Content uploaded by Lisa-ann Gershwin
Author content
All content in this area was uploaded by Lisa-ann Gershwin on May 26, 2016
Content may be subject to copyright.
RECORDS OF THE WESTERN AUSTRALIAN MUSEUM 010–019 (2014) DOI: 10.18195/issn.0312-3162.29(1).2014.010- 019
29
Two new species of box jellies (Cnidaria:
Cubozoa: Carybdeida) from the central
coast of Western Australia, both presumed
to cause Irukandji syndrome
Lisa-Ann Gershwin
CSIRO Marine and Atmospheric Research, Castray Esplanade, Hobart, Tasmania 7000, Australia.
Email: lisa-ann.gershwin@csiro.au
ABSTRACT – Irukandji jellies are of increasing interest as their stings are becoming more frequently
reported around the world. Previously only two species were known from Western Australia, namely
Carukia shinju Gershwin, 2005 and Malo maxima Gershwin, 2005, both from Broome. Two new species
believed to cause Irukandji syndrome have recently been found and are described herein. One, Malo
bella sp. nov., is from the Ningaloo Reef and Dampier Archipelago regions. It differs from its congeners
in its small size at maturity, its statolith shape, irregular warts on the perradial lappets, and a unique
combination of other traits outlined herein. This species is not associated with any particular stings, but
its phylogenetic affi nity would suggest that it may be highly toxic. The second species, Keesingia gigas
gen. et sp. nov., is from the Shark Bay and Ningaloo Reef regions. This enormous species is unique in
possessing key characters of three families, including crescentic phacellae and broadly winged pedalia
(Alatinidae) and deeply incised rhopalial niches and feathery diverticulations on the velarial canals
(Carukiidae and Tamoyidae). These two new species bring the total species known or believed to cause
Irukandji syndrome to at least 16. Research into the biology and ecology of these species should be
considered a high priority, in order to manage their potential impacts on public safety.
KEYWORDS: Malo bella, Keesingia gigas, Ningaloo Reef, Indian Ocean, taxonomy
http://www.zoobank.org/urn:lsid:zoobank.org:pub:453C214A-2544- 4DD8- BD79-5E87D0190AA D
INTRODUCTION
Irukandji jellies are becoming of increasing interest
to divers, swimmers, and occupational health and
safety managers, as their stings are becoming more
frequently reported around the world (Gershwin et
al. 2013). Irukandji syndrome is a constellation of
systemic symptoms with a delayed onset after a sting
from a jelly. Symptoms typically include severe lower
back pain, nausea and vomiting, diffi culty breathing,
profuse sweating, cramps and spasms, and a feeling of
impending doom (Williamson et al. 1996). Some cases
also have severe hypertension (high blood pressure),
which can be life threatening. Two confi rmed fatalities
from Irukandji-related hypertension occurred in
2002 (Fenner and Hadok 2002; Huynh et al. 2003).
It is generally believed that additional fatalities have
occurred, but have been masked as heart attack, stroke,
or drowning (Gershwin et al. 2009).
Irukandji are often associated with Queensland, as
that is where the syndrome was discovered and named,
and most of the research has occurred (Flecker 1952;
Barnes 1964; Southcott 1967; Gershwin et al. 2013).
However, the fi rst unequivocal report of symptoms
attributable to Irukandji syndrome in Australia
occurred in Onslow in 1927 and 1928 (Stenning 1928).
Much later, Macrokanis et al. (2004) demonstrated that
Irukandji syndrome in Broome is at least as serious a
problem as it is in Queensland. Two species of Irukandji
jellies were described from Western Australia in 2005,
namely Carukia shinju and Malo maxima, both from
the far northwest near the town of Broome (Gershwin
2005b).
Farther south, reports of Irukandji are sparse.
An unnamed species was identified from near the
Montebello Islands in the region of Dampier (Gershwin
2005a), but lack of adequately preserved material
prevented a full description and classifi cation. In the
Ningaloo Reef region, stings occur only occasionally,
with 0–2 per year being typical (P. Hannay, Exmouth
Hospital personal communication). These are not
generally linked to specimens or species. Similarly,
two Irukandji stings from unknown species have been
reported in the Perth region, one at Scarborough Beach
at a Surf Life Saving carnival in the 1980s and another
near Rottnest Island in 2003 (Gershwin et al. 2009).
TWO NEW BOX JELLIES FROM WESTERN AUSTRALIA 11
Identifi cation of these species is the fi rst step toward
understanding them and managing their stings.
It is therefore of potentially great importance to fi nd
another two new species of cubomedusae believed to
cause Irukandji syndrome in Western Australian coastal
waters. One, based on a specimen found near Exmouth
following a cluster of 23 confi rmed Irukandji stings
(Gershwin and Hannay 2014), is similar to undescribed
museum specimens from the Montebello Islands.
Despite some variation, a conservative approach is
taken herein and they are regarded as the same species,
described as Malo bella sp. nov. below.
The other is based on an enormous specimen taken
in a fi shing net near Shark Bay in late summer 2012.
