Exotic snakes in São Paulo City, southeastern Brazil: Why xenophobia?

Biodiversity and Conservation 11: 327–339, 2002. © 2002 Kluwer Academic Publishers. Printed in the Netherlands.

Exotic snakes in São Paulo City, southeastern Brazil: why xenophobia? ANDRÉ ETEROVIC1,2 and MARCELO RIBEIRO DUARTE1,3,∗ 1 Laboratório de Herpetologia, Instituto Butantan, Avenida Vital Brazil 1500, CEP 05503-900, São Paulo; 2 Laboratório de Ecologia Evolutiva, Departamento de Ecologia Geral, Instituto de Biociências, Universidade de São Paulo, Rua do Matão 321; 3 Departamento de Zoologia, Instituto de Biociências,

Universidade Estadual Paulista, Botucatu, SP, Brazil; *Author for correspondence (e-mail: mrduarte@ attglobal.net) Received 30 August 2000; accepted in revised form 2 April 2001

Abstract. Introduced exotic species cause environmental changes and threat public health in target sites. Illegal trade has enhanced this problem. To first report these risks in Brazil, exotic snakes found in São Paulo City (SPC) (23◦ 32 S, 46◦ 38 W), southeastern Brazil, and sent to Instituto Butantan between 1995 and 2000, were listed and characterized by their biological attributes. Seventy-six individuals of sixteen alien species were collected. Euriecians snakes, mainly booids, were predominant. Using multivariate techniques, their ecological niches were compared to those of 26 native species, as a way to point out the resource’s availability. To evaluate the potential of successful implantation, two species absent in SPC and considered as problem snakes are included in these analyses: the brown treesnake Boiga irregularis and the habu Trimeresurus flavoviridis. There were niche similarities between these pest snakes, exotic booids and native viperids largely due to the similarities in the chosen prey (mammals), diel activity (nocturnal), color pattern (variegated) and body size (medium to large). To avoid predictable undesirable effects of implanted pest snakes, traffic control and punishment should be improved, as well as parallel environmental education programs. Key words: animal trade, biological conservation, Brazil, ecological niche, exotic species, snakes

Introduction The major risks of human-induced introductions of exotic animals are: (1) population depletion by hunting in original countries (Dodd 1987, 1993; Adams et al. 1994; Pough et al. 1998); (2) threats to public health by zoonoses, bites or envenomation (Reid 1978; Minton 1996; Rodda et al. 1997); (3) introduction of new parasites (Reinert and Ruppert Jr 1999); and (4) changes in native fauna of target countries (Savidge 1987; Rodda and Fritts 1992; Martínez-Morales and Cuarón 1999). A critical example of snake introduction is that of the brown treesnake Boiga irregularis in Guam. Like the habu Trimeresurus flavoviridis in some Japanese islands, it needs costly management strategies (Rodda et al. 1999b). The development of global communication and international market has enhanced this problem (Rodda et al. 1997): in a few minutes, one can find a long list of species for sale by Internet, many of which

328 are regarded by CITES and IUCN as endangered. Both the introduction of exotic snakes and the commerce of native species as pets have only recently received more attention by environmental agencies in the USA (Smith and Kohler 1978; Adams et al. 1994; Pough et al. 1998; Duellman 1999). In Brazil, nothing has been published on this matter. São Paulo City (SPC; 23◦ 32 S, 46◦ 38 W; 760 m above sea level), the capital of São Paulo State, is the main commercial center in southeastern Brazil. General trade, including that of exotic pet snakes, is concentrated in this part of the country. Illegal traffic exists, as evidenced by recent IBAMA (the National Agency for Environment) apprehensions. Despite the prohibition of importing exotic snakes and of trading native ones (Federal Order no. 93/98, 7 July 1998), they are still found as pets in SPC. The illegal status leads to an absence of specialist’s care and the pet snakes frequently suffer from sick or undesirable conditions. Therefore, their owners (or ‘parents’) often release them in many parts of the city, where they also go when they escape. These areas are subject to different degrees of anthropic activities and they house at least 26 native snake species (Puorto et al. 1991). When recaptured, the destination of such alien snakes is frequently the Instituto Butantan (IB), a traditional ophiology center in SPC. Received individuals are exposed to public visiting at IBs museum and, after their death, they are incorporated into the Alphonse Hoge’s Herpetological Collection. This procedure is also applied to thousands of native specimens that arrive at IB yearly, sent by occasional collectors from all Brazilian regions, as well as by other public institutions (including IBAMA). This paper is the first quantitative survey of exotic snakes’ arrivals at IB, presented as a measure of the colonization risk of SPC and, by extent, of preservation areas in Brazil. A discussion concerning the conservation aspects of snake introduction is also made based on a qualitative approach of the ecological niche.

