cronoecologia (2)

12
© 2011 ISZS, Blackwell Publishing and IOZ/CAS 375 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 doi: 10.1111/j.1749-4877.2011.00266.x Correspondence: Emerson M. Vieira, Departamento de Ecologia, Instituto de Ciências Biológicas, Universidade de Brasília, C.P. 04457, Brasília, DF, 70919-970, Brazil. Email: [email protected] Present address: Department of Ecology, University of Brasília (UnB), Brasília, Brazil. Integrative Zoology 2011; 6: 375-386 ORIGINAL ARTICLE Temporal niche overlap among insectivorous small mammals Emerson M. VIEIRA¹ and Gabriela PAISE² ¹Laboratory of Ecology of Mammals, Zoology, University of Vale do Rio dos Sinos, São Leopoldo, Brazil and ²Graduate Program of Ecology, University of Campinas, Campinas, Brazil Abstract Being active in the same environment at different times exposes animals to the effects of very different envi- ronmental factors, both biotic and abiotic. In the present study, we used live traps equipped with timing devices to evaluate the potential role of biotic factors (competition and food abundance) on overall overlap in the tem- poral niche axis of 4 insectivorous small mammals in high-elevation grassland fields (‘campos de altitude’) of southern Brazil. Based on resources availability (invertebrates), data on animal captures were pooled in 2 sea- sons: ‘scarcity’ (June 2001−September 2001) and ‘abundance’ (November 2001–May 2002) seasons. We tested for non-random structure in temporal niche overlap among the species in each season. These species were the rodents Oxymycterus nasutus (Waterhouse, 1837), Deltamys sp., Akodon azarae (Fischer, 1829), and the marsu- pial Monodelphis brevicaudis Olfers, 1818. The studied community was mainly diurnal with crepuscular peaks. Simulations using the Pianka index of niche overlap indicated that the empirical assemblage-wide overlap was not significantly different from randomly generated patterns in the abundance season but significantly great- er than expected by chance alone in the scarcity season. All the species showed an increase in temporal niche breadth during the abundance season, which appears to be related to longer daylength and high nocturnal tem- peratures. Patterns on both temporal niche overlap and temporal niche breadth were the opposite to those that we were expecting in the case of diel activity patterns determined by competition for dietary resources. There- fore, we conclude that competition did not seem to be preponderant for determining patterns of temporal niche overlap by the studied community. Key words: activity patterns, Brazil, chronoecology, community structure, competition, niche partitioning. INTRODUCTION Being active in the same environment at different times, at both seasonal and diel scales, exposes ani- mals to the effects of very different environmental fac- tors (Kronfeld-Schor & Dayan 2003), both abiotic (e.g. photoperiod length, ambient temperatures and humidity) and biotic (e.g. predation risk, food availability, com- petitors and mates). Among the latter, timing of activi- ty might be crucial in determining encounters between neighboring territorial species (Halle 2000), in mak-

description

artigo sobre cronoecologia

Transcript of cronoecologia (2)

  • 2011 ISZS, Blackwell Publishing and IOZ/CAS 375

    123456789

    101112131415161718192021222324252627282930313233343536373839404142434445464748495051

    123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051

    doi: 10.1111/j.1749-4877.2011.00266.x

    Correspondence: Emerson M. Vieira, Departamento de Ecologia, Instituto de Cincias Biolgicas, Universidade de Braslia, C.P. 04457, Braslia, DF, 70919-970, Brazil.Email: [email protected] Present address: Department of Ecology, University of Braslia (UnB), Braslia, Brazil.

    Integrative Zoology 2011; 6: 375-386

    ORIGINAL ARTICLE

    Temporal niche overlap among insectivorous small mammals

    Emerson M. VIEIRA and Gabriela PAISELaboratory of Ecology of Mammals, Zoology, University of Vale do Rio dos Sinos, So Leopoldo, Brazil and Graduate Program of Ecology, University of Campinas, Campinas, Brazil

    Abstract Being active in the same environment at different times exposes animals to the effects of very different envi-ronmental factors, both biotic and abiotic. In the present study, we used live traps equipped with timing devices to evaluate the potential role of biotic factors (competition and food abundance) on overall overlap in the tem-poral niche axis of 4 insectivorous small mammals in high-elevation grassland fields (campos de altitude) of southern Brazil. Based on resources availability (invertebrates), data on animal captures were pooled in 2 sea-sons: scarcity (June 2001September 2001) and abundance (November 2001May 2002) seasons. We tested for non-random structure in temporal niche overlap among the species in each season. These species were the rodents Oxymycterus nasutus (Waterhouse, 1837), Deltamys sp., Akodon azarae (Fischer, 1829), and the marsu-pial Monodelphis brevicaudis Olfers, 1818. The studied community was mainly diurnal with crepuscular peaks. Simulations using the Pianka index of niche overlap indicated that the empirical assemblage-wide overlap was not significantly different from randomly generated patterns in the abundance season but significantly great-er than expected by chance alone in the scarcity season. All the species showed an increase in temporal niche breadth during the abundance season, which appears to be related to longer daylength and high nocturnal tem-peratures. Patterns on both temporal niche overlap and temporal niche breadth were the opposite to those that we were expecting in the case of diel activity patterns determined by competition for dietary resources. There-fore, we conclude that competition did not seem to be preponderant for determining patterns of temporal niche overlap by the studied community.

    Key words: activity patterns, Brazil, chronoecology, community structure, competition, niche partitioning.

    INTRODUCTION Being active in the same environment at different

    times, at both seasonal and diel scales, exposes ani-mals to the effects of very different environmental fac-tors (Kronfeld-Schor & Dayan 2003), both abiotic (e.g. photoperiod length, ambient temperatures and humidity) and biotic (e.g. predation risk, food availability, com-petitors and mates). Among the latter, timing of activi-ty might be crucial in determining encounters between neighboring territorial species (Halle 2000), in mak-

  • 2011 ISZS, Blackwell Publishing and IOZ/CAS376

    123456789

    101112131415161718192021222324252627282930313233343536373839404142434445464748495051

    E.M. Vieira and G. Paise

    ing individuals more or less subject to predation (Peter-son & Batzli 1975; Saiful et al. 2001), in reducing or in-creasing competition between species (Ziv et al. 1993), and also in influencing their foraging success (Lockard 1978).

