110 JULY 1992

ISSN 0819-6826

POSTAL ADDRESS: G.P.O. BOX 68A, HOBART, TASMANIA 7001
EDITOR: R.J. TAYLOR

THE WINTER DIETS OF FALLOW DEER (DAMA DAMA)
AND FORESTER KANGAROOS (MACROPUS GIGANTEUS)
IN THE MIDLANDS OF TASMANIA

by Anne Duncan

Forestry Commission, 199 Macquarie Street, Hobart, TasmaniajfOOl

INTRODUCTION Jj

Deer were first introduced into Australia in 1812 and the most successful
introduction was of fallow deer (Damn dama) into Tasmania in 1836 (Bentley
1978). The Forester kangaroo (Macropus giganteus tasmaniensis) is Tasmania's
largest native herbivore. It is thought that such grazing herbivores evolved to fill
a niche which became available when large areas of grasslands developed in the
dry Miocene (Russell 1974). Eutherian ruminants like the fallow deer filled this
grazing niche everywhere except Australia, where it was the marsupial
Macropodidae which evolved a similar series of adaptations (Moir, Somers and
Waring 1956). This paper details a comparison of the diet of the Forester
kangaroo and fallow deer in an area in the midlands where these two species now
both occur. The composition of their diets, their degree of selectivity and the
extent of overlap between the two species was studied.

METHODS

The study was carried out on "Beaufront", a pastoral property near Ross in the
midlands of Tasmania, from June to September 1987. The study site covered an
area of introduced pasture and a dry sclerophyll woodland community. The dry
sclerophyll community had been affected by grazing and burning and was
dominated by Eucalyptus viminalis with an upper shrub layer of species such as

Each author is responsible for the opinions and facts expressed in his or her article_ EDITOR

2

The Tasmanian Naturalist July 1992

Acacia mearnsiiand Banksia marginata over aground layer dominated by Lomandra
longifolia. The most common grasses were wallaby grass (Danthonia spp.) and
tussock grasses (Poa spp.). The pasture area consisted mainly of introduced
grasses (e.g. Holcus lanatus, Vulpia spp.) and clover (Trifolium subterraneum) with
some native wallaby grass.

Forester kangaroos and fallow deer were observed feeding in both pasture
and woodland and appeared to use them in a similar way, that is resting in
sheltered places in the woodland during the day and feeding in the open pasture
at dawn and dusk.

The diets were studied using faecal analysis. This technique is based on the
fact that plant epidermis' exhibit characteristics such as presence of silica bodies
and hairs and the size, shape and arrangement of cells and stomates, which can
be used to identify plant fragments in the faeces of animals. In order to do this
a set of reference slides was prepared from all the possible food plants in the
study area. Faeces were collected from an area of approximately one square
kilometre, including both the pasture and the woodland, every six weeks from
June to September 1987. The faeces collected were bulked each time and
prepared and analysed as set out in Duncan (1987). The diet was estimated by
recording the percent occurrence of plant fragment types in 300 fields of view for
each herbivore at each collection period.

Food preferences were investigated by comparing the percent occurrence of
plants in the faeces with an estimate of their availability on the study area.
Availability was assessed by estimating the percent cover of plant species in
quadrats laid out along transects which covered the variation in vegetation of the
study area. Transects were laid out in both the woodland and the pasture.

COMPOSITION OF THE DIETS

The mean percent occurrences of different plant types in the faeces of
kangaroos and deer are shown in Table 1. Grasses such as Holcus lanatus , Vulpia
spp., Danthonia spp. and Poa spp. made up a major component of the diet of both
species. However, they consistently formed a greater proportion of the
composition of kangaroo faeces than deer. Kangaroos also included a greater
variety of grass species in their diet and a greater proportion of native species
as opposed to introduced species. Conversely dicotyledons (e.g. low- fibre herbs
such as Trifolium spp., Viola spp. and Geranium spp. and high fibre "browse"
species such as Acacia spp., Banksia spp. and Leucopogon sp.) occurred
consistently more often in deer faeces than kangaroo faeces. Non-grass
monocotyledons such as Lomandra did not occur frequently in the faeces of either
species.

Herbivores ca n be classi fied as grazers or browsers (Hansen el al. 1985), where

The Tasmanian Naturalist July 1992

3

Table 1 Mean percent occurrence (in 300 fields of view) of plant types in
the faeces of kangaroos and deer recorded over three collection periods
from June to September 1987. Values are means ± standard deviations.

