Composition and Natural History of Snakes from Etá Farm region, Sete Barras, south-eastern Brazil

Abstract Approximately 140 snake species are known to occur in the Atlantic Forest with nearly half being endemic to this ecoregion. However, the Atlantic forest is one of the most threatened tropical ecoregions, with only 16% of its original area remaining as forest. This extensive habitat loss must have had a negative effect on its snake fauna. Indeed, 53% of the threatened snakes of Brazil occur in the Atlantic forest. Therefore, basic natural history information that can potentially contribute to the conservation of Atlantic forest snakes are urgently needed. Here the natural history of a snake assemblage at Etá Farm region, Sete Barras municipality, south-eastern Brazil is described, and a visual guide and an identification key provided that can be used by researchers and local people to identify snakes from this region. Most of the species found in the field use both open areas and forests, are primarily terrestrial, present diurnal activity, and include frogs in their diet. A higher number of enlarged follicles, eggs, and/or embryos were recorded during the warm and rainy season. Seventeen different types of defensive tactics were recorded in the species found in the field. This study provides useful information for understanding the structure of snake assemblages of the Atlantic Forest and is potentially useful for conservation assessments and for designing conservation plans.


Introduction
Natural history information, what organisms do in their respective environments, including interactions between them (Greene 1994), contributes beyond the basic refinement of science (Greene and McDiarmid 2005) but also to our understanding of how environments function and, consequently, aids in many aspects of conservation, management, and appreciation of nature (Caughley 1994, Brooks and McLennan 2002, Dayton 2003. Despite their obvious relevance, there is still a considerable knowledge gap on the ecology and behaviour of most snake species, even in well-studied regions in the world (Greene 2005). For example, Bothrops jararacussu is one of the most widespread species in the Atlantic forest, yet most of its natural history data come from only a few localities of southeastern Brazil (Marques 1998, Martins et al. 2002, Hartmann et al. 2009b). Valuable natural history information is available for only a small fraction of animal species, usually those that are large or common and relatively easy to study (Greene 1994). Neotropical snakes are no exception and despite the studies published on these animals (Strüssmann and Sazima 1993, Marques 1998, Martins and Oliveira 1998, Cechin 1999, Di-Bernardo 1999, Sawaya et al. 2008, Hartmann et al. 2009ab, Pontes et al. 2009, Gaiarsa et al. 2013, Mesquita et al. 2013, Guedes et al. 2014), many species are still only known from small portions of their distributions.
The Atlantic forest of eastern Brazil harbours a very rich snake fauna, with approximately 140 species, representing 34% of the 412 species of snakes known to occur in Brazil (Costa and Bérnils 2018;Marques et al. 2019;Nogueira et al. 2019). Furthermore, almost half (45%) of the Atlantic forest snakes are endemic to these forests. However, the Atlantic forest is one of the most threatened tropical ecoregions (Myers et al. 2000), with only 16% of its original area remaining as forest (Ribeiro et al. 2009). The extensive habitat loss to which the Atlantic forest was subject in the last decades have likely had a negative effect on its snake fauna. Indeed, 53% of the threatened snakes of Brazil occur in the Atlantic forest and there is a lack of baseline data for an additional ten snake species, from this ecoregion, making the assessment of their conservation status difficult (ICMBio, 2018). Therefore, basic natural history information that can potentially contribute to the conservation of Atlantic forest snakes are urgently needed.
Here we provide basic natural history information for an Atlantic Forest snake assemblage from south-eastern Brazil. We sampled pristine along with disturbed habitats, thus assessing the ability of the Atlantic forest snakes to persist in disturbed habitats. For each species we provide primary information on habitat and micro-habitat use, time of activity, feeding habits, reproduction and defence. We also provide a short review of the natural history of each species based on our results and on previously published accounts.

Materials and methods
The primary information used in this study was obtained between April 2013 and March 2014 at the region of Etá Farm (24°19'13"S, 48°7'3"W) in the Sete Barras municipality, São Paulo state, south-eastern Brazil. The area is located within the Atlantic forest in a hillside forest formation (Joly et al. 1992). While this region shows great variation in elevation, ranging from 45 m at Etá Farm to over 800 m at the Sete Barras Operational Centre of Carlos Botelho State Park (Forlani et al. 2010), sampling for this study was carried out within the 45-80 m range (Figure 1). We searched the literature and museum databases for additional species that might occur in the Sete Barras municipality and neighbouring areas and that we could have failed to find during our fieldwork.
Phytosociological and floristic studies have shown high richness of tree and shrub species in this type of Atlantic forest formation (Mantovani 1993). In addition to forests in an advanced stage of regeneration, areas modified by agriculture (considered "open areas"), such as peach palm (Bactris gasipaes) and banana (Musa sp.) plantations, were also sampled ( Figure 2). Field data were collected by two researchers for 14 days per month, from April 2013 to March 2014, for a total sample time of 168 days. Snakes were sampled with pitfall traps with drift fences (Greenberg et al. 1994, Cechin andMartins 2000), time constrained searches (sensu Campbell and Christman 1982, Scott et al. 1989, Martins and Oliveira 1998, and accidental encounters (Martins and Oliveira 1998), the latter including snake encounters by local people (Martins and Nogueira 2012). Three main vegetation types were sampled with pitfall traps with drift fence: banana plantation  ( Figure 2B; site 7), peach palm plantation ( Figure 2B; sites 5 and 6), and forest (Figure 2B; sites 2 and 3). Our sampling design for pitfall traps included two sampling units per vegetation type, each sampling unit comprising three Y sets (with 12 m-long branches), located 100 m from each other. Thus, we installed a total of six sampling units with a total of 18 Y sets and 72 buckets. Sampling units were located at least 500 m from each other. Each Y set had four 100 L plastic buckets (three at each branch end and one in the centre) connected by a 60 cm-high plastic fence. The buckets were perforated at the bottom to avoid accumulation of rainwater. Two additional sites (Figure 2B;sites 4 and 8) were sampled occasionally or through time constrained searches.
Habitat use was recorded through active search for individuals (made only at night), describing the habitats (e. g., open area, forest, banana plantation) and microhabitats used by each snake (fossorial, aquatic, terrestrial or arboreal) and perch height (in case of arboreal species). To characterise micro-habitats we used only information obtained during active searches; for individuals collected with pitfall traps, only vegetation cover (forest, peach palm or banana plantation) was considered.
To describe diet, collected specimens were dissected through an incision in the ventral region. Food items were identified to the lowest possible taxonomic rank using taxonomic keys, identification guides, specimens deposited in scientific collections and help from experts. Whenever the prey came from a snake captured in a pitfall trap, this information was included, given the possibility of the snake having ingested prey that had also fallen in the trap (Cechin and Martins 2000) but which is not part of the snake's usual diet. Additional specimens of the studied species from the herpetological collection of the Butantan Institute were also dissected.
To describe reproductive condition, we recorded the length of the largest follicle, egg or embryo, and number of vitellogenic follicles (> 10 mm), eggs or embryos in every month of collection. Specimens collected in the field in the study area and specimens from the herpetological collection of the Butantan Institute were dissected for this purpose.
Behavioural descriptions are based on observations made over short periods of time (ad libitum and sequence samplings; Altmann 1974). Defensive behaviours were recorded when individuals were observed in the field and when handled.

