AQUATIC CONSERVATION: MARINE AND FRESHWATER ECOSYSTEMS Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/aqc.2394 Status and the potential for extinction of the largetooth sawfish Pristis pristis in the Atlantic Ocean JOANA FERNANDEZ-CARVALHOa,b,*, JOHANNA L. IMHOFFa,c, VICENTE V. FARIAd, JOHN K. CARLSONe and GEORGE H. BURGESSa a Florida Program for Shark Research, Florida Museum of Natural History, University of Florida, Gainesville, USA b Centro de Ciências do Mar (CCMAR), Universidade do Algarve, Faro, Portugal c Department of Biological Science, Florida State University, Tallahassee, USA d Institute of Marine Sciences – Labomar, Federal University of Ceará, Fortaleza, Brazil e US National Marine Fisheries Service, Southeast Fisheries Science Center, Panama City Laboratory, Panama City, USA ABSTRACT 1. Sawfishes currently are among the most threatened elasmobranchs in the world. Only two species inhabit Atlantic waters: the largetooth sawfish (Pristis pristis) and the smalltooth sawfish (Pristis pectinata), both having suffered dramatic declines in their ranges. 2. The goal of this study was to evaluate the status of P. pristis in the Atlantic, and estimate local extinction risk based on historical and recent occurrence records. In order to accomplish these goals, a thorough search for historical and recent records of P. pristis in the Atlantic was conducted, by reviewing scientific and popular literature, museum specimens, and contacting regional scientists from the species’ historical range. 3. In total, 801 P. pristis records (1830–2009) document its occurrence in four major regions in the Atlantic: USA (n = 41), Mexico and Central America (n = 535), South America (n = 162), and West Africa (n = 48). Locality data were not available for 15 records. 4. Historical abundance centres were the Colorado-San Juan River system in Nicaragua and Costa Rica (and secondarily Lake Izabal of Guatemala), the Amazon estuary, and coastal Guinea-Bissau. 5. Currently, the species faces drastic depletion throughout its entire former range and centres of abundance. It appears to have been extirpated from several areas. The probability of extinction was highest in the USA, northern South America (Colombia to Guyane), and southern West Africa (Cameroon to Namibia). 6. Currently, the Amazon estuary appears to have the highest remaining abundance of P. pristis in the Atlantic, followed by the Colorado–San Juan River system in Nicaragua and Costa Rica and the Bissagos Archipelago in Guinea Bissau. Therefore the protection of these populations is crucial for the preservation and recovery of the species. Copyright # 2013 John Wiley & Sons, Ltd. Received 10 December 2012; Revised 4 June 2013; Accepted 1 August 2013 KEY WORDS: conservation evaluation; endangered species; estuary; coastal; brackish; fish *Correspondence to: J. Fernandez-Carvalho, Centro de Ciências do Mar, Universidade do Algarve, Campus de Gambelas, FCT Ed. 7, 8005-139, Faro Portugal. Email: jfcarvalho@ualg.pt Copyright # 2013 John Wiley & Sons, Ltd. J. FERNANDEZ-CARVALHO ET AL. INTRODUCTION Over the last century, significant exploitation and habitat destruction has led to concerns that populations of marine fish may be at risk of extinction (Dulvy et al., 2003; Monte-Luna et al., 2007). Detecting and determining local extinctions is important to prevent further loss of the species in other areas and to establish factors related to the cause of the extinction. Dulvy et al. (2003) identified 133 local, regional and global extinctions of marine populations and determined that exploitation caused most extinctions followed by habitat loss, invasive species, climate change, pollution and disease. The most common approach to inferring the extinction of a particular species is based on the record of its sightings (Solow and Roberts, 2003). For marine species, information about past occurrences has been compiled from historical sightings and museum records. Records are then analysed in a probabilistic framework to generate confidence intervals on the potential time of extinction given the sighting records (Solow, 1993a, 2005; Burgman et al., 1995; Roberts, 2006). For example, among elasmobranchs, Luiz and Edwards (2011) concluded that the Galapagos shark, Carcharhinus galapagensis has been locally extinct since 1998 at St. Paul’s Rocks in the Atlantic Ocean. Smalltooth sawfish, Pristis pectinata, were determined to be extirpated in US waters from Virginia to New York state from between 1929 and 1956 (Monte-Luna et al., 2009; NMFS (National Marine Fisheries Service), 2009b). Reliably estimating the spatial patterns and/or the timing of extinction remains a major challenge for scientists, as the danger of extinction can be overestimated because of incorrect assumptions on population size. The largetooth sawfish (Pristis pristis, Linnaeus, 1758) is one of the two species of the family Pristidae occurring in the Atlantic Ocean. It is found in tropical and subtropical waters in both the western and eastern Atlantic, with northern and southern range extremes largely dictated by seasonal water temperature regimes. The western North Atlantic distributional termini of both species occur in the USA whereas those found in the eastern Atlantic ranged as far north as Mauritania in the north and Namibia in the south (Bigelow and Schroeder, 1953; Faria et al., 2013). Copyright # 2013 John Wiley & Sons, Ltd. Like all species of sawfish, P. pristis has significantly declined owing to overfishing, chiefly as bycatch, and habitat loss (Fowler et al., 2005). As a result P. pristis is listed as Critically Endangered on the International Union for Conservation of Nature Red List (Fowler et al., 2005; Kyne et al., 2013) and as Endangered Globally on the US Endangered Species Act (76 FR 40822). Despite being listed as Critically Endangered and Endangered, respectively, the status of largetooth sawfish was determined based on anecdotal information. Thus there is a need both to qualify and quantify historical records of P. pristis in the Atlantic Ocean to provide a better determination of its current status. Here, we report on the results of an extensive search of historical and recent records of P. pristis in the Atlantic Ocean and provide a region by region analysis of its current status and risk of extinction. MATERIAL AND METHODS Data collection Sawfish records were collected between July and September 2009, and records were individually screened for accuracy. Species identification was based on taxonomic characters described in Bigelow and Schroeder (1953) for Pristis perotteti (e.g. the number of rostral teeth). Following Faria et al. (2013) all records of Pristis perotteti and Pristis microdon from the Atlantic were considered largetooth sawfish and as such are referred to as Pristis pristis. Records of P. pristis were gathered by consulting primary literature, media, museums and local expert scientists. Primary literature records and P. pristis distribution information were obtained by searching card catalogues from the libraries of the key regional universities (University of Florida, University of Texas, Texas A&M University, University of South Alabama) within the species’ historical US range. Library databases and online search engines were used for finding photographs and fishing and historical blogs pertaining to P. pristis records. In addition, two online historical newspaper databases (Proquest Historical Newspapers database and Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) STATUS OF THE LARGETOOTH SAWFISH IN THE ATLANTIC OCEAN www.NewspaperArcive.com) were used to find sawfish articles in local and regional newspapers. Positive identifications from historical newspapers were only possible when photographs were available or rostral tooth counts were provided in the story. Twenty-seven museums were contacted to obtain records of archived P. pristis specimens. Additional specimen records from museums and private collections were obtained from the International Sawfish Encounter Database (ISED) housed at Florida Museum of Natural History and from Faria V. (unpublished data). Owing to the lack of recent studies addressing the status of P. pristis in the majority of its range, scientists and fisheries managers were queried directly about any sawfish records and asked to provide their personal evaluations of the regional status of the species (only regarding Atlantic range) using a standardized set of questions. The inquiry included questions about their knowledge of the species, the existence of any records (museum, literature, research, fisheries data, evaluation of the species, etc.) in their region, their personal evaluation and possible colleagues in the region that should be contacted because of their expertise. Hundreds of researchers in the Gulf of Mexico and Caribbean regions were contacted via the list server of the Gulf and Caribbean Fisheries Institute (GCFI) in English, Spanish, Portuguese and French. Furthermore, 43 scientists (representing 13 countries: Aruba, Belize, Brazil, Colombia, Cuba, Curaçao, Guinea-Bissau, Honduras, Mexico, Nicaragua, Panama, Puerto Rico, Venezuela) were contacted directly after being identified as knowledgeable experts in their locale by colleagues, though not all replied. All the scientists who contributed sawfish information are credited in the acknowledgement section with their affiliation and country. Scientists who replied but did not provide sawfish information (e.g. only recommended other colleagues or said they had no knowledge about sawfish) were not included in the acknowledgement section. When citing the personal communications of the contributing scientists, the year when the personal communication was provided is also given. Copyright # 2013 John Wiley & Sons, Ltd. Data analysis Sawfish records were entered into a Microsoft Access database, the ISED. Duplicate records, such as museum records that were also cited within the primary literature or multiple newspaper articles referring