    In relation to the effects of competition on patterns of diel activity, it is generally recognized that the tempo-ral axis is not as common as food or space for promot-ing coexistence through specialization (Schoener 1974). However, theory suggests that diel temporal partitioning might facilitate coexistence between competitors and between predators and prey (Kronfeld-Schor & Dayan 2003), thus indicating the potential influence of such bi-otic factors on patterns of diel activity. Indeed, there are several studies demonstrating the occurrence of tempo-ral partitioning in a number of communities, including those of ants (Albrecht & Gotelli 2001), bats (Adams & Thibault 2006), lizards (Pianka 1973), raptors (Jaksic 1982) and rodents (Shkolnik 1971; Kotler et al. 1993; Ziv et al. 1993).

    The debate concerning the structuring of ecological communities and the occurrence of discernible assem-bly rules in the co-occurrence of species has been one of the most controversial issues for ecologists (Cody & Diamond 1975; Strong et al. 1984; Gotelli & Graves 1996; Weiher & Keddy 1999). In the context of tempo-ral niche overlap, the question is (following Albrecht & Gotelli 2001): how much overlap in time use would be expected in the case of species using resource states (i.e. different time intervals) randomly with respect to one another?

    The discussion in ecological literature on the evi-dence of discernible assembly rules in the co-occurrence of species has been supported by the comparison of the structure of real communities to those patterns generat-ed by null models (Strong et al. 1984; Gotelli & Graves 1996; Weiher & Keddy 1999). Although methods for null model analysis of niche overlap might also be used for analyzing the temporal structure of animal assem-blages, patterns of resource partitioning and segregation in time have not been tested in many natural communi-ties (Albrecht & Gotelli 2001; Castro-Arellano & Lach-er 2009).

    For some rodent communities, there is evidence that temporal segregation along the 24 h daily period might help in allowing coexistence between competitors (e.g. Shkolnik 1971; OFarrell 1974; Ziv et al. 1993; Viei-ra & Baumgarten 1995; Jones et al. 2001; Castro-Arel-lano & Lacher 2009). Partition of time on the diel scale might facilitate coexistence through avoidance of di-

    rect confrontation (interference competition) or through reduction of resource overlap (resource competition). Temporal partitioning is a viable mechanism for reduc-ing resource competition under either of the follow-ing conditions (according to Kronfeld-Schor & Dayan 2003): (i) if the shared limiting resources differ between activity times (e.g. for predatory species whose prey populations have different activity patterns), or (ii) if the limiting resources are renewed within the time involved in the separation. Specifically for small mammals, inter-ference interactions seem to play a more important role in causing temporal segregation than indirect resource competition (Carothers & Jaksic 1984), this being espe-cially true for insectivorous small mammals (Dickman 1991).

    Within the group of small mammals, segregation in time, to avoid or reduce resource competition, would be expected to occur mainly among species occupying the same vertical strata (e.g. ground or canopy) and belong-ing to the same feeding guild. Habitats with relatively low complexity and heterogeneity (sensu August 1983), such as grassland fields, would potentially offer more opportunities for niche segregation through divergence in time of activity. Among the diverse feeding guilds of neotropical small mammals, invertebrate eaters are po-tentially more prone to diverge along the temporal axis as a mechanism for reducing competition. This pattern is due to the fact that the available resources would be more easily replaced along a 24 h period due to possi-ble differences in daily activity peaks of distinct groups of invertebrates (e.g. Taylor & Carter 1961; Stringer & Meyer-Rochow 1997; Fernandez et al. 2001; Barrozo et al. 2004).

    In the present study, we investigated daily activi-ty patterns of insectivorous non-volant small mam-mals for evaluating the potential role of biotic factors (competition and resource availability) on the temporal niche overlap of these animals. These mammals occur in high-elevation grassland fields (campos de altitude) of southern Brazil. Campos de altitude harbour a local-ly rich small-mammal fauna (up to 13 species in a 2 ha area [EM Vieira, unpubl. data]). For most of these spe-cies, except for the abundant Oxymycterus nasutus (Wa-terhouse, 1837), their patterns of daily activity are still unknown in this ecosystem (see Paise & Vieira 2006). Grassland small mammals in South America feed more or less strictly on insects (Solari 2007) and in their hab-itat the availability of invertebrates changes markedly during the year, being high during the warmer months of the year (September to March) and lower during the

    BrunoHighlight

  • 2011 ISZS, Blackwell Publishing and IOZ/CAS 377

    123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051

    Temporal niche overlap among mammals

    cooler months of the year (May to August) (EM Vieira, pers. observ.). We addressed the following questions:1. What are the general daily activity patterns and tem-

    poral niche overlap of the insectivorous small mam-mals?

    2. Are there seasonal changes in such patterns when comparing a warmer season of high invertebrate availability with a cooler season of low invertebrate availability? Specifically for the second question, and considering

    that temporal segregation could be important for help-ing in species coexistence by reducing competition, we predict that during the season of general low-resource availability, the species would restrict their activity to the optimal times of resource acquisition. Similarly, a reduction in niche breadth in response to food scarcity (but considering the spatial dimension instead of time) is reported for sparrows (Pulliam 1986). Alternative-ly, in the absence of relevant competition, a reduction in resource availability could force the animals to be ac-tive also at sub-optimal intervals of the day, as report-ed by Lockard (1978) for the bannertail kangaroo rats (Dipodomys spectabilis). Therefore, such reduction in resource availability would result in an increase in the temporal niche breadth of the animals. Predictions about overall niche overlap and niche breadth based on classic theory of resource selection in competing species (de-scribed in Pimm & Pimm 1982; Rosenzweig 1991), as-suming that resource availability (invertebrates) might influence daily activity patterns, are summarized in Ta-ble 1.

    MATERIALS AND METHODS

    Study Site

    We conducted the study in the National Park of Aparados da Serra (NPAS) in southern Brazil. The park has a total area of 10 250 ha, ranging from 75 to 1000 m a.s.l, with a markedly seasonal climate, with 1220 C average annual temperature, moderate winters and mild, wet summers (DeForest Safford 1999; Dalmagro & Vie-ira 2005). From 1998 to 2000, the total annual precipita-tion in the region ranged from 1700 to 2000 mm. During these years, the daily extreme temperatures in summer (December to February) were 9.4 C and 24.6 C. In win-ter (June to August), these values dropped to 2.5 C and 19.3 C (Paise & Vieira 2006). We captured the animals in the campos de altitude of the higher-elevation ar-eas of the NPAS. The vegetation of such fields is high-ly diverse and characterized by non-arborescent spe-cies, mainly tall grasses. (For more detailed information on the climate and vegetation of the NPAS, see Paise and Vieira [2006].)The landscape of the area consists of low flat hills, topped with permanent drylands (i.e. with-out apparent water on the soil surface), and wet fields in lower areas.

