is white without a pattern in males, but
may be tinged with yellow in mature females.

The eye diameters and interspaces of the holo-
type and paratypes are given as a per cent of the
total width of the first row of eyes, in Table 2; all
specimens have the first row of eyes shorter than
the second row.

The male palpal organ is illustrated in Fig. 2b, d;
the epigynum of the female in Fig. 2c, and the
internal genitalia of the female in Fig. 2e.

Size Range: Mature female 11-1 mm; Mature
males 9-0 to 9-6 mm.

TABLE 1 : Measurements of Leg Segments of
Holotype of L. alteripa in mm

Leg

Femur

Patella

Tibia

Metarsus

Tarsus

1

7-7

3'3

5*6

8*7

3-7

2

7-6

3-3

5-5

9*2

3-7

3

7-2

3-2

4-6

9.4

3-8

4

90

3-2

6-3

113

4.4

Palp

40

1*8

20

—

—

Diagnosis: The mature female epigynum has a
short median guide with a recurved transverse
guide (Fig. 2c). The male palpal organ has a
sclerotised secondary conductor with a hook-like
median apophysis. All specimens are pale grey to
smoke-white without a ventral pattern on the
abdomen.

Life FIistory
Unknown.

Habitat

Claypans of southeastern Western Australia. All
specimens were collected more than 2 to 3 metres
from the margin of the claypan where the dried
grey clay was encrusted with salt crystals. The
mature males were found widely scattered over the
middle of the dry claypan and none were seen near
the samphire covered margins of the claypan where
a rather robust brown Lycosa species (?L. egena )
was very common.

Burrow

One juvenile was collected from a shallow open
burrow near the middle of the claypan but as no
other burrows were located it is possible that the
burrows of mature specimens are closed with a
tightly fitting door or lid.

Distribution

Southeastern Western Australia, but possibly
widespread on Western Australian inland and dry
coastal claypans and saltpans.

Derivation

From the aboriginal ‘alteripa’ meaning claypan.

TABLE 2: Eye Diameters and Interspaces of Lycosa alteripa converted to per cent of the
Total Width of the First Row of Eyes

Regd. No.

Sex

C.L. AM

AL

PM

PL

AM:AM

AM:AL

PM:PM

AM:PM

AL:PM

Holotype

c?M

9-6

21

19

47

36

9

7

31

8

9

WAM74-501

ZM

111

20

15

45

40

10

10

34

11

11

WAM71-41

c?M

9.4

21

18

47

41

8

8

33

11

8

WAM71-42

m

90

20

17

47

44

10

9

36

9

9

420

MEMOIRS OF THE QUEENSLAND MUSEUM

e

Fig. 2: Lycosa alteripa. a, holotype; b, male palpal organ of WAM 71-41 slightly expanded; c, epigynum of WAM 71-
501; d, male palpal organ of WAM 71-41; e, internal genitalia of WAM 71-501.

McKAY: WOLF SPIDERS OF AUSTRALIA: B

421

Lycosa salifodina sp. nov.

(Fig. 3a-g)

Material Examined

Holotype: Western Australian Museum WAM 69-
369, ?M (C.L. 14-1 mm). Lake Lefroy, Widgiemooltha,
W.A., collected by A. M. Douglas and L. Koch, 24
January 1968. In spirit.

Paratypes: Lake Lefroy, Widgiemooltha, W.A., 24
January 1968, A. M. Douglas, L. E. Koch, $M WAM 69-
104, 3$M WAM 69-366-8, 3?M 2<^M 22J WAM 69-370
-94; March 1970, A. M. Douglas, 49P WAM 70-56-9,
$M WAM 71-1747; ll.iv.1971, A. M. Douglas, 1$M 1J
QM W4543.

Description (Based on the holotype.)

Carapace brown, covered with pink-tinged white
hair; where the hair has been rubbed off some faint
brown radial lines may be discerned, but these are
not present in life; face white with a pink tinge;
paturon white with the lower part black; labium
and maxillae black; sternum and ventral surface of
coxae black. Abdomen with the dorsal surface
pink-white; a brownish longitudinal stripe reaching
half length of abdomen with two white spots on
each side at the middle; two pale-brown chevron-
shaped transverse bars posteriorly; ventral surface
pink to white with two dark brown to black-brown
longitudinal stripes tapering posteriorly and fading
before the spinnerets, the area between the two
stripes is brownish-pink in life becoming very pale
brown after preservation (Fig. 3a); area anterior to
epigastric furrow and over lung-book covers
brown, becoming dark brown after preservation.
Legs pink to ash-grey, spines black; extremities of

TABLE 3: Measurements of Leg Segments of
Holotype of L. salifodina in mm

Leg

Femur

Patella

Tibia

Metatarsus

Tarsus

1

10-3

5-3

7-5

8-7

3-5

2

9-7

50

7-5

90

3-5

3

9-8

4-5

6*5

8-8

3-9

4

112

4-8

7-9

116

4-1

Palp

5-2

2-5

3-2

- —

3-3

Variation: Males and females are similarly
coloured; juveniles lack the pattern on the ventral
surface of the abdomen, and in mature specimens
this pattern may consist of two dark-brown to
black converging broad stripes or lines, sometimes
with a faint dusky line or black bar between, or
occasionally the area between the two or three bars
is pigmented with grey, brown, or black to the
extent that the pattern appears as a dark brown
somewhat rectangular blotch with the lateral
margins darker. Some specimens have the ventral
tips of the tibiae dark brown, grey or black distally;
in large males the anterior pairs of legs may have
the tibiae, metatarsi and tarsi dark brown, greyish
or almost black-brown with a greyish tinge. In life
the spiders may be white, off-white or pink-white in
colour.

Juveniles have very prominent PM and PL eyes.
The eye diameters and interspaces of the holotype
and six paratypes are recorded as a per cent of the
total width of the first row of eyes in Table 4.

Little variation was found in the shape of the
epigynum, that of the holotype and a paratype is
the femora, tibiae and metatarsi with a dark brown
to black edge; undersurface of metatarsi and tarsi
brown, the anterior pair of legs with the distal half
of the tibiae, the whole of the metatarsi and tarsi
black below, becoming dark brown after pre-
servation.

Anterior row of eyes procurved, AM larger than

AL, PM eyes large and conspicuous measuring
about 2 mm in diameter. Ratio of eyes
AM:AL:PM:PL = 1 1:7:31:29; distance AM:AM 7,

AM. AL 4, AM:PM 5-5, AL:PM 4, PM:PM 23.
Clypeus to AM 7. Length of first eye row 5 1 , length
of second eye row 79 micrometer units.

Chelicerae with three promarginal teeth, the
middle one largest; three retromarginal teeth of
about equal size. Labium longer than wide.

Epigynum elongate with a well developed
median guide somewhat expanded anteriorly and
terminating in a moderately broad transverse guide
(Fig. 3c).

TABLE 4: Eye Diameters and Interspaces of Lycosa salifodina converted to per cent of the
Total Width of the First Row of Eyes

Regd No.

Sex

C.L.

AM

AL

PM

PL

AM:AM

AM:AL

PM:PM

AMtPM

AL:PM

Holotype

$M

14 1

22

14

61

57

14

8

45

11

8

WAM69-104

QM

13-8

22

14

62

56

12

10

46

11

11

WAM69-366

2M

12 2

22

15

60

57

14

7

44

11

10

WAM69-367

$M

13-2

22

14

63

53

13

12

44

11

8

WAM69-368

QM

12-9

22

15

64

60

11

9

45

9

6

WAM69-370

QM

13-2

23

12

63

51

10

12

43

10

8

WAM69-371

d*M

12-3

23

15

67

49

9

9

40

12

10

422

MEMOIRS OF THE QUEENSLAND MUSEUM

b

a

e

of holotype; d, burrow construction; e, internal genitalia of WAM 69- 1 04; f, median apophysis and embolic guide of
WAM 69-371; g, male palpal organ of WAM 69-371.

McKAY: WOLF SPIDERS OF AUSTRALIA: 8

423

illustrated in Fig. 3c, b. The internal genitalia of
one paratype female is shown in Fig. 3e. The male
palpal organ has a blade-like median apophysis
(Fig. 3f) and a curved pointed embolic guide (Fig.
3g), The structure of the palpal organ resembles
that of some members of the bicolor group (McKay
1973) and is a fairly common basic type found in
many of the larger Lycosa species.

Size Range: Mature females C.L. 12T to 13-8
mm. Mature males C.L. 11-8 to 12-3 mm.

Diagnosis: A large species that inhabits salt-
pans; the epigynum is elongate (Fig. 3c); the eyes
are highly developed and quite prominent; the
ventral surface of the abdomen has two converging
dark brown bars in most mature specimens.

Life History

Mature males and females are present through-
out the summer months; juveniles are common in
April.

Habitat

All specimens were collected from the open salt
surface of Lake Lefroy, many being scattered all
over the salt lake and only a few specimens near the
periphery.

Burrow

Mr A. Douglas recorded the burrows as being
open, vertical, and some 15 cm (5 to 6 inches) deep
into the salt (Fig. 3d). The spiders were found to be
"sitting out in the middle of the lake perched
directly over the burrow entrance’. Nothing is
known about their behaviour, feeding habits,
physiology, or how the spiders survive during rain
or flood conditions.

Distribution

Lake Lefroy, Widgiemooltha, Western Aus-
tralia.

Derivation

From the latin meaning ‘salt-mine’.

DISCUSSION

The salt lakes can be considered as ‘island’
habitats with infrequent or no contact with each
other. These ‘islands’ vary in size from a few
hectares to many km 2 , and may be completely
isolated or joined by dried watercourses or ribbons
of salt encrusted clay deposits to form chains of
lakes (Fig. 1). Such ‘island’ habitats provide
valuable opportunities for studying the speciation
of the salt lake Wolf Spiders, and are ideal for
constructing models for the study of population
dynamics.

Nothing is known of the biology of the salt lake
species. The physiology of an animal that occupies
a burrow constructed in salt, and is subject to
extreme dessication, occasional floods, and high
temperatures, is worthy of investigation. Much
additional collecting is necessary to record the
lycosid species on the salt lakes, and their distri-
bution.

ACKNOWLEDGMENT

I gratefully acknowledge the assistance of Mr
Athol M. Douglas of the Western Australian
Museum, in collecting specimens of Lycosa sali-
fodina and describing the burrow.

LITERATURE CITED

Hickman, V. V., 1944. The Simpson Desert Expedition,
1939. Scientific Reports No. 1: Biology, Scorpions
and Spiders. Trans. R. Soc. S. Aust. 68: 18-48.
Johns, R. K., and Ludbrook, N. H., 1963. Investigation
of Lake Eyre. Geol. Surv. S. Aust. Rept Invest. 24:
1-104.

McKay, R. J., 1973. The wolf spiders of Australia
(Araneae: Lycosidae): Two new species of the
bicolor group. Mem. Qd Mus. 16(3): 375-98.
Wopfner, H., and Twidale, C. R., 1971. Geomor-
phological history of the Lake Eyre Basin. In
Jennings, J. N., and Mabbutt, J. A., ‘Landform
Studies from Australia and New Guinea' 434 pp
(Aust. National Univ. Press: Canberra).

Mem. QdMus. 17 ( 3 ): 425-47. [1976]

THE MACROBENTHOS OF BRAMBLE BAY, MORETON BAY, QUEENSLAND

W. Stephenson
Y. I. Raphael*
and

S. D. Cook

Department of Zoology,

University of Queensland

ABSTRACT

As part of a study of the possible effects of enlarging Brisbane Airport upon the biota of
adjacent areas, the macro-benthos of Bramble Bay was investigated. Sites at c. 1 km apart were
sampled in duplicate with an 0- 1 m 2 van Veen grab and sieved through 1 mm mesh. The data gave
two three-dimensional matrices; of ca 190 spp x 27 sites x 7 seasons (at three month intervals)
and of ca 190 spp x 48 sites x 2 years respectively.

The number of species is roughly half that from a current survey of Middle Banks and roughly
half that from a past study near Peel Island, both in Moreton Bay.

Analysis of the 3D matrices followed different patterns from those of Stephenson, Williams and
Cook (1974), and these are detailed. They involved transformations using log (n + 1); also the
species were standardized by totals prior to species classification. Data were scanned for
outstandingly large values, and interactions of sites and times were explored via times
classifications within site-groups.

Site classifications gave topographically coherent groups which generally followed an
onshore-offshore sequence. The onshore sites, those most likely to be effected by airport
construction, are characterized by Spisula and Xenophthalmus and have the highest populations
but the lowest Shannon diversities of those investigated. The 48 sites data showed a site-group in
proximity to the outflow of the Brisbane River.

Season classifications showed scarcely any grouping of similar seasons from two different years
but instead groups of species persisted for c. 6-9 months and were then replaced by other groups.
These temporal changes may be related to variations in runoff from land drainage and are most
obvious in the area closest to the Brisbane River outflow.

Changes with time occur in almost all species. The magnitude of the time-changes suggests that
‘base-line’ studies for environmental impact statements in comparable situations (shallow water
with some river influence) should involve extensive chronological replication. Prediction of the
effects of airport construction would have been hazardous on the basis of present data because of
other human activities in the area. A catastrophic flood ended the present study, caused
unpredicted changes in the biota, and its effects will be described in a later paper.

This investigation was sponsored by the Aus-
tralian Department of Housing and Construction
and by the Queensland Department of the Co-
ordinator General, and is part of a study of the
possible effects of reclamation of an area east of the
present Brisbane Airport. This would destroy
mangroves, partly fill certain tidal creeks, and alter
drainage patterns into the study area. It was hoped
that a ‘before’ study of the macrobenthos of the
area would lead to predictions of ‘after’ effects.

A relatively small area of Bramble Bay was
sampled for macrobenthos at three-month in-

tervals and a larger area was sampled on successive
Septembers. Experience gained at Peel Island in
Moreton Bay by Stephenson, Williams and Cook
(1974) suggested that a prolonged investigation
might be required and the present work was
planned to extend for an initial two years. An
extreme flood in January 1974 caused dramatic
changes in the biota and these will be considered in
a later paper.

The data form two separate three dimensional
groups: I of c. 190 species x 27 sites x 7 seasons,
and II of c. 190 species x 48 sites x 2 years. To

Present address Department of Fisheries, Colombo, Sri Lanka.