This specimen poses a taxonomic puzzle, with some
characteristics of the family Alatinidae and other
characteristics more like the genus Morbakka in the
family Carukiidae. This curious species is associated
with systemic illness following a sting at Ningaloo Reef
to the north. This species is here described as Keesingia
gigas gen. et sp. nov.
Recently two methods of forecasting Irukandji jellies
occurrence have been developed. One applies to species
in the genus Alatina, which swarm on the 8th–10th
nights after the full moon in Australia or the 9th–12th
days after the full moon in Hawaii (Chiaverano et
al. 2013). The other forecasting system applies to the
diminutive Carukia barnesi in Queensland, which forms
nearshore aggregations when the along-shore winds
subside. The extent to which either of these new species
may be predictable is currently unknown.
The purpose of this paper is to describe two new
species of cubozoan jellies, presumed to cause Irukandji
syndrome. This now brings the total number of Irukandji
species described from Western Australia to four, with
nine from Australia, and at least 16 globally. The biology
and ecology of Irukandjis were recently reviewed
(Gershwin et al. 2013) highlighting our paucity of
knowledge on this remarkable group; these areas should
be considered high priority for future study and safety
management.
MATERIALS AND METHODS
All taxonomic observations and measurements were
made on preserved material unless otherwise noted.
Measurements were made with a ruler to the nearest mm
or with Max-Cal digital calipers to the nearest 0.01 mm.
Bell height (BH) was measured from the apex of the
bell to the velarial turnover. Diagonal bell width (DBW)
was measured across diagonal pedalia on a fl attened
specimen, at the height where the pedalium joins the
exumbrella of the bell. Interrhopalial width (IRW) was
measured between adjacent rhopalia, with the specimen
fl attened. Tentacle base width (TBW) was measured
at the uppermost part of the tentacle, immediately
below the pedalium. In opaque specimens, a search for
phacellae was made by making a small incision in the
upper corners of the bell, and then pulling back a small
amount of mesoglea to expose the fl oor of the stomach,
or by opening up the full length of the body wall to
expose the stomach, and then opening the stomach in
the same manner. In transparent specimens, absence of
phacellae was obvious. Nematocysts were examined and
measured with a Leica DMLB compound microscope
and Leica IM-50 Image Manager v. 1.20 for Windows;
all observations and photographs were made through a
40x objective (i.e. 400 x magnifi cation). Nematocysts
were identifi ed following the keys of Calder (1974),
Mariscal (1971), Williamson et al. (1996), and Gershwin
(2006).
Abbreviations used. Western Australia (WA); Western
Australian Museum, Perth (WAM); Museum and Art
Gallery of the Northern Territory (NTM); Western
Australian Department of Parks and Wildlife (DPaW);
Marine Information and Research Group, Perth,
Australia (MIRG Australia).
SYSTEMATIC RESULTS
Class Cubozoa Werner, 1973
Order Carybdeida Gegenbaur, 1856
(sensu Werner, 1984)
Family Alatinidae Gershwin, 2005
Genus Keesingia gen. nov.
http://www.zoobank.org/urn:lsid:zoobank.
org:act:A94C930A-4641-4CED-ABFF-DA3876A78A2A
TYPE SPECIES
Keesingia gigas sp. nov.
DIAGNOSIS
Alatinidae with a large, thick, warty bell; with deeply
incised frown-shaped rhopalial niches, divided below
at the midline; with well developed subumbrellar
mesenteries; with adradial gelatinous lappets.
REMARKS
The type species of the genus, Keesingia gigas,
is remarkable, in many ways resembling Morbakka
(Gershwin 2008) while also strongly resembling species
in the Alatinidae (Table 1). For example, the rhopalial
niche most resembles that of Morbakka or Tamoya with
the deep cavity and dumbbell-shaped ostium (Bigelow
1938: text-fi g. 20; Gershwin 2008: fi gs. 2D, E), and yet
it also has a superfi cially split lower scale (Figure 2A
herein), which is not found in any other cubozoan but
suggests an affi nity with Alatina, in which the lower
scale is fully split, creating a T-shaped ostium (Bigelow
1938: text-fi g. 11; Gershwin 2005c: fi g. 2B; Bentlage
2010: fi g. 1C). The ostium of the holotype of Alatina
grandis (Agassiz and Mayer, 1902) fi gured by Bentlage
(2010) is of the ‘normal’ Alatina type, that is to say, it is
fully and prominently T-shaped, whereas the ostium of
12 L. GERSHWIN
Keesingia is shallow and more comparable to a smile
with a hairlip or a very fl at Q, where the split is so minor
as to not disturb the integrity of the deep cavity. Another
feature suggesting an alatinid affi nity is the lack of
rhopalial horns, whereas these structures are present
in Morbakka and other carukiids (Gershwin 2005a;
Gershwin 2008).
The warty bell would appear to place it fi rmly with
Morbakka (Gershwin 2008), while its tall, barrel-
like shape is more characteristic of Alatina (Figure
1; Gershwin 2005c). So too, its crescentic phacellae
place it fi rmly within the Alatinidae (Gershwin 2005c).
However, while the pedalia are clearly of the large-
winged Alatina type, the velarial canals are clearly of
the feathered Morbakka type, but the perradial lappet, so
prominent in the latter taxon, is lacking.
