You are on page 1of 7

Journal of Arid Environments xxx (2011) 1e7

Contents lists available at ScienceDirect

Journal of Arid Environments


journal homepage: www.elsevier.com/locate/jaridenv

Historical and ecological determinants of dung beetle assemblages in two arid zones of central Mexico
G. Halffter a, *, J.R. Verd b, C.E. Moreno c, V. Halffter a
a

Instituto de Ecologa, A.C., A.P. 63, Carretera Antigua a Coatepec No. 351, El Haya, 91070 Xalapa, Veracruz, Mexico Universidad de Alicante, Centro Iberoamericano de la Biodiversidad CIBIO, Alicante, Spain c Universidad Autnoma del Estado de Hidalgo, Pachuca, Mexico
b

a r t i c l e i n f o
Article history: Received 26 May 2010 Received in revised form 2 June 2011 Accepted 17 June 2011 Available online xxx Keywords: Aphodiini Dung beetles Metztitln Scarabaeinae Tehuacn Thermoregulation

a b s t r a c t
This work focuses on coprophagous beetles (Scarabaeinae and Aphodiini) from two arid zones located in the center of Mexico: Barranca de Metztitln, Hidalgo, and Zapotitln-Salinas, in the high zone of the Tehuacn-Cuicatln depression. The rst site has a rich representation (in terms of species, individuals, and biomass) of beetle fauna from the dry Mexican Altiplano, whereas the beetles from Zapotitln correspond to a reduced subset of the fauna from the Balsas and the middle and lower zones of the Tehuacn-Cuicatln depression. Although owering plants from both arid zones show the same biogeographical afnities, there is a signicant contrast in the degree of endemism, with the highest plant endemism in the Tehuacn-Cuicatln Valley, and no a single endemic species within beetles. In this work, we propose a biogeograpical-historical explanation for this contrasting phenomenon. Beetles assemblages from both zones included in this study are characterized by the marked dominance, both in terms of individuals and biomass, of one species of the genus Canthon. Success of the dominant Canthon species can be explained by their thermoregulation mechanisms, whereas dominance of other mediumsized species could be associated with the availability of trophic resources as a result of anthropic inuence. 2011 Elsevier Ltd. All rights reserved.

1. Introduction In central Mexico, arid zones originate from an orographic shadow effect produced by the Sierra Madre Oriental that prevents humidity from the Gulf of Mexico from reaching these areas. These environmental conditions coupled with the geographic isolation of the area have resulted in a ora that, although with strong Neotropical afnities, presents a high degree of endemism. In general, owering plants in the Mexican xerophytic region (Rzedowski, 1978), including both arid zones studied in this work, have an endemism of 50e75% at the species level (Rzedowski, 1978, 1991, 1993). However the amount of endemic regional fauna does not follow the same pattern as in the plants, and cannot be generalized. The available literature examining levels of endemism for coprophagous beetles seems to indicate a priori a very different scenario from that of the owering plants. Although the composition of the assemblages also has a strong Neotropical afnity, the number of endemics is practically zero.
* Corresponding author. Tel.: 52 228 8421842; fax: 52 228 8121897. E-mail addresses: gonzalo.halffter@inecol.edu.mx (G. Halffter), jr.verdu@ua.es (J.R. Verd), cmoreno@uaeh.edu.mx (C.E. Moreno). 0140-1963/$ e see front matter 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.jaridenv.2011.06.013

It is worth noting that although throughout the Mexican Altiplano, including arid zones, there are a number of owering plants and insect groups with Neotropical afnities. In addition, there are also older elements of Septentrional origin whose speciation has occurred in the Altiplano. This dual biogeographical association (Neotropical and Septentrional) is shared between the Altiplano and the wide contact and superposition zone found between the Nearctic and Neotropical regions described by Halffter as the Mexican Transition Zone (1964, 1976 and 1987). The Biosphere Reserves of Barranca de Metztitln (Hidalgo) and Tehuacn-Cuicatln (Puebla-Oaxaca) (Fig. 1) share climatic conditions but have different biogeographic histories. Coprophagous beetles of Metztitln comprise a rich representation of fauna from the Mexican Altiplano, especially the drier northern and western zones. On the other hand, the Reserve of Tehuacn-Cuicatln (near Zapotitln-Salinas) has a far lower diversity of beetles found in the Balsas and the medium and lower zones of the Tehuacn-Cuicatln depression. The main difference between the assemblages of beetles from the Altiplano and those of the Balsas and Tehuacn-Cuicatln depressions is the former (both of Neotropical and Septentrional afnities) is comprised of species evolved in situ, whereas the dominant species of the latter are Neotropical elements having

Please cite this article in press as: Halffter, G., et al., Historical and ecological determinants of dung beetle assemblages in two arid zones of central Mexico, Journal of Arid Environments (2011), doi:10.1016/j.jaridenv.2011.06.013

G. Halffter et al. / Journal of Arid Environments xxx (2011) 1e7

Fig. 1. Map with the locations of Metztitln and Tehuacn (Zapotitln-Salinas).