426

MEMOIRS OF THE QUEENSLAND MUSEUM

analyse data of this complexity requires numerical
methods; present methods follow immediately
upon the techniques of Williams and Stephenson
(1973) and Stephenson, Williams and Cook (1974).
They also follow up on the M.Sc. thesis of Raphael
(1974) which dealt with seasonal data to March
1973 and this should be consulted for many
relevant details e.g. sediment analyses, winds,
rainfall. The present study involves three more
seasons and more derivations from the data, and
modifies certain aspects of the methods and
conclusions.

Sampling Area and Sampling Details

Moreton Bay has been described by Stephenson,
Williams and Lance (1970), Maxwell (1970) and
Newell (1971). The relevant portion is shown in
Fig. 1 and the study area in Fig. 2.

Difficulties were encountered in selecting sampl-
ing sites and sampling times. Results obtained by
Stephenson, Williams and Cook (1974) suggested
that for one area of Moreton Bay it would have
been desirable to have intersite distances of c. 0-25
km, to obtain quintuplicate catches of 0T m 2 at

Fig. 1 : Moreton Bay showing localities mentioned in text. Thick line high water mark, thin line 5-5 m (3 fm), broken line
18-2 m (10 fm). Very heavy line encloses sampling area.

STEPHENSON ET. AL.: MACROBENTHOS OF BRAMBLE BAY

427

each site, and to sample at three-monthly intervals.
Sampling at this frequency from the wide area of
interest required relatively widely spaced sites and
less replication of catches.

An initial trial was conducted with sites ca 1 km
apart and with duplicate catches obtained by an 0- 1
m 2 van Veen grab. This revealed clear patterns of
species-groups within site-groups, and was adop-
ted for subsequent sampling. For the three-month
sampling a tolerably restricted study area was
selected (sites 1-27) and this gave the series I data.
The results suggested that less frequent studies of a
wider area would be desirable and hence sites 1-48
were sampled annually — this gave the series II
data. The latter were obtained in September
because this was the season of highest populations
in the previous study (Stephenson, Williams and
Cook 1974).

Raphael (1974) should be consulted for the
source of the depth contours on Fig. 2 (these differ
from those of published charts) and for such data
as exist on tides, winds and temperatures in the
area. There are no worthwhile data on salinities but
dilution effects are available indirectly from records
of rainfall and from overflow discharges over a weir

on the Brisbane River. Details are given in Raphael
(1974), and generalities are included in the present
paper.

Information on sediments is contained in Max-
well (1970) but additional data were obtained
during the present survey.

Each sample, comprising two catches from an
0-1 m 2 van Veen grab, was wet sieved on board
through a final aperture of 1 mm. The residue was
carefully examined for small specimens, parti-
cularly of gastropods and bivalves; only if these
were present the sievings were preserved and the
biota sorted at base.

Sediments

Sediments were sampled by selecting an aliquot
of the grab contents, with precautions against loss
of biota. For series I, sediment samples were
collected thrice (June 1972, Dec. 1972, June 1973)
and for series II, at each September.

After removal of organic matter by sodium
hypochlorite solution, proportions of particle sizes
were determined by wet sieving. Sieve sizes were (in
mm): 2 00, 1 00, 0-50, 0-25, 0 125 and 0 063 and

428

MEMOIRS OF THE QUEENSLAND MUSEUM

the corresponding fractions are described as:
gravel ( = shell grit), very coarse sand, coarse sand,
medium sand, fine sand and very fine sand
respectively. The fraction not retained by the 0 063
mm sieve is referred to as mud.

Throughout there were considerable and ap-
parently random variations between successive
samples at a given site; these are believed due to the
hand selection of samples from the grab. The upper
part of the sample was generally coarser than the
lower, and an excess of either would explain the
results obtained.

The percentages of each sediment fraction at a
given site were averaged and patterns were sought.
The most meaningful are based on gradings of the
percentages of mud, and of very coarse sand plus
gravel respectively. The mud gradings adopted
were: high >75%, medium 50-75%, and low
<50%, and results are shown in Fig. 3. The coarse
sediment gradings based on breaks in the data
were: high >13%, medium 7-1 1%, and low <7%,
results are shown in Fig. 4.

The mud distribution shows two tendencies:
increase of muddiness with depth, and shorewards

extension of the more muddy areas opposite and
slightly west of the main sources of land drainage.
The latter are apparent off the Brisbane River and
Cabbage Tree Creek. Opposite Serpentine Creek
the effect is more localised and only the medium
mud zone is involved.

The distribution of the coarser sediments pre-
sents an irregular pattern. There tends to be less
coarse sediment in the offshore sites, and there are
patches, generally isolated, with high content of
coarse sediment. The coarse fractions mostly
comprise dead bivalve shells and these areas of
concentration presumably relate more to biotic
distributions in the past than to hydrographic
patterns in the present. It is of interest that
dredging of shell grit for commercial purposes
occurs in the area, but is shoreward of the main
concentrations of coarse sediment (see Raphael
1974, fig. 4).

Biotic Data

Identifications: These were made in part by
comparison with reference collections from pre-
viously published benthic studies in Moreton Bay

Fig. 3: Distribution of mud in sampled area. High proportion (mean % >75), coarse stipple; medium (50-75%),
medium stipple; and low ( < 50%), fine stipple. Scale line = 1 km.

STEPHENSON ET. AL.: MACROBENTHOS OF BRAMBLE BAY

429

Fig. 4: Distribution of coarse sediments (very coarse sand plus gravel = shell grit). High proportion ( 13%) coarse

stipple, medium (7-11%) fine stipple, and low (< 7%) medium stipple. Scale line = 1 km.

(Stephenson, Williams and Lance 1970; Stephen-
son, Williams and Cook 1974). Other reference
collections (like the above housed in the Queens-
land Museum), were made by kind assistance from
the following: Dr P. Hutchings, Australian
Museum; polychaetes (terebellides and amphare-
tids); Mr B. M. Campbell, Queensland Museum;
crabs: Dr C. R. Smalley, Zoology Department,
University of Western Australia; alpheids: Dr W.
F. Ponder, Australian Museum, Sydney; some
gastropods: Dr A. N. Baker, National Museum,
New Zealand; ophiuiroids: Dr P. Mather, Queens-
land Museum; tunicates.

Numerous species have incomplete identi-
fications and at least one taxon is known to be
polyspecific; ‘tunicate 1’ comprises both Molgula
mollis Herdman and Cnemidocarpa floccosa Slui-
ter.

Non-recordings: Grabs are inefficient col-
lectors of penaeid prawns and benthic fish, and the
few specimens obtained were not recorded.

Dead material was excluded as were empty tubes
of polychaetes. Chaetopterus variopedatus was an

exception because intact specimens were not ob-
tained; when the tubes appeared to have been
inhabited recently, two tube-ends were recorded as
one individual.

Species Obtained

Raphael (1974) listed 182 species in her shorter
survey; 4 additions were recorded in the three
seasons of extension. In the comparable Peel Island
survey of Stephenson, Williams and Cook (1974)
roughly double this number was obtained (420
species) and a current study by Stephenson, Cook
and Newlands (MS) records about 450 species
from Middle Banks in Moreton Bay. Using a small
grab in the Serpentine Creek area over five seasons
of sampling, Stephenson and Campbell (in press)
obtained ca 90 species, roughly half the present
number.

Of the 182 species listed by Raphael (1974),
polychaetes contained the largest number of spec-
ies (39%) followed in decreasing order by bivalves
(27%), arthropods (17%), echinoderms (6%), gast-

430

MEMOIRS OF THE QUEENSLAND MUSEUM

ropods (4%) and chordates (4%). These per-
centages are tolerably close to those made at Peel
Island by Stephenson, Williams and Cook (1974).

Comparisons of individual species with those
obtained in other local surveys are difficult due to
incomplete identifications. The closest available
comparison is to the Peel Island study, with 70
species known to be common.

Methods of Analyses

The account below excludes discussion of choice
between most of the alternative methods which are
available (see Clifford and Stephenson 1975). We
have not followed the pioneer 3D study by
Williams and Stephenson (1973) for reasons partly
given in Stephenson, Williams and Cook (1974)
and elaborated in the Discussion.

The data form two different three-dimensional
matrices with dimensions s (species) x q (quadrats
or sites) x t (times). The most convenient method
of handling 3D data is to summate over one of the
dimensions to produce three 2D matrices of q x t, s
x q , and s x t respectively.

The q x t matrix as derived directly contains the
summated numbers ( N ) of all species in each
sample. Various other forms of q x t data are
readily available. By reducing the recordings of
species-in-samples to binary form we obtain the
number of species (S') in each sample; this gives a
simple measure of diversity per sample. More
sophisticated measures are available and we have
also used the Shannon diversity, ( D ) per sample
(N log N - £ n log n), and also per individual
(H l ) expressing these to log base 10. For these.four
q x t matrices, recordings of all species are
incorporated.

For the remaining analyses, there are advantages
in reducing the number of species to consider.
Raphael (1974) used different numbers of species
for series I data (43) and series II data (51) basing
the reductions upon ubiquity considerations. I n the
present case, in theory we used the same species for
both analyses, employing the 81 species with
recordings of 10 or more individuals in the total
data (231 samples). In fact the species used in the
two analyses differed slightly because a few species
present in one series of data were absent in the
other. The species considered are given in the
Appendix. Where only a single species of a genus is
there listed, it is referred to in the text by generic
name only.

From the 5 x q matrix by classification we can
obtain site-groups and species-groups on data
summated (or averaged) over the times of sampl-
ing. From the s x t matrix by classification, we can

obtain time-groups and species-groups on data
summated over all sites. These groupings, based on
overall tendencies are of particular value in the
context of the present study.

The techniques used in the present case following
data reduction were: (a) prior to classification of
entities (sites or times), transform recordings by
using log io ( n + 1); (b) prior to classification of
species, standardise by totals the transformed
values; and (c) classify entities and species in both
cases using Bray-Curtis dissimilarity measure and
group-average sorting.

For easier interpretation of two-way tables,
entities (i.e. sites or times) within entity-groups and
species within species-groups were arranged by
their sequential numbers.

Perusal of the two-way tables indicates that
species-groups and their constituent species char-
acterize certain entity groups by occurring there in
greater numbers than elsewhere. In a few cases
there is ‘negative’ characterizing by the occurrence
of smaller numbers. As stated elsewhere (Stephen-
son and Dredge 1976) if these characterizations are
effected purely by visual inspection of the data
there are risks of subjectivity, while if we use
statistical tests of the significance of differences
these are open to criticism. Because data are near-
optimally grouped we are not comparing random
samples. As in the previous paper (Stephenson and
Dredge 1976) we use the mechanics of certain
statistical tests because they appear to follow
closely the conclusions reached by visual inspection
of the data. We avoid throughout use of the word
‘significantly’ and use instead ‘noticeably’ or ‘out-
standingly’.

Tests were at two levels, the first were for
‘scanning’ purposes and to determine whether or
not entity-group means appeared different. (It
should be noted that in the case of time-groups,
each entity was taken as forming a group.) In
general % 2 tests were employed using ‘raw’ (un-
transformed) summated values. The test was
extended below its legitimate limits of c. 5 per
group (Sokal and Rohlf 1969, p. 565) because it is
being used purely for indicative purposes.

When entity-groups appear different by X 2
testing, there may still be such inter-group vari-
ation that the differences are not ‘real’, hence at a
second level the more stringent F test was em-
ployed, with prior transformation of data using
logio (ff + 1).

The results are expressed in terms of noticeability
of difference, with HN the equivalent of <0 01
probability and N of <0-05 probability as these
levels would be applied in usual significance testing.
It should be stressed that the true noticeability will

STEPHENSON ET. AL .: MACROBENTHOS OF BRAMBLE BAY

431

be less than that ascribed because the data have
been summated along one of the axes of the 3D
matrix, and variation in that axis has been
suppressed.

RESULTS

The series I and series II data are separately
treated.

Series 1

(27 sites, 7 times)

Quadrats x Times Data

The four matrices involving values of N, S, D
and H 1 are conveniently approached by sum-
mations to give quadrat means and times means —
these are given in Table 1 . Heterogeneity in the data
is of interest and variances of the tabulated means
are also given in Table 1, together with the within
quadrat and within times variances of N .

Populations ( N values) and their variances are
highest in quadrats 1 , 4, 5, 6 and 14 and in times 5, 6
and 7. The high values and high variances are due
to isolated extremely high values as follows:
quadrat 1 times 6 and 7, q4 t6, q5 t5 and tl, q6 t6
and tl, and #14 1 1 and t6. While these population
values are ‘patchy’ there are clear hints of patterns
in this patchiness: the high values are in the inshore
sites and mostly in the later sampling periods.

Variances in N values in quadrats and in times
are heavily biased by the above outstandingly large
sample populations. Thus by excluding the above
nine results (ranging from 991 to 4581) the overall
variance is reduced by 98%. These high values also
contribute largely to the variance interaction of
quadrats and times which is 74% of the total
variance in the N matrix.

Spearman rank correlation coefficients were
derived for various pairs of the columns in Table 1
with results in Table 2.

The high positive correlation for quadrat values
between N and D implies that diversities per
individual are more meaningful than diversities per
site. H 1 values are positively correlated with S , the
number of species per site, and negatively cor-
related with the populations per site.

The correlations for times are not significant in
any of the cases. It should be noted that the times
interrelationships and quadrat interrelationships
did not follow similar patterns. Thus in the times
data S and D showed a high positive correlation
and also N was positively and not negatively
correlated with //'.

Considering the H ] values, variances between
sites (0-045) are much greater than the variances
between times (0-003). All inshore sites give low H l

TABLE 1: Analysis of q x t Data
M eans and variances of A; means of S , of D and of H l
in quadrats and in times. Variances of means are also
given. AH values rounded.

Quadrats

Quadrat

No.

Mean

N

Variance

N

Mean

5

Mean

D

Mean

W

1

829

2842922

7

62

0-45

2

110

5044

9

52

0-51

3

145

5452

10

63

0-55

4

215

150166

10

61

0-46

5

1002

2660803

11

109

0-28

6

452

448781

12

92

0-44

7

28

135

15

31

1 -07

8

61

3219

12

33

0-66

9

24

643

8

14

0-75

10

25

175

11

24

0*91

11

45

348

15

38

0-88

12

33

457

13

31

0-93

13

31

256

13

30

0-95

14

506

407509

10

36

0-48

15

33

333

12

25

0-84

16

43

525

15

42

I 02

17

47

907

10

26

0-62

18

53

536

13

40

0-78

19

66

4828

18

55

0-95

20

64

5166

12

38

0-73

21

82

5451

12

44

0-64

22

46

261

16

48

1-01

23

103

7356

16

81

0-83

24

87

1372

15

61

0-85

25

91

2412

14

61

0-69

26

38

1451

15

33

0-98

27

19

141

9

15

0-80

Overall

mean

158-4

12-2

46- 1

0-740

Variance
of means

62169

7-67

511

0-045

Times

Time

Mean

Variance

Mean

Mean

Mean

No.