The mesenteries are worthy of discussion. Large
species such as Morbakka and Tam o ya have well
developed mesenteries developed along the sides of the
stomach and connecting to the subumbrella (Gershwin
2008; Collins et al. 2011). It appears that they may
help keep the stomach centred and the mouth in place.
Keesingia also has well developed mesenteries (Figure
2D); however, like the Alatinids, the stomach is very
small. Therefore, the stomach is not stabilised by the
mesenteries and they seem rather without purpose.
So too, the adradial gelatinous lappets (Figure 2C) are
intriguing, as they had never been reported in any other
species until they were formally described as ’velarial
lappets’ recently (Lewis et al. 2013). In general form,
velarial lappets resemble the more familiar perradial
lappets, which are essentially just triangular thickenings
of the bell tissue extending out along the velarium,
except that perradial lappets occur over the perradii,
and velarial lappets occur between the perradii and
interradii. Moreover, perradial lappets are conspicuous
in the Carukiidae and Tamoyidae, whereas velarial
lappets were found in the Alatinidae. Curiously, velarial
lappets are also found in Keesingia, but they do not
extend down onto the velarium as they do in Alatina
alata.
Based on overall morphology, Keesingia seems to
have more of the diagnostic traits of the Alatinidae than
the Carukiidae. Characters such as the broadly winged
pedalia, the crescentic phacellae, the divided lower
rhopalial covering scale, the lack of rhopalial horns
and perradial lappets, and the very small stomach are
all consistent family features of Alatinidae, whereas
characters such as the thick warty bell, mesenteries,
broad frenula, and diverticulated velarial canals are
found in some species of Carukiidae and not others and
are also found in the Tamoyidae (Table 1).
However, recent 16S barcoding of the Keesingia
holotype confirmed the specimen’s closer affinities
with the Carukiidae (O. Berry, CSIRO, unpublished
data). The morphological and molecular confusion
between these large species suggests that the suite of
morphological characters currently used for species
Morbakka (Carukiidae) Tamoy a (Tamoyidae) A. grandis (Alatinidae) Keesingia
Bell Large, thick, warty Large, thick, wart y Large, thick, not wart y Large, thick, wart y
Rhopalial niche Dumbbell-shaped,
lower undivided
Dumbbell-shaped,
lower undivided
T-shaped with lower
strongly divided
Dumbbell-shaped,
lower divided
Rhopalial horns Present Absent Absent Absent
Pedalia Narrow -winged Narrow-winged Broad-winged Broad-winged
Pedalial canal Spike at bend,
quadrate, fl ared at end
Spike at bend,
quadrate, fl ared at end
Without spike, fl at, not
fl ared
Without spike, fl at, not
fl ared
Stomach Voluminous Voluminous Small, short Small, short
Phacellae Absent Vertical Broadly crescentic Broadly crescentic
Mesenteries Robust, fl ap-like in
upper half, cord-like in
lower half
Robust, fl ap-like in
upper half, cord-like in
lower half
[not described] Flap-like in upper half,
absent in lower half
Vel a r i a l canals Feathe r y,
diverticulated
Dendritic,
diverticulated
Short, tree-like Feathery,
diverticulated
Perradial lappets Present Absent Absent Absent
Frenula Single, hollow Double Well developed Broad, hollow
TABLE 1 Comparison of characters of large-bodied species in the Carybdeida.
TWO NEW BOX JELLIES FROM WESTERN AUSTRALIA 13
differentiation in these Cubozoa may be inadequate to
properly answer questions of species delineation.
The large specimens photographed at Ningaloo Reef
in 1987 and 2013 appear to be attributable to Keesingia
(Figure 3), as does another specimen photographed
there in the 1990s (Marsh and Slack-Smith 2010: p. 53).
The sizes, bell shape, pedalia shape, and conspicuous
orange wartiness are similar, and the gonads of the
smaller specimen appear to be immature, suggesting
that the species can grow much larger. Curiously, neither
the collected specimen nor any of the photographed
specimens appear to have tentacles; whether it is a
feature of the species that they are so fi ne as to be
overlooked, or that they are genuinely lacking, or that it
is mere coincidence that they have been broken off in all
three, is unclear.
A swarm of ~10 –15cm tall cubomedusae videotaped
at Rowley Shoals on 9 September 2013 may also be
attributable to this species (see Gershwin and Hannay
2014, fi gure 2). These medusae are tall like alatinids, but
with robust, squared off bells like Morbakka and Malo
in the Carukiidae. These are the very characters that one
would expect for Keesingia of any size.
ETYMOLOGY
The genus name, Keesingia, is named to honour
Dr John Keesing of CSIRO Marine and Atmospheric
Research, who not only provided the holotype for this
species, but has been of great assistance and inspiration
to the author for nearly 20 years.
Keesingia gigas sp. nov.
Figures 1–3
http://www.zoobank.org/urn:lsid:zoobank.
org:act:801E03E8-634D-41C1-B72B-C4D33447E84A
MATERIAL EXAMINED
Holotype
WAM Z27970, 5 nm west of Cape Inscription,
Dirk Hartog Island, Shark Bay, Western Australia,
25°36′28″S, 112°51′54″E, J. Keesing, 7 May 2012, hand
held scoop net; immature specimen, 19 cm BH, 97.65
mm DBW, 49.44 mm IRW, 4.39 mm TBW.