arrived more recently to the Mexican Transition Zone, and thus have closer afnities to Central and South American faunas. Another potential explanation for the composition and structure of both beetle assemblages is the anthropic effect, as a high human density has characterized central Mexico since Precolombine times, with a further signicant impact to the soil after European colonization (Cook, 1949). The higher zones of the Tehuacn-Cuicatln depression are a prime example of this old anthropic effect, as it is documented that beginning at least 900 years ago, the incremental increase in agricultural activity (mainly of maize) resulted in a higher demographic density causing signicant soil erosion (McAuliffe et al., 2001). Beginning with the European colonization and continuing to the present day, agricultural activities, mainly through the introduction of cattle, have continued to transform the arid and semiarid landscapes of central Mexico. Nowadays deforestation caused by carbon production, sheep overpopulation, and feralization of domestic animals such as donkeys, further modify the vegetation structure of both Biosphere Reserves, potentially leading to changes in the diversity of several groups of organisms. This study compares both arid zones of Barranca de Metztitln and Tehuacn-Cuicatln in order to compile data to aid in understanding the mechanisms behind the formation and evolution of dung beetles assemblages. Thus, our objectives were to a) describe the richness and the composition of dung beetles assemblages in both arid zones; b) compare guild structure, as well as, species abundance and biomass; c) analyze similarities and differences between both assemblages in order to nd the patterns behind the formation and evolution of faunas in the arid zones of central Mexico; d) discuss the possible anthropic factors inuencing changes in the assemblages. 2. Material and methods 2.1. Description of the study areas Barranca de Metztitln (Fig. 1) forms part of the Altiplano province within the Mexican xerophytic region (Rzedowski, 1978). The Eastern boundary is delimited by the Sierra de Zacualtipn,

forming a rain shadow that prevents humidity from the Gulf of Mexico from reaching this arid area. Barranca de Metztitln opens to the NW extreme of the arid Mexican Altiplano, a fact that may explain the great inuence this area has on the scarab fauna of Barranca de Metztitln (Halffter et al., 2008). Our study area lies between 1300 and 1800 m.a.s.l., and has an average annual precipitation of 500 mm, and an average annual temperature between 18 and 22  C. The vegetation is characterized by two main shrub types: premontane shrubland (1600e1800 m.a.s.l.) and crassicaule scrubland (1300e1700 m.a.s.l.). Premontane shrubland is characterized by Agave diformis, Agave striata, Agave lechuguilla, and several species of the genera Mimosa, Senna, Krameria, Neopringlea, Ipomoea, Cnidoscolus, as well as isolated individuals of Yucca lifera. Cassicaule scrubland (1300e1700 m.a.s.l.) is dominated by Isolatocereus dumortieri, a tall columnar cactus endemic of Mexico, plus some other species such as Fouquieria splendens, Agave xylonacanchta, A. striata, Y. lifera, Prosopis sp., Bursera sp., Opuntia leucotricha, Myrtillocactus geometrizans, and Acacia sp. In the hotter, more humid areas it is also possible to nd tropical deciduous forest, which as the name suggests, loses its leaves during the dry season (from November to March), and is dominated by Bursera morelensis and Prosopis laevigata, plus several species of cacti and succulents such as Mirtillocactus geometrizans (garambullo), Cephalocereus senilis (viejito), I. dumortieri (rgano cimarrn), Opuntia sp., Marginatocereus marginatus, and Agave xylonacantha (CONANP, 2003). Within the Mexican xerophytic region, the Tehuacn-Cuicatln depression (Fig. 1) represents a characteristic biogeographical province (Tehuacn-Cuicatln oristic province, Rzedowski, 1978). The ora of this province has afnities to the Balsas depression, and together these two provinces constitute an arid region on the leeward slopes of the Sierra Madre Oriental and Sierra Madre de Oaxaca (Montaa and Valiente-Banuet, 1998). The Tehuacn-Cuicatln depression is bordered to the east by the Sierra Mixteca, to the west by the Sierra de Zongolica, to the north and center by the Sierra Mazateca, and to the south by the Sierra de Jurez (Fig. 1). Our area of study lies between 1500 and 1830 m.a.s.l, with an annual average precipitation of 380 mm and an average annual temperature of 21  C.

Please cite this article in press as: Halffter, G., et al., Historical and ecological determinants of dung beetle assemblages in two arid zones of central Mexico, Journal of Arid Environments (2011), doi:10.1016/j.jaridenv.2011.06.013

G. Halffter et al. / Journal of Arid Environments xxx (2011) 1e7

Although our study area was located near the village of ZapotitlnSalinas, which lies at a lower altitude than the actual Tehuacn Valley, throughout the text we refer to this site as Tehuacn, as this term is used in similar studies; we use the term Tehuacn-Cuicatln when referring to the area as a whole. The vegetation of this area is characterized by crassicaule scrubland, with a dominance of tall columnar cacti such as Neobuxbaumia tetetzo (tetecho) and Cephalocereus columna-trajani (cardonal). Other type of vegetation that forms a vegetation mosaic with the crassicaule scrubland is the thorn scrubland that alternates with tropical deciduous forest. The most abundant species within the latter vegetation type are Ceiba parvifolia, Bursera copallifera, Bursera fagaroides, Lysiloma divaricada, and Plumeria rubra (Montaa and Valiente-Banuet, 1998). 2.2. Sampling design and collection of coprophagous beetles In Barranca de Metztitln we selected six sampling sites, located 500 m apart, for each of vegetation types studied. Four pitfall traps baited with a mixture of sheep and horse dung, and ethylene glycol (30%), a preservative agent, were placed at each site. Traps were separated by 50 m and remained active for six days. In order to homogenize sampling time in this site with that in Tehuacn, we analyzed data from two collection periods in June 2006, with a total of 48 traps in crassicaule scrubland and 47 in submontane shrubland (one got lost in the latter). In Tehuacn (Zapotitln-Salinas) sampling took place in June 1999 and 2007 in both vegetation types mentioned. Five collecting sites were selected for each vegetation type, with a total of 54 traps in the crassicaule scrubland and 52 traps in the thorn scrub-low tropical deciduous forest. Traps were baited with cow, goat and human feces, as well as with a mixture of sheep and horse dung. All traps remained active for 48 h. The objective of this study was to compare the faunas of coprophagous Scarabaeinae and Aphodiinae, Onthophagus rufescens Bates and Diapterna dugesi (Bates) were not considered during sample collection in Tehuacn, as both species are abundant in debris piles near the nests of leaf cutting ants (genus Atta). Similarly, within the Aphodiinae only members of the tribe Aphodiini (coprophagous) were included, as those of the tribe Eupariini are saprophagous. In contrast to more humid Neotropical environments (even in places with tropical deciduous forest), attraction to carrion-baited traps was minimal, with only a few specimens of Canthon (Gl.) leechi collected in Tehuacn. All acquired individuals were identied by the authors, except for specimens of Aphodiinae collected in Tehuacn-Cuicatln during 1999, which were identied and published by Galante et al. (2003), and those collected in 2007 identied by Francisco Cabrero (Universidad Complutense de Madrid, Spain). Furthermore, Mario Zunino (Universita di Urbino, Italy) reviewed the collected specimens of Onthophagus. 2.3. Biomass estimation of the studied species Dried weight for each species was determined following Lobo (1993), using the best-t model for the allometric relation between body length and biomass (Biomass 0.010864* Longitude3.316). This model is most appropriate when working with assemblages that differ signicantly in size, a common phenomenon when studying more than one scarab family (in this case Scarabaeidae and Aphodiidae). 2.4. Data analysis In order to evaluate the complexity of the inventory for each sampled habitat, two richness estimators, ICE (i.e. estimate of

