Food habits and resource partitioning between grey and culpeo foxes in southeastern Argentine Patagonia SONIA C. ZAPATA1, ALEJANDRO TRAVAINI1, MIGUEL DELIBES,2 & ROLANDO MARTÍNEZ-PECK1 1 Universidad Nacional de la Patagonia Austral, Puerto Deseado, Santa Cruz, Argentina, and 2Estación Biológica de Doñana (CSIC), Sevilla, Spain Abstract We studied the annual and seasonal food habits of sympatric grey fox (Lycalopex griseus) and culpeo fox (Lycalopex culpaeus) in a protected area, in southeastern Argentine Patagonia, to test the hypothesis that both foxes partition prey. Grey foxes consumed a larger proportion of rodents than culpeos (66.1% versus 37. % of occurrences in feces, respectively), and culpeos consumed a larger proportion of the introduced European hare (Lepus europaeus) (32.8% versus 7.1%, respectively). Additionally, arthropods made up a significant portion of the diet of the grey fox, and occurred in the 95.5% of the analyzed feces. Despite the great differences in size of culpeo and grey foxes, the former being significantly larger, both species consumed the same type of prey, and no significant differences in the mean weight of vertebrate prey were detected in their diets. Culpeo and grey foxes diets differed during spring – summer with low values of mean diet overlap (31%). But in autumn – winter, when prey diversity is lower in Patagonia, diet was more similar and diet overlap (73%) increased. An exploratory analysis of habitat use by culpeo and grey foxes suggested that both species utilize habitat in a similar way. Our data agree with generalizations about competitive interactions between the large and the small canid in an assemblage. Although both foxes can potentially consume the same type of prey, through interference the larger culpeo would exclude the smaller grey fox from habitats with high-quality prey, resulting in prey partitioning. The possibility that both foxes partition habitat at a finer scale in our study area should be explored. Keywords: Lycalopex griseus, L. culpaeus, diet, resource partitioning, Patagonia, Argentina Introduction The small grey fox (Lycalopex griseus) is found in central Chile and in Argentina along the Andes, from at least Salta province south to Rı́o Negro province and then across the whole country south to Tierra del Fuego (Redford & Eisenberg 1992; Novaro 1997). The culpeo fox (Lycalopex culpaeus), the largest Neotropical canid after the maned wolf (Chrysocyon brachyurus), ranges from northern Ecuador to southern Chile and Argentina, along the foothills of the Andes, including Tierra del Fuego, and throughout the Patagonian steppe of Argentina (Redford & Eisenberg 1992; Novaro 1997). However, their distributions do not completely overlap and few cases of sympatry were reported in Chile and in Argentina (see Jimé nez et al. 1996 and references therein). Nevertheless, in central-eastern Santa Cruz province, southern Argentine Patagonia, both species co-occur (Travaini et al. 2000, 2001). Early accounts indicated that culpeo foxes were found only along the Andean Cordillera and its foothills during the early 1900s (Prichard, 1902 in Novaro 1997), but at present they extend their range to the coast of Patagonia south to 438S (Novaro 1997). The increase in the culpeo range may be related to the introduction of European hares (Lepus europaeus) and sheep (Ovis aries) in the early 1900s (Crespo & De Carlo 1963; Griguera & Rapopport 1983). In their generalization of competitive interactions between the largest and the smallest canid in an assemblage, Johnson et al. (1996) argued that Correspondence: S. C. Zapata, Centro de Investigaciones de Puerto Deseado, Universidad Nacional de la Patagonia Austral, Unidad Acadé mica de Caleta Olivia, Av. Lotufo s/n. 9050- Puerto Deseado, Santa Cruz, Argentina. Fax: + 54 297 4870511. E-mail: titinazapata@yahoo.com.ar although the mechanisms for partitioning resources are complex, where little diet overlap occurs the two canids usually overlap spatially, and may or may not use different habitats. But when food habits are more similar between the two canid species, then interference competition is common, and the larger species tends to be distributed wherever there are sufficient food resources to meet its energy requirements and there are no other limiting factors. The smaller species in this scenario tends to occupy adjacent areas offering lower amounts of food resources, and thus often appears to be selecting different types of habitats or food. Sympatry in culpeo and grey foxes has been studied in central and southern Chile (Fuentes & Jaksic 1979; Jaksic et al. 1980, 1983; Johnson & Franklin 1994a, 1994b; Jiménez et al. 1996). According to Johnson et al. (1996), although both foxes can potentially consume the same type of prey, through interference the larger culpeo excludes the smaller grey fox from habitats with high-quality prey, where the introduced European hare is abundant, resulting in differences in feeding habits and habitats used. We studied the annual and seasonal food habits of sympatric grey and culpeo foxes in a protected area, the Monumento Natural Bosques Petrificados National Park (MNBP), southeastern Argentine Patagonia, to test the hypothesis that both foxes coexist by partitioning prey. According to Johnson and Franklin (1994a) and Jimé nez et al. (1996) we expect that although the two species are able to consume the same type of prey, culpeo fox will consume more energetically rewarding foods. We also present an exploratory analysis of habitat use of culpeo and grey foxes. Materials and methods Study area We searched for feces of culpeo and grey foxes in MNBP (47839.887’S; 67859.729’W), a protected area covering about 60 000 ha belonging to the Departamento Deseado, Santa Cruz province, Patagonia, in southern Argentina. Annual rainfall ranged between 100 and 300 mm. Summer temperatures averaged 178C, and winters were comparatively mild. Most of the country was covered by tussock grasses and low, dome shaped, spindly shrubs (Soriano 1983), with cover ranging from less than 10% in the most arid areas to 4 60% (Ares et al. 1990; Bertiller & Bisigato 1998). Diet analysis We collected a total of 316 feces of culpeo during spring – summer (September 1999 to March 2000, n = 166) and autumn – winter (May to August 2000, n = 150) and a total of 203 feces of grey foxes (n = 127 in spring-summer and n = 76 in autumn[-winter) by walking secondary roads. In our study area, culpeo foxes are significantly heavier and larger than grey foxes. The mean weight for culpeos is 8.03 kg and mean total length 1228.8 mm (n = 11), while for grey foxes these values are 2.6 kg and 867.6 mm (n = 17), respectively. Their feces are easily distinguished because those from culpeos are at least twice as large as those of grey foxes (diameter: x = 23.5 mm + 2.04 SD, n = 13, vs. x = 12.3 mm+ 2.29 SD, n = 20) and because of the particular pattern of scat deposition for each species. Feces of culpeos were found spread on bare soil in the roads and those of the grey foxes beside shrubs. Only fresh fecal samples were collected and care was taken to avoid collection of more than one scat that could have been from the same individual in a couple of days. We discarded feces that could not be attributed to either species. Feces were air-dried, their components separated by hand and identified using a reference collection (birds, small mammals) and a key to small mammals skulls (Pearson 1995). Mammal hairs were identified from medulla types and scale patterns following the methodology of Brunner and Coman (1974) and