OTHER MATERIAL EXAMINED
Photograph of specimen, Black Rock, Ningaloo Reef,
Western Australia, P. Baker (WAM), 31 March 1987;
approx. 50 cm BH; caused systemic sting reaction.
Photograph of specimen, Ningaloo Reef, Western
Australia, G. Taylor (physician in Busselton); did not
cause systemic reaction. Photograph of specimen,
Ningaloo Reef near Coral Bay, Western Australia, J.
Totterdell (MIRG Australia), 12 April 2013; approx. 20
cm BH, 8–9 cm wide; recorded during dense bloom of
Crambione mastigophora. No specimens retained.
DIAGNOSIS
Alatinidae with a large, thick, warty bell; with deeply
incised frown-shaped rhopalial niches, divided below
at the midline; with well developed subumbrellar
mesenteries; with adradial gelatinous lappets.
DESCRIPTION
Holotype (from preserved specimen and photographs of
live specimens)
Bell very tall and narrow (Figure 1), with rigid
jelly about 9 mm thick around middle portion of bell.
Exumbrellar surface fairly densely and evenly covered
with thousands of small gelatinous warts over most of
the body except for inside the corner furrows; the warts
do not appear to be studded with nematocysts, nor do
they appear to be abraded. Corner furrows well defi ned,
moderately deep, running the full length of the bell from
just below the apex to the top of the pedalia.
Pedalia short, with broad inner and outer wings, each
more than the width of the pedalial canal; wings warty
and fragile in life, largely abraded away in the preserved
specimen. Pedalial canal fl at, simple, about 7.5 mm wide
proximally, tapering to half its width at the tip. Pedalial
canal bend without a thorn or other adornment; fl aring
only slightly at distal end, and also fl ared slightly about
2/5 of the way toward distal end. Tentacles missing in this
specimen, presumably lost during netting.
Rhopaliar niches very deeply incised into the bell.
Ostium strongly dumbbell-shaped with a shallow vertical
split at the midline of the lower covering scale (Figure
2A). Rhopalia uninter pretable due to degradation of the
specimen. Rhopaliar horns lacking. Rhopaliar windows
nondescript. Nerve cord pronounced, running from
behind rhopalium down to base of pedalium, back up to
rhopalium, and so forth.
Velarium with three gelatinous ‘lappets’ on the adradii
of the exumbrellar side in each octant extending down
onto the proximal edge of the velarium, as if overhanging
it; these lappets are sculpted out of the mesoglea but are
not connected with velarial canals (Figure 2C). On the
subumbrellar side, approximately fi ve canal roots extend
onto velarium, all dendritic with fi rst and second order
diverticula, branching irregularly into about 6–9 canals
reaching margin, all heavily diverticulated and appearing
almost feather-like (Figure 2B). Perradial lappets lacking.
A broad, hollow frenulum marks each perradius.
Phacellae broadly crescentic in the stomach corners;
each with dozens of long, fine cirri, some distally
bifurcated. Stomach small, short. Mesenteries well
developed in upper fourth of subumbrellar interradii,
about twice the length of the lips (Figure 2D). Lips folded,
with smooth margins, only extending to about 1/8 the
length of the subumbrellar cavity (Figure 2D).
Gonads: undeveloped in this specimen.
Colour: bell is transparent and colourless, somewhat
14 L. GERSHWIN
FIGURE 1 Keesingia gigas sp. nov., freshly caught holotype specimen; note orange warts densely scattered over body.
FIGURE 2 Keesingia gigas sp. nov., holotype. A. rhopaliar niche ostium; partially divided lower covering scale is indicated
by arrow. B. velarial canals; note diverticula. C. adradial gelatinous lappets, indicated by arrows. D. mesenteries,
indicated by black arrow; mouth is visible to the upper left (white arrow pointing to one lip).
TWO NEW BOX JELLIES FROM WESTERN AUSTRALIA 15
cloudy in appearance, with orange-coloured exumbrellar
warts in life. Mesenteries and pedalial canals whitish.
Nematocysts: could not be found on the bell warts or
pedalia.
TYPE LOCALITY
Dirk Hartog Island, Shark Bay, Western Australia,
25°36′28″S, 112°51′54″E.
DISTRIBUTION
Specimen confi rmation exists only from the type
locality. However, the photographs of large alatinids
with orange warts at Ningaloo Reef in 1987 and in
2013 also appear to be attributable to this species,
and a swarm of cubomedusae at Rowley Shoals on 9
September 2013 may be the young of this species (see
Gershwin and Hannay 2014, Figure 2).
ETYMOLOGY
The species name, gigas, is in reference to the
enormous size of this species.
ASSOCIATIONS
Three juvenile leatherjacket fish (family
Monacanthidae) were captured with this specimen,
living inside the subumbrellar cavity. While the
tendency of young leatherjackets to shelter with
medusae has been previously reported (Hutchins 1994),
this is the fi rst report of fi sh sheltering in an Ir ukandji
species. Whether they were also preying upon it is
unknown, as the specimen was thought to be already
dead when captured.
