Notes

Ecology, 85(6), 2004, pp. 1740-1746 C 2004 by the Ecological Society of America

POST-MATING SURVIVAL IN A SMALL MARSUPIAL IS ASSOCIATED WITH NUTRIENT INPUTS FROM SEABIRDS

KRISTEN M. WOLFE,'13 HARRIET R. MILLS,' MARK J. GARKAKLIS,2 AND ROBERTA BENCINI'

tSchool of Animal Biology, Faculty of Natural and Agricultural Sciences, The University of Western Australia, 35 Stirling Highway, Crawley, Western Australia 6009 Australia

2School of Biological Sciences and Biotechnology, Murdoch University, Murdoch, Western Australia 6150 Australia

Abstract. We investigated a link between facultative semelparity ("male die-off") in a small, endangered marsupial, Parantechinus apicalis (the dibbler), and the availability of resources on two small islands in Western Australia. Post-mating survival of male dibblers differed between the two islands, and the input of nutrients from nesting seabirds was a possible factor affecting available resources and therefore survival of male dibblers. Sea- birds can increase significantly the concentration of nutrients in the soil and can ultimately lead to an increase in abundance of primary consumers and predators. We compared con- centrations of soil nutrients, body condition, and post-mating survival of male dibblers on both islands. The body condition and post-breeding survival were significantly better on the island with many seabirds, where soil nutrients were up to 18 times higher, than on the adjacent island with fewer seabirds. Our results indicate a potential link between male die- off and resource availability. We suggest that marine-derived nutrients transported onto islands by nesting seabirds can alter the productivity as well as the life history traits of species in an island ecosystem.

Key words: body condition; dasyurid; dibbler; male die-off; nutrients; Parantechinus apicalis; seabirds; semelparity; Western Australia.

INTRODUCTION

The phenomenon of "male die-off," found in several Australian carnivorous marsupials (dasyurids), in which males die after mating due to stress (Lee and McDonald 1985), is still poorly understood. In most species, die-off is obligate, yet it is unclear why for a few species male die-off occurs only in some years and in some populations (facultative die-off). For dasyu- rids, both with obligate and facultative male die-off (Cockburn 1997), good body condition and high fat reserves prior to the breeding season are vital for males to survive during the short, intense breeding season (2- 3 wk) that can result in stress-related deaths (Lee and Cockburn 1985, Lee and McDonald 1985). Increased levels of circulating testosterone in the breeding season lead to high levels of free cortisol (Bradley et al. 1980), which, in turn, promote gluconeogenesis from protein. Therefore males with poor body reserves cannot sustain themselves throughout the mating period (Oakwood et al. 2001). Furthermore, free cortisol depresses the im-

mune system, and males often die from stress-related diseases such as gastric ulcers or infections from bac- teria or parasites (Lee and McDonald 1985).

Within species with facultative male die-off, such as the dibbler, Parantechinus apicalis (Mills and Bencini

2000), and Dasyurus hallucatus (Dickman and Braith- waite 1992, Braithwaite and Griffiths 1994), it is un- clear why male die-off only occurs in some years and in some populations. Mills and Bencini (2000) hy- pothesized that environmental factors, such as food

availability, may play an important role. By contrast, Dickman (1993) suggested that habitat characteristics alone were not likely to be the only factors causing these differences in male die-off and that sperm com-

petition may play an important role. Studies of Ante- chinus stuartii, a dasyurid with obligate male die-off in the wild, demonstrated that males in captivity are able to survive beyond the breeding season, although exhibiting permanent spermatogenic failure (Woolley 1966). By contrast, Mills and Bencini (2000) found that male dibblers survive in captivity up to three years of age and produce sperm and sire litters up into their third year. Since resources such as food are not limited in captivity, this indicates that resource availability

Manuscript received 16 June 2003; revised 27 September 2002; accepted 3 October 2003. Corresponding Editor: C. Mar- tfnez del Rio.

3 E-mail: kwolfe @agric.uwa.edu.au

1740

June 2004 NOTES 1741

may play an important role in male survival in captiv- ity.