Results
With a sampling effort of 168 days of fieldwork, including 558 person-hours of visual search, we found 255 individuals of 17 species of snakes (14 genera, four families) at the Etá Farm region. Additionally, we included Corallus cropanii to our study because it was found previously by other researchers in our study area (Machado Filho et al. 2011). Species richness was similar between forests (13 species) and disturbed areas (banana plantation, peach palm plantation, roads, pastures, and around houses; 16 species; Table 1).

General natural history patterns
Among the species we found in the field, most used forested areas (> 70% of species), were primarily terrestrial (70%), showed diurnal activity (> 58%), and included frogs in their diet (> 50%; information supplemented with data from the literature). Only those which consumed endothermic prey and Dipsas neuwiedi showed nocturnal ac-tivity. However, there was a relatively high percentage (30%) of semi-arboreal species, observed almost exclusively in open areas or forest edges, all anuran specialists (except for Spilotes pullatus) and belonging to the family Colubridae.
In addition to species that were semi-arboreal and anuran specialists, the molluscspecialist species D. neuwiedi and T. dorsatus were also found exclusively in open areas. Only T. bilineatus proved to be the most limited to the forest habitat (N = 2), particularly to a cryptozoic micro-habitat (see Habitat Use of Taeniophallus bilineatus under Natural history accounts). The species E. miliaris and B. jararacussu showed the broadest spectrum of the assemblage in terms of resource use, as both widely used open and forested areas and included 3 and 4 different types of prey in their diet, respectively.
The viperids B. jararaca and B. jararacussu and the dipsadids D. neuwiedii and O. clathratus were the species most commonly found in disturbed areas such as plantations, around houses and even inside houses. Particularly for O. clathratus, 75% of observations occurred in disturbed habitats; the remaining took place in the forest.
Among the snake specimens whose stomach contents were analysed, 82 individuals, belonging to eleven different species, had prey in their digestive tracts. Of those, most have ingested frogs (61%), followed by mammals (18.3%), molluscs (11%), snakes (3.7%), fishes (2.4%), non-mollusc invertebrates (leech and centipede) (2.4%), and lizards (1.2%) ( Table 2). The assemblage showed a great number of species whose bulk of the diet consisted of, or included, frogs, especially leptodactylids (more than 70% of anuran records), the group that was most captured in pitfall traps in the region (see Fiorillo et al. 2018). The second item most found was small mammals (more than 18%), of which rodents (more than 80% of the mammals), particularly from the tribe Akodontini (> 40% of the rodents) were the most frequent. Almost all small mammals recorded were consumed by vipers; only two cases were reported for species from other families (S. pullatus and O. clathratus).
Although, in qualitative terms, E. miliaris and B. jararacussu showed a greater diversity of items in their diets, they may be considered specialists in frogs (> 95% of the diet of E. miliaris) and small mammals (> 58% of the diet of B. jararacussu), respectively. However, most of the records obtained for E. miliaris came from individuals caught in pitfall traps (> 87% of cases). Hence, part of the frogs may have been opportunistically consumed by this species (the finding of 14 specimens of E. miliaris captured in pitfall traps, that ingested P. spiniger, supports this assumption). Only four individuals of this species were captured by other capture methods, one of which had consumed a fish.
Of the specimens examined, 28 were reproductive females containing vitellogenic follicles, eggs or embryos. In the three families, as with activity, larger follicles were found during the Austral Spring, with the largest vitellogenic follicles, as well as eggs and embryos, occurring from September to October, except for one specimen of Erythrolamprus aesculapii that had vitellogenic follicles during the month of July (Austral Winter). Over the sampling period, only one single mating behaviour was observed for S. pullatus, in September (see Natural history accounts).
A total of 17 different defensive tactics was recorded for the assemblage studied, with some variations and combinations of them (Table 3). Most seem to be aimed at visually oriented predators (being "flatten body" most frequent among them, used by 70% of the species), but cloacal discharge had the same frequency (used by 70% of the species in the region, considering field observations and data from the literature). The numbers indicate field observations and an "X" indicates data from the literature (Hoge 1953, Marques and Sazima 2004, Menezes et al. 2015. Another defence shown by many species was cryptic colouration (82%), which was common in diurnal species (58% of the assemblage). In contrast, only three species showed aposematic or mimetic colouration (M. corallinus and their supposed mimics E. aesculapii and O. clathratus).