STING REMARKS
Two cases of stinging by Keesingia (as Tamoy a
gargantua) were reported by Williamson et al. (1996:
p. 242, and pl. 9.6A, B). One produced severe Irukandji
syndrome, whilst the other caused only local and groin
pain. These and two more cases of stinging by Keesingia
(as Alatina sp.) were reported by Marsh and Slack-Smith
(2010: pp. 52–55). Of these, one from Exmouth in 1978
was consistent with Irukandji syndrome, whilst the other
from Ningaloo in 1995 involved only intense local and
lymph pain.
FIGURE 3 Keesingia gigas sp. nov., from Ningaloo Reef A, B. lateral view (A) and subumbrellar view (B) in bloom of
Crambione mastigophora (sea tomatoes). (Image courtesy John Totterdell/MIRG Australia.) C. lateral view.
16 L. GERSHWIN
Family Carukiidae Bentlage, Cartwright,
Yanagihara, Lewis, Richards and Collins, 2010
Genus Malo Gershwin, 2005
Malo bella sp. nov.
Figure 4
http://www.zoobank.org/urn:lsid:zoobank.
org:act:A4DE4BCD-C9D5-4683-901A-AABD532650C6
“Dampier Irukandji”: Gershwin 2005a: Tables 2.1–2.12,
98, 103, 107, 119, 195, 201, Plate 4.7; Gershwin
2005b: 3, 20; Gershwin 2007: 64, 65, Table 1.
Malo n. sp. A “Dampier Irukandji”: Gershwin 2006: 10,
Table 1, Plate 19 [note that the specimen noted
“holotype” is now paratype NTM C14617].
“Undescribed species of Malo is known from the islands
off Exmouth”: Gershwin et al. 2013: 40.
MATERIAL EXAMINED
Holotype
WAM Z27971, approximately 300 m east of the
sandy beach in front of Harold E. Holt Memorial
Communications Base, Exmouth Gulf (estimated
GPS coordinates from Google Earth: 21°53′28.37″S,
114°08′57.47″E), R. De Jong (DPaW Parks and Visitors
Services Coordinator), 9 July 2013, mid afternoon, 1–2
m below surface in 4 –5 m total water depth; gravid
female, 20.04 mm BH, 18.50 mm DBW, 10.96 mm IRW,
1.43 mm T BW.
Paratypes
NTM C14617, Trimoville Island, Western Australia,
H. Larson, 21 April 1983, by nightlight at anchorage, on
surface over 5 m, fl at sandy bottom (20°24′S 115°34′E);
19.11 mm BH, 17.20 mm DBW, 8.35 mm IRW, 1.05 mm
TBW with halo-like bands, male. NTM C5143, same
data as C14617; 3 specimens, all mature but poorly
preserved.
DIAGNOSIS
Malo with a very small body size at maturity; with
long, well developed subumbrellar mesenteries; with
one tri-branched velarial canal centrally located in each
octant; perradial lappets broadly rounded, with irregular
rows of nematocyst warts; rhopalial horns short and
straight; statoliths almost perfectly hemispherical;
tentacular nematocyst bands with or without halo-like
shelves of tissue.
DESCRIPTION
Holotype
Body small, bell-shaped (Figure 4A), to about 19
mm BH, gravid at that size; with scattered nematocyst
freckles, particularly lower on the bell near the velarium
where they are slightly raised. Apex broadly rounded,
nearly hemispherical; shallowly warted, lacking
nematocysts. Bell of a robust consistency; evenly thick,
with exumbrellar sculpturing of strongly defi ned pillars
with shallow corner furrows, with rhopalial niches
somewhat raised from the body wall.
Pedalia (Figure 4E) nearly half the bell height,
unarmed, without nematocyst warts, freckles, or bars;
scalpel-shaped, with a narrow abaxial wing c. 1x canal
width, and broad adaxial wing c. 2x canal width; inner
keel not overhanging at tentacle insertion. Pedalial
canal simple, slightly diamond-shaped in cross section
throughout length; bend with slight upward-pointing
nub forming 90° corner; not fl ared at tentacle insertion.
Tentacles four, one per pedalium, with a straight
base, round in cross section; largely broken off in this
specimen, but appearing more or less evenly banded, not
of the halo form.
Gonads paired, leaf-like, nearly touching in midline
halfway down bell; connected along entire length
from stomach to velarium; narrower around and below
rhopalium, pendant below rhopalium.
Velarial canals comprising one root per octant
branching into three, one or two of which further
branch; without lateral diverticula; without warts
over the canals. Perradial lappets broadly rounded,
tongue-shaped, nearly reaching velarial edge; with an
irregular row of warts on each side (Figure 4C), without
diverticula. Frenulum comprising a single simple
sheet, lacking gelatinous buttresses, reaching about 3/4
the distance toward velarial margin, disappearing at
rhopalium.
Rhopalial niche ostium frown-shaped, with a single
well developed covering scale each above and below,
broadly concave in shape (Figure 4C). Rhopalial horns
relatively short, straight, evenly wide throughout their
length, nearly vertical in orientation (Figure 4C).