    Study species

    Previous surveys at the study area indicate that the local small-mammal community is composed of 13 species (Paise & Vieira 2006), with 4 of them feeding mainly on invertebrates. Sufficient numbers of these 4

    Table 1 Community-level predictions on degree of temporal niche overlap and temporal niche breadth for an insectivorous commu-nity of small mammals based on the assumption that resource availability (invertebrates) might influence on daily activity patterns. Predictions are presented under situations with or without relevant interspecific competition

    Resource availabilityLow High

    Low competition High competition Low competition High competition

    Overal niche overlap HighMore time necessary for acquiring resources; individuals would be forced to be active also at sub-optimal periods

    Low More specialization induced by interference competition

    Low Less time necessary for foraging

    High Less specialization

    Overall niche breadth High Low Low High

  • 2011 ISZS, Blackwell Publishing and IOZ/CAS378

    123456789

    101112131415161718192021222324252627282930313233343536373839404142434445464748495051

    E.M. Vieira and G. Paise

    species were captured for the analysis. These species were (BM = mean body mass SE, using data obtained in the present study): 3 sigmodontine rodents, O. nasu-tus (BM = 55.5 1.1 g), Akodon azarae (J. Fischer, 1829) (BM = 21.7 0.8 g) and 1 yet undescribed species of the ge-nus Deltamys (BM = 17.7 0.4 g), and 1 didelphid marsupial, Monodelphis brevicaudis Olfers, 1818 (BM = 20.0 1.7 g), fol-lowing the classification proposed by Gomes (1991) and also presented in Eisenberg and Redford (1999). Inver-tebrates comprise at least 60% of the diet of each of the 4 species studied. Such estimates on invertebrate con-sumption for these species were based on available pub-lished data for the species and also for congeneric spe-cies (Gonzlez 2001; Suarez & Bonaventura 2001; Vieira & Asta de Moraes 2003; Solari 2007) and un-published data on fecal and stomach material analy-sis obtained in the same study area (EM Vieira, unpubl. data). We are using here a broad definition of insectivo-ry, considering as insectivorous any species that feeds on invertebrates on a regular basis.

    Data collectionTrapping procedures used in this study, including lo-

    cation of trapping transects, are described in Paise & Vieira (2006) and are briefly summarized here. We con-ducted 7 trapping sessions (5 days each) in June and July (winter), September and November 2001 (spring), January 2002 (summer), and March and May 2002 (au-tumn), using standard Sherman live traps (9 8 23 cm) in 5 transects placed 50 m from each other. Transects were arranged to equally sample wet and dry areas, and comprised 14 trap stations spaced 40 m apart, totaling 70 trap stations. Censuses were made with 60 traps; in every trapping session we randomly selected 60 stations to sample (total effort of 300 trapnights per session).

    All traps were equipped with digital timers that in-dicated the time of capture to the nearest minute. Traps were baited with a mixture of peanut butter, corn meal, mashed banana, commercial cod-liver oil and vanilla es-sence. We examined traps and renewed bait if necessary between 08.30 and 12.00 hours every day. We record-ed species, location, capture time, sex and weight of ev-ery captured animal. Animals were ear tagged (National Band and Tag, monel tags no. 1005), and released at the point of capture after data collection. Trapping and han-dling conformed to guidelines sanctioned by the Amer-ican Society of Mammalogists (Animal Care and Use Committee 1998). We recorded daily sunset and sunrise times using a Garmin 12 GPS.

    Resource availability

    We estimated invertebrate availability using pit-fall traps, which consisted of 300 mL plastic cups bur-ied with the opening flush to the ground surface. Pitfall trapping has been commonly used for estimates of abun-dance of terrestrial invertebrates (e.g. Pinheiro et al. 2002; Leiner & Silva 2007). The cups were filled with 45 mL of water, 5 mL of formaldehyde and some drops of liquid detergent for breaking water tension. Those pitfall traps were arranged in 2 parallel transects 20 m apart, arranged to equally sample wet and dry areas. Each transect comprised 20 cups spaced 5 m apart, to-taling 40 traps. These traps were set monthly from May 2001 to May 2002 for 3 days per month except in Oc-tober 2001 and April 2002, when no trapping was con-ducted. Any non-functional cups (i.e. destroyed by cows or with a fallen leaf or stick present that could poten-tially allow the escape of invertebrates) were discarded. We identified captured invertebrates to the level of or-der (except for Formicidae, which was considered as a distinct group) and weighted their total biomass to the nearest 0.5 g). This value was divided by the total num-ber of functional cups and used as an index of inverte-brate availability.

    Data analysis

    We quantified daily activity patterns by assessing the number of animals captured in each 3 hour interval. Al-though this method has been used for assessing activity patterns of small mammals (Bruseo & Barry 1995; Pri-otto & Polop 1997), it could be biased because trapped animals are not available for subsequent recapture, and they occupy traps that are not available for documenting activity at later intervals (Hicks et al. 1998; Graipel et al. 2003). However, capture patterns indicated that such biases were negligible in our study (see Paise and Vieira [2006] for detailed discussion on possible drawbacks of the methods used).

    We considered only the first capture of each individ-ual in each trapping session to avoid problems with po-tential dependence of data caused by multiple captures of a few individuals. The activity pattern for each spe-cies was quantified by the proportions of total captures of the species (pooling captures of the whole season) at each of the 8 3 h time intervals of each diel cycle. Daily activity patterns of O. nasutus, the most common spe-cies in the study area, are discussed in detail elsewhere (Paise & Vieira 2006) and are presented here only for interspecies comparisons.

  • 2011 ISZS, Blackwell Publishing and IOZ/CAS 379

    123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051

    Temporal niche overlap among mammals

    Data on animal captures were pooled in 2 seasons based on the resource (invertebrates) availability in the study area (see Results section). We defined a scarci-ty season from June 2001 to September 2001 and an abundance season from November 2001 to May 2002 (see Results section below).

