Abstract
Soricid mammals and lizards are small-sized, insectivorous vertebrates that are widespread and abundant in Mediterranean habitats. Because of their broad sympatry and their diet similarity, these taxa have been suspected to compete for food. Therefore, co-occurrence patterns between these taxa were studied at 72 sites in southern Italy by means of trapping methods. The assemblages were quite distinct depending on the site bioclimate: for the Lacertidae, Podarcis siculus dominated in the thermo-Mediterranean sites and P. muralis in the temperate sites, whereas, for the soricids, Suncus etruscus and two species of Crocidura were dominant in thermo-Mediterranean sites and three Sorex species in the temperate sites. The mean number of captured soricids was statistically higher in the temperate sites, and was positively related to the first component of a Principal Component Analysis summarizing three collinear study site variables (elevation, mean annual temperature, annual rainfall), the reverse being true for lizards. Co-occurrence analysis revealed that a non-segregated structure was present in the dataset, whereas a randomization algorithm showed that the assemblage of small mammals and lizards was non-randomly structured, with the frequency distribution of shrews being non-independent by site from that of lizards. However, when we divided the sites by their bioclimatic regime (thermo-Mediterranean versus temperate), the non-randomness of the community structure disappeared, thus demonstrating that interspecific competition was not the main force driving these assemblages of species. The number of shrews captured in each sampling site was however significantly negatively related to the number of lizards, this pattern being linked to the bioclimate of the various sampling sites. Overall, our data indicated that the assemblage of shrews and lizards was likely regulated essentially by local climate and not by synecological (interspecific competition) dynamics.
Introduction
Studies of co-occurrence patterns among species are interesting in order to understand aspects of community structure across spatial scales (e.g. Wiens, 1977; Schoener, 1982; Denoël and Schabetsberger, 2003; Rychlik et al., 2006). Congruent non-random patterns can result from different, not necessarily independent, co-occurrence patterns between species (e.g. Gotelli, 2000). Firstly, negatively correlated patterns of occurrence between species can be due to interspecific competition, both currently or historically (Gotelli and Graves, 1996; Gotelli, 2000), when and where species exploit limited resources (Hanski, 1987; Shorrocks, 1990; Ray and Sunquist, 2001; Krijger et al., 2001). Competition for resources is seen as one of the primary interactions limiting diversity (Chesson and Kuang, 2008). Interspecific competition is considered to be an important structuring force for natural communities comprising either closely related (Connell, 1983; Brönmark et al., 1991) or distantly related and phenotypically dissimilar species (Capizzi and Luiselli, 1996; Mokany and Shine, 2003). Secondly, negatively correlated patterns of occurrence between species may reflect common regional histories (e.g. vicariance) and/or the deep evolutionary history of species that have followed divergent evolutionary trajectories without any direct effect of past competition (e.g. Vitt and Pianka, 2005).
Co-occurrence patterns traditionally have been studied among populations of closely related taxa (e.g., Atchison, 1987; Capula et al., 1993; Capula and Luiselli, 1994; Churchfield and Rychlik, 2006; Luiselli, 2006a; Vignoli and Luiselli, 2012). Comparatively fewer studies analysed co-occurrence patterns between unrelated taxa, although interspecific competition may occur also among phylogenetically unrelated organisms (Brown and Davidson, 1977; Capizzi and Luiselli, 1996; Luiselli, 2006b).
Contoli (1988) observed that barn owls (Tyto alba) tended to prey on shrews (family Soricidae) more frequently than on rodents at sites where lizards were particularly rare. Postulating that shrews and lizards are in competition for food (Contoli, 1988), he suggested that, given the opportunist habits of this bird of prey, intensified shrew-predation may depend on the fact that shrews predominate in sites where their lizard competitors are scarce. In addition, based on the observed apparent negative correlation in abundance between shrews and lizards in comparison of sites from southern Italy (Calabria) and Spain (Andalusia), Cagnin et al. (1998) hypothesized that this differential prevalence of shrews may have been due to the relative abundances of lizards, under the assumption of an inter-taxon competition for arthropod prey. However, the hypothesis of interspecific competition between these insectivorous vertebrates has never been adequately tested.