Extensive research on seabird colonies around the world has shown that breeding seabirds have a signif- icant effect on the productivity of island ecosystems. Seabirds can increase greatly the concentration of nu- trients in the soil (Gillham 1960, Ryan and Watkins 1989, Anderson and Polis 1999) and can ultimately lead to an increase in the abundance of primary consumers

(Ryan and Watkins 1989, Polis and Hurd 1995, 1996, Anderson and Polis 1998, Stapp et al. 1999, Sinchez- Pifiero and Polis 2000, Stapp and Polis 2003) and pred- ators (Polis and Hurd 1995, Anderson and Polis 1998, Stapp et al. 1999, Stapp and Polis 2003). In fact, the enrichment of habitat within seabird colonies can have a positive effect on the lifetime reproductive success of large mammals, such as red deer (lason et al. 1986).

In Western Australia, tropical seabirds have expand- ed their breeding range southwards due to the Leeuwin Current, a southerly flow of warm, oceanic water (Dun- lop and Wooller 1990). Indeed, the recent expansion of some tropical seabird colonies on islands in Western Australia has been almost exponential. For example, the burrow-nesting Wedge-tailed Shearwater (Puffinus pacificus) appeared on Rottnest Island in Western Aus- tralia (32?00' S, 115031' E) in the late 1880s, when a

only few burrows were recorded. In recent decades numbers have increased from 2000 burrows in the early 1980s to approximately 11 700 burrows in 2002, a five- fold increase in 20 yr (W. J. Bancroft, M. J. Garkaklis, and J. D. Roberts, unpublished manuscript).

The expansion in range of two tropical seabird spe- cies may have affected the primary productivity on two islands in Jurien Bay, Western Australia. These islands are home for two populations of dibblers, an insectiv- orous dasyurid with facultative male die-off. Dibblers once thrived along the southwest coast of Western Aus- tralia, but are currently classified as endangered as they are now restricted to a few fragmented populations, probably due to habitat destruction and introduced

predators (Short and Smith 1994). Two of these pop- ulations are found on Boullanger (25.9 ha) and Whit- lock (5.4 ha) Islands, which lie only 400 m apart and are periodically linked by a sandbar. These Mediter- ranean-climate (hot, dry summers and cool, wet win- ters), naturally nutrient-poor islands were separated from the mainland approximately 500-600 years ago and possibly separated from each other as recently as 150 years ago (Chalmers and Davies 1984). Both are aeolianite limestone islands. Sandy soils are predom- inantly found on Boullanger Island, and Whitlock Is- land, an elevated limestone shelf, has an overlying sandy substrate. Low Mediterranean coastal heath dom- inates both islands with greater species richness found on Boullanger Island while Whitlock Island is domi-

nated by three species, Nitraria billardierei, Atriplex cinerea, and Rhagodia baccata. Human access is re- stricted as they are Class A Nature Reserves (Conser- vation and Land Management Act 1984), and there are no large terrestrial predators; however both islands are

subject to aerial predators (raptor species). As well as

supporting populations of dibblers, both islands sup- port large populations of introduced house mice (Mus musculus), and Boullanger Island supports a small pop- ulation of the Boullanger Island dunnart (Sminthopsis griseoventer boullangerensis). There are six species of

reptiles on Boullanger Island and at least two species on Whitlock Island (Ford 1963). The largest of these, Egernia kingii, is present on Boullanger Island in small numbers, but is absent on Whitlock Island.

There are marked differences in the species of breed-

ing seabirds between Whitlock and Boullanger Islands. Within the last 150 years, Whitlock Island has been colonized by P. pacificus and another tropical seabird, the ground-nesting Bridled Tern (Sterna anaethetus) (Johnstone and Storr 1998), while White-faced Storm Petrels (Pelagodroma marina), a much smaller sub-

tropical burrowing seabird, are found on Boullanger Island. Intriguingly, there are also marked differences in male die-off between the islands populations of dib-

blers, with die-off reported on Boullanger Island (Dick- man and Braithwaite 1992), but never on Whitlock Is- land (Fuller and Burbidge 1987, Mills and Bencini

2000). Dibblers are primarily insectivorous. They mainly

consume the invertebrate orders Orthoptera, Blattodea, Coleoptera, Isoptera, Hymenoptera, Hemiptera, and

Dermaptera (Fuller and Burbidge 1997, Bencini et al.