presences/absence) and ACE (i.e. estimate of abundance), were calculated. To do this we determined the percentage of observed vs. expected richness as calculated by the estimators. A two-way nested analysis of variance (ANOVA) was conducted to compare average species richness between both types of vegetation within each region using StatsDirect v. 2.7.7. (StatsDirect Ltd, 2005). Accumulation curves were used to compare the accumulated species richness among sites using the Mao Tau function and 95% condence intervals from the formulas by Colwell et al. (2004). EstimateS 8.0 (Colwell, 2005) was used to calculate both the accumulation curves and the richness estimates. In order to compare the population and biomass structure of each assemblage within each habitat, we rst calculated Simpsons Ds Index (one minus Simpsons original measure of dominance, lambda) with a condence interval of 95% using StatsDirect v 2.7.7 (StatsDirect Ltd, 2005). Simpsons Ds Index (also known as PIE) expresses the probability two randomly sampled individuals are from different species. We then constructed rank-abundance curves using both the number of individuals and relative biomass for each group. These curves were quantitatively compared to explore possible differences among them. A signicant dominance of beetles belonging to the roller guild in the arid and semiarid zones of Mexico has been observed in previous studies (Verd et al., 2007a). In order to further explore the dominance of various scarab beetle guilds, we analyzed the dominance of species by guild for each of the studied habitat types. 3. Results Richness estimators showed a high sampling efciency, with high inventory levels achieved. In Barranca de Metztitln crassicaule scrubland had between 79% and 91% values of completeness (ACE and ICE, respectively), whereas the submontane shrubland had 98% values of completeness for both estimates. In Tehuacn the crassicaule scrubland had between 82% and 84% values of completeness, whereas the values of completeness for the thorn scrubland-tropical deciduous forest were 85 and 91% respectively. Regarding comparisons of average species richness (alpha diversity) among vegetation types studied, we found signicant differences between sites (Metztitln vs. Tehuacn: F 1173.35 P < 0.0001). Diversity levels also varied across habitats within each site (F 67.32, P < 0.0001), although the only signicant difference occured between the habitats of Barranca de Metztitln (Fig. 2). These results demonstrated that the habitat with the highest average species richness was the submontane shrubland. The cumulative alpha value showed submontane shrubland also represented the richest habitat (Fig. 3), followed by the crassicaule scrubland of Metztitln and the thorn scrubland-tropical deciduous forest, which do not present a signicant difference between them. The least cumulative richness was attained by the crassicaule scrubland of Tehuacn. Among the studied assemblages, the population structure shows a general pattern in which one species is always the dominant (Tables 1e4). In the crassicaule scrubland and the submontane shrubland of Barranca de Metztitln Canthon humectus hidalgoensis was the dominant species (Tables 1 and 2), whereas Canthon indigaceus chevrolati was found in the highest abundance within the crassicaule scrubland of Tehuacn (Table 3). Areas of thorn scrubland-tropical deciduous forest were unique in having two codominant species: C. indigaceus chevrolati and Blackburneus saylori (Table 4). The population structure of assemblages in the arid environments of Mexico is characterized by a signicant dominance in terms of number of individuals of one species of roller of the genus Canthon, whereas in terms of biomass, the pattern was that one species of Canthon was dominant in all the studied

Please cite this article in press as: Halffter, G., et al., Historical and ecological determinants of dung beetle assemblages in two arid zones of central Mexico, Journal of Arid Environments (2011), doi:10.1016/j.jaridenv.2011.06.013

G. Halffter et al. / Journal of Arid Environments xxx (2011) 1e7 Table 1 Number of individuals collected, % of total abundance, % of total biomass and dry weight of species collected at crassicaule scrubland in Metztitln. N number of individuals; A (%) percentage of abundance; B (%) percentage of biomass; DW (%) mean dry weight of individuals. The taxonomy of the Aphodiini follows Cabrero-Saudo et al., 2010. Species N A (%) B (%) DW (mg)

