THE CONSERVATION STATUS OF SNAKES IN CENTRAL BRAZIL

South American Journal of Herpetology, 1(1), 2006, 25-36 2006 Brazilian Society of Herpetology THE CONSERVATION STATUS OF SNAKES IN CENTRAL BRAZIL FREDERICO G. R. FRANÇA 1,3 AND ALEXANDRE F. B. ARAÚJO 2 1 Programa de pós-graduação em Ecologia, Departamento de Ecologia, Universidade de Brasília, 710-900, Brasília, Brasil. E-mail: fgrf@unb.br 2 Departamento de Biologia Animal, Universidade Federal Rural do Rio de Janeiro, 23890-000, Seropédica, Rio de Janeiro, Brasil. E-mail: araujo@ufrrj.br 3 Corresponding author. ABSTRACT: We assessed the conservation status of 61 species from a central Brazilian snake assemblage by ranking species according to their vulnerability to ten threat factors that are thought to influence the survival of snake populations. The anomalepidid snake Liotyphlops ternetzii whereas the species with the lowest risk and the colubrid Drymoluber brazili was the most threatened. Less than 15% of the Distrito Federal snake fauna was considered free from any threat. The natural-history of the species, its geographic distribution, and human attitudes were important factors in determining the levels of threat for central Brazilian snakes. Principal Component and Cluster analyses were used to classify species into vulnerability groups that were based on sharing similar threat factors. Thirteen groups were identified which classified species in categories ranging from no risk to threatened. Within Brazil, status assessments of additional snake assemblages in other biomes would be useful in generating a comprehensive conservation plan for the country. KEY WORDS: Snakes, Central Brazil, Distrito Federal, Conservation Status, Cerrado. INTRODUCTION Snakes have recently been propelled into a new position as model organisms in ecological research, as evidenced by the increasing number of publications on snake ecology (Shine and Bonnet 2000). However, research related to snake conservation is still lacking and the few existing studies have dealt primarily with species that are rare or habitat specialists (Seigel, 1986; Webb and Shine, 1997; Daltry et al., 2001). Brazil has one of the highest diversities of snake species in the world, but snake conservation efforts have been restricted to insular species (Duarte et al., 1995; Marques et al., 2002) and to controlling exotic species (Eterovic and Duarte, 2002). Snake conservation efforts are primarily hindered by the lack of basic natural history information for most species, which is essential if one is to assess species-specific threats (Dodd, 2001a). Methods for developing management strategies for snake conservation are scarce. Dodd (2001a) proposed that the conservation status of snakes should be based on factors that threaten population viability, and a combined score of these factors should be used to identify candidate taxa and set priorities for further research and management. However, few studies have been able to test the vulnerability of snake species to threat factors (Fillipi and Luiselli, 2000; Reed and Shine, 2002). Snake population declines can be related to lifehistory characteristics such as body size, reproductive potential, aggregation tendency and foraging; or can be related to distribution factors such as habitat occupation or use of altered habitats. Thus, not all species are influenced equally by certain factors (Dodd, 2001a; Fillipi and Luiselli, 2000). Herein, we use the method proposed by Fillipi and Luiselli (2000) along with data on natural history, distribution, and other factors to evaluate the conservation status of a snake assemblage in Distrito Federal, Brazil. The Distrito Federal represents an important core region of Cerrado in central Brazil (Werthein et al., 2000). The Cerrado covers approximately onequarter of Brazil s land surface and because of intense habitat alteration and the accelerated pace of destruction, it has recently been listed as one of the most threatened regions of the world (Oliveira and Marquis, 2002; Myers et al., 2000). Study Area MATERIALS AND METHODS The Distrito Federal of Brazil covers approximately 5,814 km 2 of Cerrado. The region is a mosaic of vegetation types with habitats ranging from interfluvial open areas (campos and cerrados) to forest formations (gallery forests). Other typical Cerrado formations found in the study area include humid grasslands and veredas (Eiten, 1994). The region consists of three

26 Conservation of snakes in Central Brazil major hydrogeographic basins: the Paraná River, the San Francisco River, and the Tocantins River basins (Pinto, 1994). The climate is type Aw in the Köppen classification, receiving annually 1500-2000 mm of a highly predictable and strongly seasonal precipitation, almost entirely restricted to the period from October to April (Nimer, 1989). Threat Factors Ten factors known to influence survival of snake populations in Distrito Federal were chosen. Some categories are the same as those used by Fillipi and Luiselli (2000). For each species, the factors were classified into four categories ranging from 0 (least risk) to 3 (highest risk). Factors can be divided into two groups: 1) factors directly related to natural-history of the species (factors 1 to 5) and 2) factors not directly related to natural-history of the species, but related to geographic distribution and human attitudes (factors 6 to 10). Category scores for threat factors for each snake species were based on published average conditions for snakes in the wild, or on unpublished data and personal experience of the authors (For review see França, 2003; França and Araújo, in press). Threat factors and categories were as follows: 1. Body size (BS): Smaller species are less vulnerable to the direct threats of removal or death from humans; 0 = < 50 cm length; 1 = 51-100 cm length; 2 = 101-150 cm length; 3 = > 150 cm length. 