2001, Miller et al. 2003). A significantly greater num- ber of invertebrates, including orders that are consumed

by dibblers (Orthoptera, Blattodea, Coleoptera, and Is-

optera), were observed on Whitlock than on Boullanger Island (Miller 2000, Miller et al. 2003). In addition, where the density of seabirds was greatest on Whitlock Island there was a greater number of invertebrates that are eaten by dibblers (Miller et al. 2003), including the invertebrates orders Acari, Aranae, Coleoptera, Dip- tera, Blattodea, and Isoptera (Miller 2000).

Small islands with large amounts of marine inputs via seabirds and detritus washed onto beaches (Polis and Hurd 1995, 1996) support higher population den- sities of omnivorous mice (Stapp and Polis 2003). Sim-

ilarly, we propose that seabirds have the potential to

increase the nutrient status and primary productivity of Whitlock Island, ultimately providing a greater food resource for dibblers and possibly resulting in better body condition and a greater survival of males post- mating. Within this seabird-dibbler interaction, field observation suggests that different seabird species are present on Whitlock and Boullanger Islands and that

1742 NOTES Ecology, Vol. 85, No. 6

the density of breeding seabirds is higher on Whitlock Island. Previous studies have also shown that the den-

sity of dibblers (both males and females) was greater and that more males survived after the breeding season on Whitlock Island than on Boullanger Island (Mills and Bencini 2000).

In this study, we have attempted to fill in the "miss-

ing parts" of this interaction and compared on both islands: (1) the density of breeding seabirds, (2) the concentration of soil nutrients, (3) the percentages of male dibblers in different age classes, and (4) the mean body condition of male dibblers. We expected that a

greater density of seabirds would be found on Whitlock Island and therefore that the concentration of soil nu- trients would be greater on Whitlock Island. We ex-

pected that the mean body condition of male dibblers would be better on Whitlock than on Boullanger Island due to greater abundance of invertebrates on Whitlock Island (Miller et al. 2003) and that, as a result, males would survive for longer on Whitlock than on Boul-

langer Island.

METHODS

Trapping

Dibblers were caught in baited Elliott type B metal

traps (Elliott Scientific, Upwey, Victoria, Australia). Trapping took place on both islands between June 1997 and October 2002. Trapping effort on Boullanger Island was 5074 trap nights and on Whitlock Island was 4731

trap nights. Twenty-five trapping sessions were con- ducted on Boullanger Island and 28 on Whitlock Island. In each session, trapping was undertaken for 2-4

nights. Traps were placed in transects across both is- lands in order to cover all vegetation communities.

Traps were set in the late afternoon and checked early on the following morning. Each new dibbler was marked for identification by implanting passive tran-

sponders (Trovan; Central Animal Records, Keysbor- ough, Victoria, Australia) under the skin of the dorsal neck area. Ninety-five percent of the dibblers trapped in this study were juveniles when first marked, and therefore their age was known. Standard measurements were taken for each captured dibbler: mass, sex, short

pes length, and head length.

Seabird density

The density of seabirds for Whitlock and Boullanger Islands was estimated by averaging the seabird burrow counts within 5 x 20 m quadrats along strip transects, which covered the area of each island.