    For each season we evaluated pair-wise inter-specific differences in activity patterns between species at each season by using KolmogorovSmirnov 2 sample tests (Zar 1996). For these analyses, we pooled the noctur-nal intervals between 21.00 and 03.00 hours due to the low number of captures in such intervals for most spe-cies. We used for such comparisons significance lev-el () = 0.05. We did not implement a correction when multiple tests were performed (e.g. Bonferroni correc-tion) because such corrections are excessively conser-vative and their use has been severely criticized recently (Moran 2003; Gotelli & Ellison 2004; Roback & Askins 2005), especially for the kind of exploratory analyses that we are conducting in the present study (see Roback & Askins 2005).

    We estimated pair-wise temporal niche overlap using the index suggested by Pianka (1973). This index ranges from 0 (2 species do not share resources) to 1 (complete overlap in resource use between 2 species). For all com-parisons we pooled data on all individuals of each spe-cies, assuming that such data were an overall representa-tion of use of time for the species. This assumption was supported by published data on the most common spe-cies of the studied community, O. nasutus, whose activi-ty patterns do not differ significantly between age class-es or sex (Paise & Vieira 2006), and also by the study of Castro-Arellano and Lacher (2009) on neotropical ro-dents of Mexico.

    For evaluation of the significance of temporal niche overlap among all species during each season, we per-formed Monte Carlo simulations on the daily activity matrix (species by time intervals). First, we calculated the mean value of niche overlap index between pair of species, thus considering a total of 6 values. These ob-served values were compared to the results of a null dis-tribution generated by 10 000 iterations performed by an algorithm termed Rosario, that shifts the entire activ-ity patterns of each species (keeping the same row to-tals) while maintaining the empirical structure of the data (Castro-Arellano & Lacher 2009). The obtained P-values correspond to the probability of finding, by chance alone, an equal or greater niche overlap than the observed value. Therefore, a significant niche overlap would occur when an observed mean value was great-

    er than 97.5% of the simulated values of the means and significant niche segregation would occur in the case of obtaining an observed value lower than 97.5% of the simulated mean values. We ran the analysis using the Time Overlap computer program (Windows operating system; Castro-Arellano and Lacher [2009]; program available from the authors on request).

    We used the standardized index proposed by Levins (1968) for evaluating temporal niche breadth, consider-ing each 3 h interval as a resource (as in Saiful et al. 2001). This index ranges from 1/R (R = number of re-source states available [i.e. time intervals]) when the population uses only one resource, to 1 (maximum niche breadth) when the population uses all resource states in equal proportions (Feisinger & Spears 1981). We calcu-lated this index for each species for both the abundance and the scarcity season. Due to the small sample size, we did not statistically test the results, but we would consider as an indication of an overall increase in niche breadth during the scarcity season the case of at least 3 of the 4 most common species showing higher values during this season than during the abundance season.

    RESULTSWe recorded 523 captures of 8 small-mammal spe-

    cies (300 individuals) with 1981 trapdays of effort, re-sulting in an overall trapping success of 26.4%. Thus, on average, more than 73% of traps were available (i.e. empty) at the end of any sampling night. The insectiv-orous species captured with sufficient sample size to be considered in the analysis were O. nasutus (139 individ-uals captured), Deltamys sp. (80 individuals), A. azarae (35 individuals) and M. brevicaudis (31 individuals). We also captured the following rodent species: Oligory-zomys flavescens (6 individuals), Scapteromys sp. (5 in-dividuals), Oligoryzomys nigripes (2 individuals) and Akodon montensis (2 individuals). Trapping success in the scarcity season (mean SE = 21.5 4.6 individu-als/100 traps) was higher than in the abundance season (11.5 3.14 individuals/100 traps).

    Data on pitfall trapping indicated 2 distinct periods in terms of invertebrate availability. There was a period with higher availability (abundance season), from No-vember 2001 to May 2002, when the mean invertebrate index (mean SE = 0.20 0.02 g/cup) was more than 2.5 times higher than the mean index obtained between June 2001 and September 2001 (0.08 0.02 g/cup), the peri-od with low invertebrate availability (scarcity season). Mean sunrise and sunset times differed between sea-

  • 2011 ISZS, Blackwell Publishing and IOZ/CAS380

    123456789

    101112131415161718192021222324252627282930313233343536373839404142434445464748495051

    E.M. Vieira and G. Paise

    Figure 1 Daily activity patterns of 4 in-sectivorous small mammals in the grass-land fields of the National Park of Apara-dos da Serra, southern Brazil, between June 2001 and May 2002. The left col-umn shows the results for the scarci-ty season (from June to September) and the right column shows the results for the abundance season (from November to May). Time intervals represent relative number of captures that occurred during the previous 3 h. Dashed lines indicate in-tervals that are mostly nocturnal. Also in-dicated are numbers of captures (in paren-theses) analyzed in each season.

    sons: scarcity season = 06.46 (range 06.2607.12) and 17.52 (range 17.3318.05); abundance season, 06.12 (range 05.2606.18) and 18.26 (range 18.2118.35).

    The studied small-mammal community was main-ly diurnal, with most species showing activity primarily during the day, but all species had some nocturnal activ-ity, mainly in the abundance season (Fig. 1). Akodon az-arae present high relative nocturnal activity in both sea-sons (Fig. 1), resulting in higher values of niche breadth in comparison with the other species, except with Del-tamys sp. in the abundance season (Fig. 2). In contrast, the marsupial M. brevicaudis was almost strictly diurnal (Fig. 1), and the species with the lowest temporal niche breadth (Fig. 2).

    Pair-wise contrasts in daily activity using Kol-mogorovSmirnov tests indicate only 3 significant be-tween-species comparisons. O. nasutus and M. brevi-caudis diverged significantly in their activity patterns

    in both seasons and this same rodent species diverged from A. azarae during the abundance season (Table 2). Patterns of niche overlap also changed according to year period, with an overall reduction during the abun-dance season. During the scarcity season, the mean val-ue of the Pianka index for temporal overlap between the studied small mammals was 0.829, but this mean val-ue dropped to 0.690 during the abundance season (Ta-ble 2). Results from simulations using the values of the Pianka index showed that the empirical assemblage-wide overlap was significantly greater than expected by chance alone in the scarcity season (95% range of sim-ulated values [CI] = 0.4250.718, P = 0.0012), but not significantly different from randomly generated pat-terns in the abundance season (95% range of simulat-ed values [CI] = 0.6000.768, P = 0.64). Thus, not only were insectivorous small mammals at the NPAS not temporally segregated along the 24 h cycle in any sea-

    ScarcityDeltamys sp.