Methods A checklist of native species by Puorto et al. (1991) based on individuals sent to IB in 1988–1989 was assumed to reflect the actual composition of the snake community at SPC. The majority of exotic snakes found in SPC and sent to IB from January 1995 to April 2000 were identified to species level. Translocated species and subspecies, i.e. species and subspecies that occur naturally in Brazil but not in SPC, were considered as exotic. The number of individuals in each taxon was recorded. To show the pattern of new species arrivals, a collector’s curve was designed by Sanders’ Rarefaction Method (Hurlbert 1971; Ludwig and Reynolds 1988). It consists of re-sampling with replacement (90 times, in this case) blocks of a variable number of individuals from the original data set and subsequent fitting of a logarithmic expression. Both exotic and native snakes were qualitatively classified (presence/absence) by 24 biological categories concerning reproductive mode, prey type, foraging mode,

329 Table 1. Biological categories used in a qualitative classification of the ecological niche of exotic and native snake species found in SPC, Brazil. Code

Category

Description

ovi viv mam bir liz sna

Oviparity Viviparity Mammals Birds Lizards Snakes (including amphisbaenians and caecilians) Anurans Fishes Invertebrates Active search Sit and wait Terrestrial Arboreal Aquatic Fossorial (and criptozooic) Variegated Smooth (uniform coloration) Red (mainly in annuli) Longitudinally striped Small-sized Medium-sized Large-sized Diurnal Nocturnal

Reproductive mode Reproductive mode Frequent use of this prey Frequent use of this prey Frequent use of this prey Frequent use of this prey

anu fis inv act sit ter arb acq fos var smo red str sma med big diu noc

Frequent use of this prey Frequent use of this prey Frequent use of this prey Foraging mode Foraging mode Frequent use of the ground surface Frequent use of trees Frequent activity in water Frequent activity under the ground Color pattern Color pattern Color pattern Color pattern Average body length (< 0.5 m) Average body length (0.5–1.5 m) Average body length (> 1.5 m) Frequent activity in this period Frequent activity in this period

habitat use, color pattern, body size and circadian activity of adults (Table 1). Ontogenetic changes were not considered. Data were obtained by published references or personal observations. In few cases when data were not available, attributes were inferred by that of closely related congeneric species. All of the employed attributes are dependent on environmental conditions and resources that satisfy the requirements of each species, i.e. they are linked to the n-dimensional hypervolumetric Hutchinson’s ecological niche (Reinert 1993; Begon et al. 1996; Ricklefs 1996). Based on this statement, these multivariate data were used to describe qualitatively the niche of each snake. The categories ‘oviparity’, ‘active search’, ‘terrestrial’, ‘medium-sized’ and ‘diurnal’ were presented by the majority of the snakes studied here, and they were used as default. In other words, the remaining categories were considered as ‘deviations’ from such ‘main patterns’, contributing to enhance differences between ecological niche of the snakes in further analyses. To estimate niche similarities, the taxon-categories matrix was subjected to clustering analysis (e.g. Duellman 1978, Martins and Oliveira 1998), using Jaccard’s similarity index and UPGMA method (Ludwig and Reynolds 1988; Magurran 1988). A correspondence analysis (Gauch 1982; Ter Braak 1986, 1988; Ludwig and

330 Reynolds 1988) allowed an ordination of the species by the same attributes. As these multivariate methods were employed in an exploratory way, statistical tests were not applied (Ter Braak 1986, 1988). Two pests snakes not found in SPC were included in these analyses, as model parameters of successfully implanted species: the colubrid B. irregularis and the viperid T. flavoviridis. The brown treesnake is a big variegated snake, oviparous, mainly arboreal and nocturnal, that searches actively for rodents, birds, lizards and anurans (Rodda et al. 1999a). The habu has the same attributes, but it is less arboreal than B. irregularis and it preys mainly on rodents and birds by ambushing (Mishima et al. 1999).

