Geographic patterns of genetic variation in brushtail possums trichosurus vulpecula and implications for pest control

1 School of Biological Sciences, Victoria University, P.O. Box 600, Wellington.
Present address: Science Directorate, Department of Conservation, P.O. Box 10420, Wellington.
2 Forest Research Institute, P.O. Box 31-011, Christchurch.
Present address: Advocacy and Extension Directorate, Department of Conservation, P.O. Box 10420, Wellington.
GEOGRAPHIC PATTERNS OF GENETIC VARIATION INBRUSHTAIL POSSUMS TRICHOSURUS VULPECULA ANDIMPLICATIONS FOR PEST CONTROL Summary: Two morphological types of brushtail possum (Trichosurus vulpecula) were introduced to New
Zealand: smaller, grey possums from mainland southeastern Australia, and larger, black possums from
Tasmania. Analysis of patterns of allozyme variation and allele frequencies of present-day possum populations
in New Zealand and southeastern Australia indicates that populations comprised predominantly of black
possums remain genetically similar to possums in Tasmania, whereas predominantly grey populations are
genetically closer to Victorian and New South Wales possums. The distribution of possums in New Zealand can
be accounted for at least partly by selection of stock types with respect to climate. Genetic differences between
populations may have important implications for the control of possums, because Tasmanian possums have a
greater resistance than mainland southeastern Australian possums to 1080 poison (sodium monotluoroacetate),
which is commonly used to control possums in New Zealand.
Keywords: Genetic variation; allozyme electrophoresis; brushtail possum; Trichosurus vulpecula; introduced
species; pest control; selection.
IntroductionStudies of the genetics of introduced species are often population had an overall observed heterozygosity of limited by lack of an historical context in which to 0.029 and was fixed for the common allele at three interpret patterns of genetic change, because the loci which were variable in populations from New origins, genetics, and history of introduced stock are South Wales and Victoria. A Victorian population had seldom known. The introduction of the brushtail a heterozygosity of 0.040 and was fixed at two loci possum (Trichosurus vulpecula Kerr: Marsupialia) to variable in Tasmania and New South Wales, whereas New Zealand, on the other hand, has been relatively a New South Wales population had a heterozygosity well documented, as has its morphology, distribution, of 0.048 (Triggs, 1987). Thus the level of genetic and ecology (Morgan and Sinclair, 1983). These (allozymic) variation in New Zealand populations factors provide reference points for evaluating should depend on the degree of mixing of these patterns of genetic variation, gene frequencies, and Australian stocks, as well as on any changes that have More than 200 possums were imported to New According to historical records (Pracy, 1962), Zealand from Australia between 1837 and 1924 in order to establish a fur industry (Pracy, 1962).
(Trichosurus vulpecula vulpecula) were small and grey, Although the possum is commercially important for whereas the Tasmanian stock (Trichosurus vulpecula its fur in New Zealand (Pracy, 1981), it is also a fuliginosa), although probably polymorphic for colour major pest. Possums cause damage to native and (Kean, 1971), was larger and black. In New Zealand, exotic forests (Bathgate, 1973), erosion control possums can be classified as either 'black' or 'grey', plantings (Jolly and Spurr, 1981), crops (Spurr and although a range of shades occurs from black through Jolly, 1981), pasture (Gilmore, 1965), orchards (Anon, brown, red-brown, and grey-brown to silver-grey.
1968), and nectar sources (Anon, 1973), as well as Mixed populations, having both grey and black being a reservoir for bovine tuberculosis (Ekdahl, individuals, occur in many parts of New Zealand, but Smith and Money,. 1970). The significance of the the distribution of coat colours is not even in different possum as a competitor of native birds has also been a parts of the country (Wodzicki, 1950; Kean, 1971).
cause for concern (Leathwick, Hay and Fitzgerald, Some areas, such as Westland, have almost all black 1983; Fitzgerald, 1984; Wardle, 1984).
possums, whereas other areas, such as Northland, have only grey possums. Body size also varies between from Victoria, New South Wales, and Tasmania areas (Yom Tov, Green and Coleman, 1986; our (Pracy, 1962). These Australian populations differ in both the amount of heterozygosity and in the number This mosaic of coat colours and body size may be of variable loci (Triggs, 1987). A Tasmanian the result of a) a non-random pattern of introduction New Zealand Journal of Ecology 12:New Zealand Ecological Society NEW ZEALAND JOURNAL OF ECOLOGY, VOL. 12, 1989 coupled with a subsequent lack of natural dispersal predominantly grey New Zealand populations are over long distances, b) a haphazard pattern of most similar to those of mainland Australian liberations followed by selection, c) random chance, populations. A non-random distribution of Australian or d) most likely, a combination of processes. There is stocks in New Zealand may have implications for pest some evidence that predominantly one colour morph control, as Tasmanian possums are more resistant to was liberated in certain areas. Most possums liberated 1080 poison at low temperatures than mainland in Westland, for example, were black (Pracy, 1962).