Boidae Gray, 1825
Corallus cropanii (Hoge, 1953) This large species (maximum SVL = 1510 mm; Marques et al. 2019) is rare in the Etá Farm region. Although not sampled during our study, an individual of C. cropanii was found by Machado Filho et al. (2011) in the agricultural settlement north of the Etá Farm (Fig. 2). The holotype was found on vegetation at 1.5 m above the ground; in captivity, it remained perched on branches for most of the time (Marques and Cavalheiro 1998). The only known food content is the land opossum, Metachirus nudicaudatus (Didelphimorphia, Didelphidae; Marques and Cavalheiro 1998). This is a viviparous species ), but there is no detailed information available on its reproduction. It may raise its head while opening its mouth (Hoge 1953).

Chironius exoletus (Linnaeus, 1758)
This is an aglyphous species of medium size (maximum SVL = 790 mm; N = 1). Only one individual was found, in the peach palm plantation, on the vegetation at 0.5 m above the ground, during the day, in November. The available information indicates that it is semi-arboreal and diurnal (França andAraújo 2006, Martins et al. 2008, this study), but forage mainly on the forest ground where it feeds primarily on hylid frogs (Sazima 1992, Dixon et al. 1993, França and Araújo 2006, Rodrigues 2008Marques and Sazima 2004;. Its fecundity ranges from 4-12 eggs (Dixon et al. 1993, Bernarde andAbe 2006). When handled, the individual found opened its mouth, struck, and raised its head and formed an S-coil with the anterior part of the body (Table 3). It may also perform gular inflation, lateral fattening and cloacal discharge .

Chironius fuscus Linnaeus, 1758
This is an aglyphous species of medium size (maximum SVL = 919 mm; N = 8). It was found crossing an unpaved road, always near the forest edge (N = 6), and on a trail in the forest near the abandoned banana plantation, lying coiled up on a tree at 1.5 m above the ground (N = 1). One juvenile was found while crossing a paved road close to the urban area of Sete Barras, during the day. The available information indicates that it is diurnal and semi-arboreal, but forage mainly on the ground of the forests (Martins and Oliveira 1998, Marques and Sazima 2004 where it feeds primarily on leptodactylid frogs (Strüssmann and Sazima 1993, Martins and Oliveira 1998, Marques and Sazima 2004. At Etá farm, it was found during most of the year, with a higher incidence of juveniles from April to June. One female had six vitellogenic follicles in May. In the Amazon region, females were collected with 3-8 eggs in the oviducts, from March-July and in October (Dixon and Soini 1986, Dixon et al. 1993, Martins and Oliveira 1998. The handled individuals struck (N = 2; Table 3). This snake also opens its mouth, raises its head, and forms an S-coil with the anterior part of its body, flattening and inflating the gular region in frontal display, performs cloacal discharge, rotates its body and vibrates its tail (Beebe 1946, Dixon and Soini 1986, Marques and Sazima 2004.

Chironius laevicollis (Wied, 1824)
This is an aglyphous species of large size (maximum SVL = 1650 mm; N = 4). One individual was found on the ground, during the day, moving through the peach palm plantation; another was foraging on the forest floor at the margins of the Etá River during the day. Before being captured, the individual found on the trail quickly climbed a tree to a height of approximately 3 m. One individual was observed in a shallow pond at the edge of the forest, "yawning" as if it had just ingested something, shortly before swimming away, also during the day. The available information indicates that it is diurnal and terrestrial, being arboreal during the juvenile stage (Dixon et al. 1993, this study). In the digestive tracts of the examined specimens we found frogs of the species Leptodactylus latrans (Table 2), one of which had been swallowed headfirst. Previous studies also indicate that the species feeds on frogs (Dixon et al. 1993, Marques 1998. At the Etá farm, individuals were found in January, May, June and December; one female had 38 primary follicles in May and another had ten secondary follicles in January. Previous observations indicate that it has a fecundity of 10-14 eggs and has a seasonal reproductive cycle (vitellogenic follicles in August-December and oviductal eggs in October-November; Costa et al. 2005). It may open its mouth, perform a frontal display by raising its head, forming an S-coil with the anterior part of its body, and inflating the gular region, strike, perform cloacal discharge and vibrate the tail ; Table 3).

Spilotes pullatus (Linnaeus, 1758)
This is an aglyphous species of large size (maximum SVL = 1830 mm; N = 15). It was found in a pile of wood next to a house in an operating banana plantation (N = 4), in an abandoned banana plantation (N = 1), in peach palm plantations (N = 2), on the edge of the forest (N = 1), in a pasture (N = 1) and crossing a paved road in areas surrounded by forest and pasture (N = 2). Three run-over individuals were collected on an unpaved road near the edge of forest. The available information indicates that it is semi-arboreal and diurnal (Vanzolini et al. 1980, Marques and Sazima 2004, Bernarde and Abe 2006. We found four rodent nestlings, all ingested headfirst, in the stomach of a specimen (Table 2). Previous studies with specimens from the Atlantic forest revealed that this snake feeds mainly on very small mammals and nestling birds . At Etá farm, one mating pair was found in September; two females had eight and 11 eggs in their oviducts, in September. Most adult individuals were observed in August and September (N = 7), while juveniles were observed in February (N = 3). Marques et al. (2014) reported oviposition for various localities in the Atlantic forest, restricted to the onset of the rainy season (October-November) and Bernarde and Abe (2006) observed juveniles in the same period in southwestern Amazon. Previous reports indicate that it lays 5-12 eggs (Amaral 1930. The temperament of the individuals found varied greatly, from very docile to extremely aggressive. When handled, the individuals found opened the mouth (N = 1), vibrated the tail (N = 4), inflated the gular region (N = 10) and struck (N = 4) ( Table 3). Previous studies indicate that it may also perform cloacal discharge and frontal displays by raising the head and forming an S-coil with the anterior part of its body .