Results Twenty-six species of native snakes found in SPC are listed in Table 2. Seventy-six individuals of sixteen exotic species found in SPC were sent to IB from January 1995 to April 2000. The yearly capture reaches its maximum value (32 snakes, 42% of total) in 1999, due to one of IBAMAs apprehensions of 22 illegally captive individuals (Figure 1a). The collector’s curve and parameters of the fitted equation are presented in Figure 1b. Fifty percent of the alien individuals found were from other areas of South America, 22% from North America, 17% from Africa, 5% from Asia, 1% from Oceania, whereas the origins of the remaining 4% were not determined (Table 3). Booids (Boa constrictor, Corallus caninus, Epicrates cenchria, Liasis albertisi and Python spp.) represented 67% of total individuals, and the remaining were colubrids. The order of abundance of genera was: Boa (one species, three subspecies, 25% of the individuals), Python (three species, 22%), Lampropeltis (two species, 13%), Epicrates (one species, four subspecies, 12%), Elaphe (two species, 8%), Corallus (one species, 7%), Boaedon (one species, 4%), Philodryas (one species, 4%), Lystrophis (two species, 3%), Liasis (one species, 1%) and Thamnophis (one species, 1%). Cluster analysis of interspecific similarities in biological attributes of native and exotic species (Tables 2 and 3) revealed three main groups, at 0.2 similarity level (Figure 2). Group 1 has 5 native and 10 exotic species. It includes all booids, both aborigine viperids (Bothrops jararaca and Crotalus durissus), both pseudoboines (Oxyrhopus spp.), and other medium-sized colubrids (Boaedon fuliginosus, Elaphe spp., Tropidodryas striaticeps). Both problem snakes, B. irregularis and T. flavoviridis, also compound this group. All of the Group 1 species feed mainly on mammals. Group 2 has three native (the elapid Micrurus corallinus, and the colubrids Apostolepis assimilis and Erythrolamprus aesculapii) and three exotic snakes (Philodryas psammophideus, and Lampropeltis spp.), that prey on serpentiform animals. Group 3 included 18 native and three exotic snakes (Lystrophis dorbignyi, L. semicinctus and Thamnophis sirtalis), all being medium and small colubrids that prey mainly on anurans or invertebrates (the case of Atractus reticulatus, Liotyphlops beui, Sibynomorphus mikanii, Tantilla melanocephala and Tomodon dorsatus). Rising the level

331 Table 2. Native snakes found in SPC, Brazil. This checklist was based on 309 individuals sent to IB in 1989–1990 (Puorto et al. 1991). Biological categories used to define the qualitative ecological niche of each species were coded as in Table1. The categories ‘ovi’, ‘act’, ‘ter’, ‘med’ and ‘diu’ were omitted. References are: (D) Duellman 1978; (G) Greene 1997; (J) Jordão and Bizerra 1996; (L) Leynaud and Bucher 1999; (M) Marques 1998; (Ma) Martins and Oliveira 1998; (S) Sazima and Haddad 1992; (V) Vitt and Vangilder 1983. Asterisks indicate inferences based on attributes of congeneric species. Species

Categories (References)

Apostolepis assimilis (Reinhardt 1861) Atractus reticulatus (Boulenger 1885) Bothrops jararaca (Wied 1824) Chironius bicarinatus (Wied 1820) Crotalus durissus (L. 1758) Echinantera undulata (Wied 1824) Erythrolamprus aesculapii (L. 1766) Helicops modestus (Günther 1861) Liophis almadensis (Wagler 1824) L. jaegeri (Günther 1858) L. miliaris (L. 1758) L. poecylogyrus (Wied 1825) Liotyphlops beui (Amaral 1924) Micrurus corallinus (Merrem 1820) Oxyrhopus clathratus (Dumeril, Bribon and Dumeril 1854) O. guibei (Hoge and Romano 1976) Philodryas olfersi (Lichtenstein 1823) P. patagoniensis (Girard 1857) Sibynomorphus mikani (Schlegel 1837) Simophis rhinostoma (Schlegel 1837) Tantilla melanocephala (L. 1758) Thamnodynastes strigatus (Günther 1858) Tomodon dorsatus (Dumeril and Bribon 1853) Tropidodryas striaticeps (Cope 1870) Waglerophis merremi (Wagler 1824) Xenodon neuwiedii (Günther 1863)

