Abstract: Smalltooth saw!sh (Pristis pectinata) are a marine species of high concern since they are listed as critically endangered species by the International Union for Conservation of Nature (IUCN) and listed as endangered under the U.S. Endangered Species Act (ESA). Smalltooth saw!sh are extremely vulnerable to being caught as bycatch due to their long, toothed rostrum that can easily become entangled in !shing nets, especially in shrimp trawling !sheries. Several studies have been conducted on smalltooth saw!sh since being listed under the ESA in 2003, but their vertical movements and use of depth have still not been thoroughly studied. Until recently, saw!sh were thought to stay at a depth of 10 m or less, but studies have now shown that they occupy much deeper depths. Due to these con"icting !ndings in the literature, depth was chosen to be the focus of this study. Furthermore, this study investigates whether sex or individual total length has an in"uence on the percentage of time a saw!sh spends at a particular depth. Satellite telemetry was used to track the movements of the tagged saw!sh (n=14). #e data collected from the pop-o$ archiving satellite tags (PSATs) showed smalltooth saw!sh spend most of their time in shallow water (between 0-8 m), but do frequently occupy deeper depths. Sex was found to be a signi!cant factor, while total length was not. Females were found to spend more time at deeper depths than males. Understanding how saw!sh use depth is important in order to predict population trends and dynamics. Knowing the factors that a$ect a saw!sh’s depth use would be bene!cial to e$orts that manage and conserve this species. With the decline of smalltooth saw!sh and their vulnerability to population loss, more research needs to be conducted to create more e$ective management and recovery plans. Aisthesis Volume 11, 20201 Sex and Length as Predictors of Vertical Movements in Smalltooth Saw!sh (Pristis pectinata) by Taylor Mogavero Introduction Saw!sh are large rays in the Chondrichthyes class and are known for their long, toothed “saw” or rostrum (Harrison & Dulvy, 2014). #eir rostrum is an elongated snout with horizontal rostral teeth that is used for feeding and defense (Poulakis & Seitz, 2004; Whitty et al., 2009). About 20-28% of a saw!sh’s body length comes from its rostrum, and it serves a vital role for capturing and detecting prey (Harrison & Dulvy 2014). #e rostrum is able to !nd and catch prey due to extensive sensory organs that detect minute electrical signals emitted by other animals. A saw!sh can also use its rostrum to stunt or kill !sh by slashing its toothed side into its prey (Harrison & Dulvy, 2014). #e species live in coastal tropical and subtropical waters in both estuaries and freshwater. #ey can be found across the world, but the only saw!sh species currently found in the U.S. is the smalltooth saw!sh (Pristis pectinata). #e average size at birth for a smalltooth saw!sh is 80 cm total length (TL). #e size at maturity is 370 cm TL for females and 340 cm TL for males (Brame et al., 2019). #eir age of maturity is estimated to be 7-11 years for males and females (Carlson & Simpfendorfer, 2015), with a total lifespan around 30 years. Smalltooth saw!sh are yolk-sac viviparous, meaning young are nourished in utero by an external yolk sac and are then born live. It is presumed that their reproduction occurs biennially and their average litter size is 7-14 pups (Feldheim et al., 2017). Due to their long lifespans and late maturity, saw!sh have a slow population growth rate, therefore increasing their vulnerability to and di%culty recovering from population loss. Many of the areas where saw!sh reside are highly threatened habitats such as mangroves or seagrasses, two habitats that have seen a great decline in range over the years. #eir coastal preference also tends to overlap with large cities and areas of high human population density where more activities like !shing occur (Dulvy et al., 2016). #e current geographical distribution of P. pectinata is mainly in the western Atlantic, but they have also been found in the eastern Atlantic (Harrison et al., 2014), and Sex and length as predictors of vertical movements in smalltooth saw!sh (Pristis pectinata) Aisthesis Volume 11, 20202 the largest population is found along the southwest coast of Florida (Poulakis & Seitz, 2004). Smalltooth saw!sh were historically found along the coast of the United States as far north as the Carolinas, in the Gulf of Mexico, the Caribbean Sea, and along the coast as far south as Uruguay (Carlson et al., 2014; Dulvy et al., 2016; Feldheim et al., 2017; Simpfendorfer, 2001; Wiley & Simpfendorfer, 2010). #e smalltooth saw!sh is currently found in less than 20% of its historic range (Dulvy et al., 2016). Now smalltooth saw!sh are only consistently found in the coastal waters of southern Florida with a slow increase or stable population growth, and an estimated population size of only a few thousand (Carlson et al., 2014; Feldheim et al., 2017; Wiley & Simpfendorfer, 2010). Exact population reduction rates are hard to calculate due to limited scienti!c data, but it has been estimated that the population may have declined as much as 95% from the historic stock size (Wiley & Simpfendorfer, 2010). Smalltooth saw!sh tend to stay near or within mangroves and seagrass beds as juveniles (Dulvy et al., 2016; Wiley & Simpfendorfer, 2010). Saw!sh are known to occupy shallow coastal waters typically 10 m deep or less (Carlson et al., 2014), but they can occupy depths deeper than 10 m and can be found at depths up to 122 m (Poulakis & Seitz, 2004). It has been postulated that smaller and immature saw!sh are more commonly found in shallow water (Poulakis & Seitz, 2004; Wiley & Simpfendorfer, 2010; Whitty et al., 2009; Simpfendorfer, 2001). #ey also prefer warm water temperatures, mainly 22–28°C, and their lower thermal tolerance is predicted to be around 20°C (Carlson et al., 2014). Smalltooth saw!sh o&en feed on schooling !sh such as clupeids, carangids, mugilids, elopids, sparids, and belonids, as well as dasyatids stingrays (Poulakis et al., 2017). #ey obtain their prey by slashing their rostrum sideways through a school and impaling the !sh on their rostral teeth. A&er caught, they ingest their prey whole. Most of their prey are coastal or estuarine species that are typically found at or near the surface. #ere has been evidence for sexual segregation in elasmobranchs and saw!sh may be included among those. Sexual segregation is the separation of males and females of the same species; this separation may be spatial as well as temporal in nature, for example, occurring only during the non-breeding season (Wearmouth & Sims, 2010). It is important to understand the habitat use of di$erent sexes in order to predict population trends and dynamics. #is would provide data that may be useful to the successful management and conservation of this species, as these spatial dynamics o&en overlap with area-focused human activities like !shing. #ere has never been a large-scale !shery that directly targeted smalltooth saw!sh, but it is very common for a saw!sh to get entangled in !shing nets due to their long-toothed rostrum, and therefore they are o&en caught as bycatch. #e main threat responsible for the decline in smalltooth saw!sh has been and remains commercial and recreational !sheries. Shrimp trawl !sheries present a major concern due to the high bycatch mortality of large, mature saw!sh, which could reduce the population’s reproductive potential. Fortunately, saw!sh are expected to su$er less and recover quicker when caught and released by longlines as opposed to trawls and gill nets (Brame et al., 2019). Smalltooth saw!sh are said to be one of the world’s most vulnerable marine !shes (Dulvy et al., 2016; Feldheim et al., 2017). #e United States population of smalltooth saw!sh was listed as endangered under the U.S. Endangered Species Act in April 2003 (Poulakis & Seitz, 2004). Penalties such as imprisonment or steep !nes could be given to anyone who harasses, harms, or kills any animal listed on the U.S. Endangered Species Act. Smalltooth saw!sh are classi!ed as critically endangered by the International Union for Conservation of Nature (IUCN). #is is the highest level of alert—the closest to extinction—set by the IUCN and is de!ned by them as “a species facing an extremely high risk of extinction in the wild.” #e IUCN Red List states their population trend is still decreasing (Carlson et al., 2013). Pop-o$ archiving satellite tags (PSATs) are a relatively new electronic tagging technology. PSATs detach from the tagged animal a&er a predetermined time, "oat to the surface, and transmit the archived data to satellites which provide the data to the researcher. #is technology allows tracking the movements of pelagic !sh in their natural environment to be much more accessible and economical. PSATs can sample temperature, depth, and light levels at user-de!ned time intervals, and then store and process these data (Luo et al., 2006). It has become more common to use PSATs to gather data on horizontal and vertical Sex and length as predictors of vertical movements in smalltooth saw!sh (Pristis pectinata) Aisthesis Volume 11, 20203 movements of pelagic !shes, especially since the vertical movement of PSAT-tagged !shes in sea water has a high degree of accuracy and precision (depth and temperature resolution are claimed to be 0.5 m and 0.05°C, respectively) (Luo et al., 2006). Why we need to