Scarabaeinae 1 Canthon (Canthon) humectus 25,872 95.95 98.89 68.64 hidalgoensis (Bates, 1887) 2 Canthon (Canthon) imitator 52 0.19 0.18 60.84 (Brown, 1946) 3 Canthon (Glaphyrocanthon) viridis 1 <0.01 <0.01 1.84 (Palisot de Beauvois, 1805) 4 Euoniticellus intermedius (Reiche, 1849) 8 0.03 <0.01 10.73 5 Digitonthophagus gazella 28 0.10 0.03 21.03 (Fabricius, 1787) 6 Onthophagus igualensis (Bates, 1887) 46 0.17 0.01 1.97 7 Onthophagus landolti (Harold, 1880) 126 0.47 0.02 2.74 8 Onthophagus aff. landolti Harold, 1880 53 0.20 0.01 2.74 9 Onthophagus lecontei (Harold, 1871) 20 0.07 <0.01 1.27 10 Phanaeus adonis (Harold, 1863) 186 0.69 0.80 77.11 Aphodiinae 11 Pseudagolius caballeroi (Islas, 1945) 570 2.11 0.06 1.97 12 Blackburneus guatemalensis 2 0.01 <0.01 0.42 (Bates, 1887) 13 Haroldiellus lansbergei (Harold, 1874) 1 <0.01 <0.01 0.83

Fig. 2. Fully nested/hierarchical random ANOVA of mean alpha diversity between zones and among habitats within zones. Horizontal lines within boxes represent the mean and whiskers are SD. Different letters indicate signicant differences between samples (P < 0.05).

environments (Fig. 4). In both cases the thorn scrubland-tropical deciduous forest was the one with more equity. These results were statistically supported by results of the Simpons Ds Index. Using the number of individuals as the measure of abundance, values for Barranca de Metztitln were 0.079 (bootstrap, 95% CI 0.074 to 0.083) and 0.236 (bootstrap, 95% CI 0.230 to 0.242) for the crassicaule scrubland and the submontane shrubland respectively. In Tehuacn, the probability of two sampled individuals being of the same species were 0.294 (bootstrap, 95% CI 0.253 to 0.334) for crassicaule scrubland, and 0.795 (bootstrap, 95% CI 0.784 to 0.807) for thorn scrubland-tropical deciduous forest. In terms of percentages, the values of Ds indicated a dominance of 92% and 76% for crassicaule scrubland and submontane shrubland of Barranca de Metztitln respectively, and 71% and 20%

for crassicaule scrubland and thorn scrubland-tropical deciduous forest of the Tehuacn Valley. When species abundance was measured by biomass, the results were consistent with values recovered based on absolute numbers of scarab groups/guilds/species. In Barranca de Metztitln Ds values were 0.022 (bootstrap, 95% CI 0.022 to 0.022) and 0.136 (bootstrap, 95% CI 0.135 to 0.136) for the crassicaule scrubland and the submontane shrubland, respectively. In Tehuacn, the Ds value for crassicaule scrubland was 0.402 (bootstrap, 95% CI 0.396 to 0.408), and 0.640 (bootstrap, 95% CI 0.634 to 0.646) for thorn

Table 2 Number of individuals collected, % of total abundance, % of total biomass and dry weight of species collected at submontane scrubland in Metztitln. N number of individuals; A (%) percentage of abundance; B (%) percentage of biomass; DW (%) mean dry weight of individuals. Species N A (%) B (%) DW (mg)

Fig. 3. Cumulative alpha diversity for the different habitats of Metztitln (light grey) and Tehuacn (dark grey). Vertical lines are the standard deviation of each sample.

Scarabaeinae 1 Canthon (Canthon) humectus 27,498 87.21 92.84 68.64 hidalgoensis (Bates, 1887) 2 Canthon (Canthon) humectus 4 0.01 0.02 86.28 humectus (Say, 1832) 3 Canthon (Canthon) imitator 1,503 4.77 4.50 60.84 (Brown, 1946) 4 Canthon (Glaphyrocanthon) viridis 352 1.12 0.03 1.84 (Palisot de Beauvois Beauvois, 1805) 5 Dichotomius colonicus (Say, 1835) 42 0.13 1.08 521.10 6 Euoniticellus intermedius (Reiche, 1849) 2 0.01 <0.01 10.73 7 Digitonthophagus gazella (Fabricius, 1787) 1 <0.01 <0.01 21.03 8 Onthophagus igualensis (Bates, 1887) 151 0.48 0.01 1.97 9 Onthophagus mexicanus (Bates, 1887) 156 0.49 0.06 7.57 10 Onthophagus landolti (Harold, 1880) 906 2.87 0.12 2.74 11 Onthophagus aff. landolti (Harold, 1880) 228 0.72 0.03 2.74 12 Onthophagus lecontei (Harold, 1871) 92 0.29 0.01 1.27 13 Onthophagus semiopacus (Harold, 1869) 5 0.02 <0.01 1.59 14 Phanaeus adonis (Harold, 1863) 339 1.08 1.29 77.11 Aphodiinae 15 Pseudagolius caballeroi (Islas, 1945) 18 0.06 <0.01 1.97 16 Blackburneus guatemalensis (Bates, 1887) 55 0.17 <0.01 0.42 17 Agrilinus lansbergei (Harold, 1874) 136 0.43 0.01 0.83 18 Haroldiellus sallei (Harold, 1863) 20 0.06 <0.01 1.59 19 Labarrus pseudolividus (Balthasar, 1941) 2 0.01 <0.01 1.08 20 Cephalocyclus fuliginosus (Harold, 1863) 21 0.07 <0.01 2.26

Please cite this article in press as: Halffter, G., et al., Historical and ecological determinants of dung beetle assemblages in two arid zones of central Mexico, Journal of Arid Environments (2011), doi:10.1016/j.jaridenv.2011.06.013