the keys of Chehé bar and Martin (1989). Birds were classified into large birds ( 4 1000 g), e.g. the lesser rhea (Pterocnemia pennata), and the upland goose (Chloephaga picta), medium-sized birds (200 – 1000 g), e.g. the elegant crested-tinamou (Eudromia elegans) and small birds ( 5 200 g) were all passerine birds. Reptiles were discriminated to the genus level. Scorpions, coleopterans and their larvae were grouped as arthropods. We present results as frequency of occurrence in feces of each vertebrate, arthropod prey and fruits. We tested for seasonal differences in prey items for each fox species. Occurrences of food items were grouped into nine classes: small rodents, large rodents, edentates, lagomorphs, other mammals, carrion (prey items too large to have been killed by the foxes), birds/ reptiles, arthropods and fruits. Then these categories were compared by season through a contingencytable analysis (G-test, Sokal & Rohlf 1981). We tested for interspecific differences in occurrence of vertebrate prey items in feces by chi-square analysis and then partitioning of chi-square contingency tables were made after Siegel and Castellan (1988). For this purpose, vertebrate prey items were grouped into five categories: rodents, edentates, lagomorphs, carrion and birds plus reptiles. To describe and compare the diets of both foxes we also used the parameters proposed by Jaksic et al. (1986): (1) geometric mean weight of vertebrate prey based on weight values of small mammals obtained from Pearson (1995), Redford and Eisenberg (1992) and for edentates, lagomorphs, carnivores, birds and reptiles obtained by us in the field. We did not include the guanaco (Lama guanicoe) because we considered it to be scavenged. (2) We calculated Colwell and Futuyma’s (1971) standardized form (Bsta) of Levins’ index (B) to measure food-niche breadth (Levins, 1968), an index of diet diversity: Bsta = (Bobs - Bmin)/(Bmax - Bmin), where Bobs is the observed niche breadth, Bmin = 1, is the minimum niche breadth, and Bmax is the maximum possible niche breadth (the number of taxa taken). Bobs is = 1/S pi2, where pi is the relative occurrence of prey taxon i in the diet. The Bsta allows comparison of diets with different numbers of prey categories and ranges from 0 to 1 (i.e. from narrow to broad food niche). We measured food niche breadth for each species of fox at the finest level of taxonomic resolution of prey categories. (3) Overlap in food niche, a measure of dietary similarity, was determined using Pianka’s (1963) index, where o = S pij pik /(S pij2 S pik 2) 1/2. This index ranges from 0 to 1 (i.e. from dissimilarity to complete similarity). Analysis of habitat use Habitat use data were derived by combining the data taken from Travaini et al. (2001), of the numbers of visits of the two fox species to bait stations, and environmental data taken at each station. Bait stations were established in two areas of the Monumento Natural Bosques Petrificados: El Cuadro (September 1999) and Cerro Horquetas (April 2000). At each area, we established 9 – 10 transects, each with six bait stations spaced 0.5 km apart, and transects were at least 1 km apart. On the first day, we established the stations on alternate sides of secondary unpaved roads. On the three following mornings, we checked the stations and identified tracks of station visitors (see Travaini et al. 2001 for details). We included in this approach only the stations visited by foxes (n = 32). No station included the visit of individuals of the two species simultaneously. To describe habitat characteristics within a radius of 200 m around each bait station we estimated the proportion of coverage (in %) by the following classes of vegetation and other features: (1) low cover: vegetation less than 20 cm high; (2) medium