    Akodon azarae

    Oxymycterus nasutus

    Monodelphis brevicaudis

    (65)60

    40

    20

    0

    60

    40

    20

    0

    60

    40

    20

    0

    60

    40

    20

    03.00 6.00 9.00 12.00 15.00 18.00 21.00 24.00 3.00 6.00 9.00 12.00 15.00 18.00 21.00 24.00

    Time of day

    Perc

    enta

    ge o

    f cap

    ture

    s

    (28)

    (135)

    (27)

    (50)

    (27)

    (168)

    (10)

    Abundance

  • 2011 ISZS, Blackwell Publishing and IOZ/CAS 381

    123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051

    Temporal niche overlap among mammals

    son of the year, but also showed an opposite pattern, with significant temporal overlap during the season with low resource availability. Additionally, all the 4 studied species showed an increase in temporal niche breadth during the abundance season (Fig. 2).

    DISCUSSION

    Activity patterns of the studied community

    Our results showed that the studied grassland com-munity of insectivorous small mammals is mainly di-urnal with crepuscular peaks, contrasting with the noc-turnal pattern exhibited by most neotropical small mammals (Emmons & Feer 1997; Eisenberg & Redford 1999). Nocturnality in neotropical rodents and marsu-pials seem to be related mainly to tropical and forested areas (Emmons & Feer 1997; Roll et al. 2006). In con-trast, for rodents living in cold habitats, diurnal activi-ty seems to be a common pattern (Roll et al. 2006). Di-urnality in sub-tropical grasslands, as is the case of our study area, might be related to avoidance of low noctur-nal temperatures outside tropical regions combined with protection provided by dense plant cover even during daylight hours.

    Temporal niche overlap and niche breadth

    Our results indicate a high abundance of inverte-brates during the warmest season of the year. Such a pattern was expected and is described in other studies (e.g. Bergallo & Magnusson 1999; Smale et al. 2003).

    Figure 2 Temporal niche breadth (standardized Levins index) of the 4 insectivorous small mammals for both scarcity and abundance seasons in southern Brazil.

    Deltamys sp.1.0

    0.8

    0.6

    0.4

    0.2

    0.0

    Oxymycterus nasutus Monodelphis brevicaudisAkodon azarae

    ScarcityTem

    pora

    l nic

    he b

    read

    th (L

    evin

    s)

    AbundanceSeason

    During the cooler months of the year, low temperatures might cause diapause or else permanence of insects in-side shelters to avoid heat loss (Schowalter 2000).

    For the season with relative low invertebrate avail-ability, the comparison with the null model did not in-dicate a significantly lower overall overlap in time use among species, which would be expected in the case of influence of resource abundance on use of time through competition. Therefore, our prediction on tempo-ral niche segregation under conditions of low resource availability was not confirmed by the results. Instead, we detected an inverse pattern: during the season with low invertebrate availability there was a higher and sig-nificant overlap in temporal niche. These patterns sug-gest that, for our studied community, time dimension is not relevant for helping species coexistence through re-duction in competition. Indeed, our diurnal small-mam-mal community could have converged upon the prefera-ble resource states (i.e. those intervals occurring during daylight period) during the scarcity season and, conse-quently, increased their temporal niche overlap. A sim-ilar process seemed to occur in a semiarid ecosystem in Chile, where Farias and Jaksic (2007) report an in-crease in dietary niche overlap among vertebrate preda-tors when the potential for interspecific competition in-creased. These authors suggest that if resources become scarce, all predators converge upon the few prey re-sources still available, thus increasing overlap.

    Our results are in agreement with the general state-ment that segregation in time induced by competi-tion does not occur frequently in animal assemblages (Schoener 1974). However, activity time might poten-tially be an important niche dimension when the con-sidered species are predators and predation risk is much higher than the ability to process food (Schoener 1974; Halle & Stenseth 2000b). Such conditions generally ap-ply to small mammals and, indeed, for this group, there are cases where strong evidence indicates that tempo-ral partitioning on the diel scale might facilitate coex-istence (e.g. Schoener 1974; Dickman 1991; Ziv et al. 1993). Nonetheless, temporal niche segregation might not be equally important for all rodent assemblages. Castro-Arellano and Lacher (2009) sampled 2 sub-trop-ical Mexican forests and, using the same algorithm that we used in the present study, report non-random tempo-ral niche segregation only for the most diverse rodent assemblage (5 nocturnal species).

    The lack of segregation in time as a mechanism for reducing competition that our results indicated suggests potential differences in microhabitat utilization or else

  • 2011 ISZS, Blackwell Publishing and IOZ/CAS382

    123456789

    101112131415161718192021222324252627282930313233343536373839404142434445464748495051

    E.M. Vieira and G. Paise

    Table 2 Pairwise interespecific comparisons of activity patterns of insectivorous small mammals in the grasslands fields of the National Park of Aparados da Serra, southern Brazil, for each of the 2 periods of the study (see text for details)

    Akadon azarae Deltamys sp. Oxymycterus nasutus Monodelphis brevicaudisScarcity period

    Akodon azarae 0.17 (0.63) 0.25 (0.10) 0.29 (0.21)Deltamys sp. 0.891 0.09 (0.99) 0.26 (0.17)Oxymycterus nasutus 0.822 0.979 0.33 (0.014)Monodelphis brevicaudis 0.729 0.813 0.742

    Abundance periodAkodon azarae 0.26 (0.18) 0.30 (0.028) 0.38 (0.23)Deltamys sp. 0.840 0.17 (0.20) 0.34 (0.28)Oxymycterus nasutus 0.807 0.889 0.51 (0.014)Monodelphis brevicaudis 0.462 0.685 0.454

    Values above the diagonals correspond to the results of statistical tests: KolmogorovSmirnov statistics, with correspondent P-values in parentheses (significant values are indicated in bold, P < 0.05). Values under the diagonals correspond to pairwise observed niche overlaps (Piankas index). Sample sizes are indicated in Fig. 1.

    in feeding habits of the studied species, as proposed by Rychlik (2005) for 4 sympatric species of shrews and by Albrecht and Gotelli (2001) for 7 ant species. Spatial segregation at small scales is common in small-mam-mal communities (e.g. Dueser & Shugart 1978; Dues-er & Hallet 1980; Mura & Gonzlez 1982; Dalmagro & Vieira 2005) and was suggested to occur in a grass-land small-mammal community similar to our studied community (Suarez & Bonaventura 2001). In relation to feeding habits, because most of these species do not feed exclusively on invertebrates (EM Vieira, pers. ob-serv.), the extent of feeding on plant matter might lead to a higher niche segregation in the foodresource axis and enable similar use of the temporal niche. In addi-tion, segregation in the dietary axis could occur due to a fine segregation in the type of invertebrate consumed. Unfortunately, there is no detailed information on the diet of the studied mammals for evaluating possible dif-ferences in their dietary niches within the broad catego-ry invertebrates.