study and protect saw!sh #e U.S. Endangered Species Act requires by law that critical habitat must be designated for any listed species. To date, critical habitat has only been designated for small juvenile smalltooth saw!sh. Satellite tagging of saw!sh adults was done for the following reasons: 1) to aid in de!ning critical habitat for adults and vertical space use; 2) to determine potential aggregation sites for mating and areas where males and females overlap in depth; and 3) to determine if the U.S. population is distinct from adjacent populations (e.g. Bahamas, Cuba) by determining if saw!sh frequently traverse deep water (up to 800 m depths). With the decline of saw!sh, more research needs to be conducted in order to create more e$ective status assessments, management measures, and recovery plans (Dulvy et al., 2016). Fishery management, where it does occur, focuses mostly on commercially valuable !sh populations, so populations like saw!sh are rarely the main concern. It is di%cult to develop recovery strategies for species that do not have a su%cient amount of scienti!c data concerning their distribution and habitat use, especially if they are widely dispersed (Wiley & Simpfendorfer, 2010). More research must be conducted on saw!sh so adequate recovery plans can be developed. #e purpose of this study is to assess whether sex or length has an in"uence on the percent time a smalltooth saw!sh spends at a particular depth. Conducting research on how saw!sh use depth could help predict population trends and dynamics. Knowing the factors that a$ect a saw!sh’s depth would allow for the successful management and conservation of this species, so it is important that these factors are studied and considered. Methods Pop-o$ archival satellite tagging data was collected from 2011 to 2017 and forty-seven saw!sh were tagged. #e saw!sh were caught using a research vessel and a bottom longline of 50 or 100 16/0 hooks baited with lady!sh, Elops saurus. Longline stations were selected based on being potential habitats suitable for smalltooth saw!sh and historic encounter records. Tagging was conducted in the Florida Bay and the Florida Keys. Most of the sites had coordinates around 25°N 81°W (Figure 1). Satellite telemetry was used to track the movements of the tagged saw!sh. Pop-o$ archiving satellite tags (PSATs), which record depth (m), temperature (°Celsius), time, and light levels, were attached externally to the saw!sh !rst dorsal !n. #ree di$erent models of PSATs (all manufactured by Wildlife Computers, Inc.) were used: Mini-PAT, MK10, and PATF. #ree di$erent models of PSATs were used because multiple research institutions contributed to tagging the saw!sh. #e tags were programmed to detach from the animals a&er a certain number of days (depending on the tag type) and transmit the archived data to a satellite. #e PSATs use light-based geolocation where the spatial track is based on the timing of local noon (used to estimate longitude) and day length (used to estimate latitude) and corrected for temperature (Luo et al., 2006). Depth data were collected every four hours. Each satellite tag or PTT (Platform Transmitting Terminal) was assigned a PTT number. For this study, data was available for fourteen satellite tags (Wildlife Computers, Inc.), spanning a deployment period from March 2011 to March 2017. Six of the tags were MK10 tags (programmed for 150 days), !ve were PATF tags (tracked for 60 days), and three were Mini-PAT tags (programmed for 105 days). Choosing viable data #e GPE2 program from Wildlife Computers was opened using iGOR Pro 6.37 so&ware. #e graphs that displayed viable data (daily depths measurements for at least a duration of two weeks) were chosen to examine further. Further examination was done by looking at the speci!c depth measures in the Excel sheets for each of those tags. Only depths with an error of 4 m or less were considered, since that was the most accurate reading recorded by any tag. Anything with a higher error than 4 m was considered to be inaccurate. Used data In total, forty-three satellite tags were inspected for data use. Only fourteen of those tags were analyzed due to some tags failing to report or record enough data to make it viable. Eight of these were males and Sex and length as predictors of vertical movements in smalltooth saw!sh (Pristis pectinata) Aisthesis Volume 11, 20204 six were females. Statistics for each analyzed saw!sh can be seen in Table 1. Any negative minimum depths were regarded as 0 m (at the surface). Creating histograms of percentage time-at-depth All of the recorded depth measurements, with an error of 4 m or less, for an individual saw!sh were considered. Depth bins were created in 8 m intervals (0-8, 9-16, 17-24, etc.), since