Here, we analyze the co-occurrence patterns between shrews (Soricidae; Mammalia) and lizards (Lacertidae and Scincidae; Reptilia) at several sites in southern Italy (Calabria). Shrews eat a wide variety of arthropods (beetles, larval butterflies and other insects, mites, harvestmen, woodlice, centipedes, spiders, earthworms, insect larvae) and their diets appear to reflect what is available as prey (Churchfield, 1990; Canova and Fasola, 1993). Lacertids and scincids are also mainly arthropod-eating and they are considered as active foragers (Kabisch and Engelmann, 1969; Pérez-Mellado and Corti, 1993; Corti et al., 2011), but occasionally also eat small vertebrates (Capula and Aloise, 2011) and plants (Van Damme, 1999). However, lizards and shrews differ fundamentally in their metabolism, physiology, and thermal biology. These metabolic differences may possibly generate divergent patterns in the distribution of abundance of lizards versus shrews in areas with temperate versus Mediterranean climate.
In this paper, we (1) test whether the distribution of abundance of shrews and lizards is negatively related across a set of independent sites where surveys have been conducted, and (2) try to explain the reasons behind the observed patterns, with a suite of statistical tools including null models and Monte Carlo simulations (Gotelli and Graves, 1996). The key questions analysed in the present study are: (1) what are the relationships patterns of shrews and lizards across the range of sampling stations? (2) if any non-casual pattern of distribution can be observed between shrews and lizards, what are the main reasons driving the observed trends?
Thorough evaluation of geographic patterns is often mandatory before one can design a comprehensive and compelling field experiment on co-occurrence and competition between coexisting organisms. However, geographic pattern analysis cannot substitute for actual experimentation and manipulations. Thus, we used a large number of different study sites (n = 72) in order to make our study sounder in the absence of a true experimental manipulation. However, our approach and our use of sophisticated statistical tools (Gotelli and Graves, 1996) cannot exclude complications in cases where the same process may lead to very different patterns. For example, competitors often are positively associated at a given scale because they occur only where their resources are found, but within those conditions they may be negatively associated at a smaller scale (e.g., Connell, 1983). Thus, addressing the issue of scale can be very important for competition studies (Connell, 1983; Goldberg and Barton, 1996), and the use of multiple study sites does not ensure that the scale issue is being correctly addressed. Therefore, the present study may be particularly informative because it makes use of a high number of independent experimental sites.
Materials and methods
Protocol
Lizards and shrews were collected during the course of studies on the ecological distribution of small mammal communities in northern Calabria, southern Italy (Cagnin et al., 1991, 1996, 1998; Brandmayr et al., 1997). The study was performed along a transect of about 70 km in length and running from the Tyrrhenian coast to the Ionian coast of Calabria (southern Italy) (Appendix 1), during 1989-1992, inclusive.
The transect comprised a total of 72 sampling stations, situated both in natural and human-made habitats, representative of different environmental and landscape characteristics. The sites were situated at altitudes ranging from 0-1620 m a.s.l. and were placed in all the main habitat types found in this region of Italy. At each study station we recorded the following habitat and climatic variables: (i) type of macro-habitat (ranked according to vegetation structure, i.e. woodland, ecotones between woodlands and shrublands; shrublands; grasslands; cultivations); (ii) elevation (m a.s.l.); (iii) mean annual temperature (recorded using data provided by the A.R.P.A.CAL, Agenzia Regionale per la Protezione dell’Ambiente della Calabria); and (iv) mean annual rainfall (recorded using data provided by the A.R.P.A.CAL, Agenzia Regionale per la Protezione dell’Ambiente della Calabria). The study stations belonged to two distinct bioclimatic types, i.e. thermo-Mediterranean and temperate, and were independent of each other in terms of the movement-distance capabilities of the study animals at the temporal scale of our study (minimum distance between barycentres = 500 m). Indeed, both lizards and srews are characterized by small home ranges (Amori et al., 2008; Corti et al., 2010), and in addition there are between-sites barriers such as busy roads, rivers, etc.…
To capture lizards and shrews we employed pitfall traps, commonly used for sampling almost all the species of terrestrial small mammals (Pucek, 1969; Pankakoski, 1979) but also for censusing reptiles (Morton et al., 1988; Friend et al., 1989; Blomberg and Shine, 1996). This method had been tested previously to obtain quantitative data on relative abundance of soricids and small lacertids in Mediterranean habitats (Cagnin et al., 1998).