cover: vegetation between 20 cm and 2 m high; (3) high cover: vegetation more than 2 m high; (4) bare soil: without any plant cover; (5) free water: flooded land; (6) aspect class: presence of herbaceous steppe, shrub steppe, or bare soil; (7) road: type of road, path between ranches or little used path; (8) topography: small lake border, ravines, slopes, or pampas. Then, to explore the relationship between the presence of a fox species and the range of environ- mental variables described above, a GLM model was fitted with the program GLIM (Baker 1987). Because the response variable was presence or absence of a fox species we considered that error would follow a binomial distribution and used a logit link (Crawly 1993). We started by fitting a null model so that the significance of the variables could be assessed by comparing the changes in the deviance from the null model to the model with the variable fitted (Crawly 1993). The changes in the deviance were compared with chi-square statistics. Results Seasonal differences in diets of Culpeo and Grey foxes Mammals were the main prey items in the diet of Lycalopex culpaeus in MNBP, and among these, at least six species of rodents were very important in the diet, chiefly lesser cavies (Microcavia australis) (Table I). Introduced European hares (Lepus europaeus) were also important in culpeo diet. Mammals like the pichi (Zaedyus pichiy) and the guanaco (Lama guanicoe) were moderately frequent in the feces and other carnivores were only occasionally consumed (Table I). Birds were present in 10.7% of the feces (Table I). Fruits of both native (Berberis buxifolia, Schinus molle) and introduced (apple, quince, prune) species were found in the feces while lizards and arthropods were poorly represented (Table I). Foods items recorded in feces of Lycalopex griseus included 11 species of mammals, birds, reptiles, arthropods and fruits (Table I). Arthropods and mammals, mainly rodents, were the most frequent items in grey fox diet. Among the rodents, the lesser cavy (M. australis), Akodon spp, Maguellan’s tuco (Ctenomys magellanicus) and unidentified rodents were the most frequent (Table I). Other mammals (edentates, lagomorphs, carnivores and carrion) as well as birds, reptiles and fruits, were poorly represented in grey fox feces (Table I). The standardized food niche breadth values calculated for culpeo diet during spring – summer and autumn – winter were very similar (0.27 versus 0.24, respectively). Nevertheless, birds, arthropods and fruits were more frequently consumed during spring – summer (Table I). Consequently, there were qualitative seasonal differences in the culpeo diet (Gtest, G = 21.5, df = 8, p 5 0.01). Though culpeos consumed mammals in similar proportions in the two periods studied (Table I), the geometric mean weight of vertebrate prey was higher during autumn – winter probably due to the greater consumption of European hares during these seasons (Table I). Significant seasonal differences in prey composition were detected in grey fox diet (G-test, G = 22.3, Table I. Annual and seasonal diets of culpeo (Lycalopex culpaeus) and grey foxes (Lycalopex griseus) in Monumento Natural Bosques Petrificados (MNBP) expressed as the frequency of occurrence (%) of vertebrates, arthropods and fruits found in the feces. Spring – summer: September to March; autumn – winter: May to August. Bold numbers = broad taxonomic groups. Culpeo fox Prey items Spring – summer Autumn – winter Mammals 84.4 Edentates 6.5 Chaetophractus villosus 0.5 Zaedyus pichiy 6.0 Rodents 40.2 Akodon spp 8.5 Eligmodontia typus 1.01 Phyllotis darwinii 0.5 Reithrodon auritus 3.5 Microcavia australis 19.6 Ctenomys magellanicus 2.5 Unknown rodent 4.5 Lagomorphs 32.1 (Lepus europaeus) Carnivores 1.5 Galictis cuja – Conepatus humboldtii 