N

N

5

D

W

1

130

60565

14

53

0-69

2

102

16168

14

51

0-77

3

107

12646

16

66

0-80

4

53

922

9

30

0-63

5

199

767666

10

31

0-80

6

367

908480

13

57

0-76

7

150

115420

9

37

0-73

Variance

of means

10495

7-81

193

0-003

432

MEMOIRS OF THE QUEENSLAND MUSEUM

TABLE 2: Spearman Rank Correlation Coefficients for Data in
Columns of Table 1. (H.S. <0 01, N.S. l 0-05)

N cf. D

S cf, D

S cf. W

N cf. H l

Quadrat means
Times means

+ 0-86 (HS)
+ 6-25 (NS)

+ 0-L1 (NS)
+ 0-79 (NS)

+ 0-66 (HS)
+ 038 (NS)

0-77 (HS)
+ 0-23 (NS)

values (sites 1-6 and 14) and there is also a low W
value at time 4 (March 1973). This coincides with
the lowest populations.

Further discussions of the q x / data are deferred
pending analysis of the s x q and s x t matrices.

Species x Quadrats Data

Site-groups: The dendrogram of site-

groupings (Fig. 5) shows a near trichotomy into
three unequal groups, the largest group then giving
two small and two larger groups. By accepting the
above we derive a 7-site inshore group (/), a 7-site
middle group (A/) and an 8-site offshore group (O),
plus five sites in three groups. These sites were
allocated to one or another of the three main
groups in the order of sites 1 6, 1 8, 27, 8 and 9 on the
basis of Bray-Curtis comparisons of log (n + 1)
data with group means. The three revised groups
are:

Inshore (I): sites 1, 2, 3, 4, 5, 6, 8, 9, 14.

Middle (M): sites 7, 10, 11, 13, 15, 16, 18, 19,
26, 27.

Offshore (O): sites 12, 17,20,21,22,23,24,25.

I 0 M

(1) (U (6) (1) (2) (8) (7)(1)

Fig. 5: Truncated dendrogram of classification of 27 sites
by species; sites in each group at dendrogram base. /
sites are inshore, M at middle depths and O offshore.
Sites not in marked groups were re-allocated.

These groups make a coherent topographical
picture (Fig. 6). Thus they conform to the extrinsic
attribute of proximity; they are also the groups
giving the maximum number of outstanding spec-
ies (i.e. maximum conformity to intrinsic at-
tributes).

Species-groups: The eleven groups originally
accepted are shown by dendrogram in Fig. 7. They
involved accepting lower dissimilarity levels within
ihe larger groups.

F tests were employed (on transformed data)
species by species to determine which had notice-
ably different values in the three site-groups. In
several cases where using three groups failed to
show differences one group was tested against the
two others and differences were then noticeable.
Species which conformed to these tests and those
which do not are considered separately. The former
comprise 53 of the 79 species in the analysis.

Species-groups with a majority of species confor-
ming to the 3 site-groups: Five groups are involved
and these with their constituent species are listed
below; non-conforming species are in parenthesis.
Site-group characterizations by these species-
groups are also given.

Fig. 6: Site-groups; 27 sites classified by species, after re-
allocation of sites. Scale line = I km.

STEPHENSON ET. AL.\ MACROBENTHOS OF BRAMBLE BAY

433

100 _

80 _

I I I — II — II — II — t I — II I I — I I I I I I I

1 2 3 4 5 6 7 8 9 10 11

( 2 ) ( 1 ) ( 28 ) ( 6 ) ( 1 ) ( 7 )( 12 ) ( 1 ) ( 4 + 3 ) ( 1 + 2 + 1 + 2 + 5 ) ( 1 + 1 )

Fig. 7: Series I data; truncated dendrogram of classification of species by sites (species 35 and 75 with nil records,

eliminated). At base the species-group numbers used in

Species-group 3: spp. 4, 6, 7, 8, 10, 12, 13, 14, 17,
19, 20, 21, 23, 25, 29, 30, 33, 37, 42, 45, 48, 53, 57,
58, 62, 66 (32, 47). Highest recordings (with minor
exceptions) in offshore site-group, lowest in inshore
group; designated ‘offshore’ species.

Species-group 5: sp. 26. Restricted to inshore
site-group (and only in certain of these — see later).

Species-group 6: spp. 5, 18, 28, 50, 54, (41, 55).
‘Middle’ species, highest recordings middle site-
group.

Species-group 7: spp. 1, 3, 9, 11, 15, 16, 24, 40,
(31, 51, 65, 76). ‘Inshore’ species.

Species-group 8; spp. 22, 43, 46, 49, 59, 60, 61,
63, 68, 70, 73, 77 (79). ‘Middle’ species, differences
from species-group 6 detailed below.

Summating gives 26 offshore, 17 middle, and 8
inshore species.

While the major site pattern is clearly that of the
three main groups, perusal of the data revealed that
there are smaller site-groups within them. Thus
species 26 (species-group 5) occurs in high numbers
in sites 1-4 with low numbers in sites 5-7 and is
absent from the remainder. It is a ‘sites 1-4’ species.
Similarly species 45, 47, 58 and 62 (of species-group
1) are a ‘sites 21-25’ species. These two groups of
sites are on the western side of the samples area; 1-4
being western inshore and 21-25 western offshore.

The differences between species-groups 6 and 10
are that in the latter there are proportionally fewer
specimens in the inshore site-group, and that the
species are concentrated in sites 13, 16, 18, 19, 26,
27, i.e. in the eastern portion of the middle site-
group.

Species-groups not conforming to the 3 site-
groups: These comprise: group 1 , spp. 2, 70; group

text, and in parenthesis the numbers of species in each.

2, sp. 67; group 4, spp. 27, 34, 39, 52, 56, and 69;
group 8, sp. 80; group 9, spp. 36, 38, 44, 64, 71, 74,
78; and group 1 1 , spp. 72 and 8 1 .

Only one of these species-groups characterizes
an extensive and topographically coherent area.
This is species-group 4 and all its contained species
except one have HN conformity to sites 7 12, 14,
26 and 27. (Apart from one outstandingly large
value, the remaining species, 27, also conforms at
the HN level.) The area involved approximates to
the middle site-group less the sites closest to the
Brisbane River, and this general area appears as a
discrete site-group in the series II data.

The remaining species-groups have little internal
coherence and species are separately considered.
Eight species are significantly concentrated at a
single site and are designated ‘patchy’ species. The
species with their sites of concentration in paren-
thesis are 2(14), 27(16), 36(16), 38(27), 44(16),
46(16), 64(16), and 78(16). Six of the eight species
are concentrated at site 16.

For the remaining species, it is possible to divide
the sites into two groups to obtain noticeably
different results, but to do this involves site-
groupings which approximate to random scatter
throughout the sampled area. The species involved
are: 31, 32, 51, 55, 67, 71, 72, 74, 76, 80, 81. These
probably include some having pseudo-uniform
distribution over the whole area (eg. 31, 32) and
others present in too low numbers for patterns to
show (eg. 71, 72, 74, 76, 80, 81).

Features of the Site-groups: If the main site-
groups are regarded as communities then there are
three sets of data on these groups: species com-
position, sample characteristics from the q x t

434

MEMOIRS OF THE QUEENSLAND MUSEUM

TABLE 3: Series I Data (27 Sites); Species x Sites
A. Mean Number of Specimens/hi 2 * of More
Abundant Species (in Species-groups) in Site-groups

Species Species,

Group and code No.

Site-groups
(sites in parenthesis)

1 M O

(1-6, 8, (7, 10, 11, (12, 17,
9, 14) 13, 15, 16, 20-25)
18, 19, 26,

27)

3

Paratapes , 4

1

16

158

Amphioplus sp., 6

<1

3

20

A. lobatus , 7

<1

5

29

Theora lata , 8

<1

1

10

Amphitrite , 10

3

6

11

6

Anadara , 5

3

29

6

tunicate 1, 18

2

5

1

7

Spisula , 1

1347

1

0

Xenophthalmus, 3

140

38

8

Terebellides, 9

20

3

2

Pupa , 11

15

0

0

Parcanassa, 15

12

2

1

10

oyster 1, 22

0

3

1

Area sp. 1, 43

<1

2

<1

* Mean number of specimens per sample, converted to
numbers/m 2 and rounded to nearest unit.

B. Mean Populations and Mean Individual Shannon
diversities* in Site-groups

Site-groups

I

M

O

Mean N/m 2

1858

190

346

Mean H 1

0-51

0-92

0-78

*Meaned from values per sample of two 0-1 m 2 grab
catches.

C. Mean Number of Specimens/hi 2 * of More
Abundant Species in Alternative Specified Site-
groups

Sites

specified

Species,
and code No.

Mean numbers/m 2 in:
Sites Remainder
specified

1-4

Ophelina, 26
Amphioplus

16

<1

21-25

depressus, 45

4

1

Leptomya, 47

2

<1

22-25

tunicate 3, 19

7

1

Ophiactis, 59

2

< 1

7-12,
14, 26, 27

Petaloproctus, 34

2

<1

Edwardsia, 39

2

<1

*Mean number of specimens per sample, converted to
numbers/m 2 and rounded to nearest unit.

data, and abiotic features. The first is indicated in
the 2-way coincidence table (Table 3A) in which
only the most abundant species in each species-
group are listed. More briefly the inshore group
would be described as Spisula-Xenophthalmus-
Terebellides community, the middle group as a
Xenophthalmus-Anadara-Paratapes community
and the outer group as a Paratapes-Amphioplus
lobatus community.

Comparable data on the species which charac-
terise alternative site-groupings are given in Table
3C. Data from the q x t matrices are given in Table
3B. They comprise mean values of N converted to
individuals/m 2 and mean values of individual
diversities ( W ). This table shows that the inshore
community has the highest mean population but
lowest individual diversity; the converse is true of
the middle community.

Perusal of Figs. 3 and 6 shows that there is a
partial relationship between site-groups and distri-
bution of mud in the sediments. Thus there is near
coincidence between the offshore site-group and
the area of most muddy sediments, and broadly
similar concentric patterns radiating from this
area. There are also comparable shorewards pro-
trusions of the medium mud sediments and the
middle site-group.

The correspondence between site-groups and
mud distribution breaks down in the eastern part of
the sampled area. Moreover the eastern site-group
does not coincide with the area of high con-
centration of coarse sediments (see Fig. 4).

There is a similar partial relationship between
site-groups and depths. While in the western part of
the area, the site-groups occupy different depths,
this fails to apply to the eastern part. The simplest
overall explanation is that a depth-sediment re-
lationship holds in the western part of the area, but
that this is obliterated in the eastern part by some
effect of the Brisbane River.

Species x Times Data

Time-groups: The dendrogram of time-

groupings (Fig. 8) shows two biotically isolated
times — l4 (March 1973) and tl (Dec. 1973) — and at
a lower level two more coherent groups of times 1 ,
2, 3, (June-Sept. 1972) and times 5, 6 (June-Sept.
1973) respectively. The acceptance of these groups
is at a dissimilarity level of c. 0-25 while acceptance
of the three major site-groups in Fig. 5 was at the
level of ca 0-45. Clearly while the inter-times group
heterogeneity is appreciable it is much less than the
inter-site group heterogeneity.

STEPHENSON ET. AL.: MACROBENTHOS OF BRAMBLE BAY

435

4 2 1 3 5 6 7

Fig. 8: Series I data; truncated dendrogram of
classification of times by species.

If there had been a marked seasonality of the
data, time-groupings such as 1, 5; 2, 4 and 3, 7
would have been expected. These did not occur.

Species-groups: The dendrogram gave very
unequal groupings; it is not here given because
none of the major species-groups were considered
as satisfactory from a conceptual viewpoint. ‘Satis-
factory’ species-groupings, as were obtained in the
s x q analyses, are those in which entities (sites
or times) are similarly characterized by the species
in the group.

The computer-based analyses were replaced by
visual and hand calculator analyses. They began by
visual scanning the transformed recordings of a
given species to select outstanding values. These are
usually outstandingly high, in which case times of
abundance are selected and times are positively
characterized. Occasionally the outstanding values
are low, times of scarcity are selected, and the
characterization is negative.

The recordings of each species were then divided
into two groups — the outstanding values and the
remainder — and F tests were then performed on
these two groups. Following this, species were
grouped by their characterizations of times. Re-
sults are given in Table 4.

Table 4 shows that many species (35) positively
characterized sequential time patterns, particularly

times 1-3 (14 spp.). Only six species showed
seasonal re-occurrences of high recordings. The
remaining species have been allocated to four
groups as follows:

(a) Distorted seasonal (positive) — 2 spp. Here
there is re-occurrence of high recordings to an
approximate seasonal pattern. Because there
were only four samplings per year and because
advance or delay in seasonal peaks may well
occur, these may be truly seasonal species.

(b) Single time — 9 positive spp., 6 negative spp.
These are referred to later as ‘time-patchy’
species.

(c) Nonsensical — 9 spp. These had markedly
different recordings throughout three sampl-
ing periods, and typically high, low and high,
giving a ‘saw-tooth’ graph.

(d) No times characterized — 10 spp. In these
there were no outstanding recordings and

TABLE 4: Time-groupings Characterised by Species;
Series I Data

Times

Species characterization
Positive Negative

Sequential

1-3

4, 5, 9, 10, 18, 20, 24,

32, 37, 41, 42, 56, 57, 74

2-3

28,61,74 60

5, 6

1,7, 22, 26, 38, 69, 76,81

1-6

25, 50, 52

l^t

4,9

2-4

48

3,4

29, 45, 64

2-5

15

6, 7

34, 43, 47

Seasonal

2,6

12, 31,49

3, 7

27, 36, 54

Distorted

Seasonal

1, 2, 6

2

2, 5

16

Single time

1

67, 72

2

19

3

21, 59

4

14, 33

6

71

7

8,39,44 11,23,50,52

Nonsensical

3, 16, 17, 30, 46, 51,62,

68, 73

No times

characterized

53, 55, 63, 65, 66, 70,

(Random)

77, 78, 79, 81

436

MEMOIRS OF THE QUEENSLAND MUSEUM

testing raw values with X 2 showed no notice-
able difference from randomness. They are
referred to below as random species.