G. Halffter et al. / Journal of Arid Environments xxx (2011) 1e7 Table 3 Number of individuals collected, % of total abundance, % of total biomass and dry weight of species collected at crassicaule scrubland in Tehuacn (Zapotitln-Salinas). N number of individuals; A (%) percentage abundance; B (%) percentage biomass; DW (%) mean dry weight of individuals. Species Scarabaeinae 1 Canthon (Canthon) humectus incisus (Robinson, 1948) 2 Canthon (Canthon) indigaceus chevrolati (Harold, 1868) 3 Dichotomius colonicus (Say, 1835) 4 Euoniticellus intermedius (Reiche, 1849) 5 Onthophagus alluvius (Howden & Cartwright, 1963) 6 Onthophagus igualensis (Bates, 1887) Aphodiinae 7 Labarrus pseudolividus (Balthasar, 1941) 8 Haroldiellus sallei (Harold, 1863) 9 Blackburneus saylori (Hinton, 1964) 10 Cephalocyclus villosipes (Harold, 1862) N 8 663 20 10 3 1 72 9 6 2 A (%) 1.00 83.19 2.51 1.25 0.38 0.13 9.03 1.13 0.75 0.25 B (%) 2.39 60.74 36.11 0.04 0.02 <0.01 0.40 0.07 0.22 0.01 DW (mg) 86.28 26.44 521.10 1.08 1.97 0.42 1.59 2.26 10.73 1.37

both assemblages have a separate taxonomic and biogeographical origins. The fauna of Metztitln originated in the Altiplano, particularly the central-west xerophytic areas. In contrast, the fauna of Tehuacn is a depauperated subset of widely distributed species found throughout both tropical depressions south of the Mexican Trans Volcanic Axis: the Balsas and the Tehuacn-Cuicatln. Further enhancing diversity in Tehuacn are species from the Altiplano which reached the lowlands through the southern margin of the Mexican Trans Volcanic Axis (notably Canthon (B.) puncticollis). Few species were observed in both studied arid zones in Mexico. Excluding three recently introduced species in America (Digitonthophagus gazella, Euoniticellus intermedius, and Labarus lividus, the latter possibly confused with Labarus pseudolividus from the Mexican Transition Zone), there are 26 species of Scarabaeinae and Aphodiini. A total of 17 species were recovered (17 in Metztitln and, 15 in Tehuacn) of which, with only ve species (18.5%) are shared across sites. 4.2. Adaptation to arid areas

scrubland-tropical deciduous forest. In terms of percentage, these values indicated a dominance of 98% and 86% for crassicaule scrubland and submontane shrubland respectively, whereas in Tehuacn these values were 60% and 36% for crassicaule scrubland and thorn scrubland-tropical deciduous forest, respectively. 4. Discussion 4.1. Species composition and fauna origins The coprophagous Scarab beetle fauna of both Metztitln and Tehuacn have several interesting features: 1) In contrast to patterns seen with owering plants (see Rzedowski, 1978, 1991 and 1993), there are no endemics or species with restricted distribution within Scarabaeinae and Aphodiini, 2) Although species assemblages in Metztitln and Tehuacn share several ecological parameters such as dominance (in terms of number of individuals and biomass) of a single species of medium-sized roller, however,

Table 4 Number of individuals collected, % of total abundance, % of total biomass and dry weight of species collected at xerophyllous scrubland-deciduous forest in Tehuacn (Zapotitln-Salinas). N number of individuals; A (%) percentage abundance; B (%) percentage biomass; DW (%) mean dry weight of individuals. Species Scarabaeinae 1 Canthon (Boreocanthon) puncticollis (LeConte, 1866) 2 Canthon (Canthon) humectus incisus (Robinson, 1948) 3 Canthon (Canthon) indigaceus chevrolati (Harold, 1868) 4 Canthon (Glaphyrocanthon) leechi (Martnez, Halffter & Halffter, 1964) 5 Dichotomius annae (Kolmann & Solis, 1997) 6 Dichotomius colonicus (Say, 1835) 7 Euoniticellus intermedius (Reiche, 1849) 8 Onthophagus alluvius (Howden & Cartwright, 1963) 9 Onthophagus durangoensis (Balthasar, 1939) 10 Onthophagus hoepfneri (Harold 1869) 11 Onthophagus lecontei (Harold, 1871) Aphodiinae 12 Blackburneus guatemalensis (Bates, 1887) 13 Haroldiellus sallei (Harold, 1863) 14 Blackburneus saylori (Hinton, 1964) 15 Cephalocyclus villosipes (Harold, 1862) N 30 11 A (%) 3.73 1.37 B (%) 0.27 0.05 DW (mg) 5.39 78.88 26.44 1.59

222 27.58 23.35 146 18.14 45.82 1 7 2 18 5 2 23

0.12 <0.01 514.43 0.87 14.51 521.10 0.25 0.09 10.73 2.24 0.16 2.26 0.62 10.23 0.25 0.01 2.86 0.12 2.26 1.37 1.27 1.37 0.42 0.42 1.37