the recorded data were expressed in increments of 8 m (0, 8, 16, 24, etc.). #e percent of time each saw!sh spent in each depth bin was calculated and made into a histogram in Excel. Average maximum depth Excel was used to calculate the average maximum depth of each saw!sh. #e average maximum depth of each day was found, then all of these calculations were averaged together. Model choice Akaike’s Information Criteria (AIC) is a log- likelihood that penalizes any super"uous parameters in the model. AIC estimates the quality of each entered model, relative to each of the other models. AICc has small-sample correction. A low AIC value indicates the model is a good !t; the model with the lowest AIC is considered the best !t. If the delta AIC is lower than 2, it can be said to have substantial support. Various linear mixed-e$ects models were tested. #e entered model that was the simplest was chosen by comparing AICc values to determine which model has a better !t. Due to the principle of parsimony, all the factors that did not cause a signi!cant increase in deviance were removed from the model to form a minimal adequate model. Linear mixed-e"ects model R Statistical Computing (R Core Team, 2019) and lme4 (Bates, Maechler, Bolker & Walker, 2015) were used to perform a linear mixed e$ects analysis of the relationship between percent time and depth. A mixed-e$ects model was chosen because the data had both random and !xed e$ects due to temporal pseudoreplication resulting from repeated measurements on the same individuals. As the random e$ect, the saw!sh PTT number was entered (without interaction term) into the model. Intercepts for depth and sex were !xed e$ects, as well as by- subject and by-item random slopes for the e$ect of depth. Visual inspection of residual plots did not reveal any obvious deviations from homoscedasticity or normality for sex. #e function “lmer()” was used, as it allows for non-normal errors and non-constant variance with the same error as a generalized linear model. Results Histograms of percentage time-at-depth #e results of the histograms can be seen in Figure 2. #ese histograms showed the percentage of time a saw!sh spent at a speci!c depth by analyzing all of the data recorded on that saw!sh’s tag. Every saw!sh was seen at a depth of 0-8 m for the majority of the time; however, a few saw!sh (14%) were recorded to go as deep as 88 m. Most tagged (n = 12) saw!sh spent over 60% of their time at 0-8 m, but two saw!sh (one male and one female) spent over 50% of their recorded time at depths deeper than the 0-8 m bin. As depth increases, there is no pattern of depth use. When a saw!sh is found at a depth deeper than 8 m, there is not a favored depth, and their preference for a depth did not consistently decrease with increasing depth. Sex-dependent trends were apparent, however. All of the females spent time in at least six other depth bins, while half of the males solely spent time in 0-8 m. Which sex has a greater maximum depth? #is t-test compared the average maximum depth each saw!sh occupied each day, which shows the deepest depth a saw!sh occupies on average. #e average maximum depth occupied by males was 5.736 m (standard error = 2.475) and 22.61 m (standard error = 5.118) for females. An unpaired student t-test was performed and the two-tailed p-value equaled 0.0068, suggesting there was a signi!cant di$erence between the means (t = 3.3985; df = 10 ; standard error of di$erence = 4.965). Model choice #e lowest AICc value indicated the most parsimonious model, relative to the other model !ts with a higher AICc value (best model: lmer(Percent. Time ~ Depth + Sex + (1|Saw!sh)). #e best model to explain the in"uence of percent of time at depth indicated sex was relevant, while length was not. #e minimal adequate model showed a common slope for percent time against depth with two intercepts, Sex and length as predictors of vertical movements in smalltooth saw!sh (Pristis pectinata) Aisthesis Volume 11, 20205 one for each sex. #e delta AICc of the best linear mixed-e$ects model was between 0-2 and therefore can be said to have substantial support. Linear mixed-e"ects model #e variables that were tested were depth, length, and sex to address the following question: Does total length and/or sex have an e$ect on the percent-of- time a saw!sh spent at a particular depth? Since length fell out of the model, it is not considered a signi!cant factor. #e percent time each sex stays at a speci!c depth is shown in Figure 3. In this !gure, it is clear that saw!sh spend more time at shallow depths, but de!nitely spend a considerable amount of time in deeper water. Due to the model choice conducted with AICc, both the male and female isoclines were given the same slope. Males spend a