Dipsas neuwiedi (Ihering, 1911)
This is an aglyphous species of medium size (maximum SVL = 643 mm; N = 31). It was found mostly in the peach palm plantation (N = 15), with only one individual captured in a pitfall trap. It was found also around the houses (N = 5) and crossing the unpaved road at night (N = 5). One individual was found resting under the lid of one of the pitfall trap buckets during the day. We have no information on habitat use for the other five individuals. The available information indicates that it is nocturnal and semi-arboreal (Freitas 1999, Oliveira 2001. Of the 31 specimens examined, ten had molluscs in their digestive tract ( Table 2), most of them of the genus Phyllocaulis (N = 4), endemic to South America. In two cases the snakes had eaten Limax cf. flavus snails, an exotic European species, while in another, a Meghimatium pictum slug, an exotic Chinese species, both considered invasive. The available information indicates that it is a mollusc specialist (Freitas 1999, Oliveira 2001, Marques and Sazima 2004. It was found throughout the year and one female had four oviductal eggs in March. The available information indicates that it has a seasonal reproductive cycle, with vitellogenesis occurring between July and December, and oviposition between August and February (Pizzatto et al. 2008). Barbo et al. (2011) reported a female with nine eggs. When handled, the individuals found at the Etá Farm performed cloacal discharge (N = 4), hid the head within the coiled body (N = 1), and triangulated the head (N = 19). Information available in the literature indicates that, besides being a supposed mimic of vipers such as B. jararaca, it may also raise the head and form an S-coil with the anterior part of the body   (Table 3).

Echinanthera cephalostriata Di-Bernardo, 1996
This is an aglyphous species of small size (maximum SVL = 399 mm; N = 3). Individuals were found crossing the road near forest and pasture areas during the day (N = 2), and moving in the leaf litter in the forest at night (N = 1). In a study at Serra do Medanha, Rio de Janeiro (Pontes et al. 2008), the species was found exclusively in forests. The available information indicates that it is diurnal, terrestrial, and cryptozoic , Barbo et. al. 2011, Hartmann et al. 2011, and feeds on frogs , Marques et al. 2009, Forlani et al. 2010. At the Etá farm, it was found during the months of July, November and December. One female collected in the municipality of Iguape (50 km from the Etá Farm) had eight vitellogenic follicles in November (data provided by staff from the Butantan Institute). When handled, one of the individuals found raised the anterior part of the body (Table 3). It may also flatten its body and perform cloacal discharge .

Erythrolamprus aesculapii (Linnaeus, 1766)
This is an opistoglyphous species of medium size (maximum SVL = 827 mm; N = 5). Four individuals were found on an unpaved road: two moving near a pasture, one near an operating banana plantation and one crossing an unpaved road near the forest edge, all during the day. The last individual was found inside the forest. The available information indicates that it is terrestrial, cryptozoic and primarily diurnal (Bernarde  and Abe 2006, França andAraujo 2006, Martins et al. 2008, this study). In the digestive tract of the examined specimens we found an individual of D. neuwiedi, swallowed headfirst in one specimen, and snake scales in another one (Table 2). It feeds primarily on other snakes, but juveniles also feed on very small lizards (Martins and Oliveira 1998, Bernarde and Abe 2006, França and Araujo 2006. At the Etá Farm, a female had two vitellogenic follicles and two juveniles were found in June. The available information indicates that it has a fecundity of 3-8 eggs and a continuous reproductive cycle, with vitellogenesis occurring throughout the year (Marques 1996a, Marques andSazima 2004). This snake is a supposed mimic of M. corallinus. Besides the similarity of the colour pattern, it flattens the body, hides the head, performs erratic movements, and raises the curled tail as in M. corallinus.

Erythrolamprus miliaris Linnaeus, 1758
This is an aglyphous species of medium size (maximum SVL = 1000 mm; N = 74). It was found in all sampled vegetation types and captured, in most cases, in pitfall traps (17 in the abandoned banana plantation, 22 in forest and 23 in the peach palm plantation); it was also found in other disturbed areas (N = 13). When found in the traps, all individuals submerged into the water accumulated in the buckets. Most individuals were seen in the abandoned banana plantation, moving or resting in the undergrowth (N = 4), or foraging at the edge of flooded vegetation (N = 1); in the forest, moving on the ground or on the leaf litter (N = 2); and, in the peach palm plantation, in water puddles (N = 2) or undergrowth (N = 1). In visual searches and occasional encounters, individuals were also found at the edge of the forest (N = 1), in pasture areas (N = 3), in a swamp (N = 1), and around houses (N = 2). Some individuals were captured on the unpaved road, always near flooded areas (N = 3) and one adult individual was found in a puddle in an open area near the forest edge. Two juveniles were captured in very disturbed habitats, one in the sink in a house at 12:00 h and another in an operating banana plantation, moving over rocks at 15:30 h. Additionally, four individuals were found on the unpaved road, one crossing it near a swamp during the day, and another run over by a car near the peach palm plantation at 06:00 h. Observations of individuals moving were always in daytime. The available information indicates that it is semiaquatic and both diurnal and nocturnal (Sazima and Manzini 1995, Yanosky et al. 1996, Torello and Marques 2017. More than 90% (N = 38) of specimens whose digestive tract contents were analysed contained frogs (Table  2). One snake captured in a small homemade water tank contained a specimen of Synbranchus marmoratus in its digestive tract, swallowed by the tail, while another snake, captured in a pitfall trap, had a specimen of Placosoma glabellum, also ingested by the tail ( Table 2). The available information indicates that it feeds mainly on amphibians, but also on fish, tadpoles, amphisbaenians and lizards (Achaval and Olmos 1997, Carreira 2002, Marques and Sazima 2004, Toledo et al. 2007. At the Etá Farm it was found throughout the year, with a higher incidence in hot and rainy months. Two females showed 5-12 vitellogenic follicles from April to October, and one female had 5 oviductal eggs in October. Juveniles (< 400 mm; N = 36) were mainly observed during the rainy season. Pizzatto and Marques (2006) reported different types of reproductive cycle for this species in different regions of the Atlantic Forest: a continuous reproductive cycle on the coast at the northern parts of its distribution (southern Bahia state), and a seasonal one in the southern regions, both in the interior and in coastal regions (São Paulo and Paraná states), with vitellogenesis and oviposition from September to February, and births at the end of the rainy season. It has a fecundity of 5-17 eggs and reaches sexual maturity at 12 months of age, at the earliest (Vitt 1992, Achaval and Olmos 1997. When handled, the individuals found at the Etá Farm flattened (N = 2) or rotated (N = 5) the body, raised the head while flattening the body (N = 2; see Menezes et al. 2015), performed cloacal discharge (N = 20), and triangulated the head (N = 3), as previously reported in the literature   (Table 3).