2 0.25 0.01 1 0.12 <0.01 215 26.71 4.61 120 14.91 0.76

There are several Scarabaeinae species, and even superspecic level taxa, that have adapted to life in arid regions around the world, (e.g. deserts) even though specic adaptations differ among species. The tribe Scarabaeini possesses the highest number of species adapted to arid areas, with distributions ranging across Northern Africa, the deserts of South Africa, and the Near East. Another example can be found with the tribe Eucraniini, a group exclusively adapted to the arid environments of Northeastern Argentina. Interestingly in Australia, where arid environments occupy a huge proportion of the territory, only two exclusively desert-dwelling Onthophagus species have been found (Matthews, 1972), including one roller species belonging to a genus of Canthonini. Five additional species of roller, within three genera, are distributed across Australia, each having adapted, at least marginally, to arid conditions (Matthews, 1974). In North America, including the Mexican Transition zone yet excluding the south of the Baja California peninsula discussed below, there are no species endemic to of the arid zones. Only a few species of coprophagous beetles (e.g. Canthon (Boreocanthon) puncticollis) range across the expansive arid zones. These species are widely distributed and co-occur in the dry habitats of the northern Mexican Altiplano and southwest of the United States, which were formerly occupied by prairies or xerophytic shrub. Canthon (Boreocanthon) puncticollis is distributed in the desert zones of Baja California, Arizona, and Chihuahua, arid zones of the Mexican Altiplano (even near Barranca de Metztitln), with a southern limit in Tehuacn. An exception to the general lack of endemic species in the arid zones of Mexico can be found in the southern extreme of the Baja California peninsula. Canthon obliquus Horn is the only known endemic species from an arid environment, which also exhibits apterism, and is exclusively distributed in Sierra de la Laguna. There is a marked contrast between the degree of endemism in owering plants and coprophagous beetles in the arid zones of Mexico. An explanation to this disparity could be given from a biogeographical-historical point of view. There seems to have been ample dispersion opportunities for the owering plants adapted to arid zones (even though these might have not been continuous) at the end of the Mesozoic and beginning of the Cenozoic, as evidenced by the genera and groups of plants that are common to the arid zones of North and South America (a detailed discussion on the biogeographical afnities for the owering plants of North America can be found in Rzedowski, 1972). However, these opportunities may have been non-existent for beetles adapted to

Please cite this article in press as: Halffter, G., et al., Historical and ecological determinants of dung beetle assemblages in two arid zones of central Mexico, Journal of Arid Environments (2011), doi:10.1016/j.jaridenv.2011.06.013

G. Halffter et al. / Journal of Arid Environments xxx (2011) 1e7

Fig. 4. Rank-abundance ratio and rank-biomass ratio (inset graphs) of the assemblages of scarabeids in the Metztitln Canyon (A: crassicaule scrubland; B: submontane scrubland) and at Tehuacn (Zapotitln-Salinas) (C: crassicaule scrubland; D: xerophyllous scrubland-deciduous forest). The bars on the abundance graphs indicate the logarithmic values of biomass for each species and each color represents the functional group or guild to which the species belongs (black: telecoprids or rollers; dark grey: paracoprids or tunnelers; light grey: endocoprids or dwellers). The numbers above each point on the abundance curves correspond to the reference numbers on the far left given for each species in Table 2.

similar ecological conditions. Beetles with Neotropical afnities reaching what today is the Mexican Transition Zone during the early and middle Cenozoic did not include xerophytic South American groups, such as the aforementioned Eucraniini or any other beetles from the arid zones of Argentina. As a result of these historical events, beetle colonization of the arid zones of Mexico has been marginal, with no real specialists other than the one form the Baja California peninsula. The same biogeographical afnities of the Scarabaeinae (Aphodiini are mainly of Septentrional afnities) are true for the owering plants. In general, owering plants of the arid zones of Mexico have more afnities to plants in the south (Neotropical zone) than with plants in the north (Nearctic zone) (Rzedowski, 1973). Moreover, the ora of Metztitln is highly similar to that of the Chihuahuan Desert. On the other hand the ora of Tehuacn-Cuicatln has clear afnities with that of the Balsas, which has mostly Neotropical origins and a high level of endemism (30% of the owering plants) (Dvila et al., 1995; Montaa and ValienteBanuet, 1998; Villaseor et al., 1990). El Monte biogeographic province, Argentina, located between Salta (24 350 S) and Chubut (43 260 S) (for a further description of this biogeographic province and its insects see Morello, 1958; Ocampo and Hawks, 2006; Roig-Juent et al., 2001; Stange et al., 1976) shares several genera with the deserts of North America. However, there are no characteristic genera of the Northern or Southern arid zones, as some of the shared genera are also found in the intermediate zones, and the species of these areas seem to have an independent phyletic origin. 4.3. Community structure We measured both the number of individuals and biomass of scarab species, as this variable explains in a more realistic and direct manner the amount of the trophic resource exploited, as well as the structural changes in the community (Saint-Germain et al.,

2007). For both variables there was a general pattern in which one species of Canthon was signicantly dominant in the assemblages of the four studied habitats. The only habitat in which this dominance was less intense was that of the thorn scrublandtropical deciduous forest, which had the most tropical afnity of each habitat studied. This phenomenon in which one species is signicantly more favored than the rest has not been previously described in any other environment in Mexico. Rather, community structures in other areas are generally much more equilibrated in terms of relative abundance (e.g. Halffter and Arellano, 2002). Data published for Mxico (e.g. Lobo, 1996; Verd et al., 2007b) as well as for other geographical areas (e.g. Erroussi et al., 2009; Lobo et al., 2006; Verd et al., 2000; Verd and Galante, 2004) show that there is a close relationship between aridity and dominance of heliophylic species. In the present study, species of the genus Canthon conforms to this general pattern, as they are linked to open environments such as natural pastures, xerophytic shrublands, or wide, open areas for cattle and are found in high abundance. In all cases, the dominant species are characterized for being heliophylic, with a trophic preference for cattle feces, able to reach high ying speeds (ca. 30 km/h), and to avoid overheating during ight in extreme environmental temperatures (Verd and Lobo, 2008, and unpublished data). Thermoregulation explains how the subspecies C. humectus hidalgoensis and C. humectus incisusdbeing endothermicdcan endure extreme environmental conditions by modifying the rate of body heat loss as the environmental temperature increases (Verd and Lobo, 2008). Thermoregulation can also explain the observed results in the case of C. indigaceus chevrolati of the Valley of Tehuacn, as this species presents a thermal niche characteristic of extreme warm zones (Verd et al., 2007b). On the other hand, the observed dominance in these assemblages could be the result of an anthropic effect (overgrazing), which favors certain species over others. Overgrazing has been shown to affect a few species of several invertebrate groups in