2. Litter size (LS): Eggs or younglings; 0 = > 15; 1 = 15-11; 2 = 10-5; 3 = < 5. 3. Dietary breadth (DT): Based on taxonomic order level of prey taken and percentage of main prey in the diet; 0 = generalist, main prey item < 30% of diet; 1 = low specialization, main prey 31-50% of diet; 2 = moderate specialization, main prey 51-70% of diet; 3 = highly specialized, main prey > 70% of diet. 4. Habitat breadth (HB): Based on occurrence in four Cerrado habitat types: campo, cerrado, mata, and water (river, pond and/or vereda); 0 = species found in all four habitats; 1 = species found in three habitats; 2 = species found in two habitats; 3 = species found in only one habitat. 5. Habits (HT): Based on the type of general habits exhibited by the snakes in the wild; 0 = fossorial; 1 = cryptozoic or aquatic; 2 = arboreal or semi-arboreal; 3 = terrestrial. 6. Distribution breadth in Distrito Federal (DB): 0 = present in > 80% of the region; 1 = present in 51-80%; 2 = present in 21-50%; 3 = present in < 20% of the region. 7. Adaptability to altered habitats (AH): 0 = extremely adaptable species (found even in urban centers); 1 = adaptable species (found also in suburbia if small native patches are available); 2 = scarcely adaptable species (found at best in moderately sized natural habitats); 3 = inadaptable species (found only in large areas of natural habitat). 8. Rarity in Distrito Federal (RR): Based on the percentage of specimens found in DF; 0 = > 10% of total specimens; 1 = 6-10%; 2 = 2-6%; 3 = < 2%. 9. Species color (CL): 0 = only one solid color, other than red; 1 = with stripes or dots, but not like pitvipers or Micrurus species; 2 = blotched pattern like pitvipers; 3 = coral snake triads or red-colored snakes. Species with ontogenetic color variation in life (Boiruna or Clelia species) or more than one color pattern (Waglerophis) were ranked with lesser score. 10. Road mortality (RM): 0 = never found dead on roads in Distrito Federal; 1 = < 10% of total species found dead on roads; 2 = 10-20%; 3 = > 20%. Statistical Analysis Mean scores for the 10 factors were used to determine threat levels for every snake species in the Distrito Federal as proposed by Filippi and Luiselli (2000). Mean scores < 1.2 indicate species with low risk, mean scores between 1.3 and 1.9 indicate a species vulnerable to decline, and mean scores > 1.9 indicate a seriously threatened species. We used a t test to determine if factors related to the natural-history of species and factors related to geographic distribution and human attitudes differed in importance. Differences in vulnerability, based on mean scores, among phylogenetic groups of snake species were analyzed using a Kruskal Wallis test (Zar,

França, F. G. R. and Araújo, A. F. B. 27 1999). Principal Component Analysis and Cluster Analysis were used to classify snake species in terms of their relative similarity to particular threats. The PCA was run using the factor analysis model with Varimax orthogonal rotation (Tabachnik and Fidell, 2001). The Cluster Analysis was run using the UPG- MA cluster method and Pearson correlation coefficient (Valentin, 2000). All analyses were performed in SPSS 10.0, with a set at 5%. RESULTS Analysis of scores for the threatening factors Category scores and mean scores for the 10 threat factors for the 61 snake species in Distrito Federal are presented in Table 1. The summary of those data are presented in Table 2. Based on the classification of mean scores, nine snake species (14.8%) are at low risk of decline, 34 species (55.7%) are vulnerable to decline, and 18 species (29.5%) are threatened (Table 1). Mean scores for species ranged from 0.5 for Liotyphlops ternetzii (less vulnerable) to 2.3 for Drymoluber brazili (more threatened). For the phylogenetic groups, the mean score was highest in Colubrinae (x = 1.83 ± 0.31) and lowest in Scolecophidia (Anomalepididae + Leptotyphlopidae) (x = 1.00 ± 0.71). However, there were no significant differences among means of phylogenetic groups (Kruskal-Wallis ANO- VA: H = 9.41; d.f. = 6; p = 0.15). Mean scores for the 10 threat factors were highest for the dietary breadth (2.26) and snake habits (2.03), and lowest for road mortality (0.90) and species color (1.20). Nevertheless, there was no significant difference between mean scores of factors related to natural-history of the species and factors related to geographic distribution and humans attitudes (t test = 1.6; d.f. = 8; p = 0.14). Principal Component and Cluster Analyses The model proposed by Filippi and Luiselli (2000) applying PCA to a data matrix allows classification of species into threat groups. All these groups were not easily recognized by PCA biplot graph (Fig. 1). Additionally, we used a Cluster analyses to detect the following groups of snakes in Distrito Federal (Fig. 2): 1: Snakes with small body size, black coloration, with wide distribution in DF and adapted to altered habitats (low vulnerability). Liotyphlops ternetzii. 