Soil sampling and nutrient analysis

The concentrations of nutrients in the soil on Whit- lock and Boullanger Islands were measured during

2001-2002. Approximately 250 g of surface soil, taken from four random points within a 6 x 6 m area, were combined to form one sample. All soil samples were sealed in plastic bags, stored in a cooler, and then trans- ferred to the laboratory for nutrient analysis. Available

phosphorus and potassium levels were determined by the Colwell method (Rayment and Higginson 1992). Ammonium and nitrate nitrogen were measured using a Lachat flow injection analyzer (Pro-Tech Group, Coolum Beach, Queensland, Australia). Ammonium

nitrogen concentrations were measured colorimetrical-

ly at 420 nm using the indo-phenol reaction. Nitrate

nitrogen concentrations were measured from reduced nitrate measured at 250 nm (Searle 1984). Extractable

sulphur was measured by inductively coupled plasma spectrometry following the techniques outlined in Blair et al. (1991). Reactive iron was determined spectro- photometrically at 248.3 nm after the soils had been tumbled for I h with Tamms' reagent (Garkaklis et al.

2003).

Comparison of age class structures of male dibblers

We compared seasonal changes in numbers of males in each age class (between I and 2 yr old, between 2 and 3 yr old, and greater than 3 yr old) between Whit- lock and Boullanger Islands. A comparison of age clas- ses between islands accounts for the possibility that the observed male die-off on Boullanger Island during some years (Mills and Bencini 2000) could be a re- flection of the low density of males, and therefore their low trappability, rather than the death of all males in the population.

Body condition index

For both islands, we used two methods to calculate the mean body condition of all males trapped during the four months (November to February) prior to the

breeding season when males are reaching their peak mass. With the first method (LeCren 1951) a simple index of condition is obtained by dividing body mass

by a measure of body length cubed, in this case the short pes (foot). The second method, considered by some authors to be a more accurate method than the first (Krebs and Singleton 1993), used a regression equation between body mass and a constant body mea- surement to obtain a ratio of predicted body mass over observed body mass. Using the second method, an mean body condition index equals 1.0. Body condition indices were calculated for males trapped between 1997 and 2002 using body mass and length of pes.

Statistical analysis

Body condition indices were compared between is- lands using a Student's t test. For the analysis of soil nutrient concentrations, Box's M tests showed a sig-

June 2004 NOTES 1743

TABLE 1. Concentrations of available nutrients on Boullan- ger and Whitlock Islands, two islands in Jurien Bay, West- ern Australia.

Boullanger Whitlock Island Island

Nutrients Mean SE Mean SE

Nitrate nitrogen (mg/kg) 4.2* 0.7 84.5* 18.8 Ammonia nitrogen (mg/kg) 1.2* 0.0 7.8* 1.6 Phosphorus (mg/kg) 52.8* 4.1 470.5* 39.2 Potassium (mg/kg) 32.1" 3.9 200.2* 25.7 Sulphur (%) 9.5* 0.7 60.8* 7.0 Reactive iron (mg/kg) 20.0* 2.6 218.9* 25.4

Note: An asterisk indicates significant differences; Wilks' lambda = 0.136; df = 6, 52; P < 0.001.

nificant heterogeneity of variance for nitrate, ammo- nium, and potassium. Logarithmic transformation of the data corrected for any heteroscedasticity (Tabachn- ick and Fidell 1996). Differences in soil nutrients be- tween islands were examined by multiple analysis of variance with a single independent variable being site (island) and plant-available soil nutrients the multiple dependent variables (Tabachnick and Fidell 1996). Ef- fects were assumed to be significant when the level of

probability was 5% or less. Results are presented as means ? 1 SE.

RESULTS

Comparison of seabird densities between islands

A greater density of seabirds was found on Whitlock Island with densities of 370 P. pacificus burrows per hectare and 400 S. anaethetus individuals. Only - 196 burrows per hectare of P. marina were found on Boul-

langer Island.

Comparison of soil nutrient concentrations between islands

Concentrations of plant-available soil nutrients were 5-18 times greater on Whitlock than Boullanger Island for all nutrients tested (P < 0.001, Table 1).