Please cite this article in press as: Halffter, G., et al., Historical and ecological determinants of dung beetle assemblages in two arid zones of central Mexico, Journal of Arid Environments (2011), doi:10.1016/j.jaridenv.2011.06.013

G. Halffter et al. / Journal of Arid Environments xxx (2011) 1e7

different geographic areas (e.g. Erroussi et al., 2009; Lobo et al., 2006; Seymour and Dean, 1999; Verd et al., 2007a). In both studied zones, the most abundant trophic resource are cattle, goat and horse feces, the result of intensive livestock farming (Verd et al., 2007a). Thus, the population structure and energy use of dung beetles (measured by their biomass) must be related to these resources. In general it has been found that within Scarabaeinae there is a negative relationship between body size and abundance, as the occurrence of large-sized scarabs is related to an over abundance of the trophic resource (Cambefort, 1994). Within our study area, we observed a dominance of medium-sized species, both in terms of abundance and biomass (Figs. 4e7), which is consistent with the idea medium-sized species may be favored in extreme environments (Jay-Robert et al., 2008). In conclusion, the structure and composition of current scarab beetle communities in two arid zones in Central Mexico are very different between them, even when in close geographic proximity. These singular features can only be understood in light of several important parameters, including the biological history of the region, the biogeography of the various beetle groups, in addition to current ecological determinants (e.g. farming activities and the physiological response of the species to aridity). It is also worth mentioning the contrasting patterns of endemicity found for coprophagous beetles in this study and those previously described for plants in the same areas. Thus further work to gather additional data for other groups are needed to better understand which factors inuence the composition of biodiversity in these biologeographical regions.

Acknowledgments We are thankful to Prof. Mario Zunino (Universidad de Urbino) for the identication of the collected Onthophagus, as well as to Dr. Francisco Cabrero (Universidad Complutense de Madrid) for identication of Aphodiini from the second collecting season in Tehuacn. Dr. Victoria Sosa graciously helped us with the bibliographic search on the ora of Tehuacn. Esther Quintero translated the original manuscript from Spanish to English, while Kari L. Schmidt reviewed the English version. CONABIO (FB1381/HE003/ 09), FOMIX CONACYT-Hidalgo (95828), and AECID (A/020305/08) provided support for this study.

References
Cabrero-Saudo, F.J., Dellacasa, M., Martnez, I.M., Lobo, J.M., Dellacasa, G., 2010. Distribucin de las especies de Aphodiinae (Coleoptera, Scarabaeoidea, Aphodiidae) en Mxico. Acta Zoolgica Mexicana (n.s.) 26 (2), 323e399. Cambefort, Y., 1994. Body size, abundance and geographical distribution of Afrotropical dung beetles (Coleoptera, Scarabaeidae). Acta Oecologica 15, 165e179. Colwell, R.K., Mao, Ch.X., Chang, J., 2004. Interpolating, extrapolating, and comparing incidence-based species accumulation curves. Ecology 85, 2717e2727. Colwell, R.K., 2005. Estimate S: Statistical Estimation of Species Richness and Shared Species from Samples. Version 7.5. Department of Ecology and Evolutionary Biology, University of Connecticut, Storrs. http://viceroy.eeb.uconn.edu/ estimates. CONANP, 2003. Programa de manejo Reserva de la Biosfera Barranca de Metztitln. Comisin Nacional de reas Naturales Protegidas, Mxico, DF. Cook, S.F., 1949. Soil Erosion and Population in Central Mexico. Ibero-Americana, vol. 34. University of California Press, pp. 1e86. Dvila, P., Medina, R., Ramrez, A., Salinas, A., Tenorio, P., 1995. Anlisis de la ora del Valle de Tehuacn-Cuicatln: endemismo y diversidad. In: Linares, E., Dvila, P., Chiang, F., Bye, R., Elias, T. (Eds.), Conservacin de plantas en peligro de extincin: diferentes enfoques. Instituto de Biologa, Universidad Nacional Autnoma de Mxico, Mxico, D.F., pp. 33e41. Erroussi, F., Labidi, I., Nouira, S., 2009. Seasonal occurrence and local coexistence within scarabaeid dung beetle guilds (Coleoptera: Scarabaeoidea) in Tunisian pasture. European Journal of Entomology 106, 85e94.