The time-patchy and nonsensical species (24 in
all) may either have rapid changes in populations
or, more likely, result from a patchy microtopo-
graphical distribution. The random species are all
low in the abundance hierarchy, and randomness is
likely to be due to small recordings rather than
truly stable ones.

Features of the Time-groups: As with site-
groups there are three sets of relevant data: species
composition, sample characterisation from the q x
t data, and possible abiotic ‘explanations’.

The reduced 2-way coincidence (Table 5 A)
indicates the most important positively charac-
terizing species of these time-groups, while Table
5B shows the population and individual diversity
data.

In brief, times 1-3 are a period of Paratapes ,
Anadara, Terebellides and Amphitrite ; there are no
species positively characterising time 4 only; times
5 and 6 are a period of Spisula and Amphioplus
lobatus ; while time 7 is a period of Theora lata.

TABLE 5: Series I Data (27 Sites); Species x Times

A. Mean Number of Specimens/hi 2 * of More
Abundant Species in Groups (from Table 4) in Time-
groups

Species,
and code No.

1-3

Times
4 5, 6

7

Paratapes, 4

101

57

4

5

Anadara , 5

23

10

6

5

Terebellides, 9

17

7

1

1

Amphitrite, 10

11

5

3

2

tunicate 1, 18

5

1

1

1

Spisula, 1

127

28

1082

572

Amphioplus lobatus, 1

9

7

16

8

Theora lata, 8

1

0

0

20

* Mean number of specimens per sample, converted to
numbers/m 2 and rounded to nearest unit.

B. Mean Populations and Mean Individual Shannon
Diversities* in Time-groups

1-3

Times

4 5, 6

7

Mean Njm 2

563

265

1415

750

Mean W

0-75

0-63

0-78

0-73

* Meaned from values per sample of two 0.1m 2 grab
catches.

In terms of population density times 1-3 and 7
are about average, time 4 is outstandingly low and
times 5 and 6 outstandingly high. Individual
diversities are quasi-constant throughout except
for a marked low in time 4.

Overall the most dissimilar time is time 4 (March
1973). Comparable studies near Peel Island in
Moreton Bay by Stephenson, Williams and Cook
(1974), revealed outstandingly low populations in
their two March samplings. Moreover the lowest
values were in March 1970 which followed a period
of normal climate, instead of in March 1971
following the wettest summer (Dec. -Feb.) for 24
years. In the present case March 1972 follows a dry
period.

The effect of rainfall would be primarily by run-
off from the Brisbane River, and data in Raphael
(1974) show less than normal discharge in the
Brisbane River for eight of the nine preceding
months. Assuming that river run-off is the control-
ling factor then times 1-3 fall in a period of reduced
run-off but one which follows a wetter period
(Feb. -May 1972). Conversely times 5 and 6 fall in a
period of approximately normal run-off but one
following a drier period (June 1972-May 1973). On
this basis the difference between times 3 and 7 (Dec.
1972 and Dec. 1973) reflects the low run-off
preceding the former, and the approximately
average run-off preceding the latter.

The only other continuous abiotic data of
seeming relevance to the present situation are of air
temperatures. There are no obvious relationships
between these data and the time-groups.

Further Analyses

Following the above analyses, two others sug-
gested themselves. Both are concerned with short-
comings of the technique of converting a three-
dimensional matrix into three two-dimensional
ones.

Extreme Space-time Patchiness: During sum-
mation across one of the axes of the matrix,
variation along that axis is suppressed. Thus a high
recording of a species in a site may be based upon
uniformly high recordings in all times or may be
based upon an extraordinarily high recording at a
single time. Such a value could bias both the q x t
and s x q analyses. A simple method was used to
recognise species showing extreme ‘space-time'
patchiness by way of a single outstanding record-
ing. The transformed data on each species (79 in
all) was considered within its own q x t (27 x 7)
matrix. The Ftest was then applied comparing the
largest recording with the remainder.

STEPHENSON ET. AL:. MACROBENTHOS OF BRAMBLE BAY

437

1 3 4 5 2 6 7

1 2 3 4 5 6 7

4 1 2 3 5 6 7 1 2 3 4 5 6 7

Figs. 9-12: Series 1 data; dendrogram of classification of times by species.

Fig. 9: Inshore sites. Fig. 10: Mid-eastern sites, nearest Brisbane River.

Fig. 1 1 : Mid-western sites.

Fig. 12: Offshore sites.

438

MEMOIRS OF THE QUEENSLAND MUSEUM

To determine which results show extreme
patchiness we must decide on a probability in
relation to the F level. The lowest listed in tables is
0-001, and to accept this would give 73/79 ex-
tremely patchy species. Taking five times the F level
at the 0-001 value gives the following 25 space-time
patchy species: 2, 19, 22, 27, 34, 36, 38, 43, 44, 50,
59, 60, 61, 63, 64, 65, 68, 69, 70, 71, 72, 73, 74, 77,
78, 81.

The species previously judged site-patchy are: 2,
27, 36, 38, 44, 46, 64, 78, and 7/8 of these are in the
above list. Of the 15 species previously judged time-
patchy only six appear on the above list: 19, 44, 50,
59, 7 1 , 72. It is evident that the present method does
have cautionary value, particularly regarding
space-patchiness.

The places and times of the space-time patchi-
ness give concentrations as follows: quadrats 11,13,
16, 18, and 1 9 which are all in the middle site-group
and mostly on its eastern side, and times 3, 5, 6, 7;
these are mostly towards the end of the sampling
period.

Interactions between Space and Time: The
usual calculations of variance interaction essayed
on a species by species basis, showed that most of
the interaction was due to the few, usually single,
outstandingly large values as considered above.

Further and more directly meaningful analyses
involving interactions were then undertaken and
for two reasons: (i) the main heterogeneity is
between sites; by operating within site-groups it
seems possible that times-groups would be more
coherent and (ii) within different site-groups one
might expect differences in time-groupings. If the
previous argument that variable run-off from the
Brisbane River explains some of the results is
correct, then the greatest time-dissimilarities
should occur in the sites adjacent to it.

The three original site-groups were converted to
four by arbitrary division of the middle one and by
transfer of one site (12) to the group suggested by
its topography. This gave:

Inshore: sites 1-6, 8, 9, 14.

Middle eastern (nearest river): sites 15, 16, 18,
26, 27.

Middle western: sites 7, 10-13, 19.

Offshore: sites 12, 16-18, 20 25.

The four dendrograms of time-groupings within
these site-groups are given in Figs. 9 to 12.

The mean percentages dissimilarity at which
times separate as individual entities are as follows:
inshore 32-6, mid-eastern 44-9, mid-western 39-9,
outer 30-5. As expected the site-group nearest the
river shows the greatest time heterogeneity. Pos-
sibly the river influence extends to the mid-western

region; this shows greater time-heterogeneity than
the inshore site-group.

The inshore group shows division by years into
time 1-3 and 4-7, and in the midwest there is a
somewhat similar picture (times 1-3 and 5-7). The
mid-east has a broadly annual picture, upset by a
seasonal linkage of times 2 and 6. The offshore area
differs from the remainder in having three group-
ings of successive times, viz: times 1, 2 and 3; 4 and
5; 6.

Species-groupings were examined in each of the
above cases. Those produced by the classificatory
programme failed to give conceptual sense, as
expected and discussed earlier. In the present cases,
grouping species by conformity testing likewise
failed to give meaningful conclusions. This was
because dividing the sites into groups reduced the
populations of most species to levels below those at
which differences were outstanding. The time-
groups were distinguished by summations of
tendencies which when considered individually
revealed little.

SERIES II
(48 sites, 2 times)

Quadrats x Times Data

These are only considered in relation to site-
groups and time-groups and are detailed later.

Species x Quadrats Data

Site-groups: The dendrograms of site-

groupings (Fig. 13) gives an initial dichotomy; the
site-groups so obtained lie offshore and onshore
respectively. To obtain the maximum number of
conforming species required more homogeneous
site-groups; eventually seven major groups were
adopted as shown on the dendrogram base. This
left two isolated sites, 27 and 14; the former was
allocated to the major site-group with greatest
biotic affinities, site-group E. Site 14 is very
dissimilar from the remainder, due to large num-
bers of species 2 ( Mesochaetopterus minutus ) and
was retained as an isolate. The site-groups thus
adopted are map-plotted on Fig. 14,

Site-groups A, B and C, which are tolerably
closely linked in the dendrogram, all comprise
inshore sites. Group A is the most inshore, B is
intermediate and western, C is the deepest of the
three and eastern. Site-group G, by dendrogram
more distantly related to the above three groups lies
nearest to the river mouth and tends to be more
offshore than site-group C.

STEPHENSON ET. AL .: MACROBENTHOS OF BRAMBLE BAY

439

27 D E F 14 C A B G

(1+3) (1+10) (1+7) (2+2) (1+8] (2+1+1) (1+1+1+1+1+1)

Fig. 13: Series II data (48 sites); truncated dendrogram of classification of sites by species; number of sites in each group
in parenthesis at dendrogram base.

Fig. 14: Site-groups; 48 sites, classified by species after re-allocation of site 27. Scale line = 1 km.

440

MEMOIRS OF THE QUEENSLAND MUSEUM

Site-groups D, E and F are closely linked in the
dendrogram, especially D and E. Topographically
the three form a series with D closest inshore, E
intermediate and F offshore. Site-group F extends
laterally to off the mouth of the Brisbane River.

In general there is a clear topographic pattern, in
which the most obvious relationship is to
depth/distance offshore. The separation of site-
group G seems to indicate a second influence, that
of river proximity.

Species-groups: Apart from three isolated
species (2, 45, 46) there is almost a trichotomy into
three groups. By further division of the larger
groups seven major species-groups were obtained
{Fig. 15).

F tests were performed to determine which
species had noticeably different values in the seven
site-groups. Only 30 out of the 69 species in the
analysis conformed, i.e. 43% compared with 67%
in the series I data.

This is not because there are now more ‘patchy’
species concentrated in a single site. There are seven
such species; these, with sites of occurrence in
parenthesis, are: 2(14), 6(48), 20(45), 37(31),
40(48), 45(47), 46(47). The difference is due to the
fact that there are many more ‘random’ species.
This is a reflection of the lower recordings; present
data involve four grab catches, while the series I
data involved fourteen.

Perusal of the two-way table revealed that in
several species-groups there was uniform or nearly
uniform positive characterisation of a site-group
by a species-group. Species-group I showed con-
centration in site-group A, (extreme inshore)
except for one species (41) with a single outstand-
ingly high value outside the area; species-group IV
characterised site-group E; species-group VI
showed major concentration in site-group G (near
river mouth) and minor concentration in site-group
D (adjacent). Species-groups V and VII comprised
species with low recordings mostly approximating
to random scatter amongst the sites, and there
remains species-group III, the largest of all. The
most discrete sub-unit comprises species 4, 5, 7, 12,
27, 31 and 38, and these are concentrated in site-
group E. The remainder do not consistently
characterise any of the site-groups. Perusal of the
data suggests that the remaining eleven species (9,
13, 14, 21, 22, 24, 26, 29, 32, 35, 43) characterise an
alternative grouping of sites, with concentrations in
an area approximately encompassing sites 6,11,12,
13, 17, 19, 20, 21, 22, 23, 24, 25, and 26.

Features of the site-groups: A condensed 2-
way coincidence table (Table 6A) summarises the
site-group/species-group relationships as regards
the commoner species. Perusal of the equivalent
table for the series I data (Table 3 A) shows many
differences; both site-groups and characterising

105
95 „
85 _
75 _

i j i i i i i i i i i i

I

I

in

E

I

H

m

(1) (2) (1+9+1)

(1+4+1)

(1) (2+15)

(4)

(1+1+2) (1)

(6+4+2+1)

(3+5+1)

Fig. 1 5: Series II data; truncated dendrogram of classification of species by sites. At base species-group numbers used in
text, and in parenthesis number of species in each.

STEPHENSON ET. AL.: MACROBENTHOS OF BRAMBLE BAY

441

TABLE 6: Series II Data (48 Sites); Species x Sites

A. Mean Number of Specimens/hi 2 of More Abundant Species in Main Site-groups

Species

group

Species
and code No.

A

d-6,

35-37)

B

(8, 9,
34, 38)

Main site-groups (sites in
C D

(7,10, (11,12,

13, 15) 17, 19)

parenthesis)

E

(20-
27, 33)

F

(28-32
42, 43)

G

(16, 18,
45-48)

1

Spisula, 1

2626

261

1

0

<1

0

0

Xenophthaltnus, 3

228

3

10

3

6

0

42

Amphitrite , 10

27

5

0

0

0

0

0

II

Amphioplus sp., 6

0

2

1

4

5

55

9

Parcanassa , 15

0

0

0

0

2

14

1

111

Paratapes, 4

1

6

10

53

133

58

2

Ana Aar a, 5

0

I

2

6

31

57

1

Amphioplus lobatus , 7

0

0

5

12

31

8

0

Terebellides, 9

8

4

19

22

15

5

0

IV

Pupa , 1 1

1

5

3

0

15

0

6

VI

Theora lata, 8

2

5

6

22

1

0

53

tunicate 1, 18

0

3

2

12

0

0

10

*Mean number of specimens per sample, converted to numbers/m 2 and rounded to nearest unit.

B. Mean Populations and Mean Individual Shannon Diversities* in Time-groups

1

A

B

Main site-groups
C D

E

F

G

Mean A/m 2
Mean H ]

2946

0-42

361

0-66

109 218 332

105 0-92 0-82

242

067

267

0-81

*Meaned from values per sample of two 01 m 2 grab catches.

species-groups have changed and this makes de-
tailed comparisons difficult. If a choice is to be
made between the species/site results of the two
studies, that from the more extensive series I data is
preferred.

Of the abiotic factors which might influence the
series II site-groups, sediment relationships are
obscure (compare Figs. 3 and 4 with Fig. 14); an
effect of the Brisbane River is clear; but the main
factor appears to be related to distance offshore.
Whether this is due to depth, dilution, turbidity, or
some other factor is unknown.

Table 6B shows population densities and mean
individual diversities within the site-groups, and as
before there is an inverse relationship. Again the
most inshore group (site-group A) has the highest
populations. The groups at intermediate depths
differ from each other as regards population
density and diversity: lowest populations and
highest diversities are in the central somewhat
inshore site-group C; site-group G nearest the
Brisbane River has roughly average values for
both; and the most offshore site-group F has
somewhat lower populations and higher diversity
than average.