Non-grass

Grasses Browse Herbs .. t Bark Bracken Moss

Monocots

Kangaroos 61±5 13±6 14±5 6±1 3±1 2±3 1±1

Deer 46±2 20±1 15±4 5±2 8±2 2±2 3±3

grazers eat more than 50% monocots and browsers eat more than 50% dicots.
Their faecal compositions (Table 1) showed Forester kangaroos were definitely
grazers whereas fallow deer were on the borderline between being a grazer and
a browser. In Great Britain fallow deer have a large seasonal variation in their
diet being a browser in winter and a grazer in summer (Chapman 1975). This
difference is probably a result of the milder winter climate in Tasmania, where
grasses remain accessible through winter.

Bark was found in small quantities in the faeces of both species but was
observed to occur more often in deer faeces than in that of kangaroos (Table 1).
This is probably due to their deliberately ingesting it (Chapman 1975) whereas
kangaroos may only accidentally ingest it while feeding. Deer may consume
bark for reasons such as obtaining some essential minerals, to slow digestion and
promote a feeling of satiation when food is short and as a response to stress when
disturbed (Jackson 1974).

Bracken (Pteridium esculentum ) was also found in very small quantities in the
faeces of both species (Table 1). It would not be expected to occur in large
amounts since it is a cumulative poison, containing thiaminase, an enzyme
which inactivates thiamine (Vitamin Bl) (Hungerford 1967). However, toxicity
varies with season and locality and is greatest in the young shoots.

Moss was another food type found in very small quantities in the faeces of
both species (Table 1). It has very little nutritional value (Dearden et al. 1975) and
may have been consumed accidentally. However, it is possible that it was
consumed by deer for its moisture content since according to Chapman (1975)
fallow deer rarely drink, getting their water from dew and vegetation.

DIETARY OVERLAP

Dietary overlap was examined quantitatively by calculating a similarity
index between the faecal composition of each species at each collection period.

4

The Tasmanian Naturalist July 1992

Overlap was very high during all periods being greatest (90%) in mid Winter
(June) and lowest (75%) in September. Other studies of sympatric herbivores,
both native and introduced, have also found high degrees of similarity, for
example 75% between feral goats and Yellow-footed rock wallabies (Dawson
and Ellis 1979), 66% between grey kangaroos and wallaroos (Taylor 1983). It has
been suggested that the diets of sympatric herbivores should overlap least when
food is short so as to minimise interspecific competition thus allowing continued
co-existence (Leslie et al. 1984). In studies of introduced and native herbivores
living sympatrically in Australia there are cases where this has been found to be
true (e.g. Ellis et al. 1977; Barker 1987) and examples where the opposite appears
to be the case, that is overlap is highest when food is short (e.g. Storr 1968;
Dawson and Ellis 1979). In this study the overlap was highest when there was
least food available.

FOOD PREFERENCES

When faecal composition was compared to the availability of food plants deer
were found to have a high preference for "browse" species whereas kangaroos
were found to prefer grasses, in particular native grass species such as Themeda
triandra. Both kangaroos and deer selected against Poa spp., Microlaena stipoides,
Holcus larntus, Lomandra longifolia, Juncus spp. and moss. Additionally deer
avoided the grasses Danthonia spp. and Vulpia spp. and kangaroos avoided
bracken and non-leguminous herbs. However, some of these items that were
selected against still comprised a major par t of the diet because of their dominance
in the vegetation. Thus, Poa spp. comprised almost 10% of kangaroo diets and
Holcus spp. comprised 13% and 18% of deer and kangaroo diets respectively.

DIVERSITY OF DIET

The diversity of plant species eaten by a herbivore is useful as a measure of
the breadth of its feeding niche. Deer were found to include a greater range of
items in their diets and to have a more even representation of different items than
did kangaroos, that is the kangaroos were much more selective in what they ate.
Nevertheless, Forester kangaroos appeared to be less selective than other
macropods in mainland Australia (Duncan 1987), possibly because the temperate
climate of Tasmania provides a broader range of food than the more arid areas
where most other studies have been carried out. Other studies (for example
Schwartz and Ellis 1981) have also found that introduced ruminants have a
broader feeding niche than native herbivores. It has been suggested that this is
because the native herbivores are better adapted to the available food resource.
It may also be that it is this characteristic which has enabled the introduced
herbivore, in this case fallow deer, to survive in an alien environment.