In each station 15 pitfall-traps were placed at a distance of 10 m one from another. This number of pitfalls is accurate enough for studying shrew abundance at the local scale (Churchfield et al., 1999; Nicolas et al., 2003). The pitfall traps (height: 18 cm; diameter: 8 cm) were obtained by cutting 1.5 l plastic bottles. The whole study period lasted one year, with twelve sampling sessions for each trap, that is one session in each month (Cagnin et al., 1998). Thus, all phases of the biological cycle of both lizards and shrews were investigated. During each session, traps were kept active for five days in each sampling site, and inspected every day, and three times a day, to minimize risks of death for the captured animals. Accumulation of rainfall inside the trap was avoided by appropriate placement of stones preventing water entering, and making several holes in the bottom of each plastic trap. Shrews and lizards were individually marked, and released immediately after being discovered into a trap. Shrews and lizards were individually marked by decoloration of the hair (shrews) or painting on the back with long-standing paint (lizards). Mortality rate was 1.1% in shrews, and 0% in lizards.
Statistical analyses
To calculate the species relative abundance () at the different sampling stations, the same index was used for shrews and lizards, i.e. the number of individuals captured by 100 traps/day (Pucek, 1969; Pankakoski, 1979 modified):
where N is the number of captured individuals, n is the number of trapping days, t the number of active traps (Cagnin et al., 1998).
We tested the correlation between RA and climate variables for both lizards and shrews using a Pearson’s correlation coefficient. However, because elevation, mean annual temperature and annual rainfall were significantly collinear (at least P < 0.00001), we tested the correlation between RA and PC1 of a Principal Component Analysis (PCA1) of these three variables (Appendix 2). PCA was run on a variance-covariance matrix. PC1 explained 78.78% of the total variance and was positively correlated with elevation and rainfall, and negatively with mean air temperature.
Community ecology theory suggests that, when interspecific competition is a key factor in the organization of living assemblages of species, the community data matrix should have a non-random structure (e.g., Gotelli and Graves, 1996). This random/non-random structure can be explored statistically by means of purposely designed null model algorithms (e.g., Gotelli and Graves, 1996), using (i) count data and (ii) presence/absence data. To evaluate whether the co-occurring assemblages of shrews and lizards were structured randomly using count data, we contrasted the observed data matrix with random “pseudo-communities” generated by Monte Carlo simulations (Gotelli and Graves, 1996; Luiselli, 2006a, 2008), using Pianka’s (1986) overlap formula and randomizing the original species resource utilization matrices, from which Pianka’s overlap was calculated, by shuffling the original values among resource states. Resource states were the various study stations. We used two randomization algorithms (RA2 and RA3) from Lawlor (1980), as they are particularly robust for niche overlap studies (Gotelli and Graves, 1996). RA2 keeps the zero states of the observed matrix but does not keep the observed value of niche breadth at each simulation; RA3 conserves the observed value of niche breadth for each species at each simulation but destroys the guild structure manifested by the zero structure of the resource utilization matrix (Gotelli and Graves, 1996). For each pair of species, 3 × 104 random Monte Carlo permutations were generated. This number of permutations ensures that algorithm biases are avoided (Lehsten and Harmand, 2006). Niche overlap values were calculated for each of these randomly generated matrices, and species-pair and community-summary statistics were computed (Friggens and Brown, 2005; Vignoli and Luiselli, 2012). Actual overlap values were then compared to the distributions of expected values. Structure (= interspecific competition) was determined when Pobs⩽exp = 0.05 (Gotelli and Graves, 1996). In all cases, equiprobable resource use was assumed a priori in the analyses.
To evaluate whether the co-occurring assemblages of shrews and lizards were structured randomly using presence/absence data, we analysed patterns of co-occurrence by the C-score (Stone and Roberts, 1990) to measure species co-occurrences. The C-score was calculated for all unique species pairs in the matrix and averaged as an index of community co-occurrence (Stone and Roberts, 1990; Ulrich and Gotelli, 2007). The larger the C-score, the more, on average, species pairs are segregated in their occurrences, and thus the more intensified is the competition strength. Observed C-scores from the true data matrices were compared to simulated C-scores obtained from random pseudo-communities generated by 30 000 Monte Carlo permutations. The fixed row-equiprobable column (FE) algorithm was used as the null model algorithm because this is appropriate for small sample plots (Sanders et al., 2003). FE preserved row totals (species occurrences) but columns (sample plots) were treated as equiprobable. This model allows species number per sample plot to vary in the null assemblages and also takes into account empty sample plots, which did not contain given species but which might have been occupied in a randomly assembled community. This algorithm has been demonstrated to behave well in benchmark tests for type I and type II statistical errors (Gotelli, 2000). For the observed presence/absence matrix we created 30 000 random matrices by reshuffling the elements of each row of the matrix. We then calculated the C-score of each random matrix and estimated the tail probability (one-tailed test) of the observed matrix by comparing it with the histogram of simulated values. We implemented this null model with a variation of the sequential swap algorithm (Gotelli, 2000; Ulrich and Gotelli, 2007) as done in Ulrich and Gotelli (2007), but with 30 000 rather than 25 000 randomly sampled sub-matrices with the same row and equiprobable column totals.