to a single specimen, were screened and removed. Species identifications were verified by examination of museum specimens, photographs, scientific notes, and/or interviews with relevant observers, unless already recently confirmed by an expert (Faria V., unpublished data). Dubious P. pristis records, such as records suspected to be misidentified P. pectinata or two records that were suspected (but not confirmed) to be from the same specimen (Springer and Woodburn, 1960), were included in the database. However, only confirmed (and not dubious) P. pristis records were included in the analysis. Given the nature of the data available, the probability that local scale extinction has occurred was determined using the time series of incidental observations. The probability that local extinction has occurred follows a stationary Poisson process: ¼1  t n c T where the number of time intervals in which the species was observed is n, the total number of intervals sampled is T, and the number of intervals up to the time the species was last observed is tc (Solow, 1993a, b; Burgman et al., 1995; Grogan and Boreman, 1998). Sensitivity analyses suggest using minimum probability levels of 0.75 for declaring a species to be in danger of local extinction and 0.95 for locally extinct (Grogan and Boreman, 1998). As estimates of the probability of extinction are point estimates, a Monte Carlo simulation was used to assess uncertainty in the model input parameters. Probability density functions were developed to describe the number of time intervals in which the species was observed and the number of intervals up to the time the species was last observed. The extinction risk was determined for six regions throughout the historical range of P. pristis: USA, Central America (Mexico to Panama), northern South America (Columbia to French Guinea), Brazil, Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) J. FERNANDEZ-CARVALHO ET AL. northern West Africa (Mauritania to Nigeria) and southern West Africa (Cameroon to Namibia). For those areas with a high probability of extinction (≥ 0.95), the upper bound (the upper confidence limit at which the extinction probability reaches the nominal 0.05 value) of the timing of extinction was calculated using several methods evaluated by Rivadeneira et al. (2009). The first method made the strictest assumption that sightings are a Poisson stationary process (Strauss and Sadler, 1989; Solow, 1993a; McInerny et al., 2006), with constant probabilities of occurrence and sampling through time, and an abrupt collapse to extinction. The other method (Roberts and Solow, 2003; Solow and Roberts, 2003) is least restrictive and does not make any distributional assumptions, although the independence of sightings is still required (Rivadeneira et al., 2009). In addition, the likelihood that P. pristis has become extinct was evaluated for the six distributional areas using the framework suggested by Butchart et al. (2006) and used for elasmobranchs by Luiz and Edwards (2011). This framework categorizes the level of confidence of extinction based on observational data by considering the time since the species was last reported and the evidence for and against extinction. RESULTS In total, 801 records of P. pristis throughout the Atlantic were compiled during this study. These comprised 41 US P. pristis records (from 1878 to 1961), 535 Mexico and Central America records (1877 to 1998), 162 South America records (1830 to 2009) and 48 records from the West coast of Africa (1841 to 2005) (Figure 1). Locality data were not available for 15 additional P. pristis records. United States of America Records of P. pristis in US waters (n = 41, Table 1) included 35 records from Texas, two ‘Gulf of Mexico’ records which were probably from Texas, Figure 1. Records of Pristis pristis obtained in this study. All countries that have P. pristis records are coloured and are delimited by a thick line. Underlined country names represent countries for which there are records with only general location (i.e. country or state) available, thus the records cannot be represented in the map. AM=Amazonas, AP=Amapá, BA= Bahia, ES=Espírito Santo, FL= Florida, LA=Louisiana, MA=Maranhão, PA=Pará, RN=Rio Grande do Norte, RJ=Rio de Janeiro, SE=Sergipe, SP=São Paulo,TX= Texas. Copyright # 2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) STATUS OF THE LARGETOOTH SAWFISH IN THE ATLANTIC OCEAN Table 1. List of US Pristis pristis records. Records lacking state data are listed under ‘Gulf of Mexico’ State County Year N 1910 1941 1943-1952 1 1 1 FL FL FL Monroe Monroe Martin LA N/A 1918 1 TX TX TX TX TX TX TX TX TX TX TX TX TX TX TX TX TX TX TX TX TX TX TX TX TX TX TX TX TX TX Aransas Aransas Brazoria Cameron Galveston Galveston Galveston Galveston Galveston Galveston Galveston Galveston Galveston Galveston Galveston Galveston Galveston Galveston Galveston Kleberg/Kennedy Nueces Nueces Nueces Nueces Nueces Nueces Nueces N/A N/A N/A 1940 1943 1942 1925 1929 1938 1938 1940 1940 1940 1942 1943 1943 1943 1951 1951 1957 1947 1917 1935 1938 1943 1943 1947 1961 1940 1948 - 1 1 1 1 1 1 1 1 1 1 6 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 GOM GOM N/A N/A 1878 - 1 1 TL (cm) Sex Source Mus. Cat. Num./ Reference MUS LIT LIT AMNH 11 Baughman (1943) Bigelow and Schroeder (1953) 488 MUS/MED SUI 17512 570 500 LIT LIT LIT MED MED MED MED MED MED MED LIT LIT LIT LIT MED MUS MED MUS MUS MED MED MED MED LIT LIT MED MED MUS OTH MED Baughman (1943) Baughman (1943) Baughman (1943) Hoover (2008) MUS LIT ANSP 17388 Bigelow and Schroeder (1953) 520 528 594 450 450 427 528 470 430 450 460 700 519 F M F 610 549 500 457 498 531 380 457 F Baughman (1943) (Coll. by Reid, E. F.) Baughman (1943) Baughman (1943) Baughman (1943) MRAC A4-45-P UAIC 4138.01-2 BMNH 1867.10 Hoover (2008) Baughman (1952); Baughmann (1943) Baughman (1952) TAMUCC Uncat. TX = Texas, LA = Louisiana, FL = Florida and GOM = Gulf of Mexico. ENC = records from encounter reports, LIT = Literature records, MED = Media records, MUS = specimens from museums or other collections. one record from Louisiana, and three records from Florida. All the documented records correspond to large animals ranging from 427 to 610 cm in total length (TL). An unmeasured Galveston Beach specimen documented in a 1951 home movie was estimated to have been of 700 cm TL. Several records documented sawfishes being captured by commercial and recreational fishing gear. Specifically, fishing gear information was available for 21 records. Sawfishes captured in commercial fishing gear were associated with shrimp fisheries and caught either with trawl (14%), seine (5%), shrimp net (10%) or unclassified net (9%), while sawfishes captured recreationally were caught using rod and reel (62%). It can Copyright # 2013 John Wiley & Sons, Ltd. be inferred that most captures occurred in shallow inshore waters off the coast or occasionally within bays, since most of the shrimp trawling occurred near shore, and recreational fishing was shore-based. In Texas, P. pristis was found primarily in three almost equidistant regions, from north-east to south-west: Galveston–Texas City–Freeport (n = 22), Corpus Christi–Port Aransas (n = 9) and Padre Island–Laguna Madre (n = 2). Captures occurred only in warm water months from May to October. Pristis pristis was last recorded in the USA in Texas in 1961. The estimated probability of extinction in US waters was 0.99 (±0.16; Table 2). Depending on methodology, the upper limit of time of extinction Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) J. FERNANDEZ-CARVALHO ET AL. Table 2. Probability and timing of extinction for Pristis pristis from six subregions from their historic range. Numbers in parentheses are the 95% confidence intervals for probability of extinction. Abbreviations for extinction time indicate the method used: S&S = Strauss and Sadler (1989), SOL = Solow (1993a), MCY = McInerny et al. (2006), S&R = Solow and Roberts (2003) and R&S = Roberts and Solow (2003) US Decade 1830 1840 1850 1860 1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 N Mexico and Central America Time Extinct N Northern South America N Time Extinct Brazil Northern West Africa N N N Time Extinct 2 1* 1 1 1 3 2 2 2* 1* 1 + 3* 1 3 3 2 4 23 3 1 1 + 2* 1 1 1* 1 1963 SOL,MCY 1967 S&S 9 5* 496 + 4* 1 + 2* 1994 R&S 10 6 1 (years ago) 2017 S&R 1 (nowdays) 1* 1* 1976 SOL,MCY 1 + 1* 1 + 1* 4 1 8 2 + 3* 1 49 +1* 43 10 1 6 + 3* 2014 SOL, MCY 2020 2024 S&R 2025 R&S 2030 2040 … 2080 … 2320 Probability Extinction Southern West Africa 2047 S&S 2044 S&S 2083 S&R 2326 R&S 0.99 (±0.16) 0.64 (±0.29) 0.99 (±0.06) 0.11 (±0.21) 0.25 (±0.20) 0.99(±0.77) *Only date of citation was available, thus the date of the record could be earlier. ranged from 1963 to 1994, although most methods estimated dates in the 1960s. In addition, qualitative evidence regarding population extinction did provide evidence for local extinction (Table 3). Mexico and Central America Records of P. pristis from the Mexico and Central America region (n = 535) are primarily from Nicaragua and Costa Rica, with only 11 documented records of P. pristis found outside of those two countries (Table 4). The documented records include from north to south, Mexico (n = 5), Guatemala (n = 5), Honduras (n = 1), Copyright # 2013 John Wiley & Sons, Ltd. Nicaragua (n = 483) and Costa Rica (n = 37). Additionally, four old records were found with no year information and attributed to ‘Caribbean Sea’. Four Mexican historical records of P. pristis have been documented for Tamaulipas, Veracruz, Tabasco, and Campeche, all locations in the south-western Gulf of Mexico (Zarur-Ménez, 1962 in Castro-Aguirre, 1978, 1999). Although precise dates are not available, these records (particularly those from Campeche) are from the 1950s and sawfishes are no longer registered in literature from the l970s and 1980s (J. C. P. Jiménez, pers. comm., 2009). The last record of P. pristis from Mexico was a 540 cm Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) STATUS