Australian (New South Wales) possums (McIlroy, However, liberation records reveal that both black and 1983). The analysis is complicated by any genetic grey possums were introduced to most areas of New changes that have accompanied the colonization of Zealand (Pracy, 1962), suggesting that the pattern of New Zealand by small founder populations (Triggs, distribution of possums in New Zealand has resulted 1987) and by the untestable assumption that the Australian populations that we sampled accurately The main aim of our study was to use allozyme estimate the allele frequencies of possums originally electrophoresis to determine whether the non-random distribution of colour morphs in New Zealand isaccompanied by an associated pattern of allelic distribution which can be related to the different Samples of liver, muscle, and blood were collected Australian stocks. If coat colour does reflect the from possums in four locations in southeastern origins of New Zealand populations with respect to Australia and 10 locations on New Zealand's North, Tasmanian and mainland Australian stocks, then two South and Stewart Islands (Table 1). The sample from predictions' can be made: (1) that the amount of South Australia was collected for use as an outgroup, variation (heterozygosity and polymorphism) in each because possums are not known to have been exported New Zealand population depends on the proportions to New Zealand from South Australia. Specimens of the two colour morphs in the population (in were frozen on dry ice or in liquid nitrogen in the particular, predominantly black populations should field and stored in an ultra-cold (-80°C) freezer for have a lower level of variation than grey populations the duration of the study. For electrophoretic analysis and mixed-colour populations a higher level of small sub-samples of tissues were macerated in an variation than non-mixed populations); (2) that allelic equal volume of distilled water, then centrifuged at frequencies in predominantly black New Zealand 2000 rpm for 5 minutes. The resulting supernatant populations are most similar to those of Tasmanian fractions were subjected to starch-gel electrophoresis, populations, whereas allele frequencies in Table 1: Sampling locations, sample sizes, and meteorological data (mean annual rainfall MAR and mean annual temperatureMA T). Meteorological data are from New Zealand Meteorological Service Misc. Publ. 177 (1981) or the Tasmanian YearBook (1985).*Sample collected by Ecology Division, DSIR. Australia
New Zealand
TRIGGS and GREEN: GENETIC VARIATION IN POSSUMS using gels made of 14% Electrostarch (Madison, Wisconsin, lot no. 392) and modifications of the methods of Selander et al. (1971), Harris and A total of 25 enzymes and 7 general proteins Hopkinson (1976), and Allendorf et al. (1977), as (including haemoglobin), encoding 45 loci, was described in Triggs (1987). The recommendations of resolved: aconitase (Acon 1-2, E.C. no. 3.1.3.2), Murphy and Crabtree (1985) were followed in adenylate kinase (Ak, 2.7.4.3), B-galactosidase (B-Gal, labelling enzymes, genetic loci, and alleles.
3.2.1.23), creatine kinase (Ck, 2.7.3.2), diaphorase For each population, the level of genetic variation (Dia, 1.6.2.2), erythrocyte acid phosphatase (Eap, was assessed by degree of polymorphism (P) and 3.1.3.2), esterase (list 1-6, 3.1.1.1), general proteins observed heterozygosity (H). Genetic differentiation (Gp 1-6), glucose-6-phosphate dehydrogenase (Gd, between populations was estimated using Nei's (1978) 1.1.1.49), glucose phosphate isomerase (Gpi, 5.3.1.9), unbiased genetic distance (D); populations were then clustered using the UPGMA algorithm (Sneath and dehydrogenase (Glud, 1.4.1.3), glutamate oxaloacetate Sokal, 1973). All data were analysed using the transaminase (Got 1-2, 2.6.1.1), glycerol-3-phosphate BIOSYS-l programme (Swofford and Selander, 1981).