Data normality for the used variables was tested by Kolmogorov-Smirnov test. When data variables were not normally distributed, we log-transformed them in order to achieve normality and to apply parametric tests. When variables could not achieve normality even after log-transformation, then non-parametric tests were used. Means are followed by ±standard error.
We used EcoSim software (Gotelli and Entsminger, 2011) to calculate overlap indices and generate Monte Carlo simulations. All other analyses were conducted using SPSS (SPSS 16.0 for Windows) and Statistica (Statistica 7.0 for Windows); all tests were two-tailed with alpha set at 0.05.
Results
General considerations
Overall, we captured 427 soricids (N = 427; Sorex antinorii/samniticus N = 115, S. minutus N = 111, Suncus etruscus N = 102, Crocidura leucodon N = 42, C. suaveolens N = 57), 849 lacertids (N = 849; Podarcis siculus N = 766, P. muralis N = 10, Podarcis sp. N = 69, and Lacerta bilineata N = 4) and 2 scincids (Chalcides chalcides), across the 72 sampling sites (Appendix 3). There was a considerable inter-site variation in the number of captured animals, with the peaks being 84 for soricids and 48 for lacertids. Scincids were captured only in two sites, each with just a single trapped individual. The total number of recorded species was 5/6 soricids (the uncertainty arises from the difficulty of morphologically distinguishing Sorex antinorii from S. samniticus), 3 lacertids and 1 scincids.
In terms of bioclimate (Appendix 3), the assemblages were quite distinct: for the Lacertidae, Podarcis siculus dominated in the thermo-Mediterranean sites and P. muralis in the temperate sites. As for the Soricids, Suncus etruscus and two Crocidura species were dominant in thermo-Mediterranean sites and the three Sorex species in the temperate sites.
Relationship between, respectively, number of captured animals (A = lizards, B = shrews) and PCA1 of the principal component analysis on elevation, mean annual temperature and annual rainfall. For the statistical details, see the text.
Citation: Amphibia-Reptilia 36, 3 (2015) ; 10.1163/15685381-00002998
The abundance of both lizards and shrews was not significantly correlated to any of the considered habitat variables (Spearman’s correlation coefficient, in all cases P ⩾ 0.05).
There was a statistically significant difference (Kruskal-Wallis ANOVA: H1,72 = 33.11, P < 0.0001) between the mean number of shrews captured in the thermo-Mediterranean and in the temperate sites (respectively, versus 16.33 ± 10.52 individuals), and the same was also true for lizards (respectively, versus 2.79 ± 4.83 individuals). However, the direction of these differences was opposite in the two taxa: shrews were more abundant in the temperate sites, but lizards were more abundant overall in the thermo-Mediterranean sites. However, the latter conclusion masks differences between species. More specifically, the frequency of capture of P. siculus was statistically higher in thermo-Mediterranean sites (Kruskal-Wallis , df = 1, P < 0.00001), whereas P. muralis was significantly more abundant in the temperate sites (Kruskal-Wallis , df = 1, P < 0.001).
Further analyses revealed that the number of captured lizards was negatively related to increases in PCA1 (r = −0.567, r2 = 0.322, P < 0.0001; fig. 1A), whereas the number of captured shrews was positively related to increases in PCA1 (r2 = 0.400, P < 0.05; fig. 1B). Hence, unsurprisingly, the (log + 1) of the number of soricids captured in each sampling site was significantly negatively related to the (log + 1) of the number of lizards, although with a weak predictive power (r = −0.140, n = 72, P < 0.001; fig. 2).
Relationships between (log + 1) of the number of soricids and (log + 1) of the number of lizards captured in each sampling site. For the statistical details, see the text.
Citation: Amphibia-Reptilia 36, 3 (2015) ; 10.1163/15685381-00002998
Null model analyses of community structure: all sites pooled
Our co-occurrence analysis revealed that a non-segregated structure was present in the dataset (observed index = 3.036, simulated indices = 2.970, variance of simulated indices = 0.545, Pobs⩽exp = 0.537, Pobs⩾exp = 0.482), with the number of species co-occurrences not significantly less than expected by chance (t-test, P = 0.356).