2: False-coral snakes with small body size, wide distribution in DF, fossorial habits and diet specialists (low vulnerability). Apostolepis assimilis, Tantilla melanocephala. 3: False-coral snakes, terrestrial and nocturnal, with medium to high litter size, wide distribution in DF and high mortality on the roads of DF (vulnerable). Oxyrhopus guibei, Oxyrhopus rhombifer, Oxyrhopus trigeminus. Factor 2 2 1 0-1 - 2-3 44 58 1 42 39 48 55 44 43 9 49 23 47 3 22 40 19 7 4 20 24-2 - 1 0 1 2 Factor 1 25 16 60 175431 36 45 53 2 27 8 29 34 33 61 Figure 1: Plot of factor scores from principal components for snake species of Distrito Federal, showing some clearly threatned groups. The species are: 1) Liotyphlops ternetzii; 2) Leptotyphlops koppesi; 3) Boa constrictor; 4) Epicrates cenchria; 5) Eunectes murinus; 6) Bothrops moojeni; 7) B. neuwiedi; 8) B. itapetiningae; 9) Crotalus durissus; 10) Chironius exoletus; 11) C. flavolineatus; ) C. quadricarinatus; 14) Drymarchon corais; ) Drymoluber brazili; 15) Mastigodryas bifossatus; 16) Oxybelis aeneus; 17) Simophis rhinostoma; 18) Spilotes pullatus; 19) Tantilla melanocephala; 20) Atractus pantostictus; 21) Leptodeira annulata; 22) Sibynomorphus mikanii; 23) Apostolepis assimilis; 24) A. albicolaris; 25) A. flavotorquata; 26) A. gr. taenuis; 27) Boiruna maculata; 28) Clelia plumbea; 29) C. quimi; 30) Erythrolamprus aesculapii; 31) Gomesophis brasiliensis; 32) Helicops angulatus; 33) H. leopardinus; 34) H. modestus; 35) Liophis almadensis; 36) L. maryellenae; 37) L. meridionalis; 38) L. paucidens; 39) L. poecilogyrus; 40) L. reginae; 41) Lystrophis nattereri; 42) Oxyrhopus guibei; 43) O. trigeminus; 44) O. rhombifer; 45) Phalotris nasutus; 46) Philodryas aestiva; 47) P. olfersii; 48) P. nattereri; 49) P. patagoniensis; 50) P. psammophideus; 51) Phimophis guerini; 52) Pseudablabes agassizii; 53) Pseudoboa nigra; 54) Rhachidelus brazili; 55) Echinanthera occipitalis; 56) Thamnodynastes hypoconia; 57) Thamnodynastes rutilus; 58) Waglerophis merremii; 59) Xenopholis undulatus; 60) Micrurus frontalis; 61) M. lemniscatus. 6 52 18 35 11 46 15 26 30 56 41 21 31 59 5750 37 32 10 14 5 28

28 Conservation of snakes in Central Brazil Table 1: Scores and threat categories for 10 factors affecting survival of the 61 Central Brazil snakes. Abbreviations are BS: Body size, LS: Litter size, DT: Dietary breadth, HB: Habitat breadth, HT: Habits, DB: Distribution breadth in DF, AH: Adaptability to altered habitats, RR: Rarity in DF, CL: Species color, RM: Road Mortality. Species and Threat categories BS LS DT HB HT DB AH RR CL RM Mean Threatened (> 1.9) Drymoluber brazili 2 3 3 3 3 3 3 3 0 0 2.3 Philodryas psamophidea 1 2 3 3 3 3 3 3 1 0 2.2 Apostolepis flavotorquata 1 3 3 3 0 3 2 3 3 0 2.1 Chironius exoletus 2 2 3 3 2 3 3 3 0 0 2.1 Helicops angulatus 1 2 3 3 1 3 3 3 2 0 2.1 Lystrophis nattereri 0 3 3 2 3 2 3 3 2 0 2.1 Oxybelis aeneus 2 3 2 3 2 3 2 3 0 1 2.1 Rhachidelus brazili 3 2 3 2 3 3 2 3 0 0 2.1 Simophis rhinostoma 1 2 3 2 1 2 3 3 3 1 2.1 Thamnodynastes rutilus 1 2 3 3 3 3 3 3 0 0 2.1 Apostolepis gr. taenuis 0 3 3 3 0 3 2 3 3 0 2.0 Gomesophis brasiliensis 0 3 3 3 1 3 3 3 1 0 2.0 Liophis almadensis 1 2 3 3 3 2 2 2 1 1 2.0 Liophis paucidens 1 1 3 3 3 3 2 3 1 0 2.0 Oxyrhopus trigeminus 2 2 2 2 3 1 1 1 3 3 2.0 Phimophis guerini 1 2 3 3 1 3 3 3 0 1 2.0 Pseudoboa nigra 2 2 2 2 3 2 3 3 0 1 2.0 Xenopholis undulatus 0 3 3 3 1 3 3 3 1 0 2.0 Vulnerable (1,3-1,9) Erythrolamprus aesculapii 1 2 3 2 3 1 2 1 3 1 1.9 Leptodeira annulata 1 3 3 2 2 2 2 2 2 0 1.9 Phalotris nasutus 1 2 3 2 0 2 3 2 3 1 1.9 Chironius flavolineatus 2 2 3 2 2 1 2 2 1 1 1.8 Clelia plumbea 3 0 0 3 3 3 3 3 0 0 1.8 Drymarchon corais 3 1 0 2 3 2 3 3 1 0 1.8 Liophis maryellenae 1 2 3 3 1 2 3 3 0 0 1.8 Liophis meridionalis 1 2 2 2 3 2 2 2 1 1 1.8 Micrurus lemniscatus 2 1 3 3 1 1 2 2 3 0 1.8 Thamnodynastes hypoconia 1 2 3 2 3 1 3 1 2 0 1.8 Boiruna maculata 2 1 1 2 3 2 2 3 0 1 1.7 Clelia quimi 2 2 0 1 3 2 3 3 0 1 1.7 Helicops leopardinus 0 0 3 3 1 3 3 3 1 0 1.7 Apostolepis albicolaris 0 3 3 2 0 2 1 2 3 0 1.6 Bothrops itapetiningae 1 2 1 2 3 1 3 1 2 0 1.6 Echinanthera occipitalis 1 3 3 0 3 1 1 1 1 2 1.6 Epicrates cenchria 2 1 2 2 3 0 2 1 1 2 1.6 Micrurus frontalis 2 1 3 1 1 1 2 1 3 1 1.6 Oxyrhopus rhombifer 1 2 2 1 3 1 0 0 3 3 1.6 Tantilla melanocephala 0 3 3 2 0 0 2 1 3 2 1.6 Crotalus durissus 3 0 2 1 3 0 1 0 2 3 1.5 Eunectes murinus 3 0 0 3 1 2 3 2 1 0 1.5 Helicops modestus 1 1 3 3 1 1 3 2 0 0 1.5 Leptotyphlops koppesi 0 1 3 3 0 3 2 3 0 0 1.5 Mastigodryas bifossatus 3 0 0 2 3 1 2 1 2 1 1.5 Oxyrhopus guibei 2 2 2 0 3 0 0 0 3 3 1.5 Philodryas aestiva 2 0 2 1 3 2 2 2 0 1 1.5 Spilotes pullatus 3 2 1 1 2 0 2 2 1 1 1.5 Boa constrictor 3 0 2 0 3 0 2 1 1 2 1.4 Bothrops moojeni 3 0 0 2 3 0 2 0 2 2 1.4 Chironius quadricarinatus 2 2 3 1 2 0 2 1 0 1 1.4 Atractus pantostictus 0 3 3 2 0 2 1 2 0 0 1.3 Pseudablabes agassizii 1 1 3 3 2 1 1 1 0 0 1.3 Sibynomorphus mikanii 0 2 3 1 2 0 1 0 2 2 1.3