dehydrogenase (Gpd, 1.1.1.8), haemoglobin (Hb), Correlation coefficients, r, were used to determine isocitrate dehydrogenase (Icd 1-2, 1.1.1.42), lactate the relationships between allele frequency at each dehydrogenase (Ldh 1-2, 1.1.1.27), malate locus (Table 2), coat colour (given by "% black" - dehydrogenase (Mdh 1-2, 1.1.1.37), malic enzyme the percentage of black possums in a population; (Me, 1.1.1.40), mannose phosphate isomerase (Mpi, Table 3), mean adult body length and weight in each population (Table 3), latitude (Table 1), and climate phosphogluconate dehydrogenase (Pgd, 1.1.1.44), (mean annual rainfall and mean annual temperature; phosphoglucomutase (Pgm 1-2, 2.7.5.3), purine Table 2: Allele frequencies at polymorphic loci, % polymorphic loci (P), and observed heterozygosity (H) in southeasternAustralian and New Zealand populations of Trichosurus vulpecula. (Populations numbered as in Table 1.) N = sample size. NEW ZEALAND JOURNAL OF ECOLOGY, VOL. 12, 1989 Table 3: Morphological characteristics of New Zealandpopulations of brush tail possum. Equal numbers of adult Mean estimates of allozyme variation in New females and adult males were sampled. Zealand (P = 0.155, H = 0.041) were slightly, but not significantly, greater than those of the Australian stock populations (P = 0.141, H = 0.039; Table 4).
Comparisons of polymorphism and heterozygosityamong Australian and New Zealand populations (Table 4) suggest that the amount of variation in New Zealand depends to some extent on the proportion of each colour morph in the population. New Zealand populations with more than 50% black individuals had a significantly lower level of variation than predominantly grey populations (t = 2.37, p < 0.05 for P; t = 3.4, p < 0.01 for H), in parallel with the lower level of variation in Tasmanian compared to mainland Australian populations (Table 4). Mixed nucleoside phosphorylase (Np, 2.4.2.1), sorbitol colour populations in New Zealand had a higher level dehydrogenase (Sordh, 1.1.1.14), superoxide dismutase of variation than non-mixed populations, as expected (Sod 1-2, 1.15.1.1), and unidentified dehydrogenase from the mixing of stocks fixed for different loci, although the difference in variation was not significant Allozyme variation in New Zealand populations between mixed and non-mixed populations (t = 2.07, No alleles were detected in New Zealand that were not P = 0.07 for P; t = 1.30, p = 0.2 for H). The found in at least one Australian population (Table 2).
difference in variation between mixed New Zealand TRIGGS and GREEN: GENETIC VARIATION IN POSSUMS Table 4: Mean and standard deviation of polymorphism (P) and heterozygosity (H) for brush tail possum populations ofsimilar geographic origin or coat colour, and for Australian stocks (.*excludes Adelaide sample). New Zealand mixed colour (32-66% black possums) New Zealand non-mixed (0-2%, 73-100% black possums) New Zealand black (53-100% black possums) and Australian stock populations was not significant Genetic relationships among New Zealand and (t = 1.31, p = 0.2 for P; t = 0.4, p = 0.5 for H) Two main clusters of populations were identified by The level of variation was not entirely consistent phenetic clustering, based on Nei's D (Fig. 1). New among New Zealand populations with similar Zealand populations with a high proportion of grey proportions of each colour morpho For example, possums (0-53% black) and grey Australian Waipoua 0% black) had a 50% greater heterozygosity populations were closely associated, as were than Wanganui (2% black), presumably as a result of predominantly black New Zealand populations genetic drift in small founder populations. Changes in (61-100% black) and the Tasmanian population. The levels of variation associated with colonization are inclusion of New South Wales in the former cluster may be due to either the small genetic distance Figure 1: UPGMA phenogram of genetic relationships among southeastern Australian and New Zealand populations ofbrushtail possum. NEW ZEALAND JOURNAL OF ECOLOGY, VOL. 12, 1989 colour and mean daily temperature (r = -0.77, confirmation that some possums were imported from p< 0.01), but not with latitude (r = 0.48, p > 0.05).