sna/fos/red/sma/noc (G) inv/fos/smo/sma/noc (D∗ /L∗ ) viv/mam/sit/var/noc (M/S) anu/arb/smo (M/S) viv/mam/sit/var/noc (S/V) anu/smo/sma (M) sna/red (D/M/Ma/S) viv/anu/fis/acq/smo/noc (M∗ ) anu/acq/var/red/sma (L) anu/smo/sma (L) anu/fis/acq/smo/noc (M/S) anu/smo/sma/noc (D/L/V) inv/fos/smo/sma (L∗ ) sna/fos/red (M) mam/red/noc (M) mam/red/noc (S) mam/bir/anu/arb/smo (S/V) mam/sna/anu/smo (S) inv/var/sma/noc (L∗ /M∗ ) anu/red/sma (J) inv/fos/smo/sma/noc (D/Ma) viv/anu/smo/noc (M) viv/inv/smo (M) mam/arb/var (M∗ ) anu/var/smo (V) anu/var (M/S)

of clustering up to 0.4, the greatest group (1 ) included the two pests, the majority of booids and the native viperids, all of which are variegated and nocturnal snakes. In the correspondence analysis (Figure 3), the first (horizontal) axis (eigenvalue = 0.57; variance of species data = 18.6) seems to be inversely correlated to body size: small snakes, such as L. beui and T. melanocephala, are on the right side, opposed to the boas and pythons. In relation to the second (vertical) axis (eigenvalue = 0.48; variance of species data = 15.8), several prey specialists, such as the serpentiformprey-eaters (top) and the invertebrate-eaters (bottom, e.g. the malacophagous T. dorsatus), were located far from the origin, where snakes that prey on mammals, birds and anurans were located. Again, booids appear close to the native viperids and the two pest snakes. It is important to point out that ‘body size’ and ‘prey type’ are not

332

Figure 1. Patterns in arrivals of exotic snakes at IB. (a) Number of exotic snakes (n = 76) found in SPC, Brazil, and sent to IB from January 1995 to April 2000. (b) Collector’s curve of exotic snakes found in SPC and sent to IB from 1995 to 2000. Data on increasing number of individuals were created by Sanders’ Rarefaction Method, i.e. re-sampling with replacement (90 times) from the original data set (n = 76 individuals; S = 16 species).

‘true variables’ in the sense of the analysis described above, as is the case of the categories ‘small sized’, ‘large sized’, ‘mammals’, ‘birds’, ‘lizards’, ‘snakes’, ‘anurans’, ‘fishes’, and ‘invertebrates’. The location of each category in the ordination diagram (Figure 3, left graph) indicates their relative importance to explain the variance of the species data (Ter Braak 1986, 1988).

Discussion No evidence of successful colonization by alien snakes, i.e. settlement of a reproductive population, was obtained until now in SPC. However, the risk of reaching a Minimun Viable Population (as presented in Dodd 1993) was assumed to be in direct correlation to some factors: 1. Number of individuals actually presented in the city. 2. Chance of accessing ‘outdoor environments’, by escaping (accidental) or releasing (intentional). It largely depends on the type of management in captivity (scientific institutions, legal pet shops, illegal traffic, house of aficionados). 3. Individual reproductive condition (sex, age and probability of outdoor mating, ability to store sperm, existence of parthenogenesis). 4. Demographic parameters under phylogenetic constrains (age of maturation, clutch size, number of reproductive events, lifespan). 5. Specific niche and environmental constrains (e.g. abiotic variables, food availability, predation, competition, diseases). It is out of scope of the present paper to make a detailed evaluation of the factors 2, 3 and 4, because no data are available for all the species. A risk assessment was done strictly based on the following statements: (i) Individuals of exotic species were actually present in SPC.

n

2 4d 8 2 1 2 2 1 3

1

anu/var/red/sma (L∗ ) liz/sna/str (L) mam/bir/acq/var/big/noc (Mu) mam/arb/var/big/noc (Mu∗ ) mam/bir/acq/var/big/noc (Mu) viv/anu/fis/inv/acq/str (R)

1 1

mam/arb/smo/big/noc (C) anu/var/red/sma (L)

viv/mam/bir/arb/var/big/noc (D/Ma/V) 19a mam/smo/noc (C) 3 viv/mam/sit/arb/var/big/noc (D/Ma) 5 mam/var (T) 3 mam/arb/var/smo/str (T) 1 viv/mam/bir/arb/var/big/noc (D/Ma/V) 9b mam/bir/liz/sna/var/red (T) 7c mam/bir/liz/sna/var/red (G) 1

Attributes (References)

a Individuals are 5 B.c.amarali, 11 B.c.constrictor, 1 B.c.occidentalis and 2 undetermined subspecies. Only the distribution of B.c.amarali reaches São Paulo state. b 4 E.c.alvarezi, 2 E.c.assizi, 2 E.c.crassus and 1 E.c.hygrophilus. Only the distribution of E.c.crassus reaches São Paulo state. c 5 L.g.californiae, 1 L.g.nigritus and 1 undetermined subspecies. d 3 P.m.bivittatus and 1 undetermined subspecies.