    The rarity of the occurrence of temporal partitioning as a mechanism of coexistence has been attributed to the rigidity of time-keeping mechanisms, as well as to oth-er physiological and anatomical traits that might restrict activity patterns (Kronfeld-Schor et al. 2001; Kronfeld-Schor & Dayan 2003). The evolution of such patterns is constrained by phylogeny, at least in rodents, which might also limit species ability to use the time niche axis for ecological separation (Roll et al. 2006).

    In addition to phylogenetic constraints, ecological re-quirements related to body mass might also influence the co-occurrence of mammal species and help in de-termining assembly rules (Bowers & Brown 1982). Re-source partitioning along the time dimension might help in the coexistence of similarly sized syntopic species (Halle 2000). In our study, Deltamys sp. showed a high similarity in time use with the larger O. nasutus than with the similar-sized A. azarae. Similarly, Castro-Arel-lano & Lacher (2009) report higher overlap in activity pattern for pairs of rodent species with larger differenc-es in body size.

    Competitive interactions might also be influenced by population densities of the involved species (Pimm & Pimm 1982; Rosenzweig 1991). In the present study, capture rates were higher in the scarcity season than in the abundance season. This pattern could be caused by a real increase in population numbers in the scarcity sea-son. Such an increase could be a result of a delayed re-sponse to a previous peak in resource abundance. Un-der this scenario of higher densities, we would expect an even higher competitive pressure and a reduction still more accentuated in niche overlap, which was not de-tected at all. Alternatively, the higher capture rates in the autumnwinter season could be caused by an increase in bait atractivity during this season of low resource abun-dance. If that was the case, population densities did not actually change markedly between seasons and were of least concern for our study. In any case, our discussion

  • 2011 ISZS, Blackwell Publishing and IOZ/CAS 383

    123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051

    Temporal niche overlap among mammals

    on temporal niche overlap and resource availability is still valid and yields testable and biologically interesting hypotheses.

    During the scarcity season we did not detect an over-all increase in temporal niche breadth either. Such an in-crease would we expected in the case of animals being forced to be active at sub-optimal intervals due to a lack of resource availability (as in Lockard 1978). On the contrary, during this low-resource season there was a re-duction in temporal niche breadth for the 4 species stud-ied. This reduction was caused mainly by an overall de-crease in nocturnal activity and was responsible for the significant overlap in time that we detected. A reduc-tion in activity during winter was also reported for oth-er diurnal small mammals, as for the diurnal red squirrel (Sciurus vulgaris), who changes from a long biphasic pattern in summer to a shorter unimodal pattern in win-ter (Wauters & Dhondt 1987; Wauters et al. 1992), and for the degu Octodon degus, who changes from a com-pletely bimodal pattern of surface daily activity dur-ing spring, summer and autumn to a unimodal pattern in winter (Kenagy et al. 2002).

    The relative higher activity in diurnal intervals dur-ing winter might be explained by ambient temperature, which is probably the most influential abiotic factor on time and energy budgets (Huey 1991). In the present study, the study species might have increased their diur-nal activity in an attempt to avoid the low temperatures occurring at night during the scarcity season, which cor-responded to austral autumn and winter. Such explana-tion is supported by the results presented by Paise and Vieira (2006). In Paise and Vieira (2006), a significant reduction in O. nasutus activity is reported when tem-perature goes below 3 C in the study area. The noctur-nal period is probably still less favorable for the other study species during cold months, because all of them have smaller body size and, consequently, an even larg-er ratio of surface area to volume than O. nasutus. A re-duction in nocturnal activity and consequent increase in temporal niche overlap during months of low nighttime temperatures has also been reported for ant communities (Albrecht & Gotelli 2001).

    An additional non-opposing hypothesis to explain re-duction of nocturnal activity during the scarcity season is that the small mammals would be adjusting their ac-tivity patterns to those of their prey. Activity patterns of invertebrates might be positively related with ambient temperature (Merritt & Vessey 2000; Schowalter 2000) and in the study area they could be reducing their activ-ity at night during the coolest season. Therefore, small

    mammals could be following their prey activity patterns and also reducing their nocturnal activity during such season.

    In conclusion, the insectivorous community that we studied was composed of small mammals whose diel activity was mainly diurnal with crepuscular peaks. Our observed patterns on both temporal niche overlap and temporal niche breadth were the opposite to those that we were expecting in the case of diel activity pat-terns determined by competition for dietary resources. These results suggest that resource competition did was not preponderant for determining patterns of temporal niche overlap by the studied community. Such patterns seemed to be more influenced by abiotic factors, such as daylength and temperature.

    ACKNOWLEDGEMENTSWe are grateful to the Brazilian Institute for Environ-

    mental Resources (IBAMA), especially to PCR Bastos, for the authorization and support to our fieldwork in the National Park of Aparados da Serra. Ivan Castro-Arella-no read an earlier draft of this paper. We also thank se-veral other colleagues for their help in the fieldwork and AU Christoff for confirming the species identifica-tion. This study was supported by British Ecological So-ciety (SEPG #1841), UNISINOS and FAPERGS (0112763). While conducting this research work EM Vieira received a personal research grant from the Conselho Nacional de Pesquisa - CNPq (300286/99-6). All the methods used here comply with the current Brazilian laws on wildlife studies.

    REFERENCESAdams RA, Thibault KM (2006). Temporal resource

    partitioning by bats at water holes. Journal of Zoolo-gy 270, 46672.