New South Wales. The Me-1(c) allele found in many Several significant correlations were also found New Zealand populations was also found in 'possums between allele frequencies, coat colour, latitude and from New South Wales but not in the sample from climate in New Zealand. A total of 56 correlations Victoria, although it is also possible that this allele was calculated. With a 5% chance of a type II error, was present at low frequency in Victoria. However, only 2.8 significant correlations were expected by the greater genetic similarity of New Zealand grey chance, but 7 were observed. Colour (% black) was populations to Victoria (Fig. 1) and the presence in significantly (p < 0.05) correlated with allele New Zealand of several alleles not found in the New frequency for Est-1(a), Est-1(b), Icd-2(a), Icd-2(c), South Wales sample [Est-1(c), Idh-2(c) and Me-1(b)] and Pgd-1(a). Mean annual temperature was suggest the predominance of Victorian stock in New significantly correlated with Pgd-1(a) (p < 0.05) and associated with Est-1(b), Est-5(a) and Got-1(a) (p < 0.10). Latitude was correlated with Pgd-1(a) resemble those from New Zealand. This result was (p < 0.05). No significant correlations were found expected, because no possums were imported to New between allele frequency and rainfall. Colour and Zealand from South Australia. One population, allele frequency appear to be the most closely Wanganui, did not fall into either cluster, possibly due associated; selection of one or both may be linked to to genetic drift, if the Wanganui population had a temperature. No significant correlation (r = 0.012, P > 0.05) was found between overall heterozygosity Although the genetic distances involved are very and latitude, in contrast to the result for Australia small (the 'black/grey' separation occurs at D = 0.008), our analysis suggests that the origins of New Body length and weight were also correlated with Zealand populations are still reflected in their allele coat colour in our study populations (r = 0.68 and frequencies and may be roughly estimated by coat r = 0.64 respectively; p < 0.05). Yom Tov (1984) and colour. The relationship between coat colour in New Yom Tov et al. (1986), in an extensive morphological Zealand populations and position within the genetic survey of possums in New Zealand, found significant cluster is by no means perfect. Presumably random negative correlations between many skull and body genetic drift in the relatively small founder length measurements and mean annual temperature.
populations has led to changes in gene frequency in However, they did not consider coat colour, and our many populations. However, the general pattern that reanalysis of data on body and skull measurements in emerges, grouping predominantly black populations Yom Tov (1984) with respect to coat colour produced with the Tasmanian sample and grey with mainland better correlations between body size and % black Australia, is unlikely to have arisen by chance (that is, than between body size and mean annual temperature by random drift). The implication is that either for all characters except distance between bullae interbreeding of stock types for 150 years has been insufficient to establish a panmictic unit, or that Table 5: Correlations (r) between body and skull selection has acted differentially on stock types to measurements, coat colour (% Black), and mean annual produce genetic structuring of allozymes and colour temperature (MA T). Body and skull measurements and correlation coefficients of measurements vs MAT are from Selection as an explanation for the mosaic of Yom Tov (/984). % Black (the proportion of black possums colour morph distribution in New Zealand is in a population) values are from our data. Significance levels supported by evidence of correlations between coat of r: .*p < 0.05, .**p < 0.001. colour (% black possums in a population), allelefrequencies, body size and weight, and climatic variables. A significant correlation (r = 0.642, p < 0.01) exists between coat colour and rainfall in New Zealand (R.E. Brockie, pers. comm.). Data from our 10 mainland New Zealand study areas showed a similar, although non-significant, correlation between colour (% black) and rainfall (r = 0.58, P < 0.10). A significant correlation in our data was found between TRIGGS and GREEN: GENETIC VARIATION IN POSSUMS sparrows in North America (Johnson and Selander, The introduction of possums from at least two regions 1964). In all cases, these associations have developed of Australia, followed by hundreds of largely very rapidly, within a few hundred years, suggesting undocumented liberations of New Zealand-bred stock, that selection for local adaptation may be very strong, has produced a complex pattern of genetic even in small populations in which random forces are relationships in New Zealand possums, upon which selective and random genetic changes have been Our results have important implications for the control of possums in New Zealand. Currently, large- Our results generally conform to the prediction scale control of possum numbers is by aerially-sown that if coat colour of New Zealand populations cereal baits or carrots with 1080 (sodium indicates origin, then (1) the level of genetic variation monofluoroacetate) poison. One serious concern is the in New Zealand populations should be related to the dosage level of 1080 required to kill a possum. Bell proportion of each colour morph in a population, and (1972), Rammell and Fleming (1978), and McIlroy (2) allele frequencies of New Zealand populations with (1983) experimentally determined an LD of about 0.8 different proportions of each coat colour should mg 1080 kg-1 body weight for possums. In contrast, reflect allele frequencies of the Australian stock types.