1.5 Felis catus – Carrion 4.0 (Lama guanicoe) Birds 12.0 (Pterocnemia pennata, and Chloephaga picta) 6.0 Eudromia elegans 2.5 Passerine birds 3.5 Bird eggs 2.0 Reptiles 1.5 (Liolaemus spp.) ARTHROPODS 12.0 FRUITS 36.7 GARBAGE 1.2 Number of feces 166 Food niche breadth1 0.27 Food niche breadth1 (mean + standard error) 0.25 + 0.09 Geometric mean weight of vertebrate prey (g) 472.5 Grey fox Annual Spring – summer Autumn – winter 86.5 7.3 0.3 7.0 37.7 5.5 0.9 0.3 3.0 20.8 1.8 5.2 32.8 92.7 10.0 – 10.0 71.8 25.4 5.4 – 2.7 19.0 8.1 10.9 3.6 89.0 12.3 – 12.3 57.5 9.5 4.1 1.3 – 20.5 9.5 12.3 12.3 91.2 10.9 – 10.9 66.1 19.1 4.9 0.5 1.6 19.6 8.7 11.4 7.1 2.4 0.8 0.8 0.8 7.3 1.8 0.3 1.2 0.3 6.7 1.8 – 1.8 – 5.4 1.3 1.3 – – 5.4 1.6 0.5 1.0 – 5.4 9.0 2.4 4.9 1.6 0.8 0.8 10.7 4.6 3.3 2.7 1.5 1.2 3.6 – 0.9 2.7 – 3.6 10.9 – 1.3 9.5 – – 6.5 – 1.0 5.4 – 2.1 89.3 9.02 – 9.0 35.2 0.8 0.8 – 2.4 23.7 0.8 6.5 35.2 2.6 10.6 2.0 150 0.24 7.6 24.3 1.5 316 717.1 100 1.5 – 127 0.5 0.58 + 0.15 104.6 57.8 – – 76 0.66 Annual 95.5 0.9 – 203 257.7 1 Colwell and Futuyma (1971) standardized food niche breadth. df = 6, p 5 0.01), with an increase in lagomorphs and birds during the cold season, and a decrease in the consumption of rodents and arthropods. Both standardized food niche breadth and geometric mean weight of vertebrate prey were higher during autumn – winter (Table I). Interspecific differences in the diet of Culpeo and Grey foxes Although the vertebrate prey types eaten by both culpeo and grey foxes were the same, diets were significantly different (w2 = 108.04, df = 4, p 5 0.001). Partitioning the degrees of freedom of the chi-square table revealed that grey foxes consumed higher proportions of rodents (w2 = 4.6, df = 1, p 5 0.05) and arthropods but lower propor- tions of European hares (w2 = 21.4, df = 1, p 5 0.01) and fruits (w2 = 633.12, df = 1, p 5 0.001). Grey foxes had higher seasonal indices of standardized food-niche breadths than culpeo foxes for both seasons and annual mean (these values are in Table I). Although we did not find significant differences in the mean weight of vertebrate prey taken by the two species (Kruskal – Wallis, H = 1.15, p = 0.76) the geometric mean weight was lower for grey foxes (Table I). Diet overlap between culpeo and grey foxes were low during spring – summer (0.31) and high during autumn – winter (0.73). Habitat use The exploratory analysis of habitat use by culpeo and grey foxes showed that none of the eight Table II. GLM model to explore the relationship between the presence of grey and culpeo foxes and eight environmental variables using binomial erros and logit link. None of the variables analyzed resulted significant enough to explain the presence of any of the two species. Variables Low vegetation coverage (%) Medium vegetation coverage (%) High vegetation coverage (%) Bare soil (%) Free water (%) Aspect class Road Topography Reduction Reduction in deviance in df 1.627 0.187 2.858 3.358 1.285 4.22 2.489 2.376 1 1 1 1 1 2 2 3 P ns ns ns ns ns ns ns ns Null model: deviance = 38.024, df = 31. Significance of each environmental variable was assessed by fitting to the null model and computing the changes in the deviance. Differences in deviance were compared with chi-square statistics, the degrees of freedom are shown in the table. environmental variables analyzed significantly explained the presence of either species (Table II), suggesting that both foxes utilize the habitat in a similar way in Bosques Petrificados. Discussion Feeding ecology of culpeo and grey foxes Our data on the diet of culpeo and grey foxes agree with the general pattern summarized in Medel and Jaksic (1988) and characterize them as opportunistic predators. However, they can be locally selective for certain prey as in southern Chile where European