França, F. G. R. and Araújo, A. F. B. 29 Table 1: Continued. Species and Threat categories BS LS DT HB HT DB AH RR CL RM Mean no risk (< 1,3) Bothrops neuwiedi 1 0 0 2 3 0 2 0 2 2 1.2 Liophis reginae 1 2 2 2 3 0 1 1 0 0 1.2 Philodryas nattereri 3 1 1 1 3 0 0 0 0 3 1.2 Apostolepis assimilis 0 3 3 1 0 0 0 0 3 1 1.1 Philodryas olfersii 2 1 1 2 2 0 1 0 0 2 1.1 Waglerophis merremii 2 0 3 0 3 0 0 0 0 3 1.1 Liophis poecilogyrus 1 2 2 0 3 0 0 0 0 1 0.9 Philodryas patagoniensis 2 0 1 1 2 0 1 0 0 2 0.9 Liotyphlops ternetzii 0 1 3 0 0 0 0 0 0 1 0.5 1.41 1.66 2.26 1.97 2.03 1.48 1.95 1.74 1.20 0.90 1.66 4: Snakes with small to medium body size, terrestrial and diurnal, habitat generalists, and found in almost all DF (vulnerable). Liophis poecilogyrus, Echinathera occipitalis. 5: Snakes with medium to large body size, terrestrial or semi-arboreal, with high litter size and wide distribution in DF (low vulnerability). Boa constrictor, Crotalus durissus, Epicrates cenchria, Philodryas nattereri, Philodryas olfersii, Philodryas patagoniensis, Spilotes pullatus, Waglerophis merremii. 6: Snakes with blotched patterns like pitvipers, terrestrial, with high litter size and wide distribution in DF (low vulnerability). Bothrops moojeni, Bothrops neuwiedi, Mastigodryas bifossatus. 7: Snakes that are diet specialists, and have medium litter size (vulnerable). Chironius flavolineatus, Chironius quadricarinatus, Liophis almadensis, Liophis reginae, Pseudablabes agassizii, Sibynomorphus mikanii. 8: Snakes with medium body size, terrestrial habits, and habitat and diet specialists (vulnerable). Bothrops itapetiningae, Erythrolamprus aesculapii, Thamnodynastes hypoconia. 9: Snakes with uniform color, terrestrial or semi-arboreal habits, reduced range and rare in DF, and inadaptable to altered habitats (vulnerable to threatened). Chironius exoletus, Drymoluber brazili, Liophis meridionalis, Liophis paucidens, Oxybelis aeneus, Philodryas aestiva, Philodryas psamophideus, Rhachidelus brazili, Thamnodynastes rutilus. 10: Black snakes, terrestrial and nocturnal, with reduced range and rare in DF (vulnerable to threatened). Boiruna maculata, Clelia quimi, Pseudoboa nigra. 11: Snakes with big body size, with high litter size, reduced range and rare in DF (vulnerable). Eunectes murinus, Clelia plumbea, Drymarchon corais. : Snakes with coral-snake patter of coloration, fossorial or cryptozoic habits, diet and habitat specialists, and rare in DF (vulnerable to threatened). Apostolepis albicollaris, Apostolepis flavotorquata, Apostolepis sp., Micrurus frontalis, Micrurus lemniscatus, Phalotris nasutus, Simophis rhinostoma. : Snakes that are diet and habitat specialists, rare in DF and inadaptable to altered habitats (threatened). Atractus pantostictus, Gomesophis brasiliensis, Helicops angulatus, Helicops leopardinus, Helicops modestus, Leptodeira annulata, Leptotyphlops koppesi, Liophis maryellenae, Lystrophis nattereri, Phimophis guerini, Xenopholis undulatus. In the principal component analysis, two axes explained 65% of the variation in the data (Table 3). The first axis (43% of the variation) was positively correlated to rarity and distribution breadth in Distrito Federal and negatively to road mortality. This axis is thus associated with factors related to geographic distribution and humans attitudes. The second axis (22% of the variation) was positively correlated to dietary breadth and litter size, and negatively to body size. The second axis is associated to factors related to natural-history of the snake species.

30 Conservation of snakes in Central Brazil C A S E 0 5 10 15 20 25 Label Group +---------+---------+---------+---------+---------+ Gomesophis brasiliensis Xenopholis undulatus Helicops angulatus Leptotyphlops koppesi Phimophis guerini Liophis maryellenae Helicops leopardinus Helicops modestus Atractus pantostictus Leptodeira annulata Lystrophis nattereri Apostolepis flavotorquata Apostolepis gr. taenuis Apostolepis albicollaris Simophis rhinostoma Phalotris nasutus Micrurus leminiscatus Micrurus frontalis Drymarchon corais Clelia plumbea Eunectes murinus Boiruna maculata Pseudoboa nigra Clelia quimi Chironius exoletus Drymoluber brazili Philodryas psammophidea Thamnodynastes rutilus Liophis paucidens Rhachidelus brazili Liophis meridionalis Oxybelis aeneus Philodryas aestiva Erythrolamprus aesculapii Thamnodynastes hypoconia Bothrops itapetiningae Chironius flavolinaetus Chironius quadricarinatus Sibynomorphus mikanii Liophis almadensis Pseudablabes agassizii Liophis reginae Bothrops moojeni Mastigodryas bifossatus Bothrops neuwiedi Philodryas nattereni Philodryas patagoniensis Philodryas olfersii Epicrates cenchria Boa constrictor Crotalus durissus Waglerophis merremi Spilotes pullatus Liophis poecilogyrus Echinantera occipitalis Oxyrhopus trigeminus Ocyrhopus rhombiter Oxyrhopus gribei Tantilla melanocephala Apostolepis assimilis Liotyphlops ternetzii 11 11 11 10 10 10 08 08 08 07 07 07 07 07 07 06 06 06 04 04 03 03 03 02 02 01 Figure 2: Cluster diagram showing the similarity groups for threatening factors among sixty species of Central Brazil snakes. Diamond indicates point in clustergram where significance is achieved. Numbers to the right of species indicate the significant groups.