the New Zealand Forest Service found an LD of Thus, mainland Australian and Tasmanian stocks are 1.3-2.1 mg kg-1 (Anon, 1978), requiring a toxic not distributed at random in New Zealand. Black and loading of 0.15% w/w on baits of mean weight of 4 grey populations also differ in average body size (Yom g. The higher dose not only adds to the cost of Tov et al., 1987; our study), as they do in Australia poisoning operations and the risk to non-target species, but also leads to a high aversion rate as some Selection with respect to climate appears to be an possums can detect and reject 1080 at concentrations important determinant of the distribution of possum of 0.1 % w/w or more (Morgan, 1982). At present, types in New Zealand, although deliberate flavours such as cinnamon are used as masks to introductions of possums of different stocks to disguise the poison (Morgan, Batcheler and Peters, suitable habitats probably also played a part in 1986), and baits are loaded either at 0.08% w/w or determining present distributions. In New Zealand, 0.15% w/w (D.R. Morgan, pers. comm.) as a result cold, wet areas tend to harbour large, black possums of the ambiguous data published on the possum's most similar to the Tasmanian type, whereas warm, sensitivity to 1080 poison. The cause of differences in dry areas harbour small, grey, mainland Australian- LD between laboratories is still not clear, although type possums. Areas of intermediate climate have variation in the techniques for handling 'and mixed populations. The significant correlations acclimatising possums were probably partly between temperature and allele frequency, colour and responsible (Anon, 1979). McIlroy (1983) found body size of possums in New Zealand suggest that neither acclimatisation nor stress had any effect on the these were either directly selected for or acted as LD but he did find significant differences between markers for other characteristics selected as possums possums from different regions of southeastern colonized New Zealand. The correlation between Australia. At low temperature (10°C) Tasmanian colour and rainfall has been found even within a possums were more resistant to 1080 than mainland single valley with a steep rainfall gradient (R.E.
(New South Wales) possums. The LD,. for Tasmanian Brockie, pers. comm.), emphasizing the strength of possums was 0.92 mg kg-1, while New South Wales selection. An association between coat colour and possums had an LD of 0.42 mg kg-1, (McIlroy, rainfall has also been documented in Tasmania 1983). Decreased sensitivity to 1080 at low (Guiler, 1953), black possums being more common in temperatures has also been found in racoons (Eastland areas of high rainfall. A physiological basis for and Beasom, 1986). Much higher tolerances to 1080 different climatic tolerances is suggested by the occur in brushtail possums and other mammals in differences in water metabolism between black and Western Australia, where high levels of fluoroacetates grey morphs (Williams and Turnbull, 1983).
occur naturally in some plants (King, Oliver and Other studies of introduced species have also documented associations between colour or size and The possums used in the original New Zealand climate. These studies include house sparrows (Baker, Forest Service trials (Anon, 1978) were from an area 1980), mynas (Baker and Moeed, 1979), and stoats of predominantly black possums, and were therefore (King and Moody, 1982) in New Zealand, and house likely to be of the Tasmanian type, whereas the NEW ZEALAND JOURNAL OF ECOLOGY, VOL. 12,1989 possums used in the Ministry of Agriculture and P.E. Cowan, C.H. Daugherty, P.J. Moors, R.M.
Fisheries trials (Bell, 1972) were from an area of grey Sadleir, P.J. Smith, and D.R. Towns for their advice possums and were therefore likely to be of the mainland Australian type. A higher tolerance wouldtherefore be expected in the New Zealand Forest Service results. However, a reciprocal exchange ofpossums between the two testing laboratories still Allendorf, F.W.; Mitchell, N.; Ryman, N.; Stahl, G.
resulted in different LD estimates (Anon, 1979), 1977. Isozyme loci in brown trout (Salmo trulla suggesting that some other factor, such as handling L): Detection and interpretation from population technique, was also involved. The temperature regime used in each laboratory was not given and could have Anon. 1968. Oppossum destruction. Orchardist of If Tasmanian possums are more resistant to 1080, Anon. 1973. Damage to nectar sources by opossum as found by McIlroy (1983), then there is a good case and deer. New Zealand Beekeeper 35: 75.