OF THE LARGETOOTH SAWFISH IN THE ATLANTIC OCEAN Table 3. Evaluation of the qualitative evidence for and against local extinction of Pristis pristis from six subregions from their historic range in the Atlantic Ocean, using the framework of Butchart et al. (2006) Observational data Types of evidence for extinction For species with recent last records, the decline has been well documented Severe threatening processes are known to have occurred The species possesses attributes known to predispose taxa to extinction Recent surveys have been apparently adequate given the species’ ease of detection, but have failed to detect the species Types of evidence against extinction Recent field work has been inadequate The species is difficult to detect There have been reasonably convincing recent local reports or unconfirmed sightings Suitable habitat remains within the species’ known range, and/or allospecies or congeners may survive despite similar threatening processes ? US Mexico and Central America Northern South America Brazil Northern West Africa Southern West Africa N/A Yes N/A Yes Yes N/A Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes? Yes Yes No? No? No No No? No No No Yes? No Yes Yes? No No Yes No Yes Yes No Yes Yes No No Yes Yes Yes Yes Yes Yes Existence of uncertainty owing to the scarcity of available information for the region. TL specimen landed at Mujeres Island, Quinta Roo in 1997 (J. L. Castillo-Geniz, pers. comm., 2009). All Guatemalan records (n = 5) are the result of Robert Rush Miller’s scientific collecting in Lake Izabal in 1946–1947, with specimens catalogued and curated at the National Museum of Natural History (Table 4). In addition, Lake Izabal was described by Saunders et al. (1950) as the only Guatemalan lake where sawfishes were present, being described as the most important inland fishes of the country. No records were found for any other Guatemalan Lake while conducting this study. A historical P. pristis rostrum (UF uncat.) lacking date and location information originated from Honduras, and is the only record from that country. Historically, the Lake Nicaragua–San Juan River–Colorado River system, in Nicaragua and Costa Rica, was of particular importance for P. pristis and the species was recorded from that system in several studies (Gill and Bransford, 1877; Meek, 1907; Marden, 1944; Bigelow and Schroeder, 1953; Thorson, 1982b). However, neither historical nor modern records were found from the Atlantic coast or drainages in Nicaragua, other than the records of Lake Nicaragua and San Juan River system (Montoya and Thorson, 1982; Copyright # 2013 John Wiley & Sons, Ltd. H. Guzmàn, pers. comm., 2009). Besides San Juan and Colorado Rivers, P. pristis is known in Costa Rica also from the Tempisque River and tributaries of the San Juan River basin (Bussing, 2002). In addition, one isolated rostrum examined in this study (UF uncat.) is from a specimen caught in Tortuguero during the 1960s. In total, 384 specimens caught by commercial longliners were reported captured near the source of the San Juan River at Lake Nicaragua, Nicaragua (Thorson, 1976, 1982b). Nineteen of these were large individuals, ranging in length from 324 to 394 cm TL (Thorson, 1982b). In addition, 34 specimens were reported taken by commercial longline fishers in the lower reaches of the Colorado River near Barra del Colorado (Costa Rica) between June 1966 and February 1977 (Thorson, 1982b). Two of these animals were juveniles (127 and 130 cm TL). Recent records of P. pristis in Nicaragua and Costa Rica include immature animals. One neonate was caught in the San Carlos region (near the origin of the San Juan River on the southernmost part of Lake Nicaragua), offering tangible evidence that at least some pupping or nursery areas still occur in the region. Additionally small specimens (30 to 60 cm long) have been encountered during Adpesca investigations in the western part of the Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) J. FERNANDEZ-CARVALHO ET AL. Table 4. Records of Pristis pristis from Mexico and Central America. Countries are ordered from north to south. MEX = Mexico, GTM = Guatemala, HND = Honduras, NIC = Nicaragua and CRI = Costa Rica. CAR = Caribbean Sea. ENC = records from encounter reports, LIT = Literature records, MED = Media records, MUS = museums or collections, PLIT = specimens from private collections cited in the literature N Country State / Loc Year 1 1 1 1 1 MEX MEX MEX MEX MEX Quintana Roo Campeche Tamaulipas Veracruz Tabasco 2 1 1 1 GTM GTM GTM GTM Izabal Izabal Izabal Izabal 1 HND - - 1 1 1 1 1 1 1 36 25 1 6 3 1 384 6 3 1 1 2 1 1 1 1 1 1 1 NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC NIC Granada R.San Juan/Lk. Nicaragua Lk. Nicaragua Lk. Nicaragua Lk. Nicaragua Lk. Nicaragua Rivas Granada/ Lk. Nicaragua Granada/ Lk. Nicaragua Granada/ Lk. Nicaragua Granada/ Lk. Nicaragua Granada/ Lk. Nicaragua R.San Juan/Lk. Nicaragua 1877 1942 1943 1943 1943 Late 1960s 1960 1968 1968 1968 1968 1968 1969 1966-1977 1980 1980 1985 1991 1998 1998 1998 "Nowdays" "years ago" - 1 34 2 CRI CRI CRI 2 1 1 CAR CAR CAR 1997 Pre-1962 Pre-1978 Pre-1978 Pre-1978 1946 1947 1947 1947 Rivas/ Lk. Nicaragua Rivas/ Lk. Nicaragua Rivas/ Lk. Nicaragua Chontales/ Lk. Nicaragua Granada/ Lk. Nicaragua Lk. Nicaragua Rivas/ Lk. Nicaragua Lk. Nicaragua Lk. Nicaragua Lk. Nicaragua Lk. Nicaragua Lk. Nicaragua Limon 540 Sex Source Mus. Cat. Number/Reference F ENC LIT LIT LIT LIT Castro-Aguirre Castro-Aguirre Castro-Aguirre Castro-Aguirre MUS MUS MUS MUS USNM 00111423 USNM 00111443 USNM 00393616 USNM 00146543 MUS UF Uncat. LIT LIT MUS MUS MUS PLIT MUS LIT LIT LIT LIT LIT MUS LIT LIT LIT MED PLIT PLIT PLIT PLIT MED MED LIT LIT LIT Gill and Bransford (1877) Thorson (1976) USNM 00111367 USNM 00111369 USNM 00120468 Faria et al. (2013) TU 22989 Thorson (1982b) Thorson (1982b) Thorson (1982b) Thorson (1982b) Thorson (1982b) AMNH 55624A Thorson (1976, 1982b) Thorson (1982b) Thorson (1982b) Faria et al. (2013) Faria et al. (2013) Faria et al. (2013) Faria et al. (2013) www.nicaraguafishing.com www.nicaraguafishing.com Gill and Bransford (1877) Meek (1907) Bigelow and Schroeder (1953) MUS LIT MED UF Uncat. Thorson (1976, 1982b) www.jjphoto.dk LIT MUS MUS Bigelow and Schroeder (1953) BMNH 1923.11.13.1 BMNH 1843.2.1.8 137.6 140 F 732 137.6 M Juv Juv Juv Adult Adult F M F M M 226-283 226-284 M F 100 1960 1966-1977 - lake (no adults have been observed) (Moncada, 2006). Finally, two records for Costa Rica correspond to two isolated rostra (one of them probably from a young-of-the-year). Overall for Mexico and Central America, the probability of extinction from this area is 0.64 (±0.29; Table 2). The qualitative analysis showed Copyright # 2013 John Wiley & Sons, Ltd. TL (cm) F F (1978, 1999) (1978, 1999) (1978, 1999) (1978, 1999) evidence both for and against extinction in the region, highlighting that although there are some recent local records the status of the species seems very threatened (Table 3). The existence of adequate field surveys is uncertain owing to the lack of information available for most of the countries in the region. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) STATUS OF THE LARGETOOTH SAWFISH IN THE ATLANTIC OCEAN Table 5. Records of Pristis pristis from South America. Countries are ordered geographically from north to south. COL = Colombia, VEN = Venezuela, TTO = Trinidad, GUY = Guyana, SUR = Suriname, GUF = French Guiana. ENC = records from encounter reports, LIT = literature records, MUS = specimens from museums or other collections N Country State / Loc 1 COL Atlántico 2 1 2 1 1 COL COL COL COL COL Bolívar Santander Córdoba Bolívar Magdalena Pre-1945 Pre-1945 Pre-1964 Pre-1976 Pre-1985 1 1 1 1 1 1 VEN VEN VEN VEN VEN VEN Zulia Zulia Zulia Delta Amacuro - 1894 1900 1903 1903 1962 Pre-1942 1 TTO - Pre-1940 1 GUY - 1 1 1 1 1 SUR SUR SUR SUR SUR - 1 1 GUF GUF Maroni river Cayenne river Year TL (cm) Source 1967 600 LIT LIT LIT LIT LIT LIT Martínez [1978] in Grijalba-Bendeck et al. (2009) Miles (1945) Miles (1945) Mejía and Acero (2002) Mejía and Acero (2002) Mejía and Acero (2002) ENC MUS MUS MUS LIT LIT USNM 00232690 USNM 00232688 USNM 00232689 Cervigón (1966) Bigelow and Schroeder (1953) LIT Ramjohn (1999) - MUS BMNH 1889.11.15.1 1830 1856 1878 1962 - LIT LIT MUS MUS MUS Fowler (1919) Fowler (1919) ANSP 17390 RMNH D3079 RMNH D2674 1830 1929 MUS LIT USNM 00111169 Puyo (1949) South America In total, 162 P. pristis records were found for South America, with the vast majority of records coming from Brazil (Tables 5 and 6). Much lower numbers of P. pristis were recorded in the remaining countries of the species’ range in South America: Colombia (n = 8), Venezuela (n = 6), Guyana (n = 1), Suriname (n = 5), French Guiana (n = 2), Trinidad (n = 1) and Brazil (n = 139). Eight records were found for Colombia in the Caribbean states of Córdoba (n = 2), Bolívar (n = 3), Atlántico (n = 1) and Magdalena (n = 1), and the neighbouring inland state of Santander (n = 1) (Table 5). The only record for which specific date and size are available is a 600 cm TL specimen captured in a competition in 1967 in Barranquilla (Atlántico) (Martínez, 1978 in Grijalba-Bendeck et al., 2009). Despite this low number of specific records, several texts document the species’ occurrence in Colombia (Miles, 1945, 1947; Acero et al., 1986; Flórez-González, 1986; Mejía and Acero, 2002). Listings of P. pristis (as Copyright # 2013 John Wiley & Sons, Ltd. Mus. Cat. Number/Reference 610 250 P. microdon) from the Magdalena River estuary (Miles, 1945, 1947) reported differences in rostral tooth counts and dorsal fin placement, clearly indicating that both P. pristis and P. pectinata were present. The species was present in the marine environment (Frank and Rodriguez, 1976 in Mejía and Acero, 2002; Alvarez and Blanco, 1985) and in fresh water (Dahl, 1971; Sinú River). Miles (1947) referred to a Magdalena River record ‘some 600 km from the mouth’ as P. pectinata, but a record this far upriver likely refers to P. pristis. Sawfishes (Pristis spp.) in Colombia have been described in the literature both as having low economic value (Dahl, 1971; Grijalba-Bendeck et al., 2009) and as commercial species with importance in the exportation to Asian countries (Salazar, 1978 in Caldas et al., 2009). However, no sightings or captures of the species have occurred recently (Mejía and Acero, 2002; Grijalba-Bendeck et al., 2009) and local scientists state that P. pristis has been eradicated from Colombian waters long ago (A. Acero, pers. comm., 2009; F. Gomez, pers. comm., 2009). Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) J. FERNANDEZ-CARVALHO ET AL. Table 6. Records of Pristis pristis from Brazil. States are ordered geographically from north to south and from west to east. ENC = records from encounter reports, LIT = literature records, MUS = specimens from museums or other collections. PLIT = specimens from private collections cited in the literature N State / Loc Year 48 40 13 1 2 1 1 1 1 1 1 1 1 3 1 Northern Brazil Northern Brazil Northern Brazil Amazon River Amapá Amazonas Amazonas Amazonas Amazonas Amazonas Amazonas Amazonas Amazonas Amazonas Amazonas 1999 2000 Pre-1934 Pre-1953 1955 1964 1965 1966 1967 1971 Pre-1974 - 1 1 1 1 2 1 4 Pará Pará Pará Pará Pará Pará Pará 1865 1865 1873 1878 2002 Pre-1998 1 1 1 Maranhão Maranhão Maranhão 1998 2009 1983-1986 1 Rio Grande do Norte 1 TL (cm) Sex Source Mus. Cat. Number/Reference PLIT PLIT PLIT PLIT PLIT LIT LIT LIT LIT LIT LIT LIT LIT LIT MUS Faria et al. (2013) Faria et al. (2013) Faria et al. (2013) Faria et al. (2013) Faria et al. (2013) Thorson (1974) Bigelow and Schroeder (1953) Thorson (1974) Thorson (1974) Thorson (1974) Thorson (1974) Thorson (1974) Thorson (1974) Thorson (1974) INPA Uncat. M M F MUS MUS MUS MUS MUS LIT MCZ-302 MCZ-302 MCZ 668 USNM 00110174 INPA uncat 1 to uncat 2 Carvalho and McEachran (2003) Charvet Almeida and Almeida (2008); Faria et al. (2013) F F ENC/LIT ENC LIT Almeida et al. (2006) MUS/LIT SU 34468 Sergipe MUS ZMB 32533 1 Bahia MUS/LIT ANSP 17389 1 1 Espírito Santo Espírito Santo MUS PLIT MCZ-667 Faria et al. (2013) 1 Rio de Janeiro LIT Bigelow and Schroeder (1953) 1 São Paulo LIT Bigelow and Schroeder (1953) 154 255 175 235 160 116 225 190 F 86.2 95.6 81 700 700 1913 1865 - F 1 - 1929 MUS AMNH 49528 1 - 1978 MUS NCSM 46131 The earliest of the six P. pristis records from Venezuela is a rostrum from a large adult. This specimen (believed to be from Venezuela) is estimated to have been in the possession of a US family since 1894. Three of the records were from the Maracaibo region, an area where the species was reported to have been abundant and frequently taken (Cervigón, 1966). The last documented record of P. pristis in Venezuela was an individual captured in 1962 (Cervigón, 1966), Copyright # 2013 John Wiley & Sons, Ltd. and an additional record from before 1942 has no specific locality data. Although the documented records of P. pristis in Venezuela lack detailed locality data, the species has been reported to be abundant and frequently taken in fisheries in the Gulf of Venezuela and Lake Maracaibo (north-western Venezuela), as well as the Gulf of Paria (north-eastern Venezuela) and the area south of Trinidad, but it was rare and only captured occasionally in the Margarita Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) STATUS OF THE LARGETOOTH SAWFISH IN THE ATLANTIC OCEAN region (north-eastern Venezuela) (Cervigón, 1966). Before their decline, sawfishes (Pristis spp.) were mostly caught in the mouth of the Orinoco River and nearby waters (R. Tavares, pers. comm., 2009). They were also reported as frequently caught in north-west Venezuela, specifically on the coast of Falcon state and in the Gulf of Venezuela, where interviewed fishermen still have rostra (R. Tavares, pers. comm., 2009). According to Cervigón (1993), the main threat to the species was shrimp trawling. Both species of sawfishes stopped being caught in artisanal fisheries in the early 1990s, according to fishermen, and there is no recent information about sawfishes in the country (R. Tavares, pers. comm., 2009). There is a single 19th century museum record of P. pristis from Demerara, Guyana. Regarding Suriname, three museum specimens held in public collections were located. In addition to these, Fowler (1910, 1936) noted the species from Suriname and later recorded an additional Suriname P. pristis specimen from about 1830 (Fowler, 1919). Two records from French Guiana were found, an 1830 museum specimen from the Maroni River, which forms the border between French Guiana and Suriname, and one 1929 record from the Cayenne River (Puyo, 1949). Ramjohn (1999), citing Mendes (1940), listed the P. pristis (as P. microdon) without comment from Trinidad. However, local scientists haven’t heard of any sighting of the species in the recent decades while scientific diving around the Dutch Antilles and Aruba (M. G. G. Grol, pers. comm., 2009). In Curaçao, only one sawfish specimen (Pristis sp.) was collected during scientific sampling from the country’s largest mangrove bay (Schottegat Bay) at the turn of the 20th century and another single specimen was encountered in St. Joris Bay in the early 1970s, but no species is detailed in either case (A. O. Debrot, pers. comm., 2009). Although sawfishes were probably never abundant in Curaçao, they should now be considered extirpated in that country (A. O. Debrot, pers. comm., 2009). According to qualitative analysis there seems to be more evidence supporting local extinction than against it for the northern areas of South America, although there is some uncertainty caused by the lack of information available for this subregion. The Copyright # 2013 John Wiley & Sons, Ltd. reduced number of records found resulted in the probability of extinction of 0.99 (±0.06; Table 2). Consequently, the timing of extinction varied considerably and ranged from 1976–2326. Brazil’s historical and recent records bearing location data come from the following states (west to east and north to south): Amapá (n = 2), Amazonas (n = 12), Pará (n = 11), Maranhão (n = 3), Rio Grande do Norte (n = 1), Sergipe (n = 1), Bahia (n = 1), Espirito Santo (n = 2), Rio de Janeiro (n = 1), and São Paulo (n = 1). In addition, there are 101 more records from ‘northern Brazil’ for which there is no state information (Table 6). As these numbers suggest, P. pristis has been historically more abundant in the northerly tropical fraction of its range in Brazil, a region comprising the Amazon River and the states of Amazonas, Amapá, Pará and Maranhão. In this region, neonates, juveniles and adults can be found (Santos and Val, 1998; Almeida and Carneiro, 1999; Charvet-Almeida, 2002; Nunes et al., 2005; Charvet-Almeida and Faria, 2008). The Amazon River basin area and adjacent waters have long been recognized as an area inhabited by sawfishes, and P. pristis specifically (Bates, 1964; Marlier, 1967; Furneau, 1969; Thorson, 1974). In the Amazonas state, which includes the middle section of the Amazon basin, P. pristis have been taken as far inland as 2000 km from the coast (Santos and Val, 1998) as well as several other locations downstream to the Amazonas–Pará state border. The earliest records from Brazil come from Pará, which contains the estuary and lower reaches of the Amazon River. Two relatively recent records of P. pristis were reported from Maranhão, one adult female caught by gillnet in 1998 (Almeida and Carneiro, 1999; Nunes et al., 2005) and another adult female caught in 2009, which was reported aborting 20–25 young (Jorge Nunes, pers. comm., 2009). To the south, documented reports are far more limited and historic in nature, with only one record per state. The southernmost record is a historical specimen from São Paulo (Ribeiro, 1918 in Bigelow and Schroeder, 1953; Gadig, 1998). At present, there is no quantitative information on the current status of P. pristis in northern Brazil. Fishermen have Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) J. FERNANDEZ-CARVALHO ET AL. Table 7. Records of Pristis pristis from Africa. North to south: MRT = Mauritania, SEN = Senegal, SENG = Senegambia, GMB = Gambia, GNB = Guinea Bissau, GIN = Republic of Guinea, SLE = Sierra Leone, LBR = Liberia, CIV = Côte D’Ivoire, COG = Congo, COD = Democratic Republic of the Congo, AGO = Angola. LIT = literature, MUS = museum N Country Location Year RTL (cm) Source Museum Catalogue Number Nothern West Africa 1 MRT Tidra 1 1 1 1 1 1 1 1 1 1 1 SEN SEN SEN SEN SEN SEN SEN SEN SEN SEN SEN 2 SENG Pre-1885 1 1 1 GMB GMB GMB Pre-1909 Pre-2005 - 1 1 3 2 2 1 1 1 GNB GNB GNB GNB GNB GNB GNB GNB 1 1 1 1 GIN GIN GIN GIN Soro Kamsar Bongolon 1 1 SLE SLE Bonthe 1 1 LBR LBR 1 1 CIV CIV Fadiouth Casamance Joal Joal Bissagos Arch. Bolama Bolama Bissagos Arch. Bissagos Arch. Gêba River, Gêba Sassandra Sassandra 1980s LIT Robillard and Séret (2006) LIT LIT LIT LIT LIT LIT LIT LIT LIT MUS MUS Bigelow and Schroeder (1953) Bigelow and Schroeder (1953) Bigelow and Schroeder (1953) Bigelow and Schroeder (1953) Bigelow and Schroeder (1953) Bigelow and Schroeder (1953) Robillard and Séret (2006) Robillard and Séret (2006) Ballouard et al. (2006); Robillard and Séret (2006) MNHN A-9699 IFAN R977 LIT Bigelow and Schroeder (1953) LIT LIT MUS Bigelow and Schroeder (1953) Ballouard et al. (2006); Robillard and Séret (2006) BMNH 1885.1.31.30 LIT LIT LIT/MUS LIT/MUS LIT/MUS LIT LIT MUS Ballouard et al. (2006) Robillard and Séret (2006) Ballouard et al. (2006); Robillard and Séret (2006) Ballouard et al. (2006); Robillard and Séret (2006) Ballouard et al. (2006); Robillard and Séret (2006) Ballouard et al. (2006) Ballouard et al. (2006); Robillard and Séret (2006) BMNH 1912.4.1.1 LIT LIT LIT LIT Ballouard et al. (2006); Robillard and Séret (2006) Robillard and Séret (2006) Ballouard et al. (2006); Robillard and Séret (2006) Ballouard et al. (2006); Robillard and Séret (2006) LIT MUS Ballouard et al. (2006) BMNH 2004.11.27.51 1881 1927 MUS MUS RMNH D3075 RMNH D3078 