SW Brazil, S Bolivia, C and N Argentina, N Paraguay E Bolivia, SW Brazil to Argentina and Uruguay India, Indo-China, S China, Malayan Islands W and C Africa Africa (S Sahara) S Canada and USA (coast to coast)

Mexico to Argentina, Antilles Africa (S Sahara excluding forests) Amazonian Basin C USA to Mexico E USA to NE Mexico Amazonas to Argentina and Paraguay USA (coast to coast) C and E USA to C America, NW Colombia and Ecuador New Guinea, Australian Islands SE Brazil to Argentina

Boa constrictor (L. 1758) Boaedon fuliginosus (Boie 1827) Corallus caninus (L. 1758) Elaphe guttata (L. 1766) Elaphe obsoleta (Say 1823) Epicrates cenchria (L. 1758) Lampropeltis getulus (L. 1766) Lampropeltis triangulum (Lac´ep`ede 1789)

Liasis albertisi Peters & Doria 1878 Lystrophis dorbignyi (Dumeril, Bribon & Dumeril 1854) Lystrophis semicinctus (Dumeril, Bribon & Dumeril 1854) Philodryas psammophideus Günther 1872 Python molurus (L. 1758) Python regius (Shaw 1802) Python sebae (Gmelin 1789) Thamnophis sirtalis (L. 1766) Elaphe sp. undetermined Lampropeltis sp. undetermined Philodryas sp. undetermined Python sp. undetermined

Origin

Species

Table 3. Exotic snakes found in SPC, Brazil. n denotes the number of individuals of a given species (total n = 76) sent to IB from January 1995 to April 2000. Biological categories used to define the qualitative ecological niche of each species were coded as in Table 1. The categories ‘ovi’, ‘act’, ‘ter’, ‘med’ and ‘diu’ were omitted. References are: (C) Coborn 1991; (D) Duellman 1978; (G) Greene 1997; (L) Leynaud and Bucher 1999; (Ma) Martins and Oliveira 1998; (Mu) Murphy and Henderson 1997; (R) Rossman et al. 1996; (T) Tennant 1984; (V) Vitt and Vangilder 1983. Asterisks indicate inferences based on attributes of congeneric species.

333

334

Figure 2. Dendrogram from clustering analysis (Jaccard similarity index, UPGMA, cofenetic index = 0.826) of snakes found in SPC, based on qualitative biological attributes. Asterisks indicate exotic species. Boiga irregularis and T. flavoviridis (in boxes) were not found in SPC but are problem species in Pacific Islands (see text).

(ii) There were (qualitative) niche similarities between alien and native snakes. (iii) There were (qualitative) niche similarities between these alien snakes and two other species considered as pests elsewhere.

335

Figure 3. Results of the correspondence analysis of snakes found in SPC, based on qualitative biological attributes. Left plot shows score for each of the biological categories as coded in Table 1. Right plot shows score for each of the 26 native species (Table 2) as indicated by a three-character code (the first letter of the genus plus the two first letters of species name; except for Thamnodynastes strigatus (tha) and T. striaticeps (tro) and that for each of the 16 exotic species (Table 3) by a four-character code (the first letter of the genus plus the three first letters of the species name). Boiga irregularis (birr) and T. flavoviridis (tfla) were not found in SPC but are problem species in Pacific Islands (see text). Horizontal (eigenvalue = 0.57) and vertical (eigenvalue = 0.48) axes represent 34.4% variance of species data (total inertia = 3.07).