Galante, E., Stebnicka, Z., Verd, J.R., 2003. Aphodiinae and Rhyparinae (Coleoptera: Scarabaeidae) of the southern states of Mexico (Chiapas, Oaxaca, Puebla and Veracruz). Acta Zoologica Cracoviensia 46, 283e312. Halffter, G., 1964. La entomofauna Americana, ideas acerca de su origen y distribucin. Folia Entomologica Mexicana 6, 1e108. Halffter, G., 1976. Distribucin de los insectos en la Zona de Transicin Mexicana. Relaciones con la entomofauna de Norteamrica. Folia Entomologica Mexicana 35, 1e64. Halffter, G., 1987. Biogeography of the montane entomofauna of Mexico and central America. Annual Review of Entomology 32, 95e114. Halffter, G., Arellano, L., 2002. Response of dung beetle diversity to human-induced changes in a tropical landscape. Biotropica 34, 144e154. Halffter, G., Verd, J.R., Mrquez, J., Mo reno, C.E., 2008. Biogeographical analysis of Scarabaeinae and Geotrupinae along a transect in central Mxico (Coleoptera, Scarabaeoidea). Fragmenta Entomologica 40, 273e322. Jay-Robert, P., Lumaret, J.P., Lebreton, J.D., 2008. Spatial and temporal variation of mountain dung beetles assemblages and their relationships with environmental factors (Aphodiinae: Geotrupinae: Scarabaeinae). Annals of the Entomological Society of America 101 (1), 58e69. Lobo, J.M., 1993. Estimation of dung beetle biomass (Coleoptera: Scarabaeoidea). European Journal of Entomology 90, 235e238. Lobo, J.M., 1996. Diversity, biogeographical considerations and spatial structure of a recently invaded dung beetle (Coleoptera: Scarabaeoidea) community in the Chihuahuan Desert. Global Ecology and Biogeography 5, 342e352. Lobo, J.M., Hortal, J., Cabrero-Saudo, F.J., 2006. Regional and local inuence of grazing activity on the diversity of a semi-arid dung beetle community. Diversity and Distributions 12, 111e123. Matthews, E.G., 1972. A revision of the Scarabaeine dung beetles of Australia, I. Tribe Onthophagini. Australian Journal of Zoology 9, 1e330. Matthews, E.G., 1974. A revision of the Scarabaeine dung beetles of Australia, II. Tribe Scarabaeini. Australian Journal of Zoology 24, 1e211. McAuliffe, J.R., Sundt, P.C., Valiente-Banuet, A., Casas, A., Viveros, J.L., 2001. PreColumbian soil erosion, persistent ecological changes, and collapse of a subsistence agricultural economy in the semi-arid Tehuacn Valley, Mexicos Cradle of Maize. Journal of Arid Environments 47, 47e75. Montaa, C., Valiente-Banuet, A., 1998. Floristic and life-form diversity along an altitudinal gradient in an intertropical semiarid Mexican region. Southwestern Naturalist 43 (1), 25e39. Morello, J., 1958. La provincial biogeogrca del Monte. Opera Lilloana 2, 1e155. Ocampo, F.C., Hawks, D.C., 2006. Molecular phylogenetics and evolution of the food relocation behaviour of the dung beetle tribe Eucraniini (Coleoptera: Scarabaeidae: Scarabaeinae). Invertebrate Systematics 20, 557e570. Roig-Juent, S., Flores, G., Claver, S., Debandi, G., Marvaldi, E., 2001. Monte Desert (Argentina): insect biodiversity and natural areas. Journal of Arid Environments 47, 77e94. Rzedowski, J., 1972. Relaciones geogrcas y posibles orgenes de la ora de Mxico. Boletn de la Sociedad Botnica de Mxico 29, 131e177. Rzedowski, J., 1973. Geographical relationships of the ora of Mexican dry regions. In: Graham, A. (Ed.), Vegetation and Vegetational History in Northern Latin America. Elsevier Scientic Company, Amsterdam, pp. 61e72. Rzedowski, J., 1978. Vegetacin de Mxico. Limusa, Mxico, D.F., p. 432. Rzedowsky, J., 1991. Diversidad y orgenes de la ora fanerogmica de Mxico. Acta Botanica Mexicana 14, 3e21. Rzedowsky, J., 1993. Diversity and origins of the phanerogamic ora of Mexico. In: Ramamoorthy, T.P., Bye, R., Lor, A., Fa, J. (Eds.), Biological Diversity of Mexico: Origins and Distribution. Oxford University Press, New York, pp. 129e144. Saint-Germain, M., Buddle, C.M., Larrive, M., Mercado, A., Motchula, T., Reichert, E., Sackett, T.E., Sylvain, Z., Webb, A., 2007. Should biomass be considered more frequently as a currency in terrestrial arthropod community analyses? Journal of Applied Ecology 44, 330e339. Seymour, C.L., Dean, W.R.J., 1999. Effects of heavy grazing on invertebrate assemblages in the Succulent Karoo, South Africa. Journal of Arid Environments 43, 267e286. Stange, L.A., Tern, A.L., Willink, A., 1976. Entomofauna de la provincia biogeogrca del Monte. Acta zoolgica lilloana 32, 73e120. StatsDirect Ltd, 2005. StatsDirect statistical software. StatsDirect Ltd, England. www.statsdirect.com. Verd, J.R., Crespo, M.B., Galante, E., 2000. Conservation strategy of a nature reserve in Mediterranean ecosystems: the effects of protection from grazing on biodiversity. Biodiversity and Conservation 9, 1707e1721. Verd, J.R., Galante, E., 2004. Behavioural and morphological adaptations for a lowquality resource in semi-arid environments: dung beetles Coleoptera, Scarabaeoidea associated with the European rabbit Oryctolagus cuniculus L. Journal of Natural History 38, 705e715. Verd, J.R., Moreno, C.E., Snchez-Rojas, G., Numa, C., Galante, E., Halffter, G., 2007a. Grazing promotes dung beetle diversity in the xeric landscape of a Mexican biosphere reserve. Biological Conservation 140, 308e317. Verd, J.R., Arellano, L., Numa, C., Mic, E., 2007b. Roles of endothermy in niche differentiation for ball-rolling dung beetles (Coleoptera: Scarabaeidae) along an altitudinal gradient. Ecological Entomology 32, 544e551. Verd, J.R., Lobo, J.M., 2008. Ecophysiology of thermoregulation in endothermic dung beetles: ecological and geographical implications. In: Fattorini, S. (Ed.), Insect Ecology and Conservation. Research Signpost, pp. 1e28. Villaseor, J.L., Dvila, P., Chiang, F., 1990. Biogeografa del Valle de Tehuacn-Cuicatln. Boletn de la Sociedad Botnica de Mxico 50, 135e149.

Please cite this article in press as: Halffter, G., et al., Historical and ecological determinants of dung beetle assemblages in two arid zones of central Mexico, Journal of Arid Environments (2011), doi:10.1016/j.jaridenv.2011.06.013

You might also like