Helicops carinicaudus (Wied & Neuwied, 1825)
This is an aglyphous species of medium size (maximum SVL = 623 mm; N = 12). Thirteen individuals were found in the field; of those, three in the peach palm plantation, with one individual captured in a pitfall trap, and the other three in a stream at the edge of forest, in the water, all apparently active in late afternoon. Three individuals were captured on the unpaved road, all moving during the day. The other captured individuals (N = 7) moved across floodplains on the edge of the forest, also in late afternoon. Literature records of activity were made during both day and night Sazima 2004, Hartmann et al. 2009b). Of the specimens examined, three had stomach contents: two of them had frogs and the other had a fish (Characidium sp.; Table 2). The available information indicates that the diet of this species consists mainly of fish, although it may also capture frogs (mainly leptodactylids; Albolea 1998, Marques andSazima 2004, this study). Individuals of the species were found mainly between August and November. One female had one vitellogenic follicle in August, while two others had 11 and 13 in October and November, respectively. Two juveniles (< 350 mm) were found in March and August. The available information indicates that it has a seasonal reproduction, with vitellogenesis occurring from September to December, embryos from November to March, and juvenile recruitment between February April in Atlantic forest regions (Marques 1998, Nogueira andMarques 1998, this study). Fecundity varies between 7 to 26 embryos (Nogueira and Marques 1998). When handled, individuals found at the Etá Farm struck (4) and bit (1). Previous observations indicate that it may also flatten the body, hide the head, and perform cloacal discharge   (Table 3).

Oxyrhopus clathratus Duméril, Bibron & Duméril, 1854
This is an opisthoglyphous species of medium size (maximum SVL = 710 mm; N = 12). It was found mainly around houses trying to climb walls at dusk (N = 6), run over on the unpaved road next to the forest edge (N = 2), and moving on the ground in the forest at night (N = 1). One adult male was caught while crossing the unpaved road near the edge of forest at 05:30 h. Three individuals were captured in other disturbed habitats. The available information indicates that it is terrestrial and both diurnal and nocturnal (Hartmann and Giasson 2008, Barbo et al. 2011this study). One individual regurgitated a marsupial (Monodelphis americana) (Table 2), inside one of the pitfall traps. The available information indicates that it feeds mainly on mammals, but juveniles feed primarily on lizards (Morato 2005, Hartmann et al. 2009b, Alencar 2010, Gaiarsa et al. 2013, this study). Most individuals found at the Etá Farm were juveniles (< 500 mm), found between June and September. One female collected in the municipality of Cananéia (78 km from the Etá Farm) had 17 vitellogenic follicles in April (data provided by the staff of Butantan Institute). The available information indicates that reproduction is seasonal, with the reproductive peak occurring at the onset of the rainy season in the Atlantic forest (Marques and SaziMa 2004); fecundity ranges from four to 16 eggs (Gaiarsa et al. 2013). When handled, one of the individuals found thrashed the body. This species is a supposed imperfect mimic of Micrurus spp. and the defensive behaviour also includes hiding the head and making erratic movements   (Table 3).

Sordellina punctata (Peters, 1880)
This is an aglyphous species of small size (maximum SVL = 461 mm; N = 7). It was captured in a pitfall trap in the forest (N = 1), crossing an unpaved road near the margins of the Etá River (N = 4), and in a pasture area (N = 1), at dusk (N = 3) and at night (N = 1). One individual was captured in another disturbed habitat. One individual had been ingested by an individual of B. jararacussu at the edge of the forest, at night. The available information indicates that this is a semi-aquatic, primarily diurnal species , Pereira et al. 2007, Marques et al. 2009. A leech was found in the digestive tract of one specimen from the Etá Farm (Table 2). The available information indicates that it feeds primarily on Oligochaetes (earthworms and leeches) and eventually on caecilians (Proctor 1923, Marques 1996c, Marques et al. 2009. At the Etá Farm, it was found active mainly in the hotter and rainier months, except for one individual moving across the unpaved road in June. Data obtained from preserved snakes indicate that it has a seasonal reproductive cycle (Marques, 2001). When handled, two of the individuals found performed cloacal discharge. Information available in the literature indicates that it may also flatten the body .

Taeniophallus bilineatus (Fischer, 1885)
This is an aglyphous species of small size (maximum SVL = 258 mm; N = 2). Two individuals were found, one captured in a pitfall trap in the forest, the other moving along the forest edge in the morning. The available information indicates that it is terrestrial and diurnal , Hartmann et al. 2009b, Forlani et al. 2010, this study). The second individual found was in a forest in July. It feeds on frogs and lizards (Di-Bernardo and Lema 1990, Marques andSazima 2004). Apparently, it has a seasonal reproductive cycle (Marques and Sazima 2004). It may flatten the body and perform cloacal discharge   (Table 3).

Tomodon dorsatus Duméril, Bibron & Duméril, 1854
This is an opisthoglyphous species of small-medium size (maximum SVL = 540 mm, N = 3). One adult female was found in July around a house, one adult male was found on the unpaved road in December and another adult male was found run-over near the peach palm plantation in January. The available information indicates that it is terrestrial, cryptozoic and diurnal (Marques and Sazima 2004, Araujo et al. 2010. It feeds on molluscs ) and some authors have suggested that it may show aggregation, perhaps related to food availability (molluscs; Bizerra 1998, Franco et al. 2006. One female collected in the municipality of Itariri (82 km from Etá Farm) had 13 vitellogenic follicles in November (data provided by the staff of Butantan Institute). The available information indicates that it has a seasonal reproductive cycle, with vitellogenesis occurring mainly by the onset of the rainy season, embryos throughout the rainy season and litter size ranging from 4 to 26 (Bizerra et al. 2005, Barbo et al. 2011. It may flatten the body, strike, raise the head and form an S-coil with the anterior part of the body, hide its head and display the buccal mucosa   (Table 3).