Comparison of age class structures of males between islands

From 1997 to 2002, we trapped 50 individual males on Boullanger Island. Forty-nine were between 1 and 2 yr old (98%), only one was between 2 and 3 yr old (2%), and none were more than 3 yr old. By contrast, males survived for longer on Whitlock Island, where we trapped a total of 166 individual males. Of these,

111 were between 1 and 2 yr old (67%), 45 were be- tween 2 and 3 yr old (27%), and 10 were more than 3

yr old (6%). Not only did males survive for longer on Whitlock Island, the total number of males known to be alive (KTBA) was on average six times higher on

Whitlock than on the larger Boullanger Island from 1997 to 2002 (Fig. 1).

Comparison of body condition of males between islands

The mean body condition of males was significantly better on Whitlock Island than on Boullanger Island. With the first method (LeCren 1951), the mean body condition of male dibblers was significantly better on Whitlock Island (0.242 + 0.002) than on Boullanger Island (0.235 1 0.003; P < 0.02). Using the second method (Krebs and Singleton 1993), the mean body condition of male dibblers was 1.16 + 0.02 on Whit-

A 16 11 2 1 0 0 5 4 2 1 5 3

a a

100

a> CD

E 0 50

r-

a)

a_ 0

B 29 17 8 7 17 11 19 17 13 11 10 7

a a b b c c 100-

CD

E o 50 a) 0)

W SpW SpW SpW SpW SpW Sp 1997 1998 1999 2000 2001 2002

0 1-2 yr old r[ 2-3 yr old E >3 yr old

FIG. 1. Percentages of adult (>1 yr old) male dibblers known to be alive (KTBA) on (A) Boullanger and (B) Whit- lock Islands in Jurien Bay, Western Australia, between 1997 and 2002. The letter "a" indicates that the dibblers were at least 1 yr old, the letter "b" that males in the 1-2-yr-old group were at least 2 yr old, and the letter "c" that males in the >3-yr-old group were at least 3 yr old. Numbers above the graph bars represent sample sizes. For example, the zeros above winter (W) and spring (Sp) 1999 indicate that no males were trapped during these sessions, indicating that male die- off occurred in that year.

1744 NOTES Ecology, Vol. 85, No. 6

lock Island and 1.05 ? 0.03 on Boullanger Island (P < 0.002).

Male dibblers lived longer on Whitlock than Boul-

langer Island (Fig. 1). Since this difference in the dis- tribution of age classes between the two islands could have influenced the body condition of males on Whit- lock Island, only males in the same age classes were

compared and males older than 2 yr old were eliminated from the analysis. The body condition of males between I and 2 yr old prior to the breeding season was com-

pared between islands using the regression method. The mean body condition of male dibblers was still signif- icantly better on Whitlock Island (1.14 ? 0.02) than on Boullanger Island (1.05 ? 0.03) (P = 0.013).

DIscuSSION

It is unclear why differences in male die-off exist between small dasyurid species. It has been suggested that habitat characteristics, such as seasonal food avail-

ability, are not likely to be the only factors causing these differences in male die-off (Dickman 1993). Sperm competition might play an important role in some species, in addition to, or instead of, the timing and abundance of available food resources (Dickman 1993). The stress-related deaths of males of these spe- cies may be linked to the intense reproductive effort needed to mate with many females during their short annual breeding season, thereby maximizing their re-

productive effort (Dickman 1993). This hypothesis was

suggested to explain differences found in male die-off between island and mainland populations of dibblers (Dickman and Braithwaite 1992). The observation that males die post-mating on Boullanger Island and survive within mainland populations was suggested to be a re- sult of the higher population density on Boullanger Island, which promoted more frequent interactions be- tween males. This, in turn, would promote greater lev- els of reproductive effort in competing males, increased stress, and therefore earlier death (Dickman and Braith- waite 1992). While this hypothesis has considerable merit, it does not explain the higher rates of survival of male P. apicalis on the densely populated Whitlock Island in comparison to the relatively sparsely popu- lated Boullanger Island. Therefore, the hypothesis of

sperm competition cannot adequately explain the dif- ferences in die-off between the island populations and

strongly suggests that habitat factors are important in- fluences on male die-off patterns observed in the pre- sent study.