hares were the main item for both culpeo and grey foxes (Johnson & Franklin 1994a). In our study area foxes still fed mainly on native prey as do all the other carnivore species of the assemblage (Zapata et al. 2002), whereas in northwestern Argentinean Patagonia, culpeos nowadays rely primarily on hares and grey foxes on carrion of introduced ungulates (mainly Ovis aries) (Novaro et al. 2000). In our study area and its surroundings, as in most parts of Santa Cruz province, sheep ranging was abandoned about 15 years ago (Travaini et al. 2000), allowing wildlife to restore their numbers and inter-specific interactions to a status similar to that before European settlement. During spring and summer, arthropods were a significant portion in the diet of grey foxes, and fruits in that of culpeos. As in northwestern Argentine Patagonia (Zapata et al. 1998), this study also revealed an increased consumption of arthropods by grey foxes during the reproductive season (November – January) and this finding sustains earlier conclusions that adult foxes eat the smaller and less rewarding prey on the spot and feed their cubs with the relatively bulkier items that can be carried away. This opportunistic way of exploiting resources is consistent with the central place foraging theory (CPF, Orians & Pearson 1979). Frugivory by grey foxes was reported in Patagonia (Zapata et al. 1998), but in Bosques Petrificados only culpeos feed on fruits. This is in contrast to other studies that reported increasing dietary overlap between canid when fruits were available, for example red foxes (Vulpes vulpes) and coyotes (Canis latrans) (Major & Sherburne 1987), north American grey foxes (Urocyon cinereoargenteus) and coyotes (Neale & Sakcs 2001), and maned wolves (Chrysocyon brachyurus) and crab-eating foxes (Cerdocyon thous) (Juarez & Marinho-Filho 2002). Food partitioning As expected, grey and culpeo foxes consumed the same type of prey, and no significant differences in the size of vertebrate prey species were detected. Nevertheless, they differed in the proportions of vertebrate prey in different size classes. The same pattern was reported for sympatric foxes at Aucó, in Central Chile (Jimé nez et al. 1996): European hares were more abundant than rodents in the diet of culpeo foxes and rodents in grey fox diets. Nevertheless, in Torres del Paine, southern Chile, the European hare was the most common vertebrate in both foxes diets, but culpeo ate higher proportions of these lagomorphs than grey foxes (Johnson & Franklin 1994a). This may be related to the higher abundance of European hares in Torres del Paine than in Aucó (Jimé nez et al. 1996). Rodents occurred in a similar proportion in both foxes diet and carrion was important in the diet of the grey fox (Johnson & Franklin 1994a). Culpeo and grey fox diets were different in Bosques Petrificados during spring – summer with low value of mean diet overlap (31%), but in autumn – winter diets of the foxes became more similar and an increase in mean diet overlap (73%) between them was observed. Like in northern latitudes (Litvaitis 1992), prey diversity is limited in winter in Patagonia (Terborgh 1992), which can lead to an increase in dietary overlap between sympatric predators. Arjo et al. (2002) stated that carnivore species can coexist with high overlap if diversity is greater in one species’ diet (White et al. 1995), if prey are partitioned by size (Rosenzweigg 1966) or differential use (Mills 1984), or if prey are abundant. Although mean diet overlap is high in winter in Bosques Petrificados, coexistence of both foxes can be explained by the diversity of their diets, which was different, and by the standardized food niche breadth that was higher in grey foxes diet. Finally, because they partition prey by size, although