The RA2 algorithm revealed that the assemblage of small mammals and lizards was non-randomly structured, with the frequency distribution of shrews being non-independent by site from that of lizards (observed overlap = 0.187, of simulated indices = 0.477, variance of simulated indices = 0.020, Pobs⩽exp = 0.029, Pobs⩾exp = 0.071), whereas the RA3 algorithm did not provide significant evidence of structure.
Null model analyses of community structure: sites divided by their bioclimatic regime
In the thermo-Mediterranean sites, neither the RA2 nor the RA3 algorithms provided no evidence for a structure in the shrew-lizard assemblages (observed overlap = 0.147, RA2 – of simulated indices = 0.152, variance of simulated indices = 0.00015, Pobs⩽exp = 0.351, Pobs⩾exp = 0.649; RA3 – of simulated indices = 0.120, variance of simulated indices = 0.00033, Pobs⩽exp = 0.926, Pobs⩾exp = 0.074). The same patterns also emerged in the temperate sites, where neither the RA2 nor the RA3 algorithms provided any evidence of non-random community structure (observed overlap = 0.209, RA2 – of simulated indices = 0.215, variance of simulated indices = 0.00063, Pobs⩽exp = 0.396, Pobs⩾exp = 0.604; RA3 – of simulated indices = 0.153, variance of simulated indices = 0.0011, Pobs⩽exp = 0.936, Pobs⩾exp = 0.064).
Assemblage structuring within taxa
Distribution of C-score of Soricidae was consistent with a non-random segregation by species (observed index = 228.7, of simulated indices = 202.6, variance of simulated indices = 4.22, Pobs⩽exp = 0.999, Pobs⩾exp = 0.001). Despite an overall low mean overlap among species (observed overlap = 0.233), a competitive-based structure among species was uncovered by neither RA3 ( of simulated indices = 0.153, variance of simulated indices = 0.001, Pobs⩽exp = 0.980, Pobs⩾exp = 0.02) nor RA2 ( of simulated indices = 0.264, variance of simulated indices = 0.0005, Pobs⩽exp = 0.092, Pobs⩾exp = 0.908).
Within lizards, there was no evidence of non-random co-occurrence patterns using C-score (observed index = 73.33, of simulated indices = 67.78, variance of simulated indices = 8.996, Pobs⩽exp = 0.945, Pobs⩾exp = 0.092). An absence of competitive structure was also demonstrated using RA3 (observed index = 0.082, of simulated indices = 0.091, variance of simulated indices = 0.002, Pobs⩽exp = 0.483, Pobs⩾exp = 0.517) and RA2 ( of simulated indices = 0.052, variance of simulated indices = 0.0007, Pobs⩽exp = 0.865, Pobs⩾exp = 0.135).
Discussion
Our study documented that the two groups of species (lizards and shrews) co-occur in the broad sense in a region of southern Italy, and also might broadly consume similar suites of prey or resources. This fact does not mean that they are truly syntopic, or that they compete in any meaningful way, as they may be segregated temporally or on a fine spatial scale by microhabitat. Presumably competition, where it to occur, would be driven by years of limited abundance. Anyway, we documented that there was a non-random structure in the assemblages of lizards and shrews. That is, overall, lizards (essentially Podarcis siculus) were more abundant in sites where shrews were least abundant. However, when the same null model analyses were run on the partial datasets (i.e. separating the sites by their bioclimatic regime), the above-mentioned non-randomness disappeared, and the co-occurrences of shrews and lizards appeared to be driven by chance. In addition, there was no structure within either shrews or lizards when these two taxa were considered independently in the null model analyses. Thus, the apparent avoidance pattern between shrews and lizards was driven essentially by the bioclimatic regime, given that a random assortment of species was always detected by both RA2 and RA3 algorithms when sites from different bioclimatic regimes were analysed separately. Therefore, we conclude that the abundance of either of the two groups at a given site did not influence the abundance of the other group, whereas the bioclimatic conditions did so. In any case, it should be emphasized that our lizard data concern only a single species with appropriate sample sizes (Podarcis siculus); thus our conclusions cannot be generalized to lacertids overall or in other bioclimatic contexts.
Based on the above considerations, it is clear that the significant correlational evidence (linear regression model) of the apparent avoidance pattern between shrews and lizards was not due to a true cause-effect relationship, as also shown by the fact that both taxa appeared to be present in most sites (which resulted in no non-random pattern being revealed by C-score analysis).