Prior to this study, the possibility that male die-off in a facultative species may depend on the availability of resources had been proposed (Braithwaite and Grif- fiths 1994, Mills and Bencini 2000), but not investi-

gated. We have identified differences in habitat be- tween Boullanger and Whitlock Islands, including sea-

bird density and differences in plant-available nutrients in the soil. We have also provided evidence that strong- ly suggests an association between these habitat dif- ferences and the body condition and age class structures for male dibblers on each island.

While there are limitations and a degree of error in

using a body condition index when comparing two

groups of animals (Krebs and Singleton 1993), in this

study the use of body condition indices supports the

hypothesis of a link between the availability of re- sources and the survival of male dibblers. Similar to the effect that seabirds have on the nutrient cycles and food webs of other islands (Polis and Hurd 1995, 1996, Anderson and Polis 1998, Stapp et al. 1999, Sainchez- Pifiero and Polis 2000, Stapp and Polis 2003), the ma- rine-derived nutrient inputs transported by seabirds ap- pear to play an important role in the island ecosystem of Whitlock Island by indirectly increasing the chances of post-mating survival of male dibblers. Invertebrate orders consumed by dibblers were greater in areas where seabird density is greatest on Whitlock Island

(Miller 2000). This is a similar finding to previous studies, for example Polis and Hurd (1995), who found that on islands where seabirds were breeding on only part of the island, spider densities were 12 times greater within seabird colonies compared to areas away from the colonies. Sainchez-Pifiero and Polis (2000) also re-

ported that tenebrionid beetles were six times denser within than away from seabird colonies on the same island.

While a large difference in soil nutrients (and there- fore productivity) seems a plausible explanation for the differences in body condition and post-mating survival of male dibblers on Boullanger and Whitlock Islands, other differences between the two islands may also play a role. Small populations of King's skinks (Egernia kingii) and the Boullanger Island dunnart (Sminthopsis griseoventer boullangerensis) are found on Boullanger Island, but not on Whitlock Island. The diets of dun- narts and juvenile King's skinks are similar to that of the dibbler (Dickman 1988, Bush et al. 1995), so food

competition with these species on Boullanger Island

may decrease the availability of food resources and

negatively affect the survival of dibblers. This extra element of food competition on Boullanger Island would further reduce available resources for dibblers when compared to neighboring Whitlock Island and

may contribute to the lower body condition scores of

male dibblers leading into the breeding season. How- ever, we consider this unlikely, since the population densities of these two species are very low (trap success rates of 1.6% for King's skinks and 0.04% for dun- narts).

Further support for a link between facultative die- off and resource availability comes from a study on the

June 2004 NOTES 1745

northern quoll (Dasyurus hallucatus), a species closely related to the dibbler. This species exhibits differences in male die-off between two environments in northern Australia (Braithwaite and Griffiths 1994). Males sur- vive longer, have a greater population density, and bet- ter physiological condition in the more productive rocky country (Freeland et al. 1988) than males found in relatively less productive arid savannah (Braithwaite and Griffiths 1994). This pattern, which is very similar to that of the two island populations of dibblers, sug- gests that habitat factors influence male die-off in the northern quoll.

Availability of resources is greater on islands with

higher densities of seabirds. This study strongly sug- gests that resource availability is also a major factor

determining facultative die-off in island populations of dibblers and is likely to influence facultative male die- off in other marsupials. The next step would be to test this resource availability hypothesis experimentally, for example a transplant experiment or supplementing nutrients on Boullanger Island. However, due to the

fragile nature of the islands and the high conservation status of the dibbler, at present this is not possible.

Males in captivity survive, produce sperm, and sire litters up into their third year (Mills and Bencini 2000). This study suggests that dibblers on Whitlock Island are potentially iteroparous, while dibblers on Boullan-

ger Island are potentially semelparous, a life history strategy that is imposed by limited resources within the environment.