1902 Pre-1923 MUS LIT MNHN 1902-255 Bigelow and Schroeder (1953) Pre-1841 Pre-1861 Pre-1865 Pre-1870 Pre-1882 Pre-1902 1980s 1990 Pre-2005 - 1980 1980´s 1983 2003 2004 2005 Pre-2005 1965 1980´s 1983 1989 2003 - 19 106 105 11 117 92 92 9 86 108 89 74 8 Southern West Africa 3 COG 1 Pointe-Noire 1958 CongoBasin 80-103 MUS MNHN 2003–2612 to 2613 1951 MUS MRAC 66639 2 1 1 COD COD COD Banana Banana Congo R. estuary 1951 1959 - MUS MUS MUS MRAC 68038 and 68039 MRAC A4-45-P-12 MRAC-A4-45-P-2 1 1 AGO AGO Luanda Luanda 1951 MUS MUS MRAC 74723 ZMB 16109 Copyright # 2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) STATUS OF THE LARGETOOTH SAWFISH IN THE ATLANTIC OCEAN reported a decline in the region beginning as early as the late 1980s (Charvet-Almeida, 1999, 2002). Qualitative evidence against population extinction did not support local extinction for Brazil (Table 3). It should be noted that although we acknowledge the existence of some research in the area when compared with other subregions of the Atlantic, surveys and field work were considered to be inadequate in the qualitative analyses due to the vast extent and complex morphology of the area. Regarding the quantitative analysis, given the recent records of P. pristis from the Amazon estuary, the probability of extinction is 0.11 (±0.21). Africa is 0.25 (±0.20). Accordingly, qualitative evidence against population extinction did not support local extinction for this subregion (Table 3). However, in southern areas of West Africa (Cameroon to Namibia) qualitative evidence seems to be stronger for extinction than against, although uncertainty because of the absence of data in this subregion should be acknowledged. It is likely that P. pristis has been extirpated (0.99 ± 0.77) but the upper limit of the projected time for regional extinction calculated was highly variable, probably a result of the lack of records (2014–2044). West Africa Historical museum specimens and literature records of P. pristis on the west coast of Africa exist from Mauritania (n=1), Senegal (n = 11), Gambia (n = 3), Guinea-Bissau (n = 12), Republic of Guinea (n = 4), Sierra Leone (n = 2), Liberia (n = 2), Côte d’Ivoire (n = 2), Congo (n = 3), Democratic Republic of the Congo (n = 4), and Angola (n = 2) (Table 7). In addition there is a 1951 record from the ‘Congo Basin’ that could be from either Congo, Democratic Republic of the Congo or Angola. Virtually all these records lack definitive locality and date, and specimens from this period often have been lost. That said, P. pristis was at one time relatively common along the west coast of Africa, particularly in areas with riverine estuaries. Robillard and Séret (2006) reported the unpublished historical notes gathered by Jean Cadenat, founder of the Musée de la Mer in Gorée (Senegal) in the 1950s. In addition, the same authors and Ballouard et al. (2006) examined 13 P. pristis rostra from 1965–2005 found at various locales along the coast of West Africa. Their reviews of the recent status of sawfishes from Mauritania to Republic of Guinea revealed that sawfishes, P. pectinata and P. pristis (referred to as P. microdon), were relatively common in the past but are now rarely caught or observed. The last documented regional capture the authors found was from 2005 in Nord de Caravela at Guinea-Bissau. Because of these recent records in areas around Guinea-Bissau, the likelihood that P. pristis is extirpated from areas in northern West Copyright # 2013 John Wiley & Sons, Ltd. DISCUSSION Historically, P. pristis ranged from the Gulf of Mexico to south-east Brazil in the western Atlantic, and from Mauritania to Angola in the eastern Atlantic. Over the last century this distribution has been considerably truncated and abundance levels appear to have significantly decreased, based on encounter data, museum records, literature, personal communication with scientific colleagues, and quantitative and qualitative analyses performed. Nevertheless, it should be acknowledged that even though a thorough and intensive search for records was conducted in the present study, there are catches and sightings that go unreported, especially in more isolated regions of Central and South America and of Africa. Even in the regions where sawfish are still caught or landed periodically, owing to remoteness and consequent lack of documentation very few records reach the media and very often only large specimens are reported since juveniles receive less attention. As a result, information reported in this study should be used with caution. That said, the decline of P. pristis seems to follow a similar pattern exhibited by P. pectinata in the USA, with the extremes of its distributional ranges having sequentially disappeared in the face of fishing pressure and habitat loss (NMFS, 2009a, b). Historical centres of abundance of P. pristis were generally equatorial and influenced by water temperatures, but the presence of large rivers and Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) J. FERNANDEZ-CARVALHO ET AL. riverine and lacustrine estuaries appears to have been equally essential. Pristis pristis seems to prefer freshwater and brackish water habitats more than its relative P. pectinata, although adults frequently occur in higher salinity, nearshore continental waters (Compagno and Cook, 1995; Zorzi, 1995). As is the case with P. pectinata (Poulakis and Seitz, 2004; Carlson et al., 2007), in addition to estuaries and inshore bars, the presence of mangroves seems to be an indicator of good habitat for P. pristis, since several records occurred in areas known to have mangroves (e.g. Amazon river estuary, Bissagos Archipelago). Furthermore, it is possible that the habitat of the Amazon estuary and the Bissagos Archipelago in Guinea-Bissau, might have served as refuges for the species, offering protection in the form of complex coastlines heavily indented by inlets and extensive mangroves. United States of America All P. pristis captured in US waters were large individuals occurring at the northern extremity of their distributional range only during warm water months (May to October). It is apparent that this species historically was represented in the Gulf of Mexico by seasonal transients that took advantage of Texas waters, only occasionally wandering eastward to Louisiana and Florida. Texas historically has had more documented P. pristis encounters than any geographic region outside of Nicaragua and Brazil, although this probably is partially an artefact of the maturation of regional journalism and fishing pressure (both commercial and recreational) between 1917 and 1961. Nevertheless, when considered in conjunction with its regularity of occurrence before its marked reduction in abundance, it appears that at one time the western Gulf of Mexico was an important area for this species, especially for adults and subadults. The highest number of P. pristis records in Texas was from the north-east coast, followed by the central coast, with few records from the south-western region adjacent to the Mexican border. Although it might be expected that the south-west would have more records, the lower number of encounters may correspond with the historical freshwater inflow Copyright # 2013 John Wiley & Sons, Ltd. pattern of the region. This pattern has a gradient from highest inflows in the north-east to lowest along the south-west coasts of the state (Longley, 1994). Freshwater inflow is important to productivity of the region (TDWR (Texas Department of Water Resources), 1982) and the influence of lower salinity water is especially important to P. pristis (Thorson, 1974). Thus the presence of higher-flow riverine estuaries may have played a role in attracting larger numbers of P. pristis to the Galveston–Texas City–Freeport and Corpus Christi–Aransas regions, than areas to the south-west. However, in the absence of catch-per-unit-effort data, the encounter records may simply be reflective of higher fishing efforts associated with higher human densities along the north-east and central Texas coastline. The ratio of P. pectinata to P. pristis records from Texas in the ISED is roughly 1:1, so it is likely that scientific literature and print media reports from the region currently unattributed to species represents both species in approximately equal numbers. This is consistent with the observations of fishermen who told Baughman (1943) that the two species were present in equal proportions. Perhaps half of the 178 Pristis spp. records found for Texas could represent P. pristis records, highlighting the former presence of the species in Texas waters. Unlike P. pectinata, which has a refugium in the waters of southern Florida, P. pristis apparently is extinct in US waters, with last documented records in Florida in 1941, Louisiana in 1917, and Texas in 1961. Since the evidence suggests that P. pristis encountered in US waters were probably seasonal migrants from core areas of their range, it is possible that the decline of the species in this peripheral region of its range is a result of its decline in Central America, with a smaller core population translating to a smaller number of seasonal migrants. In addition, extensive commercial shrimp trawling activities and recreational fishing may have played a role in the decline of the species in US waters. Mexico and Central America Given the habitat available along the Mexico and Central America coastlines, it might have been expected that more historical records of P. pristis Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) STATUS OF THE LARGETOOTH SAWFISH IN THE ATLANTIC OCEAN would have been found. However, the potential for taxonomic confusion with P. pectinata is high, and scientific collecting and popular reporting during times of former relative abundance may have caused P. pristis to be under-represented in museums and the media. A unifying theme presented by regional biologists is ‘no information available’. While this may be partially attributable to lack of regional data