Xenodon neuwiedii Günther, 1863
This is an aglyphous species of medium size (maximum SVL = 555 mm; N = 6). It was captured in a pitfall trap in the forest (N = 1). One juvenile was captured as it moved through the leaf litter during the day, at 08:20 h, and one adult male was captured while crossing an unpaved road at 10:50 h. One individual was found on the peach palm plantation and another on an unpaved road, near a pasture area, during the day. The last individual was caught in unpaved road close to disturbed areas. The available information indicates that it is diurnal and terrestrial (Hartmann et al. 2009b, Forlani et al. 2010. Of the specimens that had their digestive tracts examined, one had the remains of frogs: one Rhinella icterica and the legs of a R. hoogmoedi ( Table  2). The available information indicates that it feeds mainly on frogs (mainly Rhinella spp.), lizards being an occasional prey (Silva and Rodrigues 2001, Marques and Sazima 2004, Hartmann et al. 2009b, Costa et al. 2012, this study). Most individuals found at the Etá Farm were juveniles (except for one adult male) and were found between November and December, except for one juvenile found in May. The available information indicates that it reproduces throughout the year (Jordão 1996, Condez et al. 2009) and its fecundity can reach 14 eggs (Hartmann et al. 2009a). When handled, individuals found at the Etá Farm flattened the body (N = 1) or struck (N = 1). The available information indicates that, besides being a supposed mimic of vipers such as B. jararaca, it may also triangulate the head, raise the head and form an S-coil with the anterior part of the body , or elevate the head while compressing the body (Greene 1979) in a manner similar to Old World elapids (Table 3).

Micrurus corallinus (Merrem, 1820)
This is a proteroglyphous species of medium size (only juveniles were captured, with maximum SVL = 251 mm; N = 5; adults exceed 900 mm in total length; Roze 1996). One individual was captured in a pitfall trap in the peach palm plantation, another was found inside a house, and a third was caught crossing an unpaved road near the forest edge in the early morning. The other three were active during the day in unpaved roads close to disturbed habitats. It is primarily diurnal, and it forages on the ground or in underground galleries capturing caecilians, amphisbaenians, lizards and other snakes (Roze 1996, Marques and Sazima 1997, Banci et al., 2017. Only juveniles were found at the Etá Farm (< 400 mm), in April, July and August. The species has a seasonal reproductive cycle with mating and vitellogenesis occurring at the beginning of the rainy season (Marques 1996b. Fecundity ranges from 2-12 eggs (Marques 1996b). When handled, one individual from the Etá Farm raised the curled the tail. This coral snake flattens its body, hides its head, performs erratic movements and elevates its head, while compressing its body, and raises the curled tail (Greene 1979 (Table 3).

Bothrops jararaca (Wied, 1824)
This is a solenoglyphous species of large size (maximum SVL = 1220 mm; N = 23). It was found in all sampled vegetation types (eight individuals in peach palm plantation, four in forests, and one in the abandoned banana plantation), but never in pitfall traps. It was also found in operating banana plantations (N = 5). The other five individuals were caught in other disturbed habitats. Most individuals found were coiled up in the undergrowth during the day, in the peach palm plantation; one individual was found around houses and two on an unpaved road, one of them near the forest edge and the other in a pasture area. One adult male was found moving on the ground in the afternoon (15:00 h) in the abandoned banana plantation, one juvenile was found moving over a bromeliad on a fallen trunk at night (see also Marques 1998) and one adult male was found moving on the forest ground at 22:00 h. The available information indicates that it is semi-arboreal and primarily nocturnal (Sazima and Manzani 1995, Alves et al. 2000, Forlani et al. 2010. In the digestive tracts of three specimens examined we found rodents, one of them belonging to the tribe Akodontini ( Table 2). The available information indicates that it feeds mainly on rodents and amphibians, in addition to lizards, birds and centipedes, with a relatively higher consumption of ectothermic prey by juveniles and endothermic prey by adults (Sazima and Manzani 1995, Martins et al. 2002, Hartmann et al. 2009b; there have been reports of necrophagy in this species (Sazima and Strüssmann 1990). Five females from the Etá Farm had 13-35 vitellogenic follicles, throughout the year. One gravid female captured in January contained 10 fully formed embryos (SVL = 191 ± 15.23 mm, weight = 7.48 ± 1.29 g). Juveniles (< 400 mm) were mainly observed in the hotter and rainier months of the year (October to February). The available information indicates that it has a biennial, seasonal reproductive cycle, with vitellogenesis occurring between autumn and winter, ovulation probably in early spring, copulation at the beginning of the dry season and pregnant females from November to March (Janeiro-Cinquini 2004, Almeida-Santos and Salomão 2002, this study). Fecundity varies from 3-36 offspring and gestation may last 152-239 days (Alves et al. 2000, Janeiro-Cinquini 2004, Sazima 1992, Almeida-Santos and Salomão 2002, Costa et al. 2010. It was found throughout the year at the Etá Farm. When handled, individuals from the Etá Farm vibrated the tail (N = 6) and struck (N = 4). The Information available in the literature indicates that it may also flatten its body and perform cloacal discharge   (Table 3).