Species x Times Data

With only two times, the times classification is a
single dichotomy. This is at a dissimilarity level of
21%, considerably less than the dissimilarity .be-
tween the two Septembers based only on 27 sites
(31%). The decrease is due to the inclusion of sites
which are more offshore and/or further from the
Brisbane River.

With only two times, there is no variability of
species-in-times, and the F test cannot be em-
ployed. Use of the X 2 test on raw numbers
(summated over all sites) showed 22 species with
higher populations in 1 972, 1 8 with higher values in
1973, leaving 39 without noticeable differences.

Total populations of all species summated were
lower in time 1 with an average of 408
individuals/m 2 compared with 1346 in time 2.
Mean individual diversities were 0-74 and 0-67
respectively.

DISCUSSION

This involves three main topics, (a) methods,
both sampling and numerical, (b) matters related to
airport construction, and (c) general matters
related to communities, productivity, etc.

442

MEMOIRS OF THE QUEENSLAND MUSEUM

Methods

Sampling Methods: Stephenson, Williams and
Cook (1974), established certain desiderata based
on their work at Peel Island which could not be
followed in the present study. These were that
stations should be c. 0-25 km apart and that there
should be quintuplicate grab catches on each
occasion. In the present work samples were c. 1 km
apart and catches were in duplicate.

The results obtained have in general shown that
the wider spacing of sites was acceptable in that the
site-patterns obtained have been topographically
coherent and meaningful. Only one site (14) has not
been closely linked to its neighbours and this was
due to patchy distribution of species 2 ( Meso -
chaetopterus minutus).

Nevertheless problems remain over the spacing
of sites. On each sampling two catches were made
in close proximity but on a subsequent occasion the
pair were likely to be up to 25 m from the originals.
If a species is patchily distributed then on one
occasion it might be collected and on another
missed. In brief, microtopographic patterning
could give the appearance of a marked seasonal
change in numbers. A sufficient number of species
showed ‘saw-tooth’ types of seasonal change to
suggest that this was occurring. This casts doubts
on the reality of the supposed time changes of the
remaining species. Nevertheless a sufficient num-
ber of species give ‘sensible’ results for the time
changes described to be regarded as real.

Further evidence for this was obtained by
analysis of time-changes within sub-units of the
sampling area. Here the data are weaker because
summations involve fewer sites, but the time
analyses result in somewhat greater conceptual
sense.

Numerical Methods: Apart from a paper
written after but published before the present
account (Stephenson and Dredge 1975) the last
published account of classifying multidimensional
data which we are aware of was by Stephenson,
Williams and Cook (1974). This involved variance
measures of dissimilarity, whose magnitudes are
very sensitive to the type of data transformation
which is used; also it failed to give optimal species-
groupings.

Since this, work experience has been gained in a
variety of analyses of two-dimensional data by the
senior author in concert with other workers. These
include unpublished work by Godfriaux and
Stephenson, preliminary reports upon the present
work by Raphael and Stephenson (1972), and
manuscript work by Stephenson et al. on waste-
water outfalls at Los Angeles. Throughout, this

work has shown that for benthic analyses the Bray-
Curtis measure of dissimilarity has advantages.
Used on raw data dissimilarities are possibly too
strongly biased towards the numerically abundant
species and it is generally desirable to use transfor-
med data. In the earlier work quoted above
(Godfriaux and Stephenson, Raphael and
Stephenson) the \J7T transformed was used, and
this was also employed by Raphael (1974) in her
thesis. Stephenson et al. (in MS) used the cube root
transformation with the Bray-Curtis measure, but
in the present work the more stringent log (n + 1)
transformation has been employed. It follows
precedents created by Field and Robb (1970), Day,
Field and Montgomery (1971), Field and Macfar-
lane (1968), Field (1971) and Christie (1974).
However the choice was mostly influenced by
results not yet published concerning the effects of a
major flood on the present biota.

Following Boesch (personal communication),
the preliminaries to the present work, and the
manuscript work by Stephenson et al. for species-
groupings we have used the proportionality of a
species-in-sites instead of the absolute recordings.
(Actually the proportions of transformed values
were used.) This has resulted in better groupings of
species with similar site recordings and has been an
improvement compared with the variance tech-
nique of Williams and Stephenson (1973), and of
Stephenson, Williams and Cook (1974). Mean-
while from a more theoretical aspect Dale and
Anderson (1973) have already shown that optimal
groupings of sites and of species do require
different techniques.

In the present method as in the original 3D
treatment of Williams and Stephenson (1973) we
summated along a specified axis of the matrix to
produce three two-dimensional matrices of q x t,s
x q, and s x t respectively. As a consequence in
each matrix variability in the other axis is lost and a
single large recording of a species in a sample can
influence all three of the two-dimensional matrices.
A technique has been developed to locate and ‘give
warning’ of such single outstanding values. It
involves an F test (on transformed data) in which
the largest value is compared with the remainder.
In the present work ‘patchy’ species were also
sought by heuristic examination of the results of
site analyses and of time analyses. There was
general agreement between the test and the con-
clusions drawn from site analyses.

In one respect the present methods revert to
those of Stephenson, Williams and Cook (1974).
This is in the use of the F test for investigating the
conformity of species to entity-groups (i.e. site-
groups or time-groups). We appreciate that the

STEPHENSON ET. AL.. MACROBENTHOS OF BRAMBLE BAY

443

data to be tested have been grouped by near-
optimal techniques and that the basis for testing of
significance is hence destroyed. However we
suggest that this test still has conceptual value and
has a close relationship to the intuitive bases of
data scanning. Because it takes more strict account
of within-group variation, it seems preferable over
other tests. Throughout it was used at different
levels in the classificatory hierarchies, and the levels
finally selected gave the maximum number of
conforming species.

Although the present methods have proved
generally satisfactory, problems remain over
species-groups; species within a group sometimes
fail to conform in a uniform way to the entity
groups. In some cases this has proved conceptually
helpful as indicating alternative site-groupings.
Such groupings have been sought unsuccessfully in
two previous studies (Stephenson, Williams and
Cook 1974, Stephenson et al. in MS).

In general species-grouping using sites data has
proved satisfactory, but species-grouping with
times data has not. The problems have been
discussed earlier and have been partially resolved in
the present paper by intuitively based analyses. As
yet these defy formalization to the level of com-
puter programming.

Relation to Airport Construction

To predict changes in the Bramble Bay benthos
due to airport construction requires (1) that there
should be recognisable and quantifiable patterns in
the biota before construction, (2) an estimation of
which patterns are likely to change due to overall
human activities, and (3) estimation of the parti-
cular effect of airport construction.

Site-patterns have been obtained with both the
series I data (27 sites, 7 times) and the series II data
(48 sites, 2 times). As might be expected these are
not identical, but they do show broadly similar
tendencies. It is of especial interest that the
boundaries of the middle-depth site-groupings
bulge shorewards opposite Serpentine Creek. For
this there are two possible explanations, either a
specific effect of Serpentine Creek, or the fact that it
is an area roughly midway between two larger
systems of freshwater drainage. Whichever is
involved, the reduction in run-off from Serpentine
Creek following airport construction is likely to
cause changes. Present data show the area has
relatively high diversity but relatively low pop-
ulation density.

Only the series I data produced time patterns
worthy of further consideration. These show that
there are noteworthy time changes and that these

involve very little repetition from the seasons of one
year to those of the next. Instead certain species
characterise the total area for a certain period (e.g.
9 months) and are superseded by other species.
Time changes are most marked in the area nearest
the Brisbane River. They indicate an annual
change-over in biota in the inshore sites and an
approximately six-months change-over in the
offshore sites. It was suggested that variable run
off of freshwater was the major cause of these
temporal changes and if this is so, occlusion of
Serpentine Creek might lead to increased temporal
stability. The general literature on biotic diversity
suggests that this would result in an increase in
diversity. However this is far from certain and the
reverse might well be true (see Stephenson, Wil-
liams and Cook 1974; Clifford and Stephenson
1975).

It is clear that investigations of considerable
duration, much in excess of the proposed two years,
would be required before adequate predictions
could be made concerning temporal changes dur-
ing 'normal’ conditions. The flood of January 1974
terminated such 'normality’ and as will be shown in
a later paper produced dramatic effects.

Airport construction will be only one of many
human influences which will operate in the area.
Other factors include:

(a) Shell dredging. This currently occurs at site 5
and inshore of sites 6 and 14; it may well
extend in the future.

(b) Prawn trawling. This occurs seasonally
throughout the area except for the inshore
sites. Annual variations in trawling intensity
in the area have not been adequately
quantified.

(c) Port construction. Major construction at
Fisherman Island with filling of the Boat
Passage seems probable. This is likely to cause
marked changes in the suggested 'Brisbane
River’ effect.

(d) Reduction of pollution. Industrial pollution
of the Brisbane River and Cabbage Tree Creek
is currently being reduced, and discharge of
untreated sewage is likely to be reduced in the
future.

Amidst the welter of probable changes due to
man, and with the ever-present possibility of
another devastating flood, predictions of the effects
of airport construction can only be tentative.

General Matters

Most workers on benthic biotas still feel con-
strained to express their results in terms of benthic
communities, although it has been shown that the

444

MEMOIRS OF THE QUEENSLAND MUSEUM

community concept is complex and possibly con-
fusing (see Stephenson 1973; Clifford and Stephen-
son 1975). In the present case it is not possible to
accept the restraints of Petersen (1914) and limit
the species characterising the sites to those with
constantly high populations. The populations of
virtually all species show changes during the
sampling periods.

By taking averages over all times, groups of sites
characterised by groups of species can be revealed.
In the 27 sites analysis the most important of these
are: inshore a Spisula-Xenophthalmus-Terebellides
group; in the middle a Xenophthalmus-Anadara-
Paratapes group; and offshore a Paratapes-
Amphioplus lobatus group. In the 48 sites analysis
there is greater topographic resolution but charac-
terization of site-groups by species-groups is less
distinctive. The inshore species-group is now
Spisula-Xenophthalmus-Amphitrite and the
offshore area group is Paratopes- Anadara-
Amphioplus lobatus-Terebellides. These “com-
munities” bear scant relationship to that of the
adjacent area of Moreton Bay in the dredge study
by Stephenson, Williams and Lance (1970). The
difference is mainly due to the different collecting
methods, as already noted in the Peel Island study
(Stephenson, Williams and Cook 1974). There is a
somewhat closer relationship to the characterizing
species which Hailstone (1972) noted in a dredge
study of the lower Brisbane River. Hailstone
obtained large numbers of Spisula with Parcanassa
and other species in shallow sandy-mud sediments,
while Anadara characterized mid-channel sites with
muddy sand. MacIntyre (1959) in his study of Lake
Macquarie in New South Wales also obtained large
numbers of several of the present species, parti-
cularly Anadara , Paratopes, and Amphioplus loba-
tus. Black (1971) noted that Spisula is common
from sandy sites in Port Phillip Bay, Victoria, and it
is listed from three regions in that bay by Poore and
Rainer (1974). Other species of Spisula are well-
known characterizing species elsewhere (see Thor-
son 1957) and a Xenophthalmus community is
known from sandy grounds in the Persian Gulf
(Thorson 1957).

Possibly the closest parallel with the present
results is the New Guinea study by Stephenson and
Williams (1971). Here there was an Amphioplus and
a Mesochaetopterus ‘community’ both in a warm
water situation under estuarine influence. The
study by Stephenson, Williams and Cook (1974) is
closest to the present work in times of sampling and
analytical methods but the abundant species char-
acterising site-groups have little in common.

Present results show that the densest populations
are inshore, and the mean value for one data set is

1858 specimens/m 2 and for the other 2946. These
values compare with 16-764 by Chukchin (1963) in
eastern Mediterranean at depths of 100-200 m;
102 255 by Kuznetsom (1963) in the northern
Pacific at depths to 500 m; 740-5515 by Wigley and
McIntyre (1964) in the western North Atlantic at
40-366 m; and 32-1 193 by Christie (1974) in South
Africa at depths to 50 m. Values obtained by
Sanders et al. (1965) were much higher (to
21263/m 2 ) but they used a finer mesh sieve (0.42
mm). It is unfortunately not possible to compare
present densities with those of the current extensive
investigations in Port Phillip Bay, Victoria. Poore
and Rainer (1974) deal only with molluscs and give
an overall mean of 1457 individuals/m 2 . This
suggests that densities involving all species will be
distinctly higher than those of the present study.

In the inshore sites, with the densest populations,
it appears that there are marked annual changes in
populations. This must result in high productivity
of the benthic macrofauna; in absence of biomass
determinations its magnitude is unknown. In the
most offshore site-group in the 27 sites data,
populations are lower and average 346 animals/m 2 .
Here however there are indications of a marked
biotic change every six months. Again the mac-
rofauna productivity must be of a high order.

Possibly the most interesting results of the
present survey concern the time changes in biota.
They confirm the opinion stated by Stephenson,
Williams and Cook (1974) and by Clifford and
Stephenson (1975) of the extremely doubtful value
of many of the environmental impact statements
involving benthic organisms.

The time investigations, cautiously interpreted
because of possible microtopographical pattern-
ing, still stress the transitory nature of some of the
species and the marked fluctuations in the popu-
lations of others. They give pointers to matters of
both practical and fundamental importance. From
a practical viewpoint, if environmental conditions
are not too greatly disturbed, it appears that one
biotic assemblance can readily replace another. It
can hence be regarded as a buffered biotic system;
provided airport construction does not greatly
disturb conditions one might expect the buffering
to operate tolerably quickly. Another and some-
what different aspect of ‘buffering’ in benthos
has been noted in a recent paper involving a 4-year
study by Buchanan, Kingston and Sheader (1974).

ACKNOWLEDGMENTS

We wish to thank the Australian Government
for the Colombo Plan award to the second author,
and for financial support from the sponsoring

STEPHENSON ET. AL.\ MACROBENTHOS OF BRAMBLE BAY

445

authorities. The University of Queensland has also
assisted via research grants.

Our thanks are due to Messrs S. Newlands, L.

Wale and W. Hayes for assistance during field work
and to a variety of workers (listed earlier) for help
in specific identifications.

APPENDIX

Species (and their systematic group) in order of abundance in all samples (27 x 7 + 21 x 2);
only species occurring > 9 times listed. Code number is given in first column and total population
in last column.