on P. pristis, even those with good knowledge of local fauna and fisheries indicate that the species is seldom if ever encountered. The lack of flowing rivers in some areas of the region (especially the Yucatán) may have limited available habitat for the species. However, it is clear that the decline observed in Texas has also occurred to the south. The most recent sawfish capture in Mexico (1998) was the first encounter in 15 years of observing artisanal fisheries (J. L. Castillo-Geniz, pers. comm., 2009). The importance of the rivers of Costa Rica and Nicaragua to P. pristis cannot be overestimated. Clearly this was and potentially still is a core area that provides both nursery and adult habitat critical to P. pristis (Thorson, 1976, 1982a, b; Bussing, 2002; McDavitt, 2002). It also may very well have served as a population source for adult movements to the north and south-east, much as south Florida has served that function for P. pectinata movements along the US Atlantic coast (and probably north-eastern Gulf of Mexico) (NMFS, 2009b). The decline in this region has occurred as a direct result of overfishing as the species was heavily fished along this lake–river system in the 1970s, reaching population collapse in the 1980s (Thorson, 1982a; McDavitt, 2002). As a result, the Nicaraguan government imposed a temporary moratorium on targeted fishing for sawfish in Lake Nicaragua in the early 1980s (Thorson, 1982a). However, despite the ban sawfishes continued to be killed as bycatch in gillnets (McDavitt, 2002). For example, a local artisanal fisher indicated he usually captures four to six sawfishes per year (1998 interview, McDavitt, 2002). Protection was strengthened in 2006 with a Nicaraguan ban on fishing for sawfish, but only in Lake Nicaragua (Charvet-Almeida et al., 2007) and this ban still does not decrease the bycatch in other fisheries. It Copyright # 2013 John Wiley & Sons, Ltd. is probably because of the problem of bycatch in fisheries and the low productivity of sawfish (Simpfendorfer, 2000) that recovery of P. pristis will be slow in the region. South America There is little indication that P. pristis are found in anything other than low and declining numbers across the northern rim of South America from Colombia to the Guianas or in the continental islands of Aruba, Curaçao and Trinidad. One would expect that the Orinoco River in Colombia and Venezuela would have been an important area but records fail to pinpoint this area as equal in importance to higher abundance areas to the west and south, particularly the Amazon River basin. Nevertheless, the lack of recent P. pristis records in northern South America may be a partial artefact of limited reporting, but given the declines of this species throughout the remainder of its Atlantic range, and the patterns exhibited by other pristid species worldwide, it seems safe to interpret an occurrence pattern lacking in recent observations as a significant decline in abundance. However, further studies are needed to permit a more complete assessment of the area. By contrast, although information about historical and current levels of P. pristis abundance in Brazil is unavailable, recent records, personal accounts and qualitative analysis (Charvet-Almeida and Faria, 2008; Charvet, pers. comm., 2008/2009; Faria et al., 2013), indicate that the Amazon estuary is presently the location with the highest number of recent records and more information available for the species in the Atlantic, with the present distribution being possibly restricted to the states of Amapá, Pará and Maranhão (and maybe Amazonas). The remaining population of P. pristis in northern Brazil has faced several threats in past decades (between late 1990s and mid-2000s). The species has been a major component of bycatch in commercial and artisanal fisheries (Charvet-Almeida and Faria, 2008) with meat and fins traded in markets (Charvet-Almeida, 1999, 2002; McDavitt and Charvet-Almeida, 2004). In addition, small saws were sold as curios for tourists and as folk medicine Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) J. FERNANDEZ-CARVALHO ET AL. to treat asthma (Charvet-Almeida, 1999, 2002) while large saws brought high prices in international trade (Charvet-Almeida, 2002; McDavitt and Charvet-Almeida, 2004; Charvet-Almeida and Faria, 2008). Additional threats included habitat destruction and modification and the interest of the ornamental trade (aquaria) in P. pristis neonates (Charvet-Almeida and Faria, 2008). Since the mid-2000s, modifications to Brazilian legislation may have enabled some protection to P. pristis in northern Brazil. In 2004, P. pristis became a protected species under the Brazilian legislation (IN – MMA # 5), and capture or trade of any kind is prohibited. In addition, at least three subsequent new rules regulating gillnet fisheries (IBAMA # 166; 2007), and banning ornamental trade (IN- MMA # 202; 2008) and shark finning (INI - MPA # 14; 2012) may also provide further protection. Finally, several marine protected areas in northern Brazil may offer protection to the species (Charvet-Almeida and Faria, 2008). In summary, the current protection measures in place for P. pristis in Brazil seem to be stronger than in the preceding decades, although enforcement is expected to be difficult in that region. For example, recently P. pristis (as P. perotteti) meat was on sale in two fish markets of Pará labelled as ‘shark’ fillets even though its sale in Brazil is prohibited (Melo Palmeira et al., 2013). West Africa The west coast of Africa encompasses a large geographic area and though museum records and recent directed sawfish literature are available, reports are particularly rare in this region. Historically, however, certain parts of the west coast of Africa yielded higher catches of sawfishes, especially Tidra (Mauritania), Casamance (Senegal), the Bissagos Archipelago (Guinea-Bissau) and Kamsar (Republic of Guinea), although currently the species is largely restricted to the Bissagos Archipelago (Guinea-Bissau) (Ballouard et al., 2006; Robillard and Séret, 2006). Unfortunately, the lack of size data for sawfish records prevent speculation on any migrations that might occur in the eastern Atlantic. Copyright # 2013 John Wiley & Sons, Ltd. Fishers indicated that sawfishes were once common (‘a few decades ago’) on the west coast of Africa, with catches of one or two specimens per net made regularly and even up to 10 specimens captured occasionally (Robillard and Séret, 2006). Fisheries, human population increase along rivers and habitat degradation have been described as the causes of the drastic disappearance of sawfishes in the region within the last three decades (Ballouard et al., 2006; Robillard and Séret, 2006). Although there have been sporadic reports of sawfish captures in 2006 and 2008 from the islands of Orango and Formosa in Guinea-Bissau, there are no pictures or evidence to identify the sawfish species (J. Huet, pers. comm., 2009). The fact that Guinea-Bissau’s coastal populations have no strong fishing tradition when compared with the neighbouring countries may have, in part, helped protect the species in the region. However, in the last two decades an increasing number of fishermen from other nearby countries have been attracted to the rich Bissagos Archipelago, with most participating in the shark fishing fleet (Campredon and Cuq, 2001). Thus, urgent protection for sawfish is needed for Guinea-Bissau, and in particular for the Bissagos Archipelago. CONCLUSIONS Pristis pristis appears to be threatened in all of its six distribution areas in the Atlantic. Extinction risk was detected quantitatively for three of these areas: the USA, northern South America and southern West Africa. The US population is now probably extinct, since the best scenario for estimated date of extinction was the present decade and the qualitative evidence supported local extinction in this region. Regarding northern South America and southern West Africa, the results must be taken with caution, as even though quantitative analyses were robust enough to indicate extinction risks, the estimated date of extinction encompassed a wide time-span or the qualitative evidence for extinction was not strong. Although P. pristis has been extirpated from several areas, the species can still be found in some regions of the Atlantic. The Amazon Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) STATUS OF THE LARGETOOTH SAWFISH IN THE ATLANTIC OCEAN estuary in Brazil appears to have the highest remaining abundance of P. pristis in the Atlantic, although there are indications of low presence of the species in the Colorado River–San Juan River system in Nicaragua and Costa Rica, and in the Bissagos Archipelago in Guinea-Bissau. Evidence from the literature suggests that the decline of P. pristis in the Atlantic is mostly associated with fisheries activities, both direct (target) and indirect fisheries (bycatch), principally commercial (industrial) and artisanal trawling and longline fisheries. In addition, habitat degradation may also have played a role in this decline, since it has been associated as a secondary cause of depletion of other sawfish species that occur in the Atlantic (e.g. P. pectinata). Conservation and management implications The depletion of P. pristis throughout its Atlantic range is of concern. The situation is even more dramatic when considered in the context of the biological limitations of the species: long life span, late sexual maturity, and low fecundity, which in combination result in low reproductive potential and reduced capacity for recovery following population depletion (Thorson, 1976, 1982a, b; Simpfendorfer, 2000). The recovery of the related P. pectinata in the USA is estimated to require 100 years even with zero fishing mortality and concerted action to conserve critical habitat during that long time period (NMFS, 2009b). Although P. pristis shares similar biological attributes, conservation efforts are predicted