A closer examination of the data shows that shrews appeared least abundant in the thermo-Mediterranean sites with P. siculus as dominant lizard. On the other hand, shrews were abundant in the more temperate sites where the dominant lizard species was P. muralis. In particular, shrew species belonging to the genera Sorex and Crocidura rarely co-occurred, and the same was true for the lizards P. muralis and P. siculus. Although a potential interpretation is that, alternatively, the species that is a superior competitor has limited the distribution of the inferior competitor (Diamond, 1975), it is difficult to accept this hypothesis, not only because of the results of our null model analyses (see above), but also because species’ absences may actually reflect non-detection rather than absence (Cam et al., 2000; MacKenzie et al., 2004). Last but not least, species that co-occur less frequently than expected by chance may be doing so because of differences in habitat and /or other external conditions (MacKenzie et al., 2004), as seems to be the case in our study system.
Our analyses, also are not consistent with Contoli’s (1988) hypothesis, in terms of the huge differences in the metabolic rates of these two groups. Nonetheless, the available data on the food habits of Mediterranean Shrews (Churchfield, 1990; Canova and Fasola, 1993) widely confirm that their trophic composition is very similar to that of the lizards (e.g., Capula et al., 1993), and thus it might still be possible that interspecific competition between these two independently evolved taxa occurs at particular times of the year and in sites where the food resource is limited (Schoener, 1974). We tentatively anticipate that such limiting-resource-conditions may occur especially in arid sites, and especially during the summertime, when arthropod availability tends to collapse (e.g., Brown and Davidson, 1977). Further experiment is needed to test this working hypothesis.
Acknowledgements
Authorization to perform this study and to capture animals was released by the Italian Ministry of Environment. Longino Contoli is gratefully thanked for helpful discussion and for having firstly presented to us the main issue of our study. Giovanni Amori and Massimo Capula also participated to several round tables and discussions related to co-occurrence patterns in lizards and shrews, and helpfully commented upon a earlier draft of this manuscript. Two anonymous referees considerably improved the submitted draft.
References
Aitchison C.W. (1987): Review of winter trophic relations of soricine shrews. Mammal Review 17: 1-24.
Amori G., Contoli L., Nappi A., Eds (2008): Fauna d’Italia, Vol. XLIV. Mammalia II. Erinaceomorpha, Soricomorpha, Lagomorpha, Rodentia. Edizioni Calderini de Il Sole 24 ORE Editoria Specializzata S.r.l., Bologna.
Blomberg S., Shine R. (1996): Reptiles. In: Ecological Census Techniques. A Handbook, p. 218-226. Sutherland W.J., Ed., Cambridge University Press, Cambridge.
Brandmayr P., Cagnin M., Mingozzi T., Scalercio S., Pizzolotto R. (1997): Misura efficace della biodiversità animale in ambienti mediterranei e sue applicazioni. SITE Atti 18: 581-586.
Brönmark C., Rundle S.D., Erlandsson A. (1991): Interactions between freshwater snails and tadpoles: competition and facilitation. Oecologia 87: 8-18.
Brown J.H., Davidson D.W. (1977): Competition between seed-eating rodents and ants in desert ecosystems. Science 196: 880-882.
Cagnin M., Pizzolotto R., Barbieri A., Tripepi S., Rossi F., Ripolo D., Aloise P., Brandmayr P. (1991): Zoocenosi (Coleotteri Carabidi, Rettili, micromammiferi terricoli) in ambienti forestali e prativi della Catena Costiera di Calabria. SITE Atti 12: 455-459.
Cagnin M., Aloise G., Garofalo G., Milazzo C., Cristaldi M. (1996): Les communautés de petit Mammifères terrestres des trois “fiumare” de la Calabre (Italie du sud). Vie et Milieu 46: 319-326.
Cagnin M., Moreno S., Aloise G., Garofalo G., Villafuerte R., Gaona P., Cristaldi M. (1998): A comparative study of Spanish and Italian terrestrial small mammal coenoses of different biotopes in Mediterranean regions. Journal of Biogeography 25: 1105-1113.
Cam E., Nichols J.D., Hines J.E., Sauer J.R. (2000): Inferences about nested subsets structure when not all species are detected. Oikos 91: 428-434.
Canova L., Fasola M. (1993): Food habits and trophic relationships of small mammals in six habitats of the northern Po plain (Italy). Mammalia 57: 189-200.