França, F. G. R. and Araújo, A. F. B. 31 Table 2: Summary of the information of Natural History and body measurements (MEAN ± SE in mm) of the snakes in Distrito Federal. Abbreviations are: CA = Campo, CE = Cerrado, GF = Gallery Forest, VE = Vereda, semi-arb = semi-arboreal, abn = amphisbaenian, amp = amphibian, ann = annelids, arn = aranae, bi = birds, chi = chilopoda, cro = crocodylians, fi = fish, gas = gastropode, ins = insecta, li = lizards, mam = mammals, mil = millipede, sn = snakes, N = Sample sizes, SVL = snout-vent length, TL = tail length. Capital letter means principal habitats. FAMILY SUBFAMILY Tribe Species N Habits Habitats Activity Patterns Diet SVL TL ANOMALEPIDIDAE Liotyphlops ternetzii 30 Fossorial CA, CE, GF Nocturnal-Diurnal Ins 239.41 60.99 4.29 1.07 LEPTOTYPHLOPIDAE Leptotyphlops koppesi 1 Fossorial CA, CE Nocturnal-Diurnal Ins 165.00 17.00 BOIDAE Boa constrictor 14 Terrestrial, semi-arb GF, ce Nocturnal-Diurnal Mam., bi 598.86 267.53 77.77 39.67 Epicrates cenchria 21 Terrestrial, semi-arb CE, ca Nocturnal-Diurnal Mam., bi, li 520.86 260.71 67.29 34.53 VIPERIDAE Bothrops itapetiningae Terrestrial CA, ce Nocturnal-Diurnal Mam, li, amp, bi, mil, chi 304.00 116.78 39.23.81 Bothrops moojeni 45 Terrestrial GF, ve, ce Nocturnal-Diurnal Mam, li, amp, bi, sn, mil, chi 618.45 320.07 101.00 48.49 Bothrops neuwiedi 26 Terrestrial CA, ce Nocturnal-Diurnal Mam, li, amp, bi, sn, mil, chi 4.85 145.08 59.85 20.81 Crotalus durissus 50 Terrestrial CE, CA, gf Nocturnal-Diurnal Mam, bi 755.25 262.63 68.57 32.08 ELAPIDAE Micrurus frontalis 17 Cryptozoic CE, CA, gf Nocturnal-Diurnal Sn, abn 650.24 234.45 42.18 14.35 Micrurus lemniscatus 8 Cryptozoic GF Nocturnal Sn 603.00 311.57 42.50 19.96 COLUBRIDAE Colubrinae Chironius exoletus 4 Arboreal, Terrestrial GF Diurnal Amp, li, bi, mam 806.50 96.42 468.75 50.80 Chironius flavolineatus Arboreal, Terrestrial GF, ve, ce Diurnal Amp, li, bi, mam 675.08 119.46 440.54 90.50 Chironius quadricarinatus 15 Arboreal, Terrestrial CE, ca Diurnal Amp, li, bi, mam 599.40 80.90 360.40 57.43 Drymarchon corais 16 Terrestrial, semi-arb CE, gf Diurnal Mam, amp, li, bi, sn, abn 48.50 297. 304.44 84.96 Drymoluber brazili 4 Terrestrial CE Diurnal Li 493.29 187.99 8.67 49. Mastigodryas bifossatus 11 Terrestrial CE, GF, CA Diurnal Mam, amp, li, bi, sn, abn 1017.00 382.89 392.91 8.48 Oxybelis aeneus 3 Arboreal, semi-arb CE Diurnal Amp, li, bi, mam 749.33 47.82 468.67 31.94 Simophis rhinostoma 11 Terrestrial CE, CA Diurnal Amp 448.73 1.20 116.00 27.22 Spilotes pullatus 9 Arboreal, semi-arb GF, ve, ce Diurnal Bi, mam 99.33 297.87 432.00 1.72 Tantilla melanocephala 15 Fossorial, criptozoic CA, CE Diurnal Chi 228.33 61.73 73.27 21.51 Dipsadinae Dipsadini Atractus pantostictus 10 Cryptozoic GF, ce Nocturnal Ann 268.60 95.00 30.50 10.56 Sibynomorphus mikanii 36 Terrestrial, semi-arb VE, GF, ce Nocturnal Gás 269.97 108. 61.67 25.74 Leptoderini Leptodeira annulata 5 Arboreal, semi-arb GF, VE Nocturnal Amp 477.60 101.28 167.20 28.26 *Incerta sedis Echinanthera occipitalis Terrestrial CE, CA, gf Diurnal Li, Amp 323.00 69.18 107.25 24.63 Xenopholis undulatus 8 Cryptozoic GF Nocturnal Amp 268.38 66.17 47.88.11 Xenodontinae Elapomorphini Apostolepis assimilis 10 Fossorial CE, CA, gf Nocturnal-Diurnal Abn 228.30 71.92 23.00.

32 Conservation of snakes in Central Brazil Table 2: Continued FAMILY SUBFAMILY Tribe Species N Habits Habitats Activity Patterns Diet SVL TL Apostolepis gr. tenuis 5 Fossorial CA Nocturnal-Diurnal Abn 231.80 34.10 27.20 4. Apostolepis albicolaris 9 Fossorial CE, CA Nocturnal-Diurnal Abn 266.78 72.62 29.56 9.79 Apostolepis flavotorquata 2 Fossorial CE Nocturnal-Diurnal Abn 459.50 44.55 41.50 2. Phalotris nasutus 10 Cryptozoic, fossorial CA, CE Nocturnal-Diurnal Abn, Sn 517.60 160. 57.70.71 Hydropsini Helicops angulatus 4 Aquatic GF, VE Nocturnal Fi, Amp 395.25 143.84 187.25 75.87 Helicops leopardinus 1 Aquatic GF Nocturnal Fi, Amp 240.00 0.00 Helicops modestus 18 Aquatic GF, VE Nocturnal Fi, Amp 284.71 106.27 93.29 31.26 Philodriadini Philodryas aestiva 14 Terrestrial CA, ce Diurnal Mam, li 503.57 157.46 214.14 63.40 Philodryas nattereri 35 Terrestrial, semi-arb CE, ca Diurnal Mam, amp, li, bi 696.91 278.38 277.77 0.73 Philodryas olfersii 22 Semi-arb, Terrestrial GF, ce Diurnal Mam, amp, li, bi 648. 208.01 261.18 82.65 Philodryas patagoniensis 46 Terrestrial, semi-arb CE, ca, ve Diurnal Mam, amp, li, bi 677.46 251.26 252. 87.20 Philodryas psammophidea 2 Terrestrial CE Diurnal Li 453.50 86.97 155.50 43. Pseudablabes agassizii 21 Terrestrial CA, ce Diurnal Arn 260.90 75.26 85.19 24.14 Pseudoboini Boiruna maculata 3 Terrestrial CE Nocturnal Sn, mam 458.00 1.64 94.00 17.52 Clelia plumbea 1 Terrestrial GF Nocturnal Sn, li, mam 1401.00 315.00 Clelia quimi 7 Terrestrial GF, ce Nocturnal Sn, mam 493.29 187.99 8.67 49. Oxyrhopus guibei 34 Terrestrial GF, CE, CA Nocturnal Li, mam 438.74 193.80 116.82 55.11 Oxyrhopus rhombifer 40 Terrestrial CA, CE, gf Nocturnal Li, mam 3.33 5.80 77.15 36.04 Oxyrhopus trigeminus 11 Terrestrial CE, CA Nocturnal Li, mam 501.36 6.44 8.00 28.98 Phimophis guerini 4 Cryptozoic CE, CA Nocturnal Li 475.80 222.73 1.00 66. Pseudoboa nigra 7 Terrestrial CE Nocturnal Li 543.86 238.22 180.86 85.14 Rhachidelus brazili 6 Terrestrial CE Nocturnal Bi (eggs) 785.50 316. 2.33 103.52 Tachymenini Gomesophis brasiliensis 3 Aquatic VE, gf Nocturnal Ann 289.00 43.92 69.67 31.82 Thamnodynastes hypoconia 9 Terrestrial, semi-arb VE, gf Nocturnal Amp 323.44 56.82 1.67 22.77 Thamnodynastes rutilus 3 Terrestrial, semi-aqua GF Nocturnal Amp, fi 275.00 1.00 108.67 43.82 Xenodontini Erythrolamprus aesculapii 34 Terrestrial, cryptozoic GF Diurnal Sn, li 426.29 5.82 58.56 22.08 Liophis almadensis 9 Terrestrial CE, CA, gf Diurnal Amp 285.11 87.70 90.00 29.86 Liophis maryellenae Terrestrial, semi-aqua VE, gf Diurnal Fi 304.92 111.91 94.00 33.74 Liophis meridionalis 11 Terrestrial CE, CA, gf Diurnal Amp, li 443.91 119.65 167.27 49.99 Liophis paucidens 1 Terrestrial CE Diurnal Li 369.00 7.00 Liophis poecilogyrus 56 Terrestrial CE, CA, gf Diurnal-Nocturnal Amp 369. 8.87 87.23 33.19 Liophis reginae 19 Terrestrial GF Diurnal-Nocturnal Amp 417.16 99.63 159.11 48.70 Lystrophis nattereri 7 Terrestrial CA, ce Diurnal Li (eggs) 245.00 75.78 44.00 16.06 Waglerophis merremii 46 Terrestrial CE, GF Diurnal Amp 560.74 214.64 98.30 38.87