to produce an even slower recovery for P. pristis than for P. pectinata populations owing to its lower intrinsic rate of increase caused by their smaller litter size (Simpfendorfer, 2000). In addition, given that some of the remaining refugia of P. pristis are located in areas with currently minimal to no legal protection and difficult enforcement, a similar recovery time frame for this species would probably be over-optimistic. Even though it is very encouraging that international recognition of the need for conservation and management measures already exists for the species (e.g. CITES, IUCN Red List), it is crucial that measures at national or regional levels are created Copyright # 2013 John Wiley & Sons, Ltd. and that legal protection is enforced soon, if the extinction of the species in the Atlantic Ocean is to be prevented. The three areas that still have potentially viable populations of P. pristis (i.e. Amazon estuary, Colorado River–San Juan River system and Bissagos Archipelago) should be considered priority areas for conservation and law enforcement. Areas with only historical abundance records should also be taken into consideration in any conservation actions, as these may serve as refugia for P. pristis, and may be vital for the potential recovery of the species. To our knowledge, only four countries within the Atlantic range of P. pristis have national protection to date. In the USA the species has been listed under the Endangered Species Act since 2011 making it illegal to capture, harass or harm any specimen (NMFS, 2009a). In Mexico, there is a fishery ban since 2006 on all take of sawfish (Fordham, 2012). In Nicaragua, a fishery ban on targeted sawfish on Lake Nicaragua was implemented in 2006 although it does not address bycatch (Charvet-Almeida et al., 2007; Fordham, 2012). Finally, in Brazil, a federal law by the Ministry of Environment (MMA-IN05/2004) mandates no take of smalltooth and largetooth sawfish since 2004. No national laws for countries in the west coast of Africa were found. Although it is positive that laws protecting P. pristis already exist in the countries identified in this study as the areas with the largest remaining populations of the species (Brazil and Nicaragua), enforcement is still a critical problem (Charvet-Almeida et al., 2007; Melo Palmeira et al., 2013). Although protective laws are the first step in the conservation of a species, more conservation measures should follow in order to achieve the desired recovery of a slow-growing species such as P. pristis. For example, as was done in the USA for the smalltooth sawfish (NMFS, 2009b), it is advised that national recovery plans be developed and implemented for P. pristis in order to assess how to minimize human interactions (injury and mortality), identify the critical habitat for the species, and protect (or restore) those habitats to promote considerable increase in abundance to allow the Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) J. FERNANDEZ-CARVALHO ET AL. species to reoccupy areas from which it has been extirpated. Thus, research on P. pristis should be encouraged in order to gather information crucial for developing such recovery plans, namely attempt to increase the available knowledge about the basic biology of the species (reproduction, growth rates, movement patterns, etc.), assess the threats quantitatively (e.g. fishing effort, landing data) and identify possible nursery areas which, if protected, can facilitate adult recruitment. Finally, public outreach and education would be useful tools to aid the recovery of this endangered species. M. G. G. Grol (Department of Animal Ecology and Ecophysiology, Radboud University, Aruba), J. Huet ("Poisson-Scie", Noé Conservation/FIBA, Guinea-Bissau), D. Matos-Caraballo (DNER Fishery Research Laboratory, Puerto Rico), M. M. Carranza (ECOSUR, Mexico), J. Nunes (Universidade Federal do Maranhão, Brazil), F. P. Amargós (CIEC, Cuba), J. C. P. Jiménez (ECOSUR, Mexico), L. Py-Daniel (INPA, Brazil), and R. Tavares (Instituto Nacional de Investigaciones Agrícolas, Venezuela). Finally, we would like to thank the anonymous reviewers whose valuable comments greatly improved this manuscript. ACKNOWLEDGEMENTS REFERENCES Funding to support this project was provided by the National Marine Fisheries Service. During part of this study Joana Fernandez-Carvalho was supported by a grant from FCT, the Portuguese Foundation for Science and Technology (Ref: BD 60624/2009). We appreciate the assistance provided by many friends and colleagues, especially D. Bethea, S. Norton, G. Poulakis, and T. Wiley. The hard work and patience of FPSR staff members and volunteers C. Bester, S. Bühler, C. Campbell, F. Carvalho, R. Coelho, and J. Zill is gratefully acknowledged. We thank the following curators of museum, university, and private ichthyological collections who responded to our requests for information on their sawfish holdings: A. Bentley, B. Brown, D. Catania, P. Chakrabarty, P. Charvet, M. Dressler, N. Douglas, N. Ferre, A. Harold, K. Hartel, G. Hogue, B. Kuhajda, S. LeCroy, J. MacLaine, M. McDavitt, R. Norton, C. Opitz, H. Prestridge, P. Pruvost, N. Rios, M. Rogers, J. Rosales, M. Sabaj, A. Sanderson, J. Seigel, J. Seitz, B. Séret, and J. Williams. We thank the following researchers for their valuable anecdotal accounts of P. pristis in the state or country in which they work: A. Acero (Invemar-Instituto de Investigaciones Marinas y Costeras, Colombia), R. Burris (MS Dept of Marine Resources), J.L. Castillo-Geniz (INAPESCA, Mexico), A. O. Debrot (Carmabi Foundation, Curaçao), H. Guzmàn (Smithsonian Tropical Research Institute, Panama), F. Gomez (Departamento de Biologia Ponteficia Universidad Javeriana, Colombia), Acero PA, Alvarez-León R, Garzón J. 1986. Peces comerciales del caribe colombiano, sin incluir los de las islas, cayos y bajos del caribe occidental (propuesta para la unification de sus nombres vernaculars). Informe Museo del Mar, Universidad de Bogota 31: 1–15. Almeida ZS, Carneiro MC. 1999. Levantamento e ocorrência de elasmobrânquios capturados pela pesca artesanal no litoral do Maranhão. Ceuma Perspectivas 3: 122–136. Almeida ZS, Nunes JLS, Paz AC. 2006. Elasmobrânquios no Maranhão: biologia, pesca e ocorrência. In Projeto e ações em Biologia e Química, Costa da Silva A, Martins Bringel JM. (eds). EDUUEMA: São Luís; 35–57. Alvarez LR, Blanco RJ. 1985. Composition of fish communities in the lagoon and estuarine complex of Cartagena Bay, Cienaga de Tesca and Cienaga Grande of Santa Marta, Colombian Caribbean. In Fish Community Ecology in Estuaries and Coastal Lagoons: Towards an Ecosystem Integration, Yáñez-Arancibia A (ed.).UNAM Press: Mexico City; 535–656. Ballouard JM, Robillard M, Yvon C. 2006. Statut et conservation des poissons-scies et autres chondrichtyens menacés en Afrique de l’Ouest: Rapport de synthèse. Rapport Missions 2005 et 2006, Poissons-scies. Noé Conservation – CSRP. Bates M. 1964. The Land and Wildlife of South America. Time Inc: New York. Baughman JL. 1943. Notes on sawfish, Pristis perotteti Müller and Henle, not previously reported from the waters of the United States. Copeia 1943: 43–48. Baughman JL. 1952. The marine fishes of Texas. The sawfishes. Texas Game and Fish 10: 28–29. Bigelow HB, Schroeder WC. 1953. Sawfishes, guitarfishes, skates and rays. In Fishes of the Western North Atlantic, Part two., Tee-Van J, Breder CM, Parr AE, Schroeder WC, Schultz LP (eds). Memoirs of Sears Foundation for Marine Research: Yale, University, New Haven; 1–514. Burgman MA, Grimson RC, Ferson S. 1995. Inferring threat from scientific collections. Conservation Biology 9: 923–928. Bussing WA. 2002. Freshwater Fishes of Costa Rica. Universidad de Costa Rica: San José, Costa Rica. Butchart SHM, Stattersfield AJ, Brooks TM. 2006. Going or gone: defining ‘possibly extinct’ species to give a true Copyright # 2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) STATUS OF THE LARGETOOTH SAWFISH IN THE ATLANTIC OCEAN picture of recent extinctions. Bulletin of the British Ornithologists’ Club 126A: 7–24. Caldas JP; Díaz-Trujillo EM; García CB, Duarte LO. 2009. Revisión Histórica de la Pesca de Tiburones y Rayas en el Mar Caribe Continental de Colombia. In Avances en el Conocimiento de Tiburones, Rayas y Quimeras de Colombia, Puentes V, Navia AF, Mejía-Falla PA, Caldas JP, Diazgranados MC, Zapata LA (eds). Fundación Squalus, Ministerios de Ambiente, Vivienda y Desarrollo Territorial, Instituto Colombiano Agropecuario, Colciencias, Conservación Internacional, WWF; 99–130. Campredon P, Cuq F. 2001. Artisanal fishing and coastal conservation in West Africa. Journal of Coastal Conservation 7: 91–100. Carlson JK, Osbourne J, Schmidt TW. 2007. Monitoring the recovery of smalltooth sawfish, Pristis pectinata, using standardized relative indices of abundance. Biological Conservation 136: 195–202. Carvalho MR, McEachran JD. 2003. Family Pristidae (sawfishes). In Check List of the Freshwater Fishes of South and Central America, Reis RS, Kullander SO, Ferraris SJ (eds), Pontifícia Universidade Católica do Rio Grande do Sul, Porto Alegre, RS: Brasil; 17–21. Castro-Aguirre JL. 1978. Catalogo sistemático de los peces marinos que penetran a las aguas continentales de México con aspectos zoogeográficos y ecológicos. Dirección General del Instituto Nacional de Pesca, Serie Científica 19: 1–296. Castro-Aguirre JL. 1999. Ictiofauna estuarino-lagunar y vicaria de México. Editorial Limusa S.A. de C.V.: México D.F, México. Cervigón MF. 1966. Los Peces Marinos de Venezuela, Tomo II. Estación de Investigaciones Marinas de Margarita, Fundación La Salle de Ciencias Naturales: Caracas, Venezuela. Cervigón MF. 1993. Los peces marinos de Venezuela, Vol 2, 2nd edn. Edit. Fundación Científica Los Roques: Caracas. Charvet-Almeida P. 1999. General information from sawfishes in Pará state, north Brazil. Boletim SBEEL 4: 12. Charvet-Almeida P. 2002. Sawfish trade in the north of Brazil. Shark News 14: 9. Charvet-Almeida P, Faria V. 2008. Pristis