Capizzi D., Luiselli L. (1996): Feeding relationships and competitive interactions between phylogenetically unrelated predators (owls and snakes). Acta Oecologica 17: 265-284.
Capula M., Aloise G. (2011): Extreme feeding behaviours in the Italian wall lizard, Podarcis siculus. Acta Herpetologica 6: 11-14.
Capula M., Luiselli L., Rugiero L. (1993): Comparative ecology in sympatric Podarcis muralis and P. sicula (Reptilia: Lacertidae) from the historical centre of Rome: What about competition and niche segregation in an urban habitat? Bollettino di Zoologia 60: 287-291.
Chesson P., Kuang J.J. (2008): The interaction between predation and competition. Nature 456: 235-238.
Churchfield S., Rychlik L. (2006): Diets and coexistence in Neomys and Sorex shrews in Białowieża forest, eastern Poland. Journal of Zoology London 269: 381-390.
Churchfield S. (1990): The Natural History of Shrews. Christopher Helm, A & C Black, London.
Churchfield S., Nesterenko V.A., Shvarts E.A. (1999): Food niche overlap and ecological separation amongst six species of coexisting forest srews (Insectivora: Soricidae) in the Russian Far East. Journal of Zoology London 248: 349-359.
Connell J.H. (1983): On the prevalence and relative importance of interspecific competition: evidence from field experiments. American Naturalist 122: 661-696.
Connor E.F., Simberloff D. (1979): The assembly of species communities: chance or competition? Ecology 60: 1132-1140.
Contoli L. (1988): La nicchia trofica di allocco (Strix aluco) e barbagianni (Tyto alba) in Italia: acquisizioni e problemi. Naturalista Siciliano s. 4 (suppl.): 129-143.
Corti C., Capula M., Luiselli L., Razzetti E., Sindaco R., Eds (2010): Fauna d’Italia, Vol. XLV. Reptilia, Edizioni Calderini de Il Sole 24 ORE Editoria Specializzata S.r.l., Bologna.
Denoël M., Schabetsberger R. (2003): Resource partitioning in two heterochronic populations of Greek Alpine newts, Triturus alpestris veluchiensis. Acta Oecologica 24: 55-64.
Diamond J.M. (1975): Assembly of species communities. In: Ecology and Evolution of Communities, p. 342-444. Cody M.L., Diamond J.M., Eds, Harvard University Press, Cambridge, Massachusetts, USA.
Friend G.R., Smith G.T., Mitchell D.S., Dickman C.R. (1989): Influence of pitfall and drift fence design on capture rates on small vertebrates in semi-arid habitats of Western Australia. Australian Wildlife Research 16: 1-10.
Friggens M.H., Brown J.H. (2005): Niche partitioning in the cestode communities of two elasmobranches. Oikos 108: 76-84.
Goldberg D.E., Burton A.M. (1996): Patterns and consequences of interspecific competition in natural communities: a review of field experiments with plants. American Naturalist 139: 771-801.
Gotelli N.J. (2000): Null model analysis of species co-occurrence patterns. Ecology 81: 2606-2621.
Gotelli N.J., Entsminger G.L. (2011): EcoSim: null modelling software for ecology, v.7.0. Acquired Intelligence Inc. and Kesey-Bear, Jericho, http://garyentsminger.com/ecosim/index.htm.
Gotelli N.J., Graves G.C. (2006): Null Models in Ecology. Smithsonian Institution Press, Washington, D.C.
Hanski I. (1987): Colonization of ephemeral habitats. In: Colonization, Succession and Stability, p. 155-185. Gray A.J., Crawley M.J., Edwards P.J., Eds, Blackwell, London.
Kabisch K., Engelmann K.E. (1969): Zur Nahrung von Lacerta muralis (Laurenti) in Ostbulgarien. Zool. Abh. 30: 89-92.
Krijger C.L., Peters Y.C., Sevenster J.C. (2001): Competitive ability of neotropical Drosophila predicted from larval development times. Oikos 92: 325-332.
Lawlor L.L. (1980): Structure and stability in natural and randomly constructed competitive communities. American Naturalist 116: 394-408.
Lehsten V., Harmand P. (2006): Null models for species cooccurrence patterns: assessing bias and minimum iteration number for the sequential swap. Ecography 29: 786-792.
Luiselli L. (2006a): Resource partitioning and interspecific competition in snakes: the search for general geographical and guild patterns. Oikos 114: 193-211.