França, F. G. R. and Araújo, A. F. B. 33 Table 3: Factor loadings of each variable on the first two principal components and proportion of the variance explained by each component. Principal component Variable factor loadings Factor I Factor II Body size -0.157-0.810 Litter size 0.175 0.731 Dietary breadth 0.118 0.777 Habitat breadth 0.833 0.3 Habits -0.143-0.696 Distribution breadth in DF 0.887 0.186 Adaptability to altered habitats 0.847-0.1 Rarity in DF 0.932 0.4 Species color -0.277 0.434 Road Mortality -0.814-0.253 Eigenvalue 4.30 2.20 Variance explained Percent 42.95 22.08 Cumulative 42.95 65.03 DISCUSSION The lack of basic natural history information on many snake species and data on the phylogenetic relationships among species complicate evaluations of the factors that contribute to snake population declines (Greene, 1994). Our intensive studies on the snake assemblage in Distrito Federal over the last several years has generated a wealth of natural history data (França, 2003), which has permitted us to asses the conservation status for snakes in central Brazil. Filippi and Luiselli (2000) pointed out that several Italian snakes are more vulnerable to distribution-related attributes, whereas species-specific life-history traits and trade appear to be threats for a lower number of taxa. In snakes from Distrito Federal, life-history factors and distribution attributes affected the viability of snake populations similarly. Moreover, phylogenetic groups did not differ in their risk, as species in different clades are present in groups with different degrees of risk. Although the method proposed by Filippi and Luiselli (2000) was designed for a snake assemblage in the temperate zone, with less than one-third of the species richness found in the tropical region of DF, our results were similar. Filippi and Luiselli (2000) found that of 19 snake species in Italy, only 10.5% were not at risk, 52.6% were vulnerable, and 36.8% were threatened. Our results on snakes in Distrito Federal, Brazil, placed 14.8%, 55.7%, and 29.5% of species in the same categories, respectively. Mean threat scores for Italian snake species ranged from 0.7 to 2.4 (Filippi and Luiselli, 2000), whereas the mean threat scores for Brazilian species ranged from 0.5 to 2.3. Mean scores for each of the 10 factors used by Filippi and Luiselli (2000) ranged from 0.4 to 2.3, whereas mean scores ranged from 0.9 to 2.3 for Brazilian snakes. Finally, multivariate analysis produced 5 groups with similar threat factors in the Italian snake assemblage (Filippi and Luiselli, 2000), whereas we delineated groups. However, because of the higher number of snake species in DF compared to Italy (61 versus 19), the number of groups is comparable. Our results demonstrate that the method outlined by Filippi and Luiselli (2000) is consistent and applicable to snake assemblages in other regions. Of groups formed by PCA and Cluster analyses, we consider one group with no risk, three with low risk, and nine groups ranging from vulnerable to threatened. Almost all of these assessments combine life-history and distribution-related factors. Group, which includes one leptotyphlopid and 10 colubrids, had the highest threat values due to the combination of specialized natural-history traits such as diet, habitat, the inadaptability to altered habitats, and rarity in the region. Small snakes that are habitat generalist, are found in altered environments in DF, and have a wide distribution were at the lowest risk, for example, Liotyphlops ternetzii (group1), and Apostolepis assimilis and Tantilla melanocephala (group 2). Specific life-history traits increase risk for several species and groups. Species that eat exclusively a single prey type, which can be seasonally abundant, are more prone to risk during the season of lowest prey abundance (Vitt, 1983). In Distrito Federal, some snakes eat exclusively frogs (e.g., some Liophis, Thamnodynastes and Waglerophis merremii), or invertebrates (e.g., Liotyphlops ternetzii, Leptotyphlops koppesi, Gomesophis brasiliensis, Pseudablabes agassizii, and Tantilla melanocephala). The habits of snakes can also influence the degree of threat. Terrestrial and diurnal snakes are more vulnerable to predators, especially visually oriented predators such as birds (Greene, 1997; Dodd, 2001b). Finally, habitat specificity is one of the most important factors determining risk in the herpetofauna of central Brazil (Colli et al., 2002). The Cerrado has been undergoing habitat alteration at an accelerated pace (Alho and Martins, 1995) and a concomitant loss of biodiversity has occurred. Species with specific habitat requirements such as the three species of watersnake Helicops, Gomesophis brasiliensis, and

34 Conservation of snakes in Central Brazil Liophis maryellenae, forest snakes as Chironius exoletus, and snakes restricted to open areas such as Leptotyphlops koppesi and Drymoluber brazili, are more sensitive to habitat modification. Factors related to geographic distribution of species are known to be threats to snakes in all regions of the world. Since 1954, the Distrito Federal has lost more than 55% of its original vegetation and this loss continues at a rapid pace (Werthein et al., 2000). This loss is an especially serious problem for species with limited distributions in the region, such as Apostolepis flavotorquata, Chironius exoletus and Philodryas psamophidea. The first two species are found in a dry forest area that is not under any conservation protection, and the latter species was found in an area that has been entirely urbanized and has not been encountered again since the 1970s. These species also do not adapt well to altered environments, which places them at higher risk. Human attitudes also create serious threats to snakes. The traditional Judeo-Christian