The Tasmanian Naturalist July 1992

5

CONCLUSIONS

Since it was not known whether food was limiting, it was not possible to
conclude whether competition was occurring between fallow deer and Forester
kangaroos. Itappeared that the diet ofeach species was independently determined
by factors such as feeding preferences, the nutritional value of plants and their
palatability and availability. However, the high degree of overlap of diets
indicates the potential for competition if food were limiting, either due to
drought, harsh winter conditions or an increase in the populations of deer or
kangaroos. Competition between deer and red-neck wallabies (M. rufogriseus) is
also a possibility due to the greater reliance of wallabies on non-grasses in
comparison to Forester kangaroos (Jarman and Phillips 1989). Given that such
potential exists one wonders why the deer population is actively maintained, if
not encouraged, by the system of seasons and licences opera ted by the Depart men t
of Parks Wildlife and Heritage.

ACKNOWLEDGEMENTS

I would like to thank Mr Kenneth Von Bibra for allowing me l o undertake the
study on his property. M. Stoddart, R. White, R. Mawbey, J. Robin and M. Brown
provided assistance in various ways.

REFERENCES

Allison, C.D. (1985). Factors affecting forage intake by range ruminants : a
review. /. Range Manage. 38:305-311.

Barker, R.D. (1987). The diet of herbivores in the sheep rangelands. pp69-83. In
Kangaroos: their ecology and management in thesheep rangelands of Australia. Eds.
G.Caughley, N. Shepherd and J. Short. (Cambridge University Press:
Cambridge).

Bentley, A. (1978). An Introduction to the Deer of Australia. (The Koetong Trust
Service Fund, Forests Commission Victoria: Melbourne).

Chapman, D. and Chapman N. (1975). Fallow deer. (Lavenham Press : Suffolk).
Dawson, T.J. and Ellis, B.A. (1979). Comparison of the diets of Yellow-footed
rock wallabies and sympatric herbivores in Western NSW. Aust. Wtldl. Res.
6: 245-254.

Dearden,B.L.,Pegau,R.E.and FIansen,R.M.(1975). Precision of microhistological
estimates of ruminant food habits. ]. VJildl. Manage. 39:402-407.

Duncan, A.M.R. (1987). A dietary study of two sympatric herbivores, fallow deer
(Dama dama) and Forester kangaroos (Macropus giganteus tasmaniensis ).
Honours thesis. University of Tasmania.

Ellis, B.A., Russell, E.M., Dawson, T.J. and Harrop, C.J.F. (1977). Seasonal
changes in diet preferences of free-ranging red kangaroos, eu ros a nd sheep i n

6

The Tasmanian Naturalist July 1992

Western NSW. Aust. Wildl. Res. 4:127-144.

Griffiths, M., Barker, R. and Maclean, L. (1974). Further observations on the
plants eaten by kangaroosand sheep grazing together in a paddock in south¬
western Queensland. Aust. Wildl. Res. 1:27-43.

Hanley, T. A. (1982). The nutri tional basisfor food selection by ungulates. /.Range.
Manage. 35:146-151.

Hansen, R.M., Mugambi, W.M. and Bauni, A.M. (1985). Diets and trophic
ranking of ungulates in the Northern Serengeti. /. Wildl. Manage. 49:823-829.
Hungerford, T.C. (1967). Diseases of livestock. (Angus and Robertson: Sydney).
Jackson, J.E. (1974). Feeding habits of deer. Mammal. Rev. 4:93-101.

Jarman, P.J.and Phillips, C.M. (1989). Diets in a community of macropod species,
pp 143-149. In Kangaroos, wallabiesand rat-kangaroos. Eds. G.Grigg, P. Jarman
and I. Hume.fSurrey Beatty: Sydney).

Leslie, D.M., Starkey, E.E., and Vavra, M. (1984). Elk and deer diets in old growth
forests in western Washington. /. Wildl. Manage. 48:762-775.

Moir, R.J., Somers, M. and Waring, H. (1956). Studies on marsupial nutrition 1.

Ruminant-likedigestioninaherbivorousmarsupial(Sefo«i'xbrac/iyurMs,Quoy

and Gaimard). Aust. /. Biol. Sci. 9:293-304.

Russell, G.M. (1974). The biology of kangaroos (Marsupialia : Macropodidae).
Mammal Rev. 4:1-59.

Schwartz, C.C. and Ellis, J.C. (1981). Feeding ecology and niche seperation in
some native and domestic ungulates on the shortgrass prairie. ]. Appl. Ecol.
18:343-353.