1

Spisula trigonella (Lamarck)

Pelecypoda (Mactridae)

19318

2

Mesochaetopterus minutus Potts

Polychaeta (Chaetopteridae)

3367

3

Xenophthalmus pinnotheroides White

Decapoda (Pinnotheridae)

2457

4

Paratopes scordalus Iredale

Pelecypoda (Veneridae)

2238

5

Anadara trapezia Deshayes

Pelecypoda (Arcidae)

532

6

Amphioplus sp.

Echinodermata (Ophiuroidae)

520

7

Amphioplus lobatus (Ljungman)

Echinodermata (Ophiuroidae)

456

8

Theora lata Hinds

Pelecypoda (Tellinidae)

358

9

Terebellides stroemi Sars

Polychaeta (Terebellidae)

323

10

Amphitrite rubra (Risso)

Polychaeta (Terebellidae)

283

11

Pupa fumata (Reeve)

Gastropoda (Acteonidae)

221

12

Theora sp.

Pelecypoda (Tellinidae)

216

13

Ampliipholis loripes Koehler

Echinodermata (Ophiuroidae)

204

14

Leanira yhleni Malmgren

Polychaeta (Aphroditidae)

199

15

Parcanassa mangeloides Reeve

Gastropoda (Nassariidae)

185

16

Lumbrineris latreilli Audouin and

Milne Edwards

Polychaeta (Eunicidae)

117

17

Loimia medusa (Savigny)

Polychaeta (Terebellidae)

108

18

tunicate 1

Ascidiacea

105

19

tunicate 3

Ascidiacea

101

20

bivalve 1

Pelecypoda

98

21

Nucula astricta Iredale

Pelecypoda (Nuculidae)

94

22

oyster 1

Pelecypoda (Ostreidae)

94

23

Pectinaria antipoda Schmarda

Polychaeta (Pectinaridae)

90

24

Nereis jacksoni Kinberg

Polychaeta (Nereidae)

90

25

Mesochaetopterus cf. capensis (McIntosh)

Polychaeta (Chaetopteridae)

90

26

Ophelina gigantea Rullier

Polychaeta (Opheliidae)

88

27

sea anemone 1

Actinaria

87

28

Jsolda pulchella Muller

Polychaeta (Ampharetidae)

76

29

bivalve 2

Pelecypoda

74

30

Tellina texturata Sowerby

Pelecypoda (Tellinidae)

66

31

Glyeera prashadi Fauvel

Polychaeta (Glyceridae)

62

32

Cirriformia sp.

Polychaeta (Cirratulidae)

61

33

Protankyra sp.

Echinodermata (Holothuroidea)

61

34

Petaloproctus terricola Quatrefages

Polychaeta (Maldanidae)

55

35

bivalve 3

Pelecypoda

55

36

Chama fibula Reeve

Pelecypoda (Chamidae)

54

37

Cycladicama sp.

Pelecypoda (Ungulinidae)

53

38

balanid 1

Cirripedia

52

39

Edwardsia sp.

Actinaria

48

40

On up his sp.

Polychaeta (Eunicidae)

46

41

Placamen sydneyense Menke

Pelecypoda (Veneridae)

45

42

Dasybranchus caducus (Grube )*

Polychaeta (Capitellidae)

44

43

Area sp. 1

Pelecypoda (Arcidae)

41

44

oyster 3

Pelecypoda (Ostreidae)

40

45

Amphioplus depressus (Ljungman)

Echinodermata (Ophiuroidae)

40

46

Trichomya hirsuta (Lamarck)

Pelecypoda (Mytilidae)

39

47

Leptomya pura Angus

Pelecypoda (Semelidae)

36

48

Mesochaetopterus sp.

Polychaeta (Chaetopteridae)

33

446

MEMOIRS OF THE QUEENSLAND MUSEUM

49

aphroditid 1

Polychaeta (Aphroditidae)

32

50

Leonnates stephensotii Rullier

Polychaeta (Nereidae)

31

51

Clorida grand (Stephenson)

Stomatopoda

30

52

Marphysa sanguinea (Montague)

Polychaeta (Eunicidae)

30

53

Glossobalanus hedleyi Hill

Enteropneusta (Balanoglossidae)

30

54

Reticunassa paupera Gould

Gastropoda (Nassariidae)

28

55

nemertean ‘pink’

Nemertea

27

56

whip coral

Gorgonacea

27

57

Venus sp.

Pelecypoda (Veneridae)

26

58

Macoma donaciformis Deshayes

Pelecypoda (Tellinidae)

25

59

Ophiactis perplexa Koehler

Echinodermata (Ophiuroidae)

24

60

Anomia sp.

Pelecypoda (Anomiidae)

20

61

amphipod 4

Amphipoda

20

62

Elamenopsis lineata A. Milne Edwards

Decapoda (Hymenosomidae)

19

63

Fist a sp.

Polychaeta (Terebellidae)

19

64

amphipod 2

Amphipoda

18

65

Amaeana trilobata (Sars)

Polychaeta (Terebellidae)

18

66

Hexapus granuliferus Campbell and
Stephenson

Decapoda (Goneplacidae)

18

67

tunicate 2

Ascidiacea

16

68

Chaetopterus variopedatus Renier

Polychaeta (Chaetopteridae)

15

69

sabellid 1

Polychaeta (Sabellidae)

15

70

Natica sp.

Gastropoda (Naticidae)

14

71

amphipod 6

Amphipoda

14

72

Tapes watlingi I redale

Pelecypoda (Veneridae)

13

73

amphipod 1

Amphipoda

12

74

Bedeva hanleyi Angus

Gastropoda (Muricidae)

12

75

Brissopsis luzonica (Gray)

Echinodermata

12

76

nemertean ‘black’

Nemertea

11

77

Modiolus ‘ ostentatus ’ Swainson

Pelecypoda (Mytilidae)

11

78

ampharetid 1

Polychaeta (Ampharetidae)

11

79

Papilia subrugata Iredale

Pelecypoda (Veneridae)

10

80

Eurythoe parvecarunculata Horst

Polychaeta (Amphinomidae)

10

81

Polydora sp. 1

Polychaeta (Spionidae)

10

^Possible misidentification.

LITERATURE CITED

(These exclude references to authors of species de-
scriptions)

Black, J. H., 1971. Port Phillip survey 2. Benthic
communities. Mem. natn. Mus. Viet. 32: 129-70.

Buchanan, J. B., Kingston, P. F. and Sheader, M.,
1974. Long-term population trends of the benthic
macrofauna in the offshore mud of the Northumber-
land Coast. J. mar. Biol. Ass U.K. 54 : 785 95.

Christie, N. D., 1974. ‘Distribution Patterns of the
Benthic Fauna along a Transect across the Con-
tinental Shelf off Lamberts Bay, South Africa’.
Ph.D. thesis (Unpublished), Zoology Department,
University of Capetown.

Chukhchin, V. D., 1963. Quantitative distribution of
benthos in an eastern part of the Mediterranean Sea.
(In Russian). Trudy Sevastopol, biol. Sta. 16 : 215-33.

Clifford, H. T. and Stephenson, W., 1975. ‘An
Introduction to Numerical Classification’. Pp.
1-224, (Academic Press: New York).

Day, J. H., Field, J. G. and Montgomery, Mary P.,
1971. The use of numerical methods to determine the
distribution of the benthic fauna across the con-
tinental shelf of North Carolina. J. anim. Ecol. 40 :
93-123.

Dale, M. B. and Anderson, D. J., 1973. Inosculate
analysis of vegetation data. Aust. J. Bot. 21 : 253-76.

Field, J. G., 1971. A numerical analysis of changes in the
soft-bottom fauna along a transect across False Bay,
South Africa. J. exp. mar. Biol. Ecol. 7 : 215-53.

Field, J. G. and Macfarlane, G., 1968. Numerical
methods in marine ecology. I. A quantitative
‘similarity’ analysis of rocky shore samples in False
Bay, South Africa. Zool. Afr. 3: 1 19-37.

Field, J. G. and Robb, R. T., 1970. Numerical methods in
marine ecology. II. Gradient analysis of rocky shore
samples from False Bay. Zool. Afr. 5: 191-210.

Hailstone, T. S., 1972. ‘Ecological Studies on the Sub-
Tidal Benthic Macrofauna at the Mouth of the
Brisbane River.’ Ph.D. thesis (unpublished). Zool-
ogy Department, University of Queensland.

STEPHENSON ET. AL.: MACROBENTHOS OF BRAMBLE BAY

447

Kuznetsov, A. P., 1963. Benthic invertebrate fauna of
the Kamchatka waters of the Pacific Ocean and
Northern Kurile Is. (In Russian; English summary).
A had. Nauk. SSSR Inst. Okecmol. 1-271.

MacIntyre, R. J., 1959. Some aspects of Lake Mac-
quarie, N.S.W., with regard to an alleged depletion
of fish. VII, The benthic macrofauna. Aust. J. mar.
freshw. Res. 10 : 341 53.

Maxwell, W. G. H., 1970, The sedimentary framework
of Moreton Bay, Queensland. Aust. J. mar. freshw.
Res. 21 : 71—88.

Newell, B. S., 1971, The hydrological environment of
Moreton Bay, Queensland, 1967-1968. Div. Fish.
Oceanogr. Tech. Pap. No. 30, 35 pp., (C.S.I.R.O.:
Melbourne).

Petersen, C. G. J., 1914. Valuation of the sea. II. The
animal communities of the sea bottom and their
importance for marine zoogeography. Rep. Dan.
Biol. Sta. 21 : 1^44, Appendix 24 pp.

Poore, G. C. B. and Rainer, S., 1974. Distribution and
abundance of soft-bottom molluscs in Port Phillip
Bay, Victoria, Australia. Aust. J. mar. freshw. Res.
25: 371-441.

Raphael, Y. Irene, 1974. The macrobenthic fauna of
Bramble Bay, Moreton Bay, Queensland. M.Sc.
thesis (unpublished). Zoology Department, Uni-
versity of Queensland.

Raphael, Y. Irene and Stephenson, W., 1972. The
macrobenthic fauna of Bramble Bay, Moreton Bay,
Queensland. I. Preliminary studies. Rept to Comm.
Dept Works and Qd Dept Coord. General, Oct. 1972,
32 pp.

Sanders, H. L., Hessler, R. and Hampson, G. R., 1965.
An introduction to the study of deep-sea benthic
faunal assemblages along the Gay-Head Bermuda
transect. Deep Sea Res. 12 : 845-67.

Sokal, R. R. and Rohlf, F. J., 1969. 'Biometry.’ Pp.

i-xxi, 1-776. (W. H. Freeman: San Francisco).

Stephenson, W., 1973. The validity of the community
concept in marine biology. Proe. R. Soe. Qd 84(7):
73-86.

Stephenson, W., and Dredge, M. C. L., 1976. Numerical
analyses of fish catches from Serpentine Creek. Proc.
R. Soc. Qd 87 (in press).

Stephenson, W., Smith, R. W., Sarason, T. S., Greene,
C. S. and Hotchkiss, D. A. Soft bottom benthos
from an area of heavy waste discharge: hierarchical
classification of data. (MS).*

Stephenson, W., and Williams, W. T., 1971. A study of
the benthos of soft bottoms, Sek Harbour, New
Guinea, using numerical analysis. Aust. J. mar.
freshw. Res. 22: 1 1-34.

Stephenson, W., Williams, W. T. and Cook, S. D., 1974.
The benthic fauna of soft bottoms, southern More-
ton Bay. Mem. Qd Mus. 17 ( 1 ): 73-123.

Stephenson, W., Williams, W. T. and Lance, G, N,,
1970. The macrobenthos of Moreton Bay. Ecol.
Monogr. 40 : 459-94.

Thorson, G., 1957. Bottom communities (sublittoral or
shallow shelf). In Hedgpeth, J. W., (Ed.), 'Treatise
on Marine Ecology and Palaeontology’ Vol. L,
Ecology, pp. 461-534. Mem. Geol. Soc. Amer. 67 .
(1963 reprint).

Wigley, R. L. and McIntyre, A. D., 1964. Some
quantitative comparisons of offshore meiobenthos
and macrobenthos south of Marthas Vineyard.
Limnol. Oceanogr. 9 : 485 93.

Williams, W. T. and Stephenson, W., 1973. The analysis
of three-dimensional data (sites x species x times)
in marine ecology. J. exp. mar. Biol. Ecol. 1 1 : 207-27.

*Now 1975. S. Calif. Coastal Water Res. Project
TM226, 38 pp.

Mem. QdMus. 17 ( 3 ): 449-55. [1976]

INTRODUCTION OF THE NORTH ATLANTIC ASCIDIAN MOLGULA
MANHATTENSIS (DE KAY) TO TWO AUSTRALIAN RIVER ESTUARIES

Patricia Kott

Queensland Museum

ABSTRACT

Populations of the North American (Atlantic coast) Molgula manhattensis have been sampled
at Newport Power Station in the mouth of the Yarra River, Victoria; and at several stations up to
22 kilometres from the mouth of the Brisbane River. The species and its affinities are discussed in
detail, together with the implications of its distribution and its occurrence in Australian estuaries.

Eleven Stations in the Brisbane River, from its
mouth to 70 km upstream, have been sampled by
Mr R. Monroe at 3 monthly intervals from May
1974, following a major flood in January. In May
1975 two specimens (1 -5 cm and 1 0 cm in diameter,
respectively) of Molgula manhattensis were taken
off Mowbray Park, some 16 km upstream. In
August large numbers of the species were present at
this site, and even larger numbers were taken up to
6 km further up the river in the South Brisbane
Reach. By December only 3 specimens were taken
at the latter station. Downstream, the species was
taken only once, and in small numbers. The species
was not present in samples taken in January 1976.

From May to August is the winter season of low
rainfall and during that period highly seasonal
populations of Molgula spp, are present in More-
ton Bay (Kott 1972). M. manhattensis has not
been taken in Moreton Bay, however, and it is most
likely that parent stocks from which the riverine
populations were recruited were located on ship-
ping (from the western Atlantic via the Panama
Canal) in the Port of Brisbane. This extends up the
Brisbane River to within 3 km of the South
Brisbane Reach.

In the family Molgulidae, it is known that sexual
maturity is achieved early, in individuals of small
size (Berrill 1931); while the spiral arrangement of
stigmata in this family provides a means whereby
maximum filtration area is available in small
individuals to contribute to this general metabolic
efficiency. Kott (1972) suggested that these factors
would represent advantages where populations
suffered seasonal mortality resulting from the
periodic flooding, silt deposition, and temperature

fluctuations that commonly occur in sheltered
bays. The large molgulid renal organ (see Berrill
1950) could also represent an advantage in these
locations.