view of snakes is exceedingly antagonist, and the fear and lack of knowledge results in high mortality at the hands of humans (Greene, 1997; Shine and Bonnet, 2000). In central Brazil, all coral-snakes have triad patterns with red, black and white colors. All snakes with these colors are considered venomous and are killed indiscriminately, even though some species are non-venomous coral snake mimics. In Distrito Federal, we found two groups of snakes with coral-snake patterns; one group included species of Oxyrhopus, and the other group included Apostolepis, Micrurus, Phalotris and Simophis. All species are considered under threat because of their aposematic coloration, but the species of Oxyrhopus are considered less vulnerable than the other groups because they are widely distributed and have a more generalist ecology. Road mortality is a serious source of mortality for many wild animals in the world (Trombulak and Frissel, 2000). The elongated bodies and relatively slow movement of snakes, make them susceptible to vehicular mortality. Also, snakes that disperse from their birth area, make seasonal movements to and from hibernacula, track changing prey resource abundances, search for mates, or use roads for thermoregulation sites are at increased risk of being run over (Bonnet et al., 1999; Dodd, 2001b). High numbers of dead snakes have been reported on roads in several places (Dodd et al., 1989; Bush et al., 1991; Enge and Wood, 2002). On four roads in Distrito Federal, totaling 42,5 km, a one year study on road mortality found 575 dead reptiles, of which 362 were snakes (Rodrigues et al., 2002). The colubrids Oxyrhopus guibei, Philodryas nattereri, and Waglerophis merremii were the most frequently killed snakes on roads in DF, all of which are terrestrial. Conservation of snakes can be accomplished by a variety of methods. If the goal of a conservation program is to ensure the long-term survival of a snake species, its habitat needs to be conserved (Dodd, 2001b). Habitats may be protected by the implementation of parks, reserves, or conservation easements with private landowners (Sutherland, 2001). The Distrito Federal has three conservation units encompassing 50,000 ha that are integral in the protection of Cerrado flora and fauna (Werthein et al., 2000). Although these three areas represent less than 25% of the total area of the DF region, they are important to the maintenance of the local avifauna, protecting about 81% of all bird species (Braz and Cavalcanti, 2001). However, as stated previously, other important areas inside DF, such as the north dry forest, are still in need of protection. Additionally, the construction of tunnels or culverts under highways, or wildlife barriers might be useful in deterring snakes from crossing roads and minimize road mortality (Dodd, 2001b). Other potential snake conservation methods may include translocation or repatriation (Dodd and Seigel, 1991; Dodd, 2001b), although the impacts of these managements on individual snakes are not completely known. Some studies report that translocated snakes can move larger distances and show different habitat preferences to resident snakes, concluding that this practice influences individual snakes (Butler et al., 20). Finally, educational programs can be a powerful conservation tool to aid in the protection of snake species and assemblages. In DF, educational programs and materials can be presented at universities, schools, or local Zoos in an attempt to alleviate fear, correct misinformation and myths, and to teach the importance of snakes in nature. Assessing the conservation status of snake species in a specific region is an important first step in changing attitudes and generating interest in preserving snake diversity (Dodd, 2001a). However, conservation efforts are limited by the constraints imposed by the species natural history. Basic biological data are lacking for most snake species and this is problematic in as-

França, F. G. R. and Araújo, A. F. B. 35 sessing the conservation status of snakes in most regions of the world. The method proposed by Fillipi and Luiselli (2000) worked well for assessing the conservation status of snakes in central Brazil and further ecological studies on Brazilian snake assemblages should be done to test the applicability of the method to other regions. Ultimately, assessing the conservation status of snake species in different areas will allow comparisons among regions and will generate a better status assessment that will aid in the conservation of Brazilian snakes. RESUMO Nós avaliamos o status de conservação de 61 espécies de uma comunidade de serpentes do Brasil central pelo ranqueamento das espécies de acordo com suas vulnerabilidades a dez fatores de ameaça que influenciam a sobrevivência de populações de serpentes. A serpente anomalepidide Liotyphlops ternetzii foi a espécie com menor risco e o colubrídeo Drymoluber brazili foi a mais ameaçada. Menos de 15% da fauna de serpentes do Distrito Federal foi considerada livre de qualquer ameaça. A história natural das espécies, suas distribuições geográficas e atitudes humanas foram importantes fatores determinando os níveis de ameaça para as serpentes do Brasil central. Análise de Componentes Principais e Análise de Agrupamento foram usadas para classificar as espécies em grupos de vulnerabilidade. Treze grupos foram identificados variando de nenhum risco a grupos ameaçados. No Brasil, a avaliação do status de conservação para outras comunidades de serpentes em outros biomas pode ser útil para gerar um plano de conservação efetivo para o país. ACKNOWLEDGEMENTS We are grateful to D. Shepard, M. Mira, R. Constantino and V. Braz for helpful critical comments on previous version of the manuscript and for help in English and to Guarino Colli, Francisco Franco and Ronaldo Fernandes for permission to examine specimens on CHUNB, IB and MNRJ. This work was supported by a master s fellowship from Conselho Nacional de Desenvolvimento Científico e Tecnológico CNPq to FGRF. LITERATURE CITED ALHO, C. J. R. & E. S. MARTINS. 