Rhopalial windows shallowly convex with a small
indentation at the point where the rhopalial stem
meets the body wall. Lower median lensed eye round,
considerably larger than the dome-shaped upper median
lensed eye. Lateral eye spots absent (Figure 4D).
Statolith nearly perfectly hemispherical (Figure 4D).
Phacellae lacking. Stomach with numerous parallel
folds, possibly indicating great expansion capabilities,
or possibly sites of increased surface area for enzyme
activity as would normally be found on the gastric
phacellae. Mesenteries (Figure 4B) extremely well
developed, fully webbed, with fl aps extending down
about ¾ the distance to the rhopalial niches, with a
barely perceptible cord reaching the rhopalium, running
through the center of a transparent strip of tissue.
Manubrium cruciform, extending to about half the bell
height; mouth with four short, broadly quadrate lips.
Colour: bell transparent and colourless, nematocyst
warts faintly reddish, tentacles white, gonads nearly
opaque whitish, eyes dark brown.
TWO NEW BOX JELLIES FROM WESTERN AUSTRALIA 17
VARIATION
The paratypes have halo-form bands similar to those
of some Malo kingi specimens, although the holotype
does not. The origin and function of these bands remains
unclear, as does the exact relationship of the banded and
unbanded forms.
The velarial canals of paratype NTM C14617
were earlier interpreted as only two per octant, the
adperradial widely bifurcated and the adinterradial
narrow and simple (as “Dampier Irukandji” in Gershwin
2005a). These are now reinterpreted as three. However,
the canals of this paratype have lateral diverticula
whereas those of the holotype do not. One of the
paratypes in lot NTM C5143 has a single main velarial
canal root in each octant, with four fi nger-like canals
issuing from it, each with numerous lateral diverticula,
whereas the holotype has a single root with three canals
and no diverticula. These differences are interpreted as
intraspecifi c variation rather than specifi c distinction;
similar canal variation is found in the confamilial
Morbakka (Gershwin 2008).
The perradial lappets of paratype NTM C14617 are
broadly rounded with large diverticula, and with a single
large wart, whereas the holotype has an irregular row of
warts on each side and lacks diverticulating canals.
ETYMOLOGY
The species name, bella, is in triple reference to its
small bell-like shape, its beauty, and the Montebello
Islands in the Pilbara region of Western Australia, where
the species was fi rst found. Noun in apposition.
TYPE LOCALITY
Approximately 300 m east of the sandy beach in
front of Harold E. Holt Memorial Communications
Base (estimated GPS coordinates from Google Earth:
21°53′28.37″S, 114°08′57.47″E), Exmouth Gulf, Western
Australia.
FIGURE 4 Malo bella sp. nov., holotype. A. habitus. B. subumbrellar view of mesenteries (indicated by arrows) and manu-
brium. C. rhopaliar niche (rhopaliar horns are indicated by arrows); note also perradial lappets with irregular
nematocyst warts in lower centre of image. D. rhopalium (above) and statolith (below), dissected away from
specimen and rotated such that left side of image would be ‘down’ in life; note absence of lateral eye spots.
E. pedalium.
18 L. GERSHWIN
ECOLOGY
Very little is known about the ecology of this species;
however, the following details were noted at the time the
holotype was collected:
• The specimen was moving/swimming in a northerly
direction with outgoing tide past a coral bommie;
• Water temperature was reliably estimated at 19–21°;
• Many other plankton reported in the water at the same
time including several individuals of another jelly that
had same bell shape and size as the Malo specimen,
four tentacles approx. 30– 40 cm long and a very
vibrant pink purple band around the base of bell.
DISTRIBUTION
Presently only confi rmed from Exmouth Gulf and
Trimoville Island in the Montebello Islands of the
Pilbara region, WA. Photographs and video of Malo
specimens taken along Ningaloo Reef and Coral Bay
in 2013 appear to be attributable to this species, but are
unverifi ed.
NEMATOCYSTS AND STING POTENTIAL
The tentacular nematocysts of paratype NTM C14617
were figured by Gershwin (2006: pl. 19), with the
following features: rice-shaped sub-ovate microbasic
p-mastigophores: 31.67–40.47 x 14.01–16.50 μm (n = 19).
The cnidome of the holotype was not studied.
The sting potential of this species is not yet known;
however, because of its phylogenetic relationship to
M. kingi, which is apparently lethal (Gershwin 2007),
and M. maxima, which also causes severe Irukandji
syndrome (Gershwin 2005b; Li et al. 2011), this species
should be regarded as potentially dangerous.
REMARKS
Malo bella is the smallest species yet described in the
genus. It is most similar overall to M. kingi (Table 2), in
that both have a warty domed body, a 90° pedalial canal
bend, short broad rhopaliar horns, and a single palmate
velarial canal root. However, the two species differ in
several structural features. First, M. kingi has warts over
the velarial canals, whereas M. bella does not. Second,
the pedalial wings are broader relative to the pedalial
canal in M. kingi than in M. bella, and some specimens
of the former have nematocyst freckles, whereas M.
bella does not. Third, the statolith of M. kingi is globular
with a prominent apical hook, whereas that of M. bella
is almost perfectly hemispherical. Finally, M. bella is
smaller at maturity (20 mm): specimens of M. kingi at
the same size are immature.
