ArticlePDF Available

How kangaroos swim

Article

How kangaroos swim

Abstract

During extensive flooding in eastern Australia in ]974, many kangaroos were marooned on islands. A]ong the Darling River and in the Bulloo Overflow, groups of up to 150 red. and grey kangaroos were sighted from the air. When frightened, they often took to the water and one grey kangaroo was seen to swim over 300 metres. While Troughton (1967) and Breeden and Breeden (1966) described kangaroos entering water to defend themselves when chased by dogs and Gould (1863) reported a 2-mile swim by a kangaroo during a hunt, there have been no studies of the details of the manner of swimming, or the capacity of kangaroos to swim over extended distances. To elucidate the swimming action of kangaroos, underwater photographs were taken of two male red kangaroos, Mega/eia ruta, in a swimming pool, They were 3 and 4 years of age and weighed 25 and 3] kg respectively. Both had been bred in captivity and previously had not swum. Cameras used were a Bo]ex ]6-mm in a perspex housing and a Nikonos 35-JTlm. When first submerged in water the kangaroos showed a diving reflex; they did not inhale water, and then maintained their heads and necks above the surface. Initially they attempted to hop, but after] 0-15 seconds they began stroking smoothly in an instinctive manner.
research reports g
FIGURE 1
Underwater photograph of swimming red kangaroo showing alternate hindlimb movement,
and movement in unison of ipsilateral fore and hind limbs.
How Kangaroos Swim
GEORGE R. WILSON*
During extensive flooding in eastern
Australia in ]974, many kangaroos
were marooned on islands. A]ong the
Darling River and in the Bulloo
Overflow, groups of up to 150 red. and
grey kangaroos were sighted from the
air. When frightened, they often took
to the water and one grey kangaroo
was seen to swim over 300 metres.
While Troughton (1967) and Breeden
and Breeden (1966) described
kangaroos entering water to defend
themselves when chased by dogs and
Gould (1863) reported a 2-mile swim
by a kangaroo during a hunt, there
have been no studies of the details of
the manner of swimming, or the
capacity of kangaroos to swim over
extended distances.
To elucidate the swimming action of
kangaroos, underwater photographs
were taken of two male red kangaroos,
Mega/eia ruta, in a swimming pool,
They were 3 and 4 years of age and
weighed 25 and 3] kg respectively.
Both had been bred in captivity and
previously had not swum. Cameras
used were a Bo]ex ]6-mm in a perspex
housing and a Nikonos 35-JTlm.
When first submerged in water the
kangaroos showed a diving reflex;
they did not inhale water, and then
maintained their heads and necks
above the surface. Initially they
attempted to hop, but after] 0-15
seconds they began stroking smoothly
in an instinctive manner.
The forelimbs and hindlimbs were
moved ipsilaterally. TIlis is similar to
pacing, which is an unusual gait on
land; the camel and the giraffe are the
only species to use it without training
(Muybridge 1899).
TIle forelimbs of the swimming
kangaroo alternately make a posterior
power stroke, with the digits extended
and the manus at its maximum size.
During the movement anteriorly the
manus is repositioned in a flexed
condition and close to the body.
Although hindlimbs are also moved
*Dept of the Environment and Conservatiun,
Canberra.
alternately, they appear to be more
directed to treading water for upright
stability and maintaining buoyancy ,
rather than contributing much to
forward motion. The need for buoyancy
would be greatest at the hindquarters
because on land this is the site of the
kangaroo's centre of gravity (Badoux,
]965).
The tail was flexed horizontally, to the
side of the rearward hindlimb (in
contrast to the usual vertical motion of
the kangaroo tail). Although some of
this swishing tail movement may be a
reaction to hindlimb movement, it
nevertheless appears to help to drive
the kangaroo forward by sweeping
from side to side while presenting an
inclined surface in the water. This
action is similar to that of aquatic
animals as described by Gray (1957),
Tricker (1966) and Tarasoff et al .
(1972).
The speed at which the red kangaroos
were swimming was estimated at 1m
sec-1 by timing them in a 12-m pool.
The frequency of their limb and tail
contraction was 2.1-2.6 Hz. TIlis
frequency agrees well with the hopping
frequencies reported for both red
kangaroos (Dawson and Taylor, 1973),
and the grey kangaroos (Stewart and
Setchell ]974). It would seem that
there is a preferred frequency of the
cycle of limb movement, whether it
be the simultaneous motion of both
legs in hopping or their alternate
movement in swimming.