ACKNOWLEDGMENTS

Soil nutrients were analyzed by CSBP and Farmers Limited Perth (Future Farm). Associate Professor Ron Wooller pro- vided helpful comments on an earlier draft of the manuscript. Emeritus Professor David Lindsay, Professor Graeme Martin, Mr. Craig MacFarlane, and Dr. Leigh Simmons also provided helpful comments on later drafts of the manuscript.

LITERATURE CITED

Anderson, W. B., and G. A. Polis. 1998. Marine subsidies of island communities in the Gulf of California: evidence from stable carbon and nitrogen isotopes. Oikos 81:75-80.

Anderson, W. B., and G. A. Polis. 1999. Nutrient fluxes from water to land: seabirds affect plant nutrient status on Gulf of California islands. Oecologia 118:324-332.

Bencini, R., C. McCulloch, H. R. Mills, and A. N. Start. 2001. Habitat and diet of the dibbler (Parantechinus apicalis) on two islands in Jurien Bay, Western Australia. Wildlife Re- search 28:465-468.

Blair, G. J., N. Chinoim, R. D. B. Lefroy, G. C. Anderson, and G. J. Crocker. 1991. A soil sulfur test for pastures and crops. Australian Journal of Soil Research 29:619-626.

Bradley, A. J., I. R. McDonald, and A. K. Lee. 1980. Stress and mortality in a small marsupial (Antechinus stuartii, Macleay). General and Comparative Endocrinology 40: 188-200.

Braithwaite, R. W., and A. D. Griffiths. 1994. Demographic variation and range contraction in the northern quoll, Das-

yurus hallucatus (Marsupialia: Dasyuridae). Wildlife Re- search 21:203-217.

Bush, B., B. Maryan, R. Browne-Cooper, and D. Robinson. 1995. Reptiles and frogs of the Perth Region. University of Western Australia Press, Perth, Western Australia, Aus- tralia.

Chalmers, C. E., and S. M. Davies. 1984. Coastal manage- ment plan-Jurien Bay area. Bulletin Number 176. De-

partment of Conservation and Environment, Perth, Western Australia, Australia.

Cockburn, A. 1997. Living slow and dying young: senes- cence in marsupials. Pages 163-174 in N. R. Saunders and L. A. Hinds, editors. Marsupial biology: recent research, new perspectives. University of New South Wales Press, Sydney, New South Wales, Australia.

Dickman, C. R. 1988. Body size, prey size, and community structure in insectivorous mammals. Ecology 69:569-580.

Dickman, C. R. 1993. Evolution of semelparity in male das-

yurid marsupials: a critique and an hypothesis of sperm competition. Pages 25-38 in M. Roberts, J. Carnio, G. Crawshaw, and M. Hutchins, editors. The biology and man- agement of Australasian carnivorous marsupials. American Association of Zoological Parks and Aquariums, Mono- treme and Marsupial Advisory Group, Toronto, Ontario, Canada.

Dickman, C. R., and R. W. Braithwaite. 1992. Postmating mortality of males in the dasyurid marsupials, Dasyurus and Parantechinus. Journal of Mammalogy 73:143-147.

Dunlop, J. N., and R. D. Wooller. 1990. The breeding seabirds of southwestern Australia: trends in species, populations and colonies. Corella 14:107-112.

Ford, J. 1963. The reptilian fauna of the islands between Dongara and Lancelin, Western Australia. Western Austra- lian Naturalist 8:135-142.

Freeland, W. J., J. W. Winter, and S. Raskin. 1988. Australian rock-mammals: a phenomenon of the seasonally dry trop- ics. Biotropica 20:70-79.

Fuller, P. J., and A. A. Burbidge. 1987. Discovery of the dibbler, Parantechinus apicalis, on islands at Jurien Bay. Western Australian Naturalist 16:177-181.

Garkaklis, M. J., J. S. Bradley, and R. D. Wooller. 2003. The relationship between animal foraging and nutrient patchi- ness in southwest Australian woodland soils. Australian Journal of Soil Research 41:665-673.