Storr, G.M. (1968). Diet of kangaroos (Megaleia rufa and Macropus robustus) and
merino sheep near Port Hedland, Western Australia. /. Roy. Soc. W.A. 51:25-

32.

Taylor, R.J. (1983). The diet of Eastern grey kangaroos and Wallaroos in areas of
improved and native pasture in theNew England tablelands. Aust. Wildl.Res.
10:203-211.

The Tasmanian Naturalist July 1992

7

A TIGER SNAKE GOES TROUT FISHING

by Raymond N. Brereton and Ross /. Peacock

Forests Practices Unit, Forestry Commission, 30 Patrick Street, Hobart,

Tasmania 7000.

At around 6.00pm on the evening of the 9th of March 1992 we were on a
forestry road where it crosses the Wentworth Canal (AMG 4432 502243; al titude
670m), 11 km north of the township of Tarraleah on the Central Plateau. The
water in the canal was shallow (20 cm) and fast flowing. An open slashed strip
is maintained either side of the canal which passes through alpine ash ( Eucalyptus
delegatensis) dominated scrubby montane forest. The canal batter is formed from
Jurassic dolerite rock.

We heard splashing from amongst the rocks forming the canal batter about
10m upstream from where the road crosses the canal. On closer investigation a
small rainbow trout ( Oncorhynchus mykiss), approximately 15cm in length, was
found poking out from under a rock near the edge of the canal. On attempting
to shift the fish back out into the canal, it was found difficult to dislodge. It was
soon discovered why; a snake had the tail and part of the rear of the fish in its
mouth. We let go of the fish and the snake disappeared back under the rock. The
snake had been holding the fish while submerged under the rock. No part of the
snake was visible from above. The rock was dislodged and a black Tiger snake
(Notechis ater), approximately l-2m in length, emerged and swam under some
rocks nearby. It emerged again and swam downstream for 10m or so where it
sought refuge in rocks on the opposite bank.

Tiger snakes inhabit the margins of watercourses and swamps. They prey
mainly on vertebrates associated with these habitats, namely frogs, lizards and
small mammals (Cogger 1983; Wilson and Knowles 1988). They have not been
recorded preying on fish in the wild although they are known to take eels (P.
Whittaker, Zoology Department, University of Tasmania, pers. comm.). The
closely related mainland tiger snake (Notechis scutatis) preys on fish (Wilson and
Knowles 1988). Worrell (1963) reported thatChapell Island tiger snakes (Notechis
aterserventyi) kept in captivity ate "mice, fish, rats, strips of liver and horse meat,
and sausages". Populations of tiger snakes which inhabit islands with mutton
bird colonies feed on the young chicks (Worrel 1963). They also prey on petrel
chicks on some islands off the coast of South Australia (Wilson and Knowles
1988). The King Island race of tiger snake (Notechis ater humphreysi) is
cannibalistic (Worrell 1963).

The temperature of the water in the canal was measured on the 11 th of March
at 11.00am and found to be 11°C. Daily fluctuations in water temperatures of

8

The Tasmanian Naturalist July 1992

streams on the Central Plateau can be of the order of 5°C in summer (Peter Davies,
Inland Fisheries Commission, pers. comm.). Thus it is likely that the water was
warmer on the evening of the incident after a warm day. Tiger snakes have been
recorded active at temperatures as low as 12°C on the Central Plateau (P.
Whittaker, pers. comm.). However, they were generally active during the
middle of the day when they had had the opportunity to bask. Tiger snakes are
shuttling heliotherms (Rawlinson 1974), that is they will bask to raise their body
temperature until it reaches their normal activity range. Once they have
achieved this they can forage in shade, or the water as described above, where
ambient temperatures are lower than their normal activity temperatures.

ACKNOWLEDGMENTS

We would like to thank Pat Whittaker for his observations on tiger snakes and
Peter Davies for information on water temperatures of Central Plateau waterways.

REFERENCES

Cogger, H.G. (1986). Reptiles and amphibians of Australia. (A.H. and A.W.Reed:
Sydney).

Rawlinson, P. (1974). Biogeography and ecology of the reptiles of Tasmania and
the Bass Strait area. In Biogeography and Ecology in Tasmania. Ed. W.D.
Williams. (Dr. W. Junk Publishers: The Hague).

Wilson, S.K. and Knowles, D.G. (1988). Australia's Reptiles. A photographic
reference to the terrestrial reptiles of Australia. (Collins:Sydney).

Worrell, E. (1963). Reptiles of Australia. (Angus & Robinson: Sydney).

1 JJ

3 o
J </|