Luiselli L. (2006b): Food niche overlap between sympatric potential competitors increases with habitat alteration at different trophic levels in rain-forest reptiles (omnivorous tortoises and carnivorous vipers). Journal of Tropical Ecology 22: 695-704.
Luiselli L. (2008): Do lizards communities partition the trophic niche? A worldwide meta-analysis using null models. Oikos 117: 321-330.
MacKenzie D.I., Bailey L.L., Nichols J.D. (2004): Investigating species co-occurrence patterns when species are detected imperfectly. Journal of Animal Ecology 73: 546-555.
Mokany A., Shine R. (2003): Competition between tadpoles and mosquito larvae. Oecologia 135: 615-620.
Morton S.R., Gillam M.W., Jones K.R., Fleming M.R. (1988): Relative efficiency of different pit-trap systems for sampling reptiles in spinifex grasslands. Australian Wildlife Research 15: 571-577.
Nicolas V., Barriere P., Colyn M. (2003): Impact of removal pitfall trapping on the community of shrews (Mammalia: Soricidae) in two African tropical forest sites. Mammalia 67: 133-138.
Pankakoski E. (1979): The cone trap – a useful tool for index trapping of small mammals. Annales Zoologici Fennici 16: 144-150.
Pérez-Mellado V., Corti C. (1993): Dietary adaptations and herbivory in lacertid lizards of the genus Podarcis from western Mediterranean islands (Reptilia: Sauria). Bonner Zoologische Beitraege 44: 193-220.
Pianka E.R. (1986): Ecology and Natural History of Desert Lizards. Analyses of the Ecological Niche and Community Structure. Princeton University Press, Princeton, New Jersey.
Pucek Z. (1969): Trap response and estimation of numbers of shrews in removal catches. Acta Theriologica 14: 403-426.
Ray J.C., Sunquist M.E. (2001): Trophic relations in a community of African rainforest carnivores. Oecologia 127: 395-408.
Rychlik L., Ramalhinho G., Polly P.D. (2006): Response to environmental factors and competition: skull, mandible and tooth shapes in Polish water shrews (Neomys, Soricidae, Mammalia). Journal for Zoological and Systematic and Evolutionary Research 44: 339-351.
Sanders N.J., Gotelli N.J., Heller N.E., Gordon D.M. (2003): Community disassembly by an invasive species. Proceedings of the National Academy of Sciences of the USA 100: 2474-2477.
Schoener T.W. (1974): The compression hypothesis and temporal resource partitioning. Proceedings of the National Academy of Sciences of the USA 71: 4169-4172.
Schoener T.W. (1982): The controversy over interspecific competition. American Scientist 70: 586-595.
Shorrocks B. (1990): Coexistence in a patchy environment. In: Living in a Patchy Environment, p. 91-106. Shorrocks B., Swingland I.S., Eds, Oxford Univ. Press, Oxford.
Stone L., Roberts A. (1990): The checkerboard score and species distributions. Oecologia 85: 74-79.
Toschi A. (1959): Insectivora. In: Mammalia. Generalità, Insectivora, Chiroptera. Fauna d’Italia. Vol. IV. Toschi A., Lanza B., Eds, Calderini ed., Bologna.
Ulrich W. (2004): Species co-occurrences and neutral models: reassessing J.M. Diamond’s assembly rules. Oikos 107: 603-609.
Ulrich W., Gotelli N.J. (2007): Disentangling community patterns of nestedness and species co-occurrence. Oikos 116: 2053-2061.
Van Damme R. (1999): Evolution of herbivory in lacertid lizards: effects of insularity and body size. Journal of Herpetology 33: 663-674.
Vignoli L., Luiselli L. (2012): Dietary relationships among coexisting anuran amphibians: a worldwide quantitative review. Oecologia 169: 499-509.
Vitt L.J., Pianka E.R. (2005): Deep history impacts present-day ecology and biodiversity. Proceedings of the National Academy of Sciences of the USA 102: 7877-7881.
Wiens J.A. (1977): On competition and variable environments. American Scientist 65: 590-597.
Footnotes
Associate Editor: Gabriel Blouin-Demers.
Geographic (UTM) coordinates of the sampling sites in southern Italy.
Scatter plot of a PCA of elevation, mean annual temperature, and annual rainfall data from the 72 study stations. For the statistical details, see the text.
Citation: Amphibia-Reptilia 36, 3 (2015) ; 10.1163/15685381-00002998
Raw data on the number of soricids and lizards captured by sampling site. Bioclimate: M = thermo-Mediterranean; T = temperate.