    Albrecht M, Gotelli NJ (2001). Spatial and temporal niche partitioning in grassland ants. Oecologia 126, 13441.

    Animal Care and Use Committee (1998). Guidelines for the capture, handling, and care of mammals as ap-proved by the American Society of Mammalogists. Journal of Mammalogy 79, 141631.

    August PV (1983). The role of habitat complexity and heterogeneity in structuring tropical mammal com-munities. Ecology 64, 1495507.

    Barrozo RB, Schilman PE, Minoli SA, Lazzari CR (2004). Daily rhythms in disease-vector insects. Bio-logical Rhythm Research 35, 7992.

  • 2011 ISZS, Blackwell Publishing and IOZ/CAS384

    123456789

    101112131415161718192021222324252627282930313233343536373839404142434445464748495051

    E.M. Vieira and G. Paise

    Bergallo HG, Magnusson WE (1999). Effects of climate and food availability on 4 rodent species in south-eastern Brazil. Journal of Mammalogy 80, 47286.

    Bowers MA, Brown JH (1982). Body size and coexistence in desert rodents: chance or community structure? Ecol-ogy 63, 391400.

    Bruseo JA, Barry RE Jr (1995). Temporal activity of syntopic Peromyscus in the Central Appalachians. Journal of Mammalogy 76, 7882.

    Carothers JH, Jaksic FM (1984). Time as a niche differ-ence: the role of interference competition. Oikos 42, 4036.

    Castro-Arellano I, Lacher TE Jr (2009). Temporal niche segregation in two rodent assemblages of subtropical Mexico. Journal of Tropical Ecology 25, 593603.

    Cody ML, Diamond JM, eds (1975). Ecology and Evo-lution of Communities. Harvard University Press, Cambridge.

    Dalmagro A, Vieira EM (2005). Patterns of habitat utili-zation of small rodents in an area of Araucaria forest in Southern Brazil. Austral Ecology 30, 35362.

    DeForest Safford H (1999). Brazilian Pramos I. An in-troduction to the physical environment and vegeta-tion of the campos de altitude. Journal of Biogeogra-phy 26, 693712.

    Dickman CR (1991). Mechanisms of competition among insectivorous mammals. Oecologia 85, 46471.

    Dueser RD, Hallet JG (1980). Competition and habitat selection in a forest-floor small mammal fauna. Oikos 35, 2937.

    Dueser RD, Shugart, HH Jr (1978). Microhabitats in a forest-floor small mammal fauna. Ecology 59, 8998.

    Eisenberg JF, Redford KH (1999). Mammals from the Neotropics: The Central Neotropics Volume 3: Ecua-dor, Peru, Bolivia, Brazil. The University of Chicago Press, Chicago.

    Emmons LH, Feer F (1997). Neotropical Rainforest Mammals: A Field Guide. The University of Chicago Press, Chicago.

    Farias AA, Jaksic FM (2007). El Nio events, the lean versus fat scenario, and long-term guild dynamics of vertebrate predators in a South American semiarid ecosystem. Austral Ecology 32, 22538.

    Feisinger P, Spears EE, Poole RW (1981). A simple measure of niche breadth. Ecology 62, 2732.

    Fernandez JL, Garces G, Portuondo E, Valdes P, Exposi-to I (2001). Insects associated with flowers of weeds of the Botanical Garden of Santiago de Cuba, with emphasis on Hymenoptera. Revista de Biologia Trop-ical 49, 101326.

    Gomes NF (1991). Reviso sistemtica do gnero Monodelphis (Didelphidae: Marsupialia) (MSc the-sis). Universidade de So Paulo, So Paulo, Brazil.

    Gonzlez EM (2001). Gua de campo de los mamferos de Uruguay. Introduccin al estudio de los mamfer-os. Vida Silvestre, Montevideo.

    Gotelli NJ, Ellison AM (2004). A Primer of Ecological Statistics. Sinauer Associates, Inc., Sunderland, MA.

    Gotelli NJ, Graves GR (1996). Null Models in Ecology. Smithsonian Institution Press, Washington, DC.

    Graipel ME, Miller PRM, Glock L (2003). Padro de atividade de Akodon montensis e Oryzomys russatus na Reserva Volta Velha, Santa Catarina, Sul do Bra-sil. Mastozoologia Neotropical 10, 25560.

    Halle S (2000). Ecological relevance of daily activity patterns. In: Halle S, Stenseth NC, eds. Activity Pat-terns in Small Mammals: An Ecological Approach. Springer-Verlag, Heidelberg, pp. 6790.

    Halle S, Stenseth NC (2000a). Introduction. In: Halle S, Stenseth NC, eds. Activity Patterns in Small Mam-mals: An Ecological Approach. Springer-Verlag, Hei-delberg, pp. 317.

    Halle S, Stenseth NC (2000b). Chronoecology: new light through old windows a conclusion. In: Halle S, Stens-eth NC, eds. Activity Patterns in Small Mammals: An Ecological Approach. Springer-Verlag, Heidelberg, pp. 27584.

    Hicks NG, Menzel MA, Laerm J (1998). Bias in the de-termination of temporal activity patterns of syntopic Peromyscus in the Southern Appalachians. Journal of Mammalogy 79, 101620.

    Huey RB (1991). Physiological consequences of habitat selection. American Naturalist 137, S91S115.

    Jaksic FM (1982). Inadequacy of activity time as a niche difference: the case of diurnal and nocturnal raptors. Oecologia 52, 1715.

    Jones ME, Mandelik Y, Dayan T (2001). Coexistence of temporally partitioning spiny mice: roles of habitat structure and foraging behavior. Ecology 82, 216476.

  • 2011 ISZS, Blackwell Publishing and IOZ/CAS 385

    123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051

    Temporal niche overlap among mammals

    Kenagy GJ, Vsquez RA, Nespolo RF, Bozinovic F (2002). Daily and seasonal limits of time and temper-ature to activity of degus. Revista Chilena de Histo-ria Natural 75, 56781.

    Kotler BP, Brown JS, Subach A (1993). Mechanisms of species coexistence of optimal foragers: tempo-ral partitioning by two species of sand dune gerbils. Oikos 67, 54856.

    Kronfeld-Schor N, Dayan T (2003). Partitioning of time as an ecological resource. Annual Review of Ecology and Evolution 34, 15381.