both foxes prey upon rodents and lagomorphs (the most rewarding prey) in a similar way, they eat different proportions of these vertebrate prey. Accordingly, the hypothesis that both foxes coexist by partitioning prey would be valid in Bosques Petrificados. Habitat partitioning We did not find significant differences in habitat use by both foxes in Bosques Petrificados, possibly due to our small sample size. On the other hand, our study area is more homogeneous than those studied in Chile and we must test whether the foxes partition habitats at a finer scale in Bosques Petrificados than in other places. A previous study in Bosques Petrificados (Travaini et al. 2001) showed that only one fox species (i.e. grey or culpeo fox) visited bait stations during the first night after they were established. No station was visited simultaneously by both fox species suggesting avoidance and probably habitat partitioning at a finer scale. Acknowledgments We thank C. Zoratti, F. Escobar, P. Collavino, G. Aguilera and G. Soria for all their help during field work and C. Chehebar and the Administración de Parques Nacionales for authorizing us to work inside the protected area. Funds were provided by the Universidad Nacional de la Patagonia Austral (P.I. 29/B013, 29/B050), the Secretarı́a de Agricultura, Ganaderı́a y Pesca de la Nació n (SAGPyA Nos. 613/ 96, 716/97 and 925/97) and the Consejo Agrario Provincial. Javier Bustamante helped SCZ with the GLM model and Pablo Ferreras, Hubert Hendrichs and two anonymous reviewers substantially improved a previous version of this manuscript. References Ares J, Beeskow AM, Rostagno CM, Irisarri MP, Anchorena J, Defossé GE, Meroni CA. 1990. Structural and dynamic characteristics of overgrazed lands of northern Patagonia, Argentina. In: Bremeyer A, editor. Managed Grasslands: Regional Studies. Amsterdam: Elsevier Science. pp. 149 – 175. Arjo WM, Pletscher DH, Ream RR. 2002. Dietary overlap between wolves and coyotes in northwestern Montana. J Mammal 83:754 – 766. Baker RJ. 1987. GLIM 3.77 Reference Manual, 2nd ed. Oxford: Numerical Algorithms Group. 379 pp. Bertiller MB, Bisigato A. 1998. Vegetation dynamics under grazing disturbance. The state-and-transition model for the Patagonia steppes. Ecol Austral 8:191 – 199. Brunner H, Coman BJ. 1974. The Identification of Mammalian Hair. Melbourne: Inkata Press. 407 pp. Chehé bar C, Martı́n S. 1989. Guı́a para el reconocimiento microscó pico de los pelos de los mamı́feros de la Patagonia. Doñ ana Acta Vert 16:247 – 293. Colwell RR, Futuyma DJ. 1971. On the measurement of niche breadth and overlap. Ecology 52:567 – 572. Crawly MJ. 1993. GIM for Ecologists. Methods in ecology. Oxford, Blackwell. 379 pp. Crespo JA, De Carlo JM. 1963. Estudio ecoló gico de una població n de zorros colorados Dusicyon culpaeus culpaeus en el oeste de la provincia de Neuquén. Rev Mus Arg Cien Nat ‘‘Bernardino Rivadavia’’ 1:1 – 155. Fuentes ER, Jaksic FM. 1979. Latitudinal size variation of Chilean foxes: tests of alternative hypotheses. Ecology 60:43 – 47. Grigera DE Rapoport EH. 1983. Status and distribution of the European hare in South America. J Mammal 64:163 – 166. Jaksic FM, Schlatter RP, Yañ ez JL. 1980. Feeding ecology of central Chilean foxes. Dusicyon culpaeus and Dusicyon griseus. J Mammal 61:254 – 260. Jaksic FM, Yañ ez JL, Rau JR. 1983. Trophic relations of the southernmost populations of Dusicyon in Chile. J Mammal 64:697 – 700. Jaksic FM, Yañ ez JL, Rau JR. 1986. Prey and trophic ecology of Great Horned Owls in western South America: an indication of latitudinal trends. J Raptor Res 20:113 – 116. Jiménez JE, Yañ ez JL, Tabilo EL, Jaksic FM. 1996. Nichecomplementarity of South American foxes: reanalysis and test of a hypothesis. Rev Chil Hist Nat 69: 113 – 123. Johnson WE, Franklin WL. 