perotteti. In Livro Vermelho da Fauna Brasileira Ameaçada de Extinção (Série Biodiversidade), Machado ABM, Drummond GM, Paglia AP (eds). Belo Horizonte: Fundação Biodiversitas II; 33–35. Charvet-Almeida P, Faria V, Furtado M, Cook SF, Compagno LJV, Oetinger MI. 2007. Pristis perotteti. In IUCN 2010. IUCN Red List of Threatened Species. Version 2010.4. (Available from: http//www.iucnredlist.org). Compagno LJV, Cook SF. 1995. The exploitation and conservation of freshwater elasmobranches: status of taxa and prospects for the future. Journal of Aquariculture and Aquatic Sciences 7: 62–90. Dahl G. 1971. Los Peces del Norte de Colombia. Ministerio de Agricultura, Instituto de Desarrollo de los Recursos Naturales Renovables: Bogota, Colombia. Dulvy NK, Sadovy Y, Reynolds JD. 2003. Extinction vulnerability in marine populations. Fish and Fisheries 4: 25–64. Faria VV, McDavitt MT, Charvet P, Wiley TR, Simpfendorfer CA, Naylor GJP. 2013. Species delineation and global population structure of critically endangered sawfishes (Pristidae). Zoological Journal of the Linnean Society 167: 136–164. Flórez-González L. 1986. Lista preliminar de las especies ícticas marinas para el Caribe colombiano. Informe Museo del Mar 32: 1–101. Copyright # 2013 John Wiley & Sons, Ltd. Fordham S. 2012. Conservation status of sawfishes. Global Sawfish Conservation Strategy Workshop, London, May 2012. Fowler HW. 1910. Notes on batoid fishes. Proceedings of the Academy of Natural Sciences of Philadelphia 62: 468–475. Fowler HW. 1919. Notes on tropical American fishes. Proceedings of the Academy of Natural Sciences of Philadelphia 71: 128–155. Fowler HW. 1936. The marine fishes of West Africa: based on the collection of the American Museum Congo Expedition, 1909–1915. Bulletin of the American Museum of Natural History 70: 1–606. Fowler SL, Cavanagh RD, Camhi M, Burgess GH, Cailliet GM, Fordham S, Simpfendorfer CA, Musick JA. 2005. Sharks, Rays and Chimaeras: The Status of the Chondrichthyan Fishes. IUCN/SSC Shark Specialist Group, IUCN: Gland, Switzerland. Furneau R. 1969. The Amazon. G. P. Putnam’s and Sons: New York. Gadig OBF. 1998. Peixes cartilaginosos da costa do estado de São Paulo. Revista Ceciliana 8: 41–51. Gill T, Bransford JF. 1877. Synopsis of the fishes of Lake Nicaragua. Proceedings of the Academy of Natural Sciences of Philadelphia 1877: 175–191. Grijalba-Bendeck M, Acero A, Díaz-Trujillo EM, Gómez F. 2009. Estado del conocimiento de los peces cartilaginosos del Caribe continental de Colombia. In Avances en el Conocimiento de Tiburones, Rayas y Quimeras de Colombia, Puentes V, Navia AF, Mejía-Falla PA, Caldas JP, Diazgranados MC, Zapata LA (eds). Fundación Squalus, Ministerios de Ambiente, Vivienda y Desarrollo Territorial, Instituto Colombiano Agropecuario, Colciencias, Conservación Internacional, WWF; 41–96. Grogan CS, Boreman J. 1998. Estimating the probability that historical populations of fish species are extirpated. North American Journal of Fisheries Management 18: 522–529. Hoover JJ. 2008. Searching for sawfish: a history of the hunt. American Currents 34: 1–15. Kyne PM, Carlson J, Smith K. 2013. Pristis pristis In: IUCN 2013. IUCN Red List of Threatened Species. Version 2013.1. www.iucnredlist.org. Downloaded on 29 August 2013. Longley WL. 1994. Freshwater inflows to Texas bays and estuaries: ecological relationships and methods for determination of needs. Texas Water Development Board and Texas Parks and Wildlife Department: Austin, TX. Luiz OJ, Edwards AJ. 2011. Extinction of a shark population in the Archipelago of Saint Paul’s rocks (equatorial Atlantic) inferred from the historical record. Biological Conservation 144: 2873–2881. Marden L. 1944. A land of lakes and volcanoes. National Geographic 86: 161–192. Marlier G. 1967. Hydrobiology in the Amazon region. In Atas do Simpósio sobre a Biota Amazónica Vol. 3: Limnologia, Lent H (ed.). Conselho Nacional de Pesquisas: Rio de Janeiro. McDavitt MT. 2002. Lake Nicaragua revisited: conversations with a former sawfish fisherman. Shark News 14: 5. McDavitt M, Charvet-Almeida P. 2004. Quantifying trade in sawfish rostra: two examples. Shark News 16: 10–11. McInerny GJ, Roberts DL, Davy AJ, Cribb PJ. 2006. Significance of sighting rate in inferring extinction and threat. Conservation Biology 20: 562–567. Meek SE. 1907. Synopsis of the fishes of the great lakes of Nicaragua. Field Columbia Museum Publication 121, Zoology Series 7: 97–132. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013) J. FERNANDEZ-CARVALHO ET AL. Mejía LS, Acero AP. 2002. Libro rojo de peces marinos de Colombia. INVEMAR, Universidad Nacional de Colombia: Bogota. Melo Palmeira CA, da Silva Rodrigues-Filho LF, de Luna Sales JB, Vallinoto M, Schneider H, Sampaio I. 2013. Commercialization of a critically endangered species (largetooth sawfish, Pristis perotteti) in fish markets of northern Brazil: authenticity by DNA analysis. Food Control 34: 249–252. Miles C. 1945. Some newly recorded fishes from the Magdalena River system. Caldasia 3: 453–464. Miles C. 1947. Los peces del río Magdalena. Ministerio de la Economía Nacional, Sección de Piscicultura, Pesca y Caza, Bogotá. Moncada MJ. 2006. Prohíben pesca de tiburón y pez sierra. La Prensa (17 January 2006). Monte-Luna P, Lluch-Belda D, Serviere-Zaragoza E, Carmona R, Reyes-Bonilla H, Aurioles-Gamboa D, Castro-Aguirre JL, del Proo SAG, Trujillo-Millan O, Brook BW. 2007. Marine extinctions revisited. Fish and Fisheries 8: 107–122. Monte-Luna P, Castro-Aguirre JL, Brook BW, Cruz-Aguero J, Cruz-Escalona VH. 2009. Putative extinction of two sawfish species in Mexico and the United States. Sociedade Brasiliera de Ictiologia: Neotropical Ichthyology 7: 59–512. Montoya RV, Thorson TB. 1982. The bull shark (Carcharhinus leucas) and largetooth sawfish (Pristis perotteti) in Lake Bayano, a tropical man-made impoundment in Panama. Environmental Biology of Fishes 7: 341–347. NMFS (National Marine Fisheries Service). 2009a. Listing Endangered and Threatened Species and Designating Critical Habitat: Notice of Finding on a Petition To List the Largetooth Sawfish (Pristis perotteti) as an Endangered or Threatened Species Under the Endangered Species Act. Federal Register 74(144): 37671–37674. NMFS (National Marine Fisheries Service). 2009b. Recovery Plan for Smalltooth Sawfish (Pristis pectinata). Prepared by the Smalltooth Sawfish Recovery Team for the National Marine Fisheries Service. Silver Spring: Maryland. Nunes JLS, Almeida ZS, Piorski NM. 2005. Raias capturadas pela pesca artesanal em águas rasas do Maranhão - Brasil. Arquivos de Ciências do Mar, Fortaleza 38: 49–54. Poulakis GR, Seitz JC. 2004. Recent occurrence of the smalltooth sawfish, Pristis pectinata (Elasmobranchiomorphi: Pristidae), in Florida Bay and the Florida Keys, with comments on sawfish ecology. Florida Scientist 67: 227–235. Puyo J. 1949. Poissons de la Guyane Française. XII: Faune de l’Empire Français. Office de la Recherche Scientific Outre-Mer, Librairie Larose: Paris, France. Ramjohn DD. 1999. Check list of coastal and marine fishes of Trinidad and Tobago. Marine Fishery Analysis Unit, Fisheries Division, Ministry of Agriculture, Land and Marine Resources, Trinidad and Tobago, Fisheries Information Series 8. Copyright # 2013 John Wiley & Sons, Ltd. Rivadeneira MM, Hunt G, Roy K. 2009. The use of sighting records to infer species extinctions: an evaluation of different methods. Ecology 90: 1291–1300. Roberts DL. 2006. Extinct or possibly extinct? Science 312: 997–998. Roberts DL, Solow AR. 2003. Flightless birds - when did the dodo become extinct? Nature 426: 245–245. Robillard M, Séret B. 2006. Cultural importance and decline of sawfish (Pristidae) populations in West Africa. Cybium 30: 23–30. Santos GM, Val AL. 1998. Peixe-serra: o falso tubarão da Amazônia. Ciência Hoje 23: 66–67. Saunders GB, Holloway AD, Handley CO. 1950. A fish and wildlife survey of Guatemala. US Fish and Wildlife Service, Special Scientific Report 5: 1–162. Simpfendorfer CA. 2000. Predicting population recovery rates for endangered western Atlantic sawfishes using demographic analysis. Environmental Biology of Fishes 58: 371–377. Solow AR. 1993a. Inferring extinction from sighting data. Ecology 74: 962–964. Solow AR. 1993b. Inferring extinction in a declining population. Journal of Mathematical Biology 32: 79–82. Solow AR. 2005. Inferring extinction from a sighting record. Mathematical Biosciences 195: 47–55. Solow AR, Roberts DL. 2003. A nonparametric test for extinction based on a sighting record. Ecology 84: 1329–1332. Springer VG, Woodburn KD. 1960. An ecological study of the fishes of the Tampa Bay area. Florida State Board of Conservation, Professional Paper Series 1: 1–104. Strauss D, Sadler PM. 1989. Classical confidence intervals and Bayesian probability estimates for ends of local taxon ranges. Mathematical Geology 21: 411–421. TDWR (Texas Department of Water Resources). 1982. The influence of freshwater inflows upon the major bays and estuaries of the Texas Gulf Coast: Executive Summary. Texas Department of Water Resources: Austin, TX. Thorson TB. 1974. Occurrence of the sawfish, Pristis perotteti, in the Amazon River, with notes on P. pectinatus. Copeia 1974: 560–564. Thorson TB. 1976. Observations on the reproduction of the sawfish, Pristis perotteti, in Lake Nicaragua, with recommendations for its conservation. In Investigations of the Ichthyofauna of Nicaraguan Lakes, Thorson TB (ed.), University of Nebraska, School of Life Sciences: Lincoln, NE; 641–650. Thorson TB. 1982a. The impact of commercial exploitation on sawfish and shark populations in Lake Nicaragua. Fisheries 7: 2–10. Thorson TB. 1982b. Life history implications of a tagging study of the largetooth sawfish, Pristis perotteti, in the Lake Nicaragua-Río San Juan system. Environmental Biology of Fishes 7: 207–228. Zorzi GD. 1995. The biology of freshwater elasmobranchs: an historical perspective. Journal of Aquariculture & Aquatic Sciences 7: 10–31. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013)