Bothrops jararacussu Lacerda, 1884
This is a solenoglyphous species of large size (maximum SVL = 1150 mm; N = 47). It was found in all sampled vegetation types (23 individuals in the peach palm plantation, six in forest and four in banana plantations), coiled up in the undergrowth, often at the base of the peach palms (N = 17), in open areas or on the leaf litter in the forest (N = 4); only two juveniles were captured in pitfall traps, in the abandoned banana plantation and the peach palm plantation. Nine individuals were found on the unpaved road, three of them at the edge of forest, one of which had just ingested an individual of Sordellina punctata, and three near pasture areas. The other five individuals were caught in other disturbed habitats. The available information indicates that it is terrestrial and frequently found close to watercourses; it has mostly nocturnal activity, although juveniles and eventually adults may hunt during the day (Martins et al. 2002, Marques andAraujo 2011, M. Martins, personal observation). Approximately 63% of the specimens (N = 12) whose digestive tracts were examined had rodents ( Table 2) that, in all cases where it was possible to assess, had been ingested headfirst. One specimen contained the bones of an opossum. The population of the Etá Farm seems to show an ontogenetic change in diet, with juveniles feeding mainly on ectothermic animals, including a snake (Sordellina punctata), frogs of the families Hylidae and Leptodactylidae, and a centipede, and adults feeding primarily on small mammals. However, adult individuals may occasionally consume ectothermic prey, as in the case of one adult male containing a rodent and a leptodactylid frog in its digestive tract ( Table 1). The available information indicates that it feeds on centipedes, frogs, lizards and mammals, with ontogenetic variation in diet, with juveniles mainly feeding on ectothermic animals (especially frogs) and adults mainly feeding on small mammals (Lema et. al. 1983, Martins et. al. 2002, Hartmann et al. 2009b. At the Etá Farm it was found throughout the year; however, most individuals were found in the hotter and rainier months (October-April). Two females had vitellogenic follicles in June and July and one had secondary follicles in January. Most juveniles were observed in December and February. The available information indicates that the reproductive cycle of this species is seasonal; mating occurs in May and June, and births are concentrated in March, with 13-37 hatchlings (Marques 1998, Almeida-Santos andSalomão 2002). When handled, individuals from the Etá Farm struck (N = 2) and vibrated the tail on the ground (N = 5). Information available in the literature indicates that it may also flatten its body and perform cloacal discharge   (Table 3).