Discussion
Kangaroos swim so well that it raises
questions of whether their ability is an
adaption to environment, or an
indication of phylogeny, or a
coincidence due to a morphology
amenable to the dynamics of swim-
ming.
Any animal swimming forward meets
drag or resistance which is a complex
function of shape, size and speed, and
to transport itself it must supply a
force, by propelling water backward,
equal to this drag. To this end, swim-
ming animals (like digging animals)
have short powerful limbs, in contrast
to the long fast-moving limbs of
cursorial animals (Hilderbrand 1960).
The kangaroo has short forearms and
long hindlimbs, of which the former
are used in swimming and digging
while the latter are essentially
cursorial in type and function. The
tail motion resembles the tail motions
of fish, and it has been established
that fish (as would be expected) are
more efficient swimmers in energy
temlS than are land animals paddling
at the surface (Schmidt-Nielson,
]972); the difference would be largely
due to the turbulence generated,
which in case of the ka~jgaroo tail
would be relatively low.
598 Search Vol. 5 No. 11-12, Nov-Dec 1974
N 5rlJtrOV:Jfl5UR51Tb51J~t51QBzi 5grlJ
ever 1 es I s Ile 01 th se. or 0-
I1!J5Nt5lw t~flgrn5v,omdf(1j~o t~'{:Joglc a nDues Wlllcn ass].';!ilie rea
k fl Od 0 flOUOITIJp05flBJ1U.fTU25f115d
angaroo In 'f.ater" e occaslOnion
. Bf115u; ] ':J 10 IU 0w
whIch they need t9rfWfo/,ar~~n requent
in an environnltfurvwrn~ rarely floods
and then only along the limited area
ofwit er courses. It is interesting to
note, however, that Euros, whose
normal habitat is mountain ranges,
were seen in Kinchega National Park
to be poor swimmers and flounder
awkwardly in water. Possibly the
kangaroo's ability to swim developed
as an adaptation to predators; there is
anecdotal evidence of the use to which
kangaroos will put water when
pursued.
The ability of macropods to swim
must have an effect upon migration
and dispersion, and its significance in
the formation of subspecies on islands
is a subject warranting further
attention. One kangaroo marked by
Bailey (197 I) was recorded on the
other side of the Darling River,
presumably having swum across;
Main (196 I), describing the distribu-
tion of macropods on offshore islands
in Western Australia, attributed their
presence to movement over land
bridges, not to swimming.
It seems unlikely that the 'pacing'
swimming action of kangaroos has
any phylogenetic significance such as
that attributed to human infants.
Under 4 months of age human infants
show a reflex swimming behaviour
which bears striking testimony to the
phylogenesis of man (McGraw, 1939).
ll1eir trotting action is comparable
to that of most other animals such as
the horse and the dog which Kolb
(I 962) described as the most apt of
domestic animal swimmers.
ll1e separate-hindlimbs action of
swimming kangaroos is reminiscent of
phalangerids from whose aboreal
progression Marshall (I 974) has
reviewed the development of the
kangaroo's cursorial specializations
and the evolution of hopping. Windsor
and Dagg (I 97 I) found that the walk
of the tree kangaroo dendrolagus was
the only gait of macropods in which
pairs of limbs were not used
synchronously, and represen ted a
return to aboreal habitats by a wallaby.
The separate hind-leg action of the
swimming red kangaroo may therefore
represent a reversion to earlier times,
and the phylogenesis of the animal.
Nevertheless, the only aquatic
Search Vo!. 5 No. 11-12, Nov-Dec 1974
a-··-
rsu is still the South America
opossum Chironectes minimus,
and although kangaroos are com-
petent swimmers they swim only
under unusual circumstances.
Acknowledgements
The National Parks and Wildlife
Service of NSW employed the author
during the period in which the investi-
gation was carried out.
Mr G. Steer photographed and edited
the film which was provided by Mr L.
Hudson. Mr K. Coles allowed us to
use his swimming pool. The kangaroos
were housed in facilities provided by
the Faculty of Veterinary Science,
University of Sydney and they were
attended by Mr G. Robertson and
Miss C. Coles from the National Parks
and Wildlife Service of NSW.
FIGURE 2
Below: Kangaroo's tail exhibits a wave-like
motion similar to that of a fish; right: a
later stage in the cycle of motion of the tail.
References
BADOUX, D.M (1965) Acta anat. 62418.
BAILEY, P.T. (1971) J. CSIRO Wildl. Res.
16 11.
BREEDEN, S. and BREEDEN, K. (1966)
The Life of the Kangaroo. Sydney: Angus
& Robertson.
DAWSON, T.F. and TAYLOR, C.R. (1973)
Nature 246 313.
GOULD, J. (1863) (1973) Kangaroos.
Melbourne: Macmillan.
GRAY, J. (1957) Scient. Am. 197 248.
HILDEBRAND, M. (1960) Scient. Am.
202 148.