Gillham, M. E. 1960. Destruction of indigenous heath veg- etation in Victorian sea-bird colonies. Australian Journal of Botany 8:278-317.

Iason, G. R., C. D. Duck, and T. H. Clutton-Brock. 1986.

Grazing and reproductive success of red deer: the effect of local enrichment by gull colonies. Journal of Animal Ecol-

ogy 55:507-515. Johnstone, R. E., and G. M. Storr. 1998. Handbook of West-

ern Australian birds. Volume 1. Western Australian Mu- seum, Perth, Western Australia, Australia.

Krebs, C. J., and G. R. Singleton. 1993. Indices of condition for small mammals. Australian Journal Zoology 41:317- 323.

LeCren, E. D. 1951. The length-weight relationship and sea- sonal cycle in gonad weight and condition in the perch (Perca fluviatilis). Journal of Animal Ecology 20:201-219.

Lee, A. K., and A. Cockburn. 1985. Evolutionary ecology of marsupials. Cambridge University Press, Cambridge, UK.

Lee, A. K., and I. R. McDonald. 1985. Stress and population regulation in small mammals. Pages 261-304 in C. A. Finn, editor. Oxford reviews of reproductive biology. Volume 7. Clarendon Press, Oxford, UK.

1746 NOTES Ecology, Vol. 85, No. 6

Miller, S. 2000. Food availability on Boullanger and Whit- lock Islands for the dibblers, Parantechinus apicalis. The- sis. University of Western Australia, Perth, Western Aus- tralia, Australia.

Miller, S., R. Bencini, H. Mills, and D. Moro. 2003. Food availability for the dibbler, Parantechinus apicalis, on Boullanger and Whitlock Islands, Western Australia. Wild- life Research 30:649-654.

Mills, H. R., and R. Bencini. 2000. New evidence for fac- ultative male die-off in island populations of dibblers, Par- antechinus apicalis. Australian Journal of Zoology 48:501- 510.

Oakwood, M., A. J. Bradley, and A. Cockburn. 2001. Se- melparity in a large marsupial. Proceedings of the Royal Society of London B 268:407-411.

Polis, G. A., and S. D. Hurd. 1995. Extraordinarily high spider densities on islands: flow of energy from the marine to terrestrial food webs and the absence of predation. Pro- ceedings of the National Academy of Science USA 92: 4382-4386.

Polis, G. A., and S. D. Hurd. 1996. Linking marine and terrestrial food webs: allochthonous input from the ocean supports high secondary productivity on small islands and coastal land communities. American Naturalist 147:396- 423.

Rayment, G. E., and E R. Higginson. 1992. Australian lab- oratory handbook of soil and water chemical methods. In- kata Press, Melbourne, Victoria, Australia.

Ryan, P. G., and B. P. Watkins. 1989. The influence of phys- ical factors and ornithogenic products on plant and arthro-

pod abundance at an inland nunatak group in Antarctica. Polar Biology 10:151-160.

Sanchez-Pifiero, E, and G. A. Polis. 2000. Bottom-up dy- namics of allochthonous input: direct and indirect effects of seabirds on islands. Ecology 81:3117-3132.

Searle, P. L. 1984. The Bethelot of indophenol reaction and its use in the analytical chemistry of nitrogen. A review. Analyst 109:549-568.

Short, J., and A. Smith. 1994. Mammal decline and recovery in Australia. Journal of Mammalogy 75:288-297.

Stapp, P., and G. A. Polis. 2003. Marine resources subsidize insular rodent populations in the Gulf of California, Mex- ico. Oecologia 134:496-504.

Stapp, P., G. A. Polis, and E Sinchez-Pifiero. 1999. Stable isotopes reveal strong marine and El Niflo effects on island food webs. Nature 401:467-469.

Tabachnick, B. G., and L. S. Fidell. 1996. Using multivariate statistics. Harper Collins, New York, New York, USA.

Woolley, P. 1966. Reproduction in Antechinus spp. and other dasyurid marsupials. Symposium of the Zoological Society of London 15:281-294.