The sample of exotic snakes sent to IB is an underestimate of ‘who’ and ‘how many’ are released in SPC and, therefore, is an underestimate of alien introduction in Brazil. Note that exotic living individuals in scientific collections were considered here ‘with no risk’ of escaping from captivity or being released (see ‘factor 2’ above), and were not included in this discussion. Using a reasonable extrapolation of the adjusted collector’s curve in Figure 1b, one can estimate 20 accumulated species (77% of the present native richness, an increase of 25% in the ‘exotic richness’) when the number of exotic snakes reaches 182 individuals. This scenario of four new exotic species arriving at SPC, besides those 16 reported here, could be expected for the year 2008 if both (1) the present average arrival of 14.3 individuals/year and (2) the dominance pattern of ‘captured alien community’ are fixed. At least the first assumption was not supported by the annual capture plot in Figure 1a: no significant tendency was evident (y = 2.47 + 3.20x; R 2 = 0.31; P > 0.05). Moreover, the second assumption is linked to oscillations of international trade and to the chance of importing new species, legally or not. Therefore, arrivals of exotic snakes at any place should be considered primarily a stochastic, unpredictable event.

336 The different conditions under which each individual was captured restrict the direct correlation between the number of species effectives and the chance of colonization. However, it constitutes a reasonable (and available) first approach. Out of the exotic snakes analyzed, booids composed the majority of the sample. These euriecians snakes have an extreme ability of adaptation in many ecotopes (as in MartínezMorales and Cuarón 1999). The same could possibly be true of Lampropeltis and Elaphe species, which show broad geographical distributions (Coborn 1991). There is an example of an established population of B. constrictor in Cozumel Island (Martínez-Morales and Cuarón 1999), the first snake in the SPCs exotic abundance rank (19 individuals, see Table 3). Besides ecological factors, phylogeny also contributes to natural history patterns (as in Martins and Oliveira 1998), resulting in niche similarities between closely related species. Therefore, once Elaphe taeniura was introduced to Ryukyu Archipelago (Ota 1999), the presence of congeneric species in SPC also needs attention. The performed multivariate exploratory analysis shows ‘where’, in a multidimensional space among the native snakes, the exotic snakes have their potential niche. As an example, the booids would be found ‘somewhere’ near the viperids but far from L. beui and T. melanocephala, the invertebrate-eaters, fossorial, small-sized snakes. No assumptions on competition or predation involving exotic and native species were done. The similarities found by these analysis only indicate a chance that alien species would find in SPC some conditions that correspond to their needs. Additions of B. irregularis and T. flavoviridis to the analyses allowed a comparison with effective colonizers. Therefore, niche similarities between these two snakes and the booids indicate that boas and pythons may be as successful as the known pest snakes, probably using the same type of resources available to the native pitvipers. The introduction of B. irregularis to Guam is largely reputed as a factor that contributed to depress the local avifauna (Savidge 1987). On the other hand, T. flavoviridis is native from Japanese Islands (despite the possible human introduction to Minnajima) (Katsuren et al. 1999), reaching high population densities but with no ecological damage attributed to them. However, due to the risk of bites, substantial efforts to reduce those populations have been made. This measure itself (removal of a top predator) may affect the food chain. Quantitative research on the intensity of resource exploitation and environmental requirements by both native and exotic species (not only snakes) are needed to allow more robust comparisons between habitat use (Vitt 1987; Reinert 1993), as well as between niches. While this knowledge is not a current device, a little bit of xenophobia is a salutary prophylactic way to prevent drastic changes in local communities by introduced fauna (but see Smith and Kohler 1978 for alternative views). Importing, keeping and commercializing dangerous or even poisonous snakes lead to a new public health problem (Reid 1978; Minton 1996). Some opistogliphodont and agliphodont snakes also present toxic secretions from Duvernoy’s glands (e.g. Philodryas, Assakura et al. 1992). No antiserum for alien species is available for the