The halo-form tentacle armament is worthy of
discussion. Both M. kingi and M. bella have it, whereas
this feature is not known in other species. Whether
this represents ontogenetic change or unelucidated
phylogenetic distinction or mere individual variation is
not clear.
The asymmetrical occurrence of the perradial lappet
nematocyst warts are curious. In specimen NTM
C14617, these warts occur on the animal’s left side of
the lappets only, though the warts in the holotype occur
asymmetrically on both sides; in the poorly preserved
other paratypes, the velarium could not be examined
adequately to determine their wart arrangement. Velarial
warts are typically either symmetrical on the lappets
(e.g. Gerongia rifkinae and Malo kingi, in which the
warts occur in rows on both lappets), symmetrical on the
velarial canals (e.g. Carukia barnesi, in which the warts
occur in a patterned manner on the canals only), or
completely random (e.g. the Morbakka species complex,
Species Body Pedalia Rhopalia Velarium
Malo maxima ≈ 50 mm BH, warty,
with fl attened apex;
moderate MS
3 nematocyst freckles;
PCB with short blunt
oblique projection
RH short, thick, straight;
statolith globular with
basal indentation
VC 1 root, 4 canals, 4-6
tips; PL with one wart
Malo kingi ≈ 30 mm BH, warty
apically, with domed
apex; large MS
With or without
nematocysts; PCB 90°
RH short, broad, curved
inward; statolith globular
with basal indentation
and apical tooth
VC 1 root, palmate, 4-5
tips; PL with 2 rows of
3-4 patches
Malo fi lipina ≈ 40 mm BH, warty,
large MS
1 row of round warts;
PCB with ‘spike’
RH short, broad, blunt;
statoliths unknown
VC palmate, 3-4 tips; PL
with 2 rows of 2-4 warts
Malo bella sp. nov. ≈ 20mm BH, wart y, with
domed apex; large MS
Lacking nematocysts;
PCB 90°
RH short, thick, straight;
statolith nearly perfectly
hemispherical
VC 1 root, 3-4 canals,
ca 6 tips; PL with 2
irregular rows of warts
TABLE 2 Comparison of main diagnostic characters of species in the genus Malo. Abbreviations: Pedalial canal bend
(PCB); rhopalial horns (RH); velarial canals (VC); perradial lappets (PL); mesenteries (MS).
TWO NEW BOX JELLIES FROM WESTERN AUSTRALIA 19
in which the warts are scattered among lappets, canals,
and intermediate tissues). In this current case, the
arrangement of warts is inconsistent with those known
thus far.
ACKNOWLEDGMENTS
Spectacular photos and vivid specimen and sting
descriptions were kindly provided by Pat Baker (WAM)
and John Totterdell (MIRG Australia) – thank you
both! Sincere thanks also to Peter Barnes, Ray De
Jong, and Arvid Hogstrom of DPaW for access to the
Malo holotype, and to John Keesing for access to the
Keesingia holotype. Sincere acknowledgments also
to Drs Joanna Strzelecki and Tim Ryan of CSIRO for
lab assistance when desperately needed. And a hearty
thanks to Jane Fromont of WAM for encouraging and
facilitating this paper. This paper benefi ted from the
comments of two reviewers. Part of this research was
funded by Environment Australia/Australian Biological
Resources Study (ABRS grants 207-63 and 208-82).
REFERENCES
Agassiz, A. and Mayer, A.G. (1902). Reports of the scientifi c
research expedition to the tropical Pacifi c. U.S. Fish Comm.
St. Albatross, 1899–1900. III. The Medusae. Memoirs of the
Museum of Comparative Zoology, Harvard 26(3): 139–176,
14 pla t e s.
Barnes, J.H. (1964). Cause and effect in Irukandji stingings.
Medical Journal of Australia 1 (51st Year)(24): 897–904.
Bentlage, B. (2010). Carybdea alata auct. (Cubozoa): rediscovery
of the Alatina grandis type. Zootaxa 2713: 52–54.
Bigelow, H.B. (1938). Plankton of the Bermuda Oceanographic
Expeditions. VIII. Medusae taken during the years 1929 and
1930. Zoologica, N.Y. 23 (part 2)(5–9): 99–189.
Calde r, D. R. 1974. Nematocysts of the coronate scyphomedusa,
Linuche unguiculata, with a brief reexamination of
scyphozoan nematocyst classifi cation. Chesapeake Science
15: 170 –173.
Chiaverano, L.M., Holland, B.S., Crow, G.L., Blair, L. and
Yanagihara, A.A. (2013). Long-term f luctuations in
circalunar beach aggregations of the box jellyfi sh Alatina
moseri in Hawaii, with links to environmental variability.
PLoS ONE 8(10): e77039.
Collins, A.G., Bentlage, B., Gillan, W.B., Lynn, T.H., Morandini,
A.C. and Marques, A.C. (2011). Naming the Bonaire banded
box jelly, Tamoya ohboya, n. sp. (Cnidaria: Cubozoa:
Carybdeida: Tamoyidae). Zootaxa 2753: 53–68.