much greater time at shallower depths than females. #is is consistent with the average maximum depth t-test results that show females have a signi!cantly deeper average depth than males. #is shows that females are expected to be found at deeper depths than males. Discussion #e goal of this study was to assess whether sex or length in"uences the percent of time a smalltooth saw!sh spends at a speci!c depth. Knowing what factors determine the depth of saw!sh will help further understand the habits of saw!sh and could help with conservation management for this critically endangered species. #e results concluded that both male and female saw!sh spend more time in shallow depths over deeper depths, as expected from previous saw!sh literature (Carlson et al., 2014; Poulakis & Seitz, 2004; Simpfendorfer, 2001; Wiley & Simpfendorfer, 2010). #is knowledge is important to comprehend in order to understand their habitats and apply them to conservation practices. Expecting shallower depths to be favored, the main focus of this study was to see which sex spent more time occupying deeper depths. #e results showed females tend to spend more time at deeper depths than males, and males spend more time at shallower depths than females. Two males did go relatively deep (65-72 m), and two females only occupied shallow water (0-8 m), but the general trend showed females spending more time at deeper depths than males. #is was seen in the histograms of percentage time- at-depth (Figure 2), the t-test conducted for greatest maximum depth, and the linear mixed-e$ects model (Figure 3). Trends in the histograms (Figure 2) reveal that despite sex di$erences, all saw!sh analyzed spent the majority of their time at 0-8 m. #erefore, this area of the water column should still be the main concern for conservation and management practices. However, it is important to know that saw!sh do also frequently occupy deeper depths, and deeper waters still need to be considered for management measures to support population increases of this critically endangered species. #ese results mean we can expect to !nd females at greater depths than males and could potentially a$ect the way female, especially pregnant, saw!sh are protected. Females are very important for population dynamics since they give birth to pups. In order to ensure the saw!sh population grows, females must be protected. If females have di$erent depth preferences than males, this could change the way the conservation of female saw!sh is managed. It may be more important to ensure female survival, so this information can help tailor saw!sh conservation more towards females. Further research on whether sexual maturity or reproductive state a$ects the depth a saw!sh occupies would add to scienti!c knowledge and management, helping to ensure these individuals are protected, therefore promoting the growth of the saw!sh population. Improvements to this model could include adding factors such as seasonality. Month was considered when sorting the data, but seasonality was not analyzed. Upon visual inspection of the data, it appears saw!sh mostly occupied deeper depths during the late summer months. #e deepening of the thermocline during the summer months may be a reason for saw!sh to go deeper. Further analysis regarding month and the breeding season would need to be calculated to see if sex is only segregating at certain times of the year. Di$erences in depth between males and females could also be due to the reproductive cycle of females (Carlson et al., 2014). #is could be counterintuitive since females are more o&en found at nurseries, which occur in shallow water areas like estuaries, and would not support the data showing females spend more time at deeper depths than males (Feldheim et al., 2017). However, this could potentially be evidence of sexual Sex and length as predictors of vertical movements in smalltooth saw!sh (Pristis pectinata) Aisthesis Volume 11, 20206 segregation, where females are actively avoiding males during non-reproductive seasons. Further studies that track the exact seasons during which female saw!sh mate would have to be conducted in order to see if this could be a factor. Other factors that may a$ect the distribution of sexes could be due to feeding behavior or thermoregulation (Carlson et al., 2014). #ere has been evidence of sex-speci!c dietary requirements, such as females eating more than males of the same size. Sex-speci!c temperature preferences may also occur if a female is pregnant, as warm waters can help an embryo develop (Wearmouth & Sims, 2010). Feeding behavior and temperature preference could be di$erent among sexes, but more data collection and analysis would need to be done to see if these factors have an e$ect on percent of time