337 general public in SPC. Great constrictors, 67% of the present findings, can also cause human death (Branch and Hacke 1980; Chiszar et al. 1993). As pointed out by Fritts and Rodda (1999), some biological features of snakes require management strategies distinct from those applied to other vertebrates. They are well-hidden predators, with the absence of legs and elongation of body contributing both to locomotion and to crypsis. They forage infrequently, and are able to store sperm for a long period. Because of this, snakes are generally difficult to capture, and their population sizes are difficult to assess, making it difficult to eradicate them, once they are established as a result of accidental (or intentional) transportation. Management programs of introduced snakes may employ many public resources (Rodda et al. 1999b). As many researchers commonly suggest eradication of established populations of exotic snakes (Rodda and Fritts 1992; Katsuren et al. 1999; Martínez-Morales and Cuarón 1999), the exotic trade prohibition in Brazil has to be sustained. Improved control and punishment may restrict illegal traffic only with parallel implementation of environmental education programs. Therefore, preventive measures should be preferred to corrective ones. Acknowledgements Thanks to our colleagues in IB Francisco L Franco, Hebert Ferrarezzi, and Otavio AV Marques for taxonomic and biological information; Amauri F da Silva, Circe C de Albuquerque and Giuseppe Puorto for making records on exotic species under their responsibility available to us. OAVM, Cristiano C Nogueira, Leila L Longo, and Paula H Valdujo (IBUSP) read the drafts. Pérsio S Santos-Filho (IBUSP) made important suggestions. References Adams CE, Thomas JK, Strnadel KJ and Jester SL (1994) Texas rattlesnake roundups: implications of unregulated commercial use of wildlife. Wildlife Society Bulletin 22: 324–330 Assakura MT, Salomão MG, Puorto G and Mandelbaum FR (1992) Hemorragic, fibrinogenolytic and edema-forming activities of the venom of the colubrid snake Philodryas olfersii (green snake). Toxicon 30: 427–438 Begon M, Harper JL and Townsend CR (1996) Ecology, 3rd Edition. Blackwell Science, Oxford ter Braak CJF (1986) Canonical correspondence analysis: a new eigenvector technique for multivariate direct gradient analysis. Ecology 67: 1167–79 ter Braak CJF (1988) CANOCO – a FORTRAN Program for Canonical Community Ordination by (partial) (detrended) (canonical) Correspondence Analysis, Principal Components Analysis and Redundancy Analysis (version 2.1). Groep Landbouwwiskunde, Wageningen Branch WR and Hacke D (1980) A fattal attack on a young boy by an african rock python, Python sebae. Journal of Herpetology 14: 305–307 Chiszar D, Smith HM, Petkus A and Dougherty J (1993) A fattal attack on a teenage boy by a captive Burmese python (Python molurus bivitattus) in Colorado. Bulletin of the Chicago Herpetological Society 28: 261–262 Coborn J (1991) The Atlas of Snakes of the World. TFH Publications, Neptune City