Fenner, P.J. and Hadok, J.C. (2002). Fatal envenomation by
jellyfi sh causing Irukandji syndrome. Medical Journal of
Australia 177( 7 ): 362–363.
Flecker, H. (1952). Irukandji sting to North Queensland bathers
without production of weals but with severe general
symptoms. Medical Journal of Australia July 19: 89–91.
Gershwin, L. (2005a). Taxonomy and Phylogeny of Australian
Cubozoa. Unpublished PhD thesis, School of Marine
Biology and Aquaculture. James C ook University:
Tow ns v i l l e. 221 pp., 49 plates.
Gershwin, L. (2005b). Two new species of jellyfi shes (Cnidaria:
Cubozoa: Carybdeida) from tropical Western Australia,
pres u med to cause Irukandji Syndrome. Zootaxa 1084: 1–30.
Gershwin, L. (2005c). Carybdea alata auct. and Manokia
stiasnyi, reclassifi cation to a new family with descripti on
of a new genus and two new species. Memoirs of the
Queensland Museum 51(2): 501–523.
Gershwin, L. (2006). Nematocysts of the Cubozoa. Zootaxa
1232: 1–57.
Gershwin, L. (2007) . Malo kingi: A new species of Irukandji
jellyfi sh (Cnidaria: Cubozoa: Carybdeida), possibly lethal to
humans. Zootaxa 1659: 55– 68.
Gershwin, L. (2008). Morbakka fenneri: a new genus and
species of I rukandji jellyfi sh (Cnidaria: Cubozoa). Memoirs
of the Queensland Museum 54(1): 23–33.
Gershwin, L., De Nardi, M., Fenner, P.J. and Winkel, K.D.
(2009). Marine stingers: Review of an under-recognized
global coastal manage ment issue. Journal of Coastal
Management 38(1): 22–41.
Gershwin, L. and Hannay, K.D. (2014). An anomalous cluster
of Irukandji jelly stings (Cnidaria : Cubozoa: Carybdeida) at
Ningaloo Reef. Records of the Western Australian Museum
29(1): 78–81.
Gershwin, L., Richardson, A.J., Winkel, K.D., Fenner, P.J.,
Lippmann, J., Hore, R., Avila-Soria, G., Brewer, D., Kloser,
R., Steven, A. and Condie, S. (2013). Biology and ecology of
Irukandji jellyfi sh (Cnidaria: Cubozoa). Advances in Marine
Biology 66: 1–85.
Hutchins, B. (1994). Jellyfi sh and the little leatherjacket. Geo
Australia 16(1 (Jan / Feb)): 66–75.
Huynh, T.T., Seymour, J., Pereira, P., Mulcahy, R., Cullen, P.,
Carrette, T. and Little, M. (2003). Severity of Irukandji
syndrome and nematocyst identification from skin
scrapings. Medical Journal of Australia 178 (1): 38–41.
Lewis, C., Bentlage, B., Yanagihara, A., Gillan, W., van
Blerk , J., Keil, D.P., Bely, A.E. and Collins, A.G. (2013).
Redescr iption of Alatina alata (Reynaud, 1830) (Cnidaria:
Cubozoa) from Bonaire, Dutch Caribbean. Zootaxa 3737(4):
473– 487.
Li, R., Wright, C.E., Winkel, K.D., Gershwin, L.A. and Angus,
J.A. (2011). The pharmacology of Malo maxima jellyfi sh
venom extract in isolated cardiovascular tissues: A probable
cause of the Ir ukandji syndrome in Western Australia.
Tox icol og y Let t e rs 201: 221–229.
Macrokanis, C.J., Hall, N.L. and Mein, J.K. (2004). Irukandji
syndrome in northern Western Australia: an emerging
health problem. Medical Journal of Australia 181(11/12):
699–702.
Mariscal, R.N. (1971). Effect of a disulfi de reducing agent
on the nematocyst capsules from some coelentera tes,
with an illustrated key to nematocyst classifi cation. In:
Lenhoff, H.M., Muscatine L. and Davis L. V., Experimental
Coelenterate Biology: 157–168. University of Hawaii Press:
Honolulu.
Marsh, L. and Slack-Smith, S. (2010). Field Guide to Sea
Stingers. Western Australian Museum: Welshpool, Western
Australia.
Southcott, R.V. (1967). Revision of some Carybdeidae
(Scyphozoa: Cubomedusae), including a description of
the jellyfish responsible for the "Irukandji syndrome".
A ustralian Journal of Zoology 15(3): 651–671.
Stenning, A.E. (1928). Poisoning by Trachymedusae. Medical
Journal of Australia 1(18 (May 5, 1928)): 568–569.
Williamson, J.A., Fenner, P.J., Burnett, J.W. and Rif kin, J.
(eds) (1996). Venomous and Poisonous Marine Animals: A
Medical and Biological Handbook. NSW University Press:
Sydney.
MANUSCRIPT RECEIVED 1 NOVEMBER 2013; ACCEPTED 25 MARCH 2014.