for using a higher dose rate of 1080 in cold, wet areas, Anon. 1978. Bait development and toxicology. New where Tasmanian-type possums predominate, than in Zealand Forest Service, Forest Research Institute warm, dry areas, where mainland-type possums predominate. Tasmanian possums are also heavier and Anon. 1979. Bait development and toxicology. New larger (Yom Tov and Nix, 1986; Triggs, 1987, Zealand Forest Service, Forest Research Institute Appendix II). In order to receive a lethal dose they would need to consume more baits at the same toxic Baker, A.J. 1980. Morphometric differentiation in loading than would smaller, lighter, mainland-type New Zealand populations of the house sparrow possums. As a hypothetical example, assuming that a (Passer domesticus). Evolution 24: 638-653.
totally grey population has an equivalent LD and Baker, A.J.; Moeed, A. 1979. Evolution in the mean body weight to mainland Australian possums introduced New Zealand populations of the (i.e. 0.42 mg 1080 kg-1 and 2.3 kg) and a black common myna, Acridotheres tristis (Aves: population is equivalent to Tasmanian possums (0.92 Sturnidae). Canadian Journal Zoology 57: mg 1080 kg-1 and 3.1 kg), then 50070 of the grey population would be killed with 0.97 mg 1080 per Bathgate, J.L 1973. Summary of questionnaire possum, whereas the black population would require returns. In: Assessment and management of an average of 2.9 mg 1080 per possum.
introduced animals in New Zealand forests, pp.
102-116. New Zealand Forest Service, Forest Research Institute Symposium No. 14.
Bell, J. 1972. The acute toxicity of four common This research was supported by a Postgraduate Grant poisons to the opposum, Trichosurus vulpecula. to S. Triggs from Ecology Division, Department of New Zealand Veterinary Journal 20: 212-214.
Scientific and Industrial Research. Technical support Eastland, W.G.; Beasom, S.L 1986. Effects of and additional finance were provided by Victoria University of Wellington. For assistance with sample racoons. Wildlife Society Bulletin 14: 234-235.
collections we are indebted to staff of Forest Research Ekdahl, M.a.; Smith, B.L.; Money, B.L. 1970.
Institute, Christchurch; Ecology Division, DSIR; Tuberculosis in some wild and feral animals in Rotorua-Taupo, Western, and Wairarapa Pest New Zealand. New Zealand Veterinary Journal Destruction Boards; Wildlife Service; David Crockett of Taiko Camp; staff of Macquarie University, Fitzgerald, A.E. 1984. Diet overlap between kokako University of Adelaide, Victoria Fisheries and Wildlife and the common brushtail possum in central Service, Monash University, and Tasmanian National North Island, New Zealand. In: Smith, A.P. and Parks and Wildlife Service; and the Dunbabbin Hulme, J.D. (Editors), Possums and gliders, pp.
family. Permits for collecting, exporting and 569-573. Australian Mammal Society, Sydney.
importing possum tissues were supplied by the Australian National Parks and Wildlife Service of Gilmore, D.P. 1965. Opposums eat pasture. New New South Wales, South Australia, Victoria and Zealand Journal of Agriculture 110: 284-286.
Tasmania, and the Ministry of Agriculture and Guiler, E.R. 1953. Distribution of the brush possum Fisheries, New Zealand. We thank Drs R.E. Brockie, in Tasmania. Nature 172: 1091-1093.
TRIGGS and GREEN: GENETIC VARIATION IN POSSUMS rattlesnake, Crotalus viridis viridis. Herpetologia Harris, H.; Hopkinson, D.A. 1976. Handbook of enzyme electrophoresis in human genetics. North- Nei, M. 1978. Estimation of average heterozygosity and genetic distance from a small number of Johnson, R.F.; Selander, R.F. 1964. House sparrows: individuals. Genetics 89: 583-590.
rapid evolution of races in North America.
New Zealand Meteorological Service. 1981. Summaries of climatological observations to 1980. New Jolly, J.N.; Spurr, E.B. 1981. Damage by possums Zealand Meteorological Service Miscellaneous (Trichosurus vulpecula) to erosion-control plantings. In: Bell, B.D. (Editor), Proceedings of the First Symposium on Marsupials in New Pracy, LT. 1962. Introduction and liberation of the Zealand, pp. 205-210. Zoology Publication No.
opossum (Trichosurus vulpecula) into New 74, Victoria University, Wellington.