In fact, 7 of 12 free living ascidian species in
Moreton Bay are molgulid species (Kott 1972); and
M. mollis and M . sabulosa commonly occur in Port
Phillip Bay (Kott, in press). Of 29 species reported
on from America (Van Name 1945) 7 are recorded
from harbours, estuaries, and river mouths; 6 are
recorded from intertidal or shallow waters; and 1 1
occur in shallow waters in polar regions where
melting ice causes seasonal dilution of sea water.

Molgula manhattensis (de Kay) and the closely
related M. tubifera (Orsted) appear to have de-
veloped the capacity to withstand brackish con-
ditions to a surprising extent. The combined
records of both species (> M. manhattensis :
Thompson 1930; Berrill 1950) range ‘from the
White Sea to the tropics in water whose salinity
varies from 16 to over 30% o ’ (Thompson 1930, p.
23). Berrill (1950, p. 248) has also commented on
their tolerance of the ‘diluted and polluted waters
typical of estuaries and harbours’. Van Name
(1945, p. 388) refers to the western Atlantic M.
manhattensis as ‘one of the few ascidians that will
live in water of somewhat diminished salinity’.

It is not altogether surprising, therefore, that it is
M. manhattensis which has been introduced into
Brisbane River, some 22 km from its mouth, where
the bottom salinity (registered at the Port Office, in
the Town Reach of the river, between the stations
sampled off Mowbray Park and in the South
Brisbane Reach) was in the vicinity of 16% 0 in

450

MEMOIRS OF THE QUEENSLAND MUSEUM

May 1975; 19% 0 in August 1975; 25% 0 in Septem-
ber 1975; and 11% 0 in December 1975.

It is not impossible that there have been earlier
introductions that were eradicated by the January
1974 flood. It is also possible that introduced
riverine populations do not withstand the summer
rainfall period and that, unless the species has
become established in refuges in Moreton Bay,
subsequent recruitment (if any) will also be from
ships hulls.

Apart from the fact that Molgula manhattensis of
1-2 cm diameter from Woods Hole in September
1927 had attained sexual maturity (Berrill, 1931),
little is known of the growth rate and breeding
season of M. manhattensis and this must be largely
inferred from what is known of related species. M.
tubifera (closely related to M. manhattensis) has
been found breeding at Plymouth in all seasons
other than winter (March to October; Berrill 1931,
1935). Molgula mollis (> M. sabulosa: Kott, 1972)
is a species of similar size which reaches sexual
maturity before reaching 2 cm in diameter. After
settling in Moreton Bay at the end of winter, this
species apparently produced at least one generation
of offspring which grew to at least 2 cm in diameter
before the populations disappeared in summer
(Kott 1972). Therefore, sexual maturity is ap-
parently attained within 2 months. If the growth
rate and breeding season of the introduced pop-
ulations of M. manhattensis are comparable with

these, juveniles settling on ships in the western
Atlantic in autumn could have been transported to
the southern hemisphere, where their offspring
settled in the Brisbane River following the period of
summer rain (March 1975). Then the large mature
populations sampled in August would have been
adults of the second generation, progeny of the
parent generation that settled in the river to reach
sexual maturity in May; and the individuals taken
in October would represent a third generation.

Molgula manhattensis (de Kay, 1843)

(Figs. 2-4)

Ascidea manhattensis De Kay, 1843, p. 259.

Molgula manhattensis : Van Name, 1945, p. 385 and
synonymy.

Molgula plater. Arnback, 1 928, p. 22, plate 1 , figs. 31-4.
New Records

Brisbane R.: QM G8976, G8977, G8979 Mowbray
Park (32 specimens, R. Monroe, 7. viii. 1 975); QM G8978,
mud channel, mouth of Norman Creek (2 specimens, R.
Monroe, 12.V.1975); South Brisbane Reach, below Exec-
utive Building (numerous specimens, R. Monroe,
7.viii.l975; 50 specimens including juveniles, R. Monroe,
30.x. 1975; 3 specimens, P. Davies, 5.xii. 1975); Bulimba
corner (1 specimen, R. Monroe, 30.x. 1975).

Yarra R.: NMV H301, Newport Power Station
(9. i. 1967).

(T) Norman Creek
(D Mowbray Park
(3) City Reach
(?) South Brisbane
Reach

© Inner City area
© Bulimba corner

km

Fig. 1: Lower Reaches of the Brisbane River from the Inner City area to Moreton Bay.

KOTT: INTRODUCTION OF ASCIDIAN TO AUSTRALIA

451

Other Material Examined

Molgula tubifera: AM Y1949, Duke Rock, Plymouth,
U.K, (2 specimens, P. Kott, July 1951).

Distribution

On the Atlantic coast of North America from
Portland to Louisiana (Van Name 1945). M.platei
Hartmeyer, 1914; Van Name, 1945, known only
from a single specimen from Chile, is similar to the
present species. If closer examination should prove
them synonymous, the possibility that it had been
transported between the Pacific and Atlantic coasts
of America, by ship, should not be overlooked.

The species is commonly taken from shallow
waters, the deepest reliable record being at 30 m
(Van Name 1945).

Description

Individuals are almost spherical, to ovoid,
slightly laterally flattened, and up to 2-5 cm in
diameter. Both apertures are on short siphons a
short distance apart on the upper surface. The
branchial aperture is turned ventrally and the atrial
aperture diverges slightly in the opposite direction.
The test is whitish, thin, papery and transparent. It
is covered with sparse, short hairs to which mud
and fine sand adhere. On the basal half of the body
the hairs are longer and form root-like processes
anchoring the animal into the substrate. (Fig. 2).

The body wall is very thin with muscles con-
spicuous only around the siphons. It is very closely
adherent to the test. At the base of the branchial
siphon there is a straight edged velum on both sides
which reduces the opening to a longitudinal slit.
Longitudinal muscles from the body wall extend
into these velar folds of the siphonal lining. The
branchial tentacles are very bushy. The dorsal
tubercle is a circular cushion with a U-shaped
opening turned posteriorly, the horns turned
inwards. The dorsal lamina is very short and is
joined by 3 broad transverse vessels from each side
of the body.

The branchial sac has 6 narrow deeply curved
folds on each side of the body. There are no internal
longitudinal vessels between the folds. Those on the
folds are extremely broad and project out from the
ventral surface of the fold as a flat membrane. The
internal longitudinal vessels become progressively
wider toward the base of each fold. There are no
internal longitudinal vessels on the dorsal surface
of the fold. Longitudinal vessels on the ventral
surface of each fold are arranged according to the
formula DL 0(3) 0(4) 0(4) 0(4) 0(4) 0(3)0 E. In the
very extensive spaces between the folds there are
very numerous, and irregular, interstitial in-

fundibula, with interrupted and irregular vessels
extending across them (Fig. 3). The primary
infundibula in each fold are subdivided into two
and each subdivision is again divided at its apex in
the margin of the fold. However, this arrangement
is much obscured by the many irregular accessory
or interstitial coils that are present, especially near
the apex of the coils. In older specimens there is an
unperforated area along either side of the en-
dostyle.

The gut forms a very narrow deeply curved loop
enclosing the gonad on the left side of the body.
The stomach is long, with internal longitudinal
glandular folds. The anal border is divided into
about 12 shallow rounded lobes. On the right side
the molgulid kidney occupies the usual postero-
ventral position. It is long and slightly curved and
increases in length as the individual becomes larger.
The right gonad extends along parallel to the dorsal
border of the kidney. The gonads consist of an
elongate or flask-shaped ovary terminating
postero-dorsally in a short oviduct. Very dense
clumps of arborescent testis follicles are arranged
continuously along the proximal end and the
ventral border of each ovary. Only occasionally
there are small isolated clumps of testis follicles on
the dorsal margin of the ovary. Vasa efferentia
extend from the testis follicles onto the mesial
surface of the ovary where they unite into one short
vas deferens on the right gonad, but on the left
gonad there are up to six short vas deferens
arranged along the length of the ovary. (Fig. 4).

Occasionally eggs, with follicle cells, are found
in the peribranchial cavity, but no larvae were
found suggesting that the species is oviparous. The
eggs are 1 • 1 mm in diameter, excluding the follicle
cells. Juveniles were present attached to the test of
adult specimens in August.

Relations

Hartmeyer (1923) and Berrill (1950) believe the
European M. tubifera to be synonymous with M.
manhattensis. Hartmeyer’s synonymy is based on
the fact that the gut loop of both M. tubifera and
M. manhattensis is equally narrow and deeply
curved and encloses the left ovary. Berniks view is
supported by the small eggs and oviparous habit
and similar development in both species (Berrill
1928).

Despite similarities between the species, neither
Arnback (1928) nor Van Name (1945) accepted the
synonymy of M. tubifera and M. manhattensis.
Arnback points out that the branchial sac of
European specimens has fewer accessory spirals
than M. manhattensis.

452

MEMOIRS OF THE QUEENSLAND MUSEUM

Fig. 2: Molgula manhattensis; external appearance (specimen from the Brisbane River).

Fig. 3: Molgula manhattensis ; portion of branchial sac between the folds (specimen from the Brisbane River).
Fig. 4: Molgula manhattensis', kidney, gut and gonads on inner body wall (specimen from the Brisbane River).

KOTT: INTRODUCTION OF ASCIDIAN TO AUSTRALIA

453

Specimens from the English Channel in the
collection of the Australian Museum (AM Y1949,
Duke Rock, Plymouth, July 1951) have been re-
examined. They resemble the present specimens of
M. manhattensis from the Brisbane River in size
and general appearance, in the number of branchial
folds, in the deeply curved gut loop, in the position
of the right gonad parallel to the kidney, and in the
size of the eggs and testis follicles. The differences
between the American M. manhattensis and the
European M. tubifera are confirmed, however, and
the details of these differences are set out in Table 1 .

The multiplicity of short testis ducts in the
present specimens from the Brisbane and Yarra
Rivers is similar to that inferred for M. manhat-
tensis by Huntsman, 1922, and that known for
some specimens of the European species (see
Arnback 1928). However, Arnback ( loc . cit.) has
drawn attention to variations in the condition of
the vas deferens in a range of specimens from
European locations.

The multiple short vas deferens found in M.
manhattensis and related species was used by
Huntsman (1922) to characterise the genus Gym-
nocystis. Based on this criterion M. ampulloides (a
synonym of M. tubifera) was excluded from that
genus.

Huntsman’s subdivision of the genus Molgula
was not adopted, however, since variations in
arrangement of testis follicles relative to the ovary
and variations in the length and disposition of the
vas deferens occur throughout the genus, and are
not considered to be of more than specific

significance. The significance of this reported
variation in the condition of the vas deferens in M.
tubifera and its synonyms has yet to be resolved.

In the Australian Museum specimens from the
English Channel, referred to above, a single vas
deferens extends along the mesial surface of the
ovary and opens near the opening of the oviduct.

The Brisbane and Yarra River specimens con-
form exactly in all respects with the American
species M . manhattensis (see Van Name 1945).

Table 2 sets out the principal characters which
distinguish those species in which more than a
single vas deferens associated with the right or left
gonad has been reported.

Filtration Rate

The capacity of the individual to filter large
amounts of water and deal effectively with the
sediments filtered from the water could also be
relevant to its occurrence in shallow estuaries and
in other locations where there is an unusual amount
of suspended matter and other pollutants.

The pronounced branchial folds of M. manhat-
tensis and the complex arrangement of primary and
accessory infundibula on the folds and in the
interspace have developed the area available for
filtration to a maximum degree.

MacGinitie (1939) and Day (1974) have shown
that in Ascidia californica and Pyura stolonifera
respectively, mucous moves over the pharynx in a
continuous sheet. Jorgensen’s (1939) assessment of
filtration rate of Molgula manhattensis and Ciona
intestinalis suggests that all particulate matter

TABLE 1 : Differences between M. manhattensis and M. tubifera

Species

Apertures

Internal

longitudinal

vessels/fold

Testis

follicles

Dorsal

tubercle

Stigmata

Anal

border

M. man-

close together on

never exceeds

not

cup-shaped

numerous

lobed

hattensis

the upper surface;
long when
extended (see
Arnback 1928, fig.
34)

4; present on
dorsal side of
fold only

present on
dorsal
border of
ovary

with unrolled
horns turned
to the right or
posteriorly

accessory
spirals; long
stigmata (see
also Van
Name 1945)

M.

a little distance

up to 7

present all

slit-like, S-

only

smooth

tubifera

apart on the upper
surface; moderate
length when
extended (see
Thompson 1930,
pi. 3, figs. 1, 2)

(Thompson
1930, p. 21; ‘3
to 6 mostly’ 5
to 6); present
on both sides
of fold

around

ovary

shaped, C-
shaped or U-
shaped
turned to the
right, or left
or posteriorly
(Arnback
1928)

occasional

accessory

spirals;

stigmata

short (see

also

Thompson
1930, pi. 3,
fig. 6)

454

MEMOIRS OF THE QUEENSLAND MUSEUM

2 3

4J

0>

<D

8-2
3 CO

o

£

Cd

£

cd

£

cd

M

o

o

o

SSj-c

cd

Z

z

z

cd

cd

JO

(£ £
QJ 3

Arnt

1928

Van

1945

Van

1945

Van

1945

Arnt

1928

Arnt

1928

3

8

C/5

a

3

06

tu

35

s ,o

.3 3 3 jsg
X S3 D
3 3 00 M

s c s s

<-< o >

>1

£

i-i

o

z

Cfl

CJ

r

G

u

o

g

jd

3

3

on

o

£

3

♦G

cd

s.

C/5

cd

3

z

<

<

£

33

.a o

^z

<

w

s

z

z

PQ

z

■§ &

§S

fc; o

-o ^

§

£ >

£ o

3

M

G

JS

<

£

z

>1

5 — i

cd

Cd >

53 o

3 t «
D O

> l_

3) «

3 V

° o

3 JO

tO •'t

<0%

3 « </5

3 3

g<<
-3 tlj ^
2Zvi

* L'

P

£ o

33

3 )

TO

33

"O

-j

3

3

o

rj

3

O

o

o

O

3 .

3

3

3

G

.5

3

3

3

3

G

!S tn

O

O

O

O

O

O 33

3

3

3

3

C

3 'o

O

O

O

is

£

O

o

3 i£

L.

t-H

co

S-i

ft

5 -

ft

_o

o

—

ft

S-.