Discussion
The snake assemblage of the Etá Farm region has a species composition similar to those of other studied assemblages in the Ribeira River Valley (e.g., Marques 1998, Forlani et al. 2010. The new records for Sete Barras Municipality (C. fuscus, C. laevicollis, E. cephalostriata, Spilotes pullatus, and T. bilineatus) were already expected to occur in the region, because they are typical Atlantic Forest species and their distributions overlap Sete Barras region (Nogueira et al. 2019). All species that potentially occur in the Etá Farm region are typical forest species , Nogueira et al. 2019) and most of them are semi-arboreal (Marques and Cavalheiro 1998, Antunes and Haddad 2009, Hartmann et al. 2009ab, Marques et al. 2019) and though most species found in the field present diurnal activity, when species of potential occurrence are added, most species of the assemblage are nocturnal , Antunes and Haddad 2009, Hartmann et al. 2009ab, Marques et al. 2019. The main prey of these species are amphibians (C. bicarinatus, C. foveatus, E. cyanopleura, E. undulata, T. persimilis, T. nattereri, and T. paucisquamis; Dixon et al. 1993 Of the species recorded in the study area, most used both forested (or were at least observed in forest edges) and open areas, except for C. exoletus, D. neuwiedi, S. pullatus, and T. dorsatus, which were observed only in disturbed areas (banana and peach palm plantations). The peach palm plantations, surrounded by forested areas may be functioning as routes from one edge to the other of the forested areas, and as foraging sites (as they have large quantities of frogs and molluscs; see Fiorillo et al. 2018). However, the edges of a given habitat tend to be hostile to organisms adapted to living in its interior and may contain both competitors and predators (Andrén and Angelstam 1988, Chalfoun et al. 2002). Even so, some snakes from the region of the Etá Farm seem to benefit from using these areas.
The colubrids of the assemblage, as well as the xenodontines, are mostly anuran specialists , this study), except for S. pullatus, which differed from other colubrids by showing a diet based mainly on small mammals and nestling birds (Marques 1998. The anuran prey and substrate use differ among the diurnal frog-eating species. Chironius spp. prey on leptodactylid and hylid frogs at various substrate heights, X. neuwiedii search mainly by Rhinella spp. in the forest ground and in disturbed areas and E. cephalostriata and T. bilineatus consume mainly small frogs commonly found amid the leaf litter, such as Adenomera cf. marmorata, Haddadus binotatus, Ischnochnema sp., and Physalaemus spiniger (Fiorillo et al. 2018), and occasionally their eggs (e.g., E. cephalostriata, see Moura-Leite et al. 2003).
The nocturnal and terrestrial species, B. jararaca, B. jararacussu, and O. clathratus show similar diet and were all found in both open, disturbed areas and in forested areas. Although marsupials (e.g., Monodelphis americana) were restricted to forested areas, rodents were abundant in both habitat types. However, juveniles of these species feed on ectothermic prey (O. clathratus feeds on lizards and Bothrops spp. feeds mainly on frogs). It is known that O. clathratus is occasionally found in open and disturbed areas (Di-Bernardo et al. 2007, Hartmann and Giasson 2008, Hartmann 2009b. Being an almost strictly terrestrial species (Hartmann and Giasson 2008, Hartmann et al. 2009a, Barbo et al. 2011), probably it is not as restricted to forested habitats as other species that use arboreal substrates more often (e.g., Chironius spp.). However, it is important to consider that even the disturbed habitats in which the species was found, were surrounded by forest. In addition, the disturbed habitats sampled (banana and peach palm plantations) presented a considerable abundance of frogs (see Fiorillo et al. 2018), which in turn would favour the foraging of Bothrops species. The malacophagous species, D. neuwiedii and T. dorsatus, show terrestrial behaviour, but distinct daily activities (nocturnal for the former, diurnal for the latter), and both are found in open areas where the molluscs are abundant (personal observation). Two of the slug species consumed by D. neuwiedii are invasive species, the European Limax cf. flavus and the Chinese Meghimatium pictum. Thus, these malacophagous snakes can potentially control the populations of invasive molluscs.
Although annelids have been previously reported for the diet of S. punctata (Marques 1998, we here provided the first report of consumption of leeches by this species. This prey, earthworms (including aquatic giant earthworm) and one caecilian (previously reported by Proctor, 1923) confirm that this snake forages in aquatic habitats.
Most individuals were found during the hot and rainy season from September to March, when most species show reproductive activity, as may be seen by the presence of vitellogenic follicles for some species and mating (e.g., S. pullatus). This seasonal activity peak has been reported for other assemblages of Neotropical snakes (Strüssmann and Sazima 1993, Martins and Oliveira 1998, Marques 1998, Sawaya et al. 2008b, Pontes et al. 2009) and seems to reflect the effect of environmental variables favouring snake metabolism for the development of eggs or offspring. The exceptions were E. aesculapii, which had vitellogenic follicles in July; E. miliaris, which also presented secondary vitellogenic follicles from April to August, although smaller in the hottest period of the year; and the vipers B. jararaca and B. jararacussu, which showed vitellogenic follicles in both the hot and rainy season and during the month of June. Females of E. aesculapii and X. neuwiedii show vitellogenic follicles in every month of the year (Jordão 1996, Marques 1996a, Pizzatto et al. 2008. These two species belong to the tribe Xenodontini and continuous reproductive cycles may be conservative in this lineage of snakes (Pizzatto et al. 2008). Additionally, the type of resource used by these species may enable a continuous reproductive cycle (Vitt andVangilder 1983, Seigel andFord 1987), as these species feed on prey that are abundant throughout the year (Seigel andFord 1987, Roberto et al. 2011).
The reproductive cycle of another member of the tribe Xenodontini, E. miliaris, varies along its distribution and, although vitellogenic follicles were observed from April to August (two individuals), the population of the Etá Farm region is characteristic of the southern coastal region of the Atlantic Forest, where the reproductive cycle of this species may be seasonal . This reproductive peak during the hotter and rainier months, as well as with other species in the assemblage (members of the genus Chironius), may be related to the temporal distribution of frogs, because, as in other species of the genus (e.g., E. poecilogyrus, see Alencar and Nascimento 2014), female E. miliaris apparently do not stop feeding during vitellogenesis until the deposition of the eggs (two females containing secondary follicles and one female with eggs in the oviducts had stomach contents), a characteristic that may be related to the possibility of foraging during the reproductive period (Winne et al. 2006, Dyke et al. 2012. Alternatively, Oliveira and Martins (2001) suggested that snakes, as well as frogs, may simply be responding to the same environmental variables (or set of variables) and, thus, their activity peaks coincide in time.
The results described herein for the reproduction of B. jararaca and B. jararacussu corroborate the results of other studies that describe the reproductive phenology for the genus Bothrops Salomão 2002, Janeiro-Cequini 2004). These studies suggest that copulation occurs from April to September, when the animals are in vitellogenesis, there is a reduction of ovarian mass from October to March, prob-ably due to ovulation and advanced pregnancy occurs during the rainy season (see the account of B. jararaca in Results). These patterns occur due to the storage of sperm in females and late fertilisation, which allows mating to occur in one season and follicular development, fertilisation and parturition to occur in another. Moreover, it provides females with the possibility of repeated fertilisation in a single mating event (Birkhead andMoller 1993, Marques 1996b).
Most defensive tactics observed at the Etá Farm were apparently aimed at visually oriented predators such as birds (especially birds of prey), important predators of Neotropical snakes, and some mammals (Sazima 1992, Martins and Oliveira 1998, Martins 1996. However, in the case of mammals, these tend to show nocturnal activity and, thus, to use mainly their sense of smell and hearing as the main ways to locate their prey (Martins 1996). Thus, a defensive tactic such as cloacal discharge (employed by 70% of the species in the region, considering field observations and literature data; see Table 3) could be more effective. Alternatively, Martins (1996) suggest that defensive tactics seem to correlate with phylogeny, therefore, cloacal discharge (a defensive behaviour with apparently low energy costs) may have been the most frequent behaviour observed simply because it is a common, well established, widespread behaviour among snakes in their evolutionary history.
In contrast, only one species showed aposematic colouration (M. corallinus) and two (E. aesculapii and O. clathratus) are supposed mimics of the coral snakes (Greene and McDiarmid 2005;Martins et al., 2008). Many of the trends in defensive tactics observed in this study are similar to those found by previous works in the Central Amazon (Martins 1996, Martins and Oliveira 1998, including: (1) inaccessibility (e.g., see the account of C. laevicollis); (2) many species employ tactics like fleeing, compressing their bodies, and biting; (3) all diurnal species are cryptic, except for the aposematic species, M. corallinus, and the supposed coral snake mimic, E. aesculapii (additionally, although not observed in this study, there are reports in the literature of the same defensive behaviours of M. corallinus performed by E. aesculapii); and (4) head triangulation was commonly used by supposed mimics of pitvipers (e.g., D. neuwiedii, X. neuwiedii; Greene and McDiarmid 2005), but also by E. miliaris.
A poorly documented visual defensive behaviour was reported for E. miliaris at Fazenda Etá (Menezes et al. 2015;this study). Two individuals (one juvenile and one adult) were observed simultaneously raising and compressing the anterior part of their body, a behaviour commonly performed by Old World elapids and previously reported for other Neotropical snakes, for example, species of the genera Thamnodynastes, Hydrodynastes, and Xenodon (Franco et al. 2003, Young and Kardong 2010, Kahn 2011. It is likely that this behaviour has the same goal of intimidating the predator as other frontal displays previously described, such as raising the head and inflating the glottis, which alter the predator's perception of the size of the individual that performs them (Greene 1988, 1997, Young and Kardong 2010.

Conclusions
The region of the Etá Farm harbours a rich snake fauna that is similar in composition to those of other snake assemblages in the Ribeira River Valley and includes one threatened species. This study contributed to the knowledge of the snake fauna of this region also by providing five new records for the Sete Barras Municipality. The detailed natural history information provided herein may be used in the assessment of the conservation status of these snakes and in the definition of action plans aiming to conserve this rich and biologically diverse fauna.
Keys to families and species of snakes from Sete Barras region, southeastern Brazil