338 Dodd Jr CK (1987) Status, conservation and management. In: Seigel RA, Collins JT and Novak SS (eds) Snakes: Ecology and Evolutionary Biology. Macmillan Publishing Company, New York Dodd Jr CK (1993) Strategies for snake conservation. In: Seigel RA and Collins JT (eds) Snakes: Ecology and Behavior, pp 363–393. McGraw Hill, New York Duellman WE (1978) The biology of an equatorial herpetofauna in Amazonian Ecuador. The University of Kansas Museum of Natural Hystory Miscellaneous Publication 65: 1–352 Duellman WE (1999) Perils of permits: procedures and pitfalls. Herpetological Review 30: 12–16 Fritts TH and Rodda GH (1999) Contributions of brown treesnakes and habu to science and society. In: Rodda GH, Sawai Y, Chiszar D and Tanaka H (eds) Problem Snake Management, (Epilogue) pp 481– 485. Comstock Publishing Associates, Ithaca Gauch HG (1982) Multivariate Analysis in Community Ecology. Cambridge University Press, New York Greene HW (1997) Snakes: The Evolution of Mistery in Nature. University of California Press, Los Angeles Hurlbert SH (1971) The nonconcept of species diversity: a critique and alternative parameters. Ecology 52: 577–586 Jordão RS and Bizerra AF (1996) Reprodução, dimorfismo sexual e atividade de Simophis rhinostoma (Serpentes, Colubridae). Revista Brasileira de Biologia 56: 507–512 Katsuren S, Yoshida C and Nishimura M (1999) A ten-year trapping program to eradicate Habu (Trimeresurus flavoviridis) from Minnajima, a small island in the Okinawa Islands, Japan. In: Rodda GH, Sawai Y, Chiszar D and Tanaka H (eds) Problem Snake Management, pp 340–347. Comstock Publishing Associates, Ithaca Leynaud GC and Bucher EH (1999) La fauna de serpientes del chaco sudamericano: diversidad, distribuicion geografica y estado de conservacion. Academia Nacional de Ciencias de Cordoba, Argentina, Miscelanea 98: 3–46 Ludwig JA and Reynolds JF (1988) Statistical Ecology: A Primer on Methods and Computing. John Wiley and Sons, New York Magurran A (1988) Ecological Diversity and its Measurement. Princeton University Press, New Jersey Marques OAVM (1998) Composição faunística, história natural e ecologia de serpentes da Mata Atlântica, na região da Estação Ecológica Juréia-Itatins, São Paulo, SP. PhD thesis, Universidade de São Paulo Martínez-Morales MA and Cuarón AD (1999) Boa constrictor, an introduced predator threatening the endemic fauna on Cozumel Island, Mexico. Biodiversity and Conservation 8: 957–963 Martins M and Oliveira ME (1998) Natural history of snakes in forests of the Manaus region, Central Amazonia, Brazil. Herpetological Natural History 6: 78–150 Minton SA (1996) Bites by non-native venomous snakes in the United States. Wilderness and Environmental Medicine 4: 297–303 Mishima S, Tanaka H and Saway Y (1999) The biology of the Habu (Trimeresurus flavoviridis). In: Rodda GH, Sawai Y, Chiszar D and Tanaka H (eds) Problem Snake Management, pp 29–43. Comstock Publishing Associates, Ithaca Murphy JC and Henderson RW (1997) Tales of Giant Snakes: A Historical Natural History of Anacondas and Pythons. Krieger Publishing Company, Malabar Ota H (1999) Introduced amphibians and reptiles of the Ryukyu Archipelago, Japan. In: Rodda GH, Sawai Y, Chiszar D and Tanaka H (eds) Problem Snake Management, pp 439–452. Comstock Publishing Associates, Ithaca Pough FH, Andrews RM, Cadle JE, Crump ML, Savitsky AH and Wells KD (1998) Herpetology. PrenticeHall, New Jersey Puorto G, Laporta-Ferreira IL and Sazima I (1991) Serpentes na selva de pedra. Ciência Hoje 13: 66–67 Reid HA (1978) Bites by foreign venomous snakes in Britain. British Medical Journal 1: 1598–1600 Reinert HK (1993) Habitat selection in snakes. In: Seigel RA and Collins JT (eds) Snakes: Ecology and Behavior, pp 201–240. McGraw Hill, New York Reinert HK and Ruppert Jr RR (1999) Impacts of translocation on behavior and survival of Timber rattlesnakes, Crotalus horridus. Journal of Herpetology 33: 45–61 Ricklefs RE (1996) Ecology, 3rd Edition. WH Freeman and Company, New York

339 Rodda GH and Fritts TH (1992) The impact of introduction of the colubrid snake Boiga irregularis on Guam’s lizards. Journal of Herpetology 26: 166–174 Rodda GH, Fritts TH and Chiszar D (1997) The disappearance of Guam’s wildlife. BioScience 47: 565– 574 Rodda GH, Fritts TH, McCoid MJ and Campbell III EW (1999a) An overview of the biology of the brown treesnake (Boiga irregularis), a costly introduced pest on Pacific Islands. In: Rodda GH, Sawai Y, Chiszar D and Tanaka H (eds) Problem Snake Management, pp 44–80. Comstock Publishing Associates, Ithaca Rodda GH, Sawai Y, Chiszar D and Tanaka H (1999b) Snake management. In: Rodda GH, Sawai Y, Chiszar D and Tanaka H (eds) Problem Snake Management, pp 1–23. Comstock Publishing Associates, Ithaca Rossman DA, Ford NB and Seigel RA (1996) The Garter Snakes: Evolution and Ecology. University of Oklahoma, Norman Savidge JA (1987) Extinction of an island forest avifauna by an introduced snake. Ecology 68: 660–668 Sazima I and Haddad CFB (1992) Répteis da Serra do Japi: notas sobre história natural. In: Morellato LPC (org) História natural da Serra do Japi. Unicamp/Fapesp, Campinas Smith HM and Kohler AJ (1978) A survey of herpetological introductions in the United States and Canada. Transactions of the Kansas Academy of Science 80: 1–24 Tennant A (1984) The Snakes of Texas. Texas Monthly Press, Austin Vitt LJ (1987) Communities. In: Seigel RA, Collins JT and Novak SS (eds) Snakes: Ecology and Evolutionary Biology. Macmillan Publishing Company, New York Vitt LJ and Vangilder LD (1983) Ecology of a snake community in Northeastern Brazil. Amphibia-Reptilia 4: 273–296