KOLB, E. (1962) Lehrubuch der
Physiologie der Haustiere. Jena: Gustav
Fischer.
MAIN, A.R. (1961) J. Roy. Soc. W. Aust.
44 84.
MARSHALL. loG. (1974) Nature 248174.
MUYBRIDGE, E. (1899) Animals in
motion. London: Chapman &Hall 1957.
McG RAW, M.B. (1939) J. Pediatrics 15485.
SCHMIDT-NIELSEN, K. (1972) Science
177 222.
Continued
599
~... ,.
"'. ~:'''>••,
''110'- , !.
"~~ "
,.~,
FIGURE 3 .
Red kangaroo swimming past a diver with an underwater camera.
STEWART. I.C.F. and SETCHELL, P.J.
(1974) Search 5107.
TARASOFF. F.J.• BISAILLON. A.,
PIERARD. J. and WHITT, A.P. (1972) Can.
J. Zool. 50915.
TRICKER, R.A.R. (1966) The Science of
Movement. London: Mills &Boon Ltd.
TROUGHTON. E. (1967) Furred Animals
of Australia. Sydney: Halstead Press.
WINDSOR, D.E. and DAGG, A.1. (1971)
J. Zool .• Lond. 163 165.
Submitted 29 Mav. 1974
600 Search Vol. 5 No. 11-12, Nov-Dec 1974
Article
Full-text available
Background: Most animal studies of spinal cord injury are conducted in quadrupeds, usually rodents. It is unclear to what extent functional results from such studies can be translated to bipedal species such as humans because bipedal and quadrupedal locomotion involve very different patterns of spinal control of muscle coordination. Bipedalism requires upright trunk stability and coordinated postural muscle control; it has been suggested that peripheral sensory input is less important in humans than quadrupeds for recovery of locomotion following spinal injury. Methods: We used an Australian macropod marsupial, the tammar wallaby (Macropuseugenii), because tammars exhibit an upright trunk posture, human-like alternating hindlimb movement when swimming and bipedal over-ground locomotion. Regulation of their muscle movements is more similar to humans than quadrupeds. At different postnatal (P) days (P7–60) tammars received a complete mid-thoracic spinal cord transection. Morphological repair, as well as functional use of hind limbs, was studied up to the time of their pouch exit. Results: Growth of axons across the lesion restored supraspinal innervation in animals injured up to 3 weeks of age but not in animals injured after 6 weeks of age. At initial pouch exit (P180), the young injured at P7-21 were able to hop on their hind limbs similar to age-matched controls and to swim albeit with a different stroke. Those animals injured at P40-45 appeared to be incapable of normal use of hind limbs even while still in the pouch. Conclusions: Data indicate that the characteristic over-ground locomotion of tammars provides a model in which regrowth of supraspinal connections across the site of injury can be studied in a bipedal animal. Forelimb weight-bearing motion and peripheral sensory input appear not to compensate for lack of hindlimb control, as occurs in quadrupeds. Tammars may be a more appropriate model for studies of therapeutic interventions relevant to humans.
Article
Because most animals are able to survive much longer when in water than most humans and apes, it is likely that at some time during human phylogeny the-innate-ability to swim like animals has been lost. Paleontology and comparative zoology provide indications for selective pressures and adaptations explaining this phenomenon. Among these are (a) buoyancy having become reduced slightly due to a smaller lung air/body weight ratio, less air trapped in the body covering hairs and reduction in gastrointestinal gas, (b) the caudally directed nasal entrance cranially to which there is the heavy neurocranium, (c) the low position of the laryngeal entrance, (d) the ineffective propulsion by the human innate crawling and walking movements, and (e) the well developed cerebrum being present cranially to the airway entrance, being sensitive for anoxia and hypothermia, and causing behaviour adding to drowning risks.
  • Raw Mcg
McG RAW, M.B. (1939) J. Pediatrics 15485.
The Life of the Kangaroo
  • S Breeden
BREEDEN, S. and BREEDEN, K. (1966) The Life of the Kangaroo. Sydney: Angus
  • A R Main
MAIN, A.R. (1961) J. Roy. Soc. W. Aust. 44 84.
  • D Badoux
BADOUX, D.M (1965) Acta anat. 62418. BAILEY, P.T. (1971) J. CSIRO Wildl. Res.
  • J Gray
GRAY, J. (1957) Scient. Am. 197 248.
  • M Hildebrand
HILDEBRAND, M. (1960) Scient. Am.
Lehrubuch der Physiologie der Haustiere
  • E Kolb
KOLB, E. (1962) Lehrubuch der Physiologie der Haustiere. Jena: Gustav Fischer.
1899) Animals in motion
  • E Muybridge
MUYBRIDGE, E. (1899) Animals in motion. London: Chapman & Hall 1957. McG RAW, M.B. (1939) J. Pediatrics 15485.
  • K Schmidt-Nielsen
SCHMIDT-NIELSEN, K. (1972) Science 177 222. ...,. "'.~:'''> ••, ''110'-, !. "~~" ,.~, FIGURE 3. Red kangaroo swimming past a diver with an underwater camera.