1994a. The role of body size in the diets of sympatric grey and culpeo foxes. J Mammal 75:163 – 174. Johnson WE, Franklin WL. 1994b. Spatial resource partitioning by sympatric grey fox (Dusicyon griseus) and culpeo fox (Dusicyon culpaeus) in southern Chile. Can J Zool 72:1788 – 1793. Johnson WE, Fuller TK, Franklin WL. 1996. Sympatry in canids: a review and assessment. In: Gittleman JL, editor. Carnivore Behaviour, Ecology and Evolution, Vol. 2. Ithaca: Cornell University Press. pp. 189 – 219. Juá rez KM, Marinho-Filho J. 2002. Diet, habitat use, and home range of sympatric canids in central Brazil. J Mammal 83:925 – 933. Levins R. 1968. Evolution in changing environments. Princeton: Princeton University Press. 120 pp. Litvaitis JA. 1992. Niche relations between coyotes and sympatric carnivora. In: Boer AH, editor. Ecology and Management of the Eastern Coyote. Canada: Wildlife Research Unit, University of New Brunswick. pp. 73 – 85. Major JT, Sherburne JA. 1987. Interspecific relationships of coyotes, bobcats and red foxes in western Maine. J Wildl Manage 51:606 – 616. Medel RG, Jaksic FM. 1988. Ecologı́a de los cánidos sudamericanos: una revisión. Rev Chil Hist Nat 61:67 – 79. Mills MGL. 1984. Prey selection and feeding habits of the large carnivores in the southern Kalahari. Koedoe 27 supplement:281 – 294. Neale J, Sakcs BN. 2001. Food habits and space use of grey foxes in relation to sympatric coyotes and bobcats. Can J Zool 79:1794 – 1800. Novaro AJ. 1997. Pseudalopex culpaeus. Mammalian Species 558: 1 – 8. Novaro AJ, Funes MC, Walker RS. 2000. Ecological extinction of native prey of a carnivore assemblage in Argentine Patagonia. Biol Conserv 92:25 – 33. Orians GH, Pearson NE. 1979. On the theory of central place foraging. In: Horn DJ, Stairs GR, Mitchell RD, editors. Analysis of Ecological Systems. Columbus: Ohio State University Press. pp. 155 – 177. Pearson OP. 1995. Annotated keys for identifying small mammals living in or near Nahuel Huapi National Park or Lanı́n National Park, southern Argentina. Mastozool Neotrop 2:99 – 148. Pianka ER. 1963. The structure of lizard communities. Ann Rev Ecol Syst 4:53 – 74. Redford KH, Eisenberg JF. 1992. Mammals of the neotropics. The southern cone, Vol. 2, Chicago: The University of Chicago Press. 430 pp. Rosenzweigg ML. 1966. Community structure in sympatric carnivora. J Mammal 47:602 – 612. Siegel S, Castellan NJ. 1988. Nonparametric statistics for the behavioural sciences. New York: McGraw-Hill International Editions. 399 pp. Sokal RR, Rohlf FJ. 1981. Biometry. 2nd. edn. New York: W.H. Freeman and Co. 832 pp. Soriano A. 1983. Deserts and semi-deserts of Patagonia. In: West NE, editor. Temperate desert and semi-deserts. Amsterdam: Elsevier. pp. 440 – 454. Terborgh J. 1992. Diversity and the tropical rain forest. New York: Scientific American Library. 256 pp. Travaini A, Martı́nez-Peck R, Zapata SC. 2001. Selection of odour attractants and meat delivery methods to control Culpeo Foxes (Pseudalopex culpaeus) in Patagonia. Wildl Soc Bull 29: 1089 – 1096. Travaini A, Zapata SC, Martı́nez-Peck R, Delibes M. 2000. Percepció n y actitud humanas hacia la predació n de ganado ovino por el zorro colorado (Pseudalopex culpaeus) en Santa Cruz, Patagonia Argentina. Mastozool Neotrop 7:117 – 129. White PJ, Ralls K, Vanderbilt White. 1995. Overlap in habitat and food use between coyotes and San Joaquin kit foxes. Southwest Natur 40:342 – 349. Zapata SC, Travaini A, Delibes M. 1998. Neither large nor small: intermediate-sized food items for the Patagonian grey fox (Pseudalopex griseus). Can J Zool 76:2281 – 2284. Zapata SC, Travaini A, Delibes M. 2002. Relaciones troficas de un ensamble de carnı́voros en el sudeste de la Patagonia. XVII Jornadas Argentinas de Mastozoologı́a. Mar del Plata, Buenos Aires Argentina.