    Kronfeld-Schor N, Dayan T, Elvert R, Haim A, Zisapel N, Heldmaier G (2001). On the use of the time axis for ecological separation: diel rhythms as an evolu-tionary constraint. American Naturalist 158, 4517.

    Leiner NO, Silva WR (2007). Seasonal variation in the diet of the Brazilian slender opossum (Marmosops paulensis) in a montane Atlantic forest area, south-eastern Brazil. Journal of Mammalogy 88, 15864.

    Levins R (1968). Evolution in Changing Environments: Some Theoretical Explorations. Princeton University Press, New Jersey.

    Lockard RB (1978). Seasonal change in activity pattern of Dipodomys spectabilis. Journal of Mammalogy 59, 5638.

    Merritt FJ, Vessey SH (2000). Shrews small insecti-vores with polyphasic patterns. In: Halle S, Stens-eth NC, eds. Activity Patterns in Small Mammals: An Ecological Approach. Springer-Verlag, Heidelberg, pp. 23551.

    Moran MD (2003). Arguments for rejecting the sequen-tial Bonferroni in ecological studies. Oikos 100, 4035.

    Mura R, Gonzlez LA (1982). Microhabitat selection in two Chilean cricetid rodents. Oecologia 52, 125.

    OFarrell MJ (1974). Seasonal activity patterns of ro-dents in a sagebrush community. Journal of Mam-malogy 55, 80923.

    Paise G, Vieira EM (2006). Daily activity of a neotrop-ical rodent (Oxymycterus nasutus): seasonal chang-es and influence of environmental factors. Journal of Mammalogy 87, 7339.

    Peterson RM, Batzli GO (1975). Activity patterns in nat-ural-populations of brown lemming Lemmus trimu-cronatus. Journal of Mammalogy 56, 71820.

    Pianka ER (1973). The structure of lizard communities. Annual Review of Ecology and Systematics 4, 5374.

    Pimm SL, Pimm JW (1982). Resource use, competition, and resource availability in Hawaiian honeycreepers. Ecology 63, 146880.

    Pinheiro F, Diniz IR, Coelho D, Bandeira MPS (2002). Seasonal pattern of insect abundance in the Brazilian cerrado. Austral Ecology 27, 1326.

    Priotto J, Polop J (1997). Space and time use in syntopic populations of Akodon azarae and Calomys venustus (Rodentia, Muridae). Zeitschrift fr Sugetierkunde 62, 3036.

    Pulliam HR (1986). Niche expansion and contraction in a variable environment. American Zoologist 26, 719.

    Roback PJ, Askins RA (2005). Judicious use of multiple hypothesis tests. Conservation Biology 19, 2617.

    Roll U, Dayan T, Kronfeld-Schor N (2006). On the role of phylogeny in determining activity patterns of ro-dents. Evolutionary Ecology 20, 47990.

    Rosenzweig ML (1991). Habitat selection and popula-tion interactions: the search for mechanism. Ameri-can Naturalist 137, S5S28.

    Rychlik L (2005). Overlap of temporal niches among 4 sympatric species of shrews. Acta Theriologica 50, 17588.

    Saiful AA, Rashid YN, Idris AH (2001). Niche segre-gation among three sympatric species of squirrels in-habiting a lowland dipterocarp forest, Peninsular Ma-laysia. Mammal Study 26, 13344.

    Schoener TW (1974). Resource partitioning in ecologi-cal communities. Science 185, 2738.

    Schowalter TD (2000). Insect Ecology. An Ecosystem Approach. Academy Press, San Diego.

    Shkolnik A (1971). Diurnal activity in a small desert ro-dent. International Journal of Biometeorology 15, 11520.

    Smale L, Lee T, Nunez AA (2003). Mammalian diur-nality: some facts and gaps. Journal of Biological Rhythms 18, 35666.

    Solari S (2007). Trophic relationships within a highland rodent assemblage from Manu National Park, Cusco, Peru. In: Kelt DA, Lessa EP, Salazar-Bravo J, Pat-ton JL, eds. The Quintessential Naturalist: Honoring the Life and Legacy of Oliver P. Pearson. University of California Publications in Zoology, Berkeley, pp. 22540.

    Stringer IAN, Meyer-Rochow VB (1997). Flight activi-ty of insects within a Jamaican cave: in search of the zeitgeber. Invertebrate Biology 116, 34854.

  • 2011 ISZS, Blackwell Publishing and IOZ/CAS386

    123456789

    101112131415161718192021222324252627282930313233343536373839404142434445464748495051

    E.M. Vieira and G. Paise

    Strong DR Jr, Simberloff D, Abele LG, Thistle AB (1984). Ecological Communities, Conceptual Issues and the Evidence. Academic Press, Princeton.

    Suarez OV, Bonaventura SM (2001). Habitat use and diet in sympatric species of rodents of the low Parana delta, Argentina. Mammalia 65, 16775.

    Taylor LR, Carter CI (1961). The analysis of numbers and distribution in an aerial population of macrolepi-doptera. Transactions of the Royal Entomological So-ciety of London 113, 36986.

    Vieira EM, Asta De Moraes D (2003). Carnivory and insectivory in Neotropical marsupials, In: Jones M, Dickman CR, Archer M, eds. Predators with Pouch-es: The Biology of Carnivorous Marsupials. CSIRO, Melbourne, pp. 26780.

    Vieira EM, Baumgarten LC (1995). Daily activity pat-terns of small mammals in a cerrado area from cen-tral Brazil. Journal of Tropical Ecology 11, 25562.

    Wauters LA, Dhondt AA (1987). Activity budget and foraging behavior of the red squirrel Sciurus vulgaris (Linnaeus, 1758) in a coniferous habitat. Zeitschrift fr Sugetierkunde 52, 34153.

    Wauters L, Swinnen C, Dhondt AA (1992). Activity budget and foraging behavior of red squirrels Sciurus vulgaris in coniferous and deciduous habitats. Jour-nal of Zoology 227, 7186.

    Weiher E, Keddy P, eds (1999). Ecological Assembly Rules: Perspectives, Advances, Retreats. Cambridge University Press, Cambridge.

    Zar JH (1996). Biostatistical Analysis, 3rd edn. Prentice Hall, Upper Saddle River, New Jersey.

    Ziv Y, Abramsky Z, Kotler BP, Subach A (1993). Inter-ference competition and temporal and habitat parti-tioning in two gerbil species. Oikos 66, 23746.