1995. De Grão em Grão, o Cerrado Perde Espaço. WWW-Fundo Mundial para Natureza. Brasília. 66 p. BONNET, X., G. NAULLEAU & R. SHINE. 1999. The dangers of leaving home: dispersal and mortality in snakes. Biological Conservation, 89:39-50. BRAZ, V. S. & R. B. CAVALCANTI. 2001. A representatividade de áreas protegidas do Distrito Federal na conservação da avifauna do Cerrado. Ararajuba, 9:61-69. BUSH, B., R. BROWNE-COOPER & B. MARYAN. 1991. Some suggestions to decrease roadkills in reserves with emphasis on the W.A. Wheatbelt. Herpetofauna, 21:23-24. BUTLER, H., B. MALONE & N. CLEMANN. 20. Wildlife Reasearch, 32:157-163. COLLI, G. R., R. P. BASTOS & A. F. B. ARAÚJO. 2002. The character and dynamics of the cerrado herpetofauna; pp 223-241. In: P. S. Oliveira & R. J. Marquis (Eds.), The Cerrados of Brazil: Ecology and Natural History of a Neotropical Savanna. Columbia University Press, New York. DALTRY, J. C., Q. BLOXAM, G. COOPER, M. L. DAY, J. HARTLEY, M. HENRY, K. LINDSAY & B. E. SMITH. 2001. Five years of conserving the world s rarest snake, the Antiguan racer Alsophis antiguae. Oryx, 35:119-7. DODD, C. K., JR. 2001a. Status, Conservation, and Management; pp. 478-5. In: R. A. Seigel, J. T. Collins & S. S. Novak (Eds.), Snakes: Ecology and Evolutionary Biology. Reprint of First Ed. The Blackburn Press, Caldwell, New Jersey. DODD, C. K., JR. 2001b. Strategies for Snake Conservation; pp. 363-393. In: R. A. Seigel & J. T. Collins (Eds.), Snakes: Ecology and Behaviour. Reprint of First Ed. The Blackburn Press, Caldwell, New Jersey. DODD, C. K., JR. & R. A. SEIGEL. 1991, Relocation, repatriation, and translocation of amphibians and reptiles: Are they conservation strategies that work? Herpetologica, 47:336-350. DODD, C. K., JR., K. M. ENGE & J. N. STUART. 1989. Reptiles on highways in north-central Alabama, USA. Journal of Herpetology, 23:197-200. DUARTE, M. R., G. PUORTO & F. L. FRANCO. 1995. A biological survey of the pitviper Bothrops insularis Amaral (serpentes, viperidae): an endemic and threatened offshore island snake of southeastern Brazil. Studies on Neotropical Fauna and Environment, 30:1-. EINTEN, G. 1994. Vegetação do Cerrado; pp. 9-65. In: M. N. Pinto (Ed.), Cerrado: Caracterização, Ocupação e Perspectivas. Editora Universidade de Brasília, Brasília. ENGE, K. M. & K. N. WOOD. 2002. A Pedestrian Road Survey of an Upland Snake Community in Florida. Southeastern Naturalist, 1:365-380. ETEROVIC, A. & M. R. DUARTE. 2002. Exotic snakes in São Paulo City, southeastern Brazil: why xenophobia? Biodiversity and Conservation, 11:327-339. FILIPPI, E. & L. LUISELLI, L. 2000. Status of the Italian fauna and assessment of conservation threats. Biological Conservation, 93:219-225. FRANÇA, F. G. R. 2003. Ecologia de uma Comunidade de Serpentes no Cerrado do Brasil Central: Composição, Estrutura e Status de Conservação. M.Sc. Dissertation. Universidade de Brasília, Brasília, 64 pp. FRANÇA, F. G. R. & A. F. B. ARAÚJO. In press. Are There Cooccurrence Patterns That Structure Snake Communities In Central Brazil? Brazilian Journal of Biology. GREENE, H. W. 1994. Systematics and Natural History, Foundations for Understanding and Conserving Biodiversity. American Zoologist, 34:48-56.

36 Conservation of snakes in Central Brazil GREENE, H. 1997. Snakes: the evolution of mystery in nature. University of California Press, California, 351 pp. MARQUES, O. A. V., M. MARTINS & I. SAZIMA. 2002. A new insular species of pitviper from Brazil, with comments on evolutionary biology and conservation of the Bothrops jararaca group (Serpentes, Viperidae). Herpetologica, 58:303-3. MYERS, N., R. A. MITTERMEIER, C. G. MITTERMEIER, G. A. B. FONSECA & J. KENT. 2000. Biodiversity hotspots for conservation priorities. Nature, 403:853-858. NIMER, E. 1989. Climatologia do Brasil (2nd ed.). Fundação Instituto Brasileiro de Geografia e Estatística IBGE, Rio de Janeiro, 421 pp. OLIVEIRA, P. S. & R. J. MARQUIS. 2002. The Cerrados of Brazil: Ecology and Natural History of a Neotropical Savanna. Columbia University Press, New York. 424 pp. PINTO, M. N. 1994. Caracterização geomorfológica do Distrito Federal; pp. 277-308. In: M. N. Pinto (Ed.), Cerrado: Caracterização, Ocupação e Perspectivas. Editora Universidade de Brasília, Brasília. REED, R. N. & R. SHINE. 2002. Lying in wait for extinction: ecological correlates of conservation status among Australian elapid snakes. Conservation Biology, 16:451-461. RODRIGUES, F. H. G., A. HASS, L. M. REZENDE, C. S. PEREIRA, C. F. FIGUEIREDO, B. F. LEITE, & F. G. R. FRANÇA. 2002. Impacto de rodovias sobre a fauna da Estação Ecológica de Águas Emendadas, DF. Anais do III Congresso Brasileiro de Unidades de Conservação, 585-593. SEIGEL, R. A. 1986. Ecology and conservation of an endangered rattlesnake Sistrurus catenatus in Missouri USA. Biological Conservation, 35:333-346. SHINE, R. & X. BONNET. 2000. Snakes: a new model organism in ecological research? Trends in Ecology and Evolution, 15:221-222. SUTHERLAND, W. J. 2001. The Conservation Handbook: research, management and policy. Iowa State University Press, Blackwell Science, Iowa, 278 pp. TABACHNICK, B. G. & L. S. FIDELL. 2001. Using Multivariate Statistics (4 th ed.). Allyn and Bacon, Boston, 932 pp. TROMBULAK, S.C. & C. A. FRISSEL. 2000. Review of ecological effects of roads on terrestrial and aquatic communities. Conservation Biology, 14:18-30. VALENTIN, J. L. 2000. Ecologia Numérica: uma introdução a analise multivariada de dados ecológicos. Editora Interciência, Rio de Janeiro. 117 pp. VITT, L. J. 1983. Ecology of an anuran-eating guild of terrestrial tropical snakes. Herpetologica, 39:52-66. WEBB, J. K. & R. SHINE. 1997. Out on a limb: conservation implications of tree-hollow use by a threatened snake species (Hoplocephalus bungaroides: Serpentes, Elapidae). Biological Conservation, 81:21-33. WERTHEIN, J., M. D. A. BORGES & C. CUNHA. 2000. Vegetação no Distrito Federal tempo e espaço. UNESCO, Brasília, 74 pp. ZAR, J. H. 1999. Biostatistical Analysis (4 th ed.). Prentice-Hall, Inc., New Jersey, 929 pp. Submitted January 2006 Accepted 15 March 2006