Cu

02

JO

'ft

ft

ft

Oo

tn 3

° 3

3 §>

3

3 _

.o t:
35 O
3 ,c

CJ C/5

o

3

£ &o

3 3

£ 3

3

3

T3

-j

a

a

3

JO JO

in c/5

u 6

CO

tvo

CO

^Z

ft §
£>

2

I «

■ &3

^ Cfl

~3 3

I £

3 3

:sz

o c

^ ^ ^

& '£ t2

b| a
3 3, P
£ 'sT;Z

^-g
3 2
So

| 3

1*
3- 3
3 -O

a n
3 2

■%*

3 zn

KOTT: INTRODUCTION OF ASCIDIAN TO AUSTRALIA

455

(down to 1 micron at least) is strained from the
water by this sheet of mucous.

It was possible, therefore to estimate an
approximate rate at which Molgula manhattensis
filtered the muddy water of the Brisbane River.

The filtration rate was assessed according to the
formula:

(log cone — log cone ) x M

m 0 —

loge x t

where m is the quantity of water filtered in time t;
conc o is the concentration of particles in the water
at the beginning of the experiment and cone is the
concentration of particles in the water at the end of
the experiment; t is the duration of the experiment;
and M is the amount of water in the experimental
vessel (see Jorgensen 1943).

The present experiment was conducted for a
period of 12 hours with duplicate specimens of M.
manhattensis , each in 1-3 litres of Brisbane River
water with a heavy suspension of fine mud. No
sedimentation had occurred in the control vessel at
the end of the experiment. The concentration of
particles in suspension after 12 hours is expressed
as a percentage of the concentration at the
beginning of the experiment, and was estimated by
measured dilution of the unfiltered water to match
opacity observed in the experimental vessels at the
end of the experiment. The dilution achieved by the
removal of suspended particles by the ascidian,
over this 12 hour period was of the order of 1 in
1000 .

The filtration rate of a single specimen of
Molgula manhattensis was thus shown to be in the
vicinity of T25 mls/minute (750 mls/hour). This
rate confirms Jorgensen’s (1952) values of 8 to 18
litres per hour for 1 5 specimens.

These small individuals are, therefore, very
efficient filter feeders. The capacious and long gut
loop with its pronounced typhlosolar fold un-
doubtedly contributes to the accommodation of
large amounts of sediment from which nutriment is
extracted.

LITERATURE CITED

Arnback-Christie-Linde, Augusta, 1928. Northern
and Arctic invertebrates in the collection of the
Swedish State Museum. IX. Tunicata. 3 Molgulidae
and Pyuridae. K. svenska Vetensken-Akad. Handl.
14(a): 1-101.

Berrill, N. J., 1928. The identification and validity of
certain species of ascidians. J. mar. biol Ass. U.K. 15:
159-75.

1931. Studies in tunicate development. Part II. Abbrev-
iation of development in the Molgulidae. Phil. Tram
R. Soc. (B) 219: 281-346.

1935. Studies in tunicate development. Part III.
Differential retardation and acceleration. Phil. Trans
R. Soc. (B) 225: 255-326.

1950. The Tunicata. Ray Soc. Pubis 133: 1-354.

Day, R. W., 1974. An investigation of Pyura stolonifera
(Tunicata) from the Cape Peninsula. Zoologica
Africana 9(1): 35-58.

De Kay, J. E„ 1843. ‘Zoology of New York, or the New
York fauna’, pt 5, Mollusca. viii + 271 pp., 40 pis.
(Albany).

Hartmeyer, R., 1914. Diagnosen einige neuer Molgu-
lidae aus der sammlung des Berliner Museums nebst
bemerkungen fiber die systematik und nomenklatur
dieser familie. Sber. Ges. naturf. Freunde Berl. 1914:
1-27.

1923. Ascidiacea, part I. Zugleich eine ubersicht fiber
die Arktische und boreale ascidienfauna auf tier-
geographischer grundlage. Ingolf-Exped. 2(6):
1-365.

Huntsman, A. G., 1922. The ascidian family Caesiridae.
Proc. Trans. R. Soc. Can. (5) 16(3): 21 1-34.

Jorgensen, C., 1943. On the water transport through the
gills of bivalves. Acta Phys. Scand. 5: 297-304.

1949. Feeding-rates of sponges, lamellibranchs and
ascidians. Nature 163 (4154): 912.

1952. On the relation between water transport and food
requirements in some marine filter feeding in-
vertebrates. Biol. Bull. mar. biol. Lab., Woods Hole
103: 356-63.

Kott, Patricia, 1964. Stolidobranch and phlebobranch
ascidians of the Queensland Coast. Pap. Dep. Zool.
Univ. Qdl(l)\ 127 52.

1969. Antarctic Ascidiacea. A monographic account of
the known species based on specimens collected
under U.S. Government auspices 1947 to 1963.
Antarct. Res. Ser. American Geophysical Union, 13:
i-xv, 1-239.

1972. Some sublittoral ascidians in Moreton Bay and
their seasonal occurrence. Mem. Qd Mus. 16(2):
233-60.

In press. Ascidian fauna of Western Port Bay, Victoria
and a comparison with that of Port Phillip Bay.
Mem. Natn. Mus. Viet.

Macginitie, G. E., 1939. The method of feeding of
tunicates. Biol. Bull. mar. biol. Lab., Woods Hole
77(3): 443-7.

Thompson, H., 1930. The Tunicata of the Scottish area.
Sclent. Invest. Fishery Bd. Scot!. 3: 1-45, 8 pis., 13
charts.

Van Name, W. G., 1945. The North and South American
ascidians. Bull. Am. Mus. nat. Hist. 84: 1-476.

Mem. QdMus . 17 ( 3 ): 457-9, pis. 61-2. [1976]

UPPER MOLAR ALVEOLAR PATTERNS OF SOME MURIDAE IN QUEENSLAND

AND PAPUA NEW GUINEA

Elizabeth Knox

Queensland National Parks and Wildlife Service

SUMMARY

The upper molar alveolar patterns of 33 species in 16 genera of Muridae are illustrated. The
eight patterns recognized are each related to a genus or a group of genera; the taxonomic value of
the patterns is limited because of inconsistent, though infrequent, variation. Although patterns do
not distinguish among native Rattus species, they provide a useful additional criterion for
separating Melomys cervinipes (Gould) from Melomys littoralis (Lonnberg).

Biological entities within the Muridae remain ill-
defined despite a voluminous literature in the
medical, agricultural and pure zoological fields.
During long-term studies of Rattus and Melomys in
Queensland, attention was given to the taxonomic
value of alveolar patterns in the maxillae. Jones
(1922) described the alveolar patterns of five
species of murids from South Australia, and
Ellerman (1942) discussed the roots of M 1 of
Australasian Muridae. With the expansion of field
work throughout Queensland to Cape York Penin-
sula, exploratory efforts were made with other
genera in Queensland and some species from Papua
New Guinea.

Material Examined

Rattus and Melomys specimens were available in
numbers (see Table 1) from recent field collections
throughout the State and from consequent breed-
ing colonies. The collections of the Department of
Forestry and the Queensland Museum (QM) were
examined also. Skulls were selected from the
available material. Hydromys specimens were
readily available; skulls of Conilurus albipes (Lich-
tenstein) were used because a Queensland specimen
of C. penicillatus (Gould) was not available.

The Australian Museum (AM) provided a skull
of Xeromys myoides Thomas from Mackay, Q.,
and a skull of Melomys lutillus (Thomas) from
Papua New Guinea. The University of Queensland
supplied specimens of M. rufescens (Alston) (type
species), Pogonomelomys sevia (Tate and Arch-
bold), and Pogonomys mollipilosus Peters and

Doria, also from Papua New Guinea. All other
specimens, except those from recent field col-
lections presently retained by the Queensland
National Parks and Wildlife Service, Brisbane, are
in the Queensland Museum.

Nomenclature is based on Iredale and Trough-
ton (1934).

Alveolar Patterns

Alveolar patterns of the 33 species examined are
shown in Plates 6D2. The distribution of alveoli
among molars falls into eight patterns which, with
related genera, are set out in Table 1. Melomys
(Group A) consists of M. lutillus , M. littoralis, and
M. australius; Melomys (Group D) of M. cer-
vinipes, M. rufescens , M. rubicola , and Melomys sp.
The patterns for Rattus , Hydromys, Mus, and
Notomys agree with those of Jones (1922).

The plates also illustrate variations seen in the
patterns recorded in Table 1 . The irregular shape of
a large alveolus is frequently the result of a union
with an adjacent small one; this becomes apparent
when the third molar is fully erupted and a
permanent pattern is discernible, as is illustrated by
Plate 62Da. Occasionally an alveolus of compara-
tively small size and variable location may occur
(Plate 6 IE); such aberrations are rare, being
present for example in less than 1% of the series of
M. cervinipes.

The taxonomic value of the alveolar patterns is
limited because of the inconsistent variation that
occurs infrequently within these. This study thus
did not assist in distinguishing species of native

458

MEMOIRS OF THE QUEENSLAND MUSEUM

Rattus, but the occurrence of two patterns in
Melomys (Table 1 A and D) provides a morphologi-
cal character, independent of age and of measure-
ments, which is a useful additional criterion for
separating M. cervinipes and M. lit (oral is.

ACKNOWLEDGMENTS

Field officers of the Research' and Planning
Branch, Queensland National Parks and Wildlife
Service, supplied most of the field specimens; Dr P.
D. Dwyer, Senior Lecturer in Zoology, University
of Queensland, provided the Papua New Guinea
material; Dr F. H. Talbot, Director of the Aus-
tralian Museum, Sydney, supplied material as
requested; Dr A. Bartholomai, Director of the

Queensland Museum, permitted the use of its
reference collection. Mr P. Fry, formerly Photo-
graphy Branch, Department of Primary Industries,
provided all of the illustrations. This assistance is
gratefully acknowledged.

LITERATURE CITED

Ellerman, J. R., 1942. The Families and Genera of
Living Rodents’, Vol. 11 (British Museum (Natural
History): London).

Iredale, T. and Troughton, E. Le G., 1934. 'A Check
List of the Mammals Recorded from Australia’.
(Australian Museum: Sydney).

Jones. F. Wood, 1922. On the dental characters of certain
Australian rats. Proc. Zool. Soc. Lond. 1922: 587-98.

TABLE 1: Molar Alveolar Patterns among 16 Murid Genera

Pattern

Distribution of
molar alveoli

Genus (specimens examined)

M 1

M 2

M- 1

A

5

5

3

Melomys (120)

B

5

4

3

Rattus (90)

C

5

3

Hydromys (5), Xeromys ( 1 )

D

4

4

3

Uromys (6), Melomys ( 1 80),
Pogonomys (6)

E

4

3

2

Conilurus (2)

F

3

3

3

Mus (3), Pseudomys (18),

Mesembriomys (2), Zyzomys (2),
Thetomys (2), Pogonomelomys (6)

G

3

3

3*

Gyomys (2)

H

3

3

2

Leggadina (5), Notomys (7),
( Leporillusf )

* arrangement different from Pattern F (cf. Plate 62 F a)
f after Jones (1922)

460

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 61

Right maxillae, showing alveolar patterns A, B, C, E, and G.
Pattern A

a, Melomys lutillus (Thomas) AM M6925

b, Melomys littoralis (Lonnberg)

c, Melomys australius Thomas

Pattern B

a, Rattus norvegicus (Erxleben)

b, Rattus rattus Linnaeus

c, Rattus lutreolus (Gray)

d, Rattus assimilis (Gould)

e, Rattus leucopus (Gould)

f, Rattus villosissimus Waite QM J22613

g, Rattus culmorum (Thomas and Dollman)

h, Rattus conatus Thomas

Pattern C

a, Hydromys chrysogaster Jeffroy

b, Hydromys longmani Thomas QM J3784, Paratype

c, Xeromys myoides Thomas AM M6529

Pattern E

Conilurus albipes (Lichtenstein) QM J3348
Pattern G

Gyomys berneyi Troughton QM J14755

KNOX: MOLAR PATTERNS OF MURIDAE

Plate 61

MEMOIRS OF THE QUEENSLAND MUSEUM

Plate 62

Right maxillae, showing alveolar patterns D, F, and H.
Pattern D

a, Uromys caudimaculatus (Krefft) QM J22540

b, Uromys sherrini Thomas QM J3785 Paratype

c, Melomys cervinipes (Gould)

d, Melomys rufescens (Alston)

e, Melomys rubicola Thomas QM J20169

f, Melomys sp.

g, Pogonomys mollipilosus Peters and Doria

Pattern F

a, Mus musculus Linnaeus QM J2991

b, Pseudomys novaehollandiae (Waterhouse) QM J 1 7920

c, Pseudomys oralis Thomas

d, Pseudomys minnie Troughton QM J5944

e, Mesembriomys gouldii (Gray) QM J 16978

f, Zyzomys argurus (Thomas) QM J22401

g, Thetomys gracilicaudatus (Gould)

h, Pogonomelomys sevia (Tate and Archbold)

Pattern H

a, Leggadina delicatula (Gould) QM J 1 6468

b, Notomys filmeri Mack QM J 10009

KNOX: MOLAR PATTERNS OF MURIDAE

Plate 62

CONTENTS

Archer, M.

Revision of the marsupial genus Planigale Troughton (Dasyuridae)

Archer, M.

Phascolarctid origins and the potential of the selenodont molar in the evolution of
diprotodont marsupials
Bartholomai, Alan

The genus Wallabia Trouessart (Marsupialia: Macropodidae) in the Upper

Cainozoic deposits of Queensland

Archer, M. and Wade, Mary

Results of the Ray E. Lemley Expeditions, Part 1. The Allingham Formation and
a new Pliocene vertebrate fauna from northern Queensland
Davies, Valerie Todd

Dardurus , a new genus of amaurobiid spider from eastern Australia, with

descriptions of six new species

McKay, R. J.

The wolf spiders of Australia (Aranae: Lycosidae): 7. Two new species from

Victoria

McKay, R. J.

The wolf spiders of Australia (Aranae: Lycosidae): 8. Two new species inhabiting

salt lakes of Western Australia

Stephenson, W., Raphael, Y. I., and Cook, S. D.

The macrobenthos of Bramble Bay, Moreton Bay, Queensland
Kott, P. '

Introduction of the North Atlantic ascidian Molgula manhattensis (De Kay) to

two Australian river estuaries

Knox, Elizabeth

Upper molar alveolar patterns of some Muridae in Queensland and Papua New
Guinea

Page

341

367

373

379

399

413

417

425

449

457