Zealand. New Zealand Forest Service Information Kean, R.I. 1971. Selection for melanism and low Series No. 45, New Zealand Forest Service.
reproductive rate in Trichosurus vulpecula Pracy, LT. 1981. Opossum survey. Counterpest 5: (Marsupialia). Proceedings of the New Zealand Ecological Society 18: 42-47.
Rammell, E.G.; Fleming, P.A. 1978. Compound King, C.M.; Moody, J.E. 1982. The biology of the 1080. Properties and use of sodium stoat (Mustela erminea) in the National Parks of monofluoroacetate in New Zealand. Animal New Zealand. III. Morphometric variation in Health Division, Ministry of Agriculture and relation to growth, geographical distribution, and colonisation. New Zealand Journal of Zoology 9: Selander, R.K.; Smith, M.H.; Yang, S.Y.; Johnson, King, D.R.; Oliver, A.J.; Mead, R.J. 1978. The polymorphisms and systematics in the genus adaptation of some Western Australian mammals Peromyscus. I. Variation in the old-field mouse, to food plants containing fluoroacetate.
Peromyscus polionotus. Studies in genetics, Australian Journal of Zoology 26: 699-712.
University of Texas Publication 7103: 49-90.
Leathwick, J.R.; Hay, J.R.; Fitzgerald, A.E. 1983.
Sneath, P.H.A.; Sokal, R.R. 1973. Numerical taxonomy. W.H. Freeman, San Francisco. 359 p.
mammals on the decline of the North Island kokako. New Zealand Journal of Ecology 6: possumsTrichosurus vulpecula to farm crops and pasture.In: Bell, B.D. (Editor), Proceedings of the McIlroy, J.C. 1983. The sensitivity of the brushtail FirstSymposium on Marsupials in New Zealand, pp.
possum (Trichosurus vulpecula) to 1080 poison.
197-203. Zoology Publication No. 74, Victoria New Zealand Journal of Ecology 6: 125-131.
Morgan, D.R. 1982. Field acceptance of non-toxic and Swofford, D.L; Selander, R.B. 1981. BIOSYS-I: toxic baits by populations of the brushtail possum (Trichosurus vulpecula Kerr). New Zealand analysis of electrophoretic data in population genetics and systematics. Journal of Heredity 72: Morgan, D.R.; Batcheler, C.L; Peters, J.A. 1986.
Tasmanian Year Book. 1985. Australian Bureau of operations? In: Salmon, J.P. (Editor), Proceedings 12th Vertebrate Pest Conference, pp.
Triggs, S.J. 1987 (unpublished). Population and 210-214. University of California, Davis, ecological genetics of the brush-tailed possum Trichosurus vulpecula in New Zealand. PhD Morgan, D.R.; Sinclair, M.J. 1983. A bibliography of Thesis, Victoria University of Wellington. 146 p.
the brush-tailed possum (Trichosurus vulpecula Wardle, J. 1984. The New Zealand beeches. Ecology, Kerr). Forest Research Institute Bulletin No. 25, utilisation and management. Caxton Press, Protection Forestry Division, Forest Research Williams, C.K.; Turnbull, H.L 1983. Variations in Murphy, R.W.; Crabtree, C.B. 1985. Evolutionary seasonal nutrition, thermoregulation and water aspects of isozyme patterns, number of loci, and balance in two New Zealand populations of the tissue-specific gene expression in the prairie NEW ZEALAND JOURNAL OF ECOLOGY, VOL. 12, 1989 common, brush-tail possum, Trichosurusvulpecula (Phalangeridae). Australian Journal ofZoology 31: 333-343.
Wodzicki, K.A. 1950. Introduced mammals of New Zealand (an ecological and economic survey).
DSIR Bulletin 98: 19-49.
Yom Tov, Y. 1984 (unpublished). Morphological trends in the brush tail possum, Trichosurusvulpecula, in New Zealand. Report to EcologyDivision, Department of Scientific and IndustrialResearch, Lower Hutt. 18 p.
Yom Tov, T.; Green, W.Q.; Coleman, J.D. 1986.
Morphological trends in the common brushtailpossum, Trichosurus vulpecula, in New Zealand.
Journal of Zoology 208: 583-593.
Yom Tov, Y.; Nix, H. 1986. Climatological correlates for body size of five species of Australianmammals. Biological Journal of the LinnaeanSociety 29: 245-262.

Source: http://nzes-nzje.grdev.co.nz/free_issues/NZJEcol12_1.pdf