at a particular depth. Depth could also depend on the time of day. Saw!sh have been recorded to move into shallower waters at night and deeper waters during the day, so diurnal movements could be a factor (Carlson et al., 2014). Total length was not a signi!cant factor when considering what depth a saw!sh occupies, but modeling length and depth could still be useful to see if any visual trends are spotted, especially if this work was extended to include juveniles as well as adult saw!sh. It’s important to note that the analyzed saw!sh in this study did not have much size variety (ranging from 279-428 cm); therefore, length may be a signi!cant factor, but the study may not have had enough statistical power to show any variation. Further testing with saw!sh of greater size variety could be done to see if increasing the statistical power of length creates a di$erence. Overall, there are several potentially confounding variables that could have in"uenced the results of this study. Additional data and analyses would need to be collected and completed in order to see if these variables cause a signi!cant di$erence. Understanding how saw!sh use depth is important in order to predict population trends and dynamics. Knowing the factors that a$ect a saw!sh’s depth preferences would allow for the successful management and conservation of this species, since these spatial dynamics o&en overlap with area- focused human activities like !shing. Until recently, saw!sh were thought to stay at a depth of 10 m or less (Simpfendorfer, 2001), so data on their depth use is limited, but important to know in order to fully understand their full vertical range. Understanding their depth range could help with the conservation and recovery e$orts of this critically endangered species. With the decline of saw!sh, more research needs to be conducted in order to create more e$ective status assessments, management measures, and recovery plans (Dulvy et al., 2016). With the knowledge from this study, hopefully more e$ective recovery plans can be developed and key areas of research that still need to be conducted can be identi!ed. Acknowledgements I want to thank Jasmin Graham, Dr. Dean Grubbs, Dr. Janie Wul$ and Dr. Ian MacDonald for their guidance and assistance throughout my entire thesis project. References Brame, A. B., Wiley, T. R., Carlson, J. K., Fordham, S. V., Grubbs, R. D., Osborne, J., Scharer, R. M., Bethea, D. M., Poulakis, G. R. (2019). Biology, ecology, and status of the smalltooth saw!sh Pristis pectinata in the United States, Endangered Species Research, 39, 9-23. Carlson, J. K., Gulak, S. J. B., Simpfendorfer, C. A., Grubbs, R. D., Romine, J. G., & Burgess, G. H. (2014). Movement patterns and habitat use of smalltooth saw!sh, Pristis pectinata, determined using pop?up satellite archival tags. Aquatic Conservation: Marine and Freshwater Ecosystems, 24(1), 104-117. Carlson, J., Wiley, T., & Smith, K. (2013). Pristis pectinata. #e IUCN Red List of #reatened Species. http://dx.doi.org/10.2305/IUCN. UK.2013-1.RLTS.T18175A43398238.en. Carlson, J. K., & Simpfendorfer, C. A. (2015). Recovery potential of smalltooth saw!sh, Pristis pectinata, in the United States determined using population viability models. Aquatic Conservation: Marine and Freshwater Ecosystems, 25(2). Dulvy, N. K., Davidson, L. N., Kyne, P. M., Simpfendorfer, C. A., Harrison, L. R., Carlson, Sex and length as predictors of vertical movements in smalltooth saw!sh (Pristis pectinata) Aisthesis Volume 11, 2020 J. K., & Fordham, S. V. (2016). Ghosts of the coast: Global extinction risk and conservation of saw!shes. Aquatic Conservation: Marine and Freshwater Ecosystems, 26(1), 134-153. Feldheim, K. A., Fields, A. T., Chapman, D. D., Scharer, R. M., & Poulakis, G. R. (2017). Insights into reproduction and behavior of the smalltooth saw!sh Pristis pectinata. Endangered Species Research, 34, 463-471. Harrison, L. R., & Dulvy, N. K. (eds). (2014). Saw!sh: A global strategy for conservation. IUCN Species Survival Commission’s Shark Specialist Group. Luo, J., Prince, E. D., Goodyear, C. P., Luckhurst, B. E., & Serafy, J. E. (2006). Vertical habitat utilization by large pelagic animals: a quantitative framework and numerical method for use with pop-up satellite tag data. Fisheries Oceanography, 15(3), 208-229. National Marine Fisheries Service (2009). Recovery plan for smalltooth saw!sh (Pristis pectinata). Prepared by the Smalltooth Saw!sh Recovery Team for the National Marine Fisheries Service. https://repository.library.noaa.gov/view/ noaa/15983. Papastamatiou, Y. P., Grubbs, R. D., Imho$, J. L., Gulak, S. J., Carlson, J. K., & Burgess, G. H. (2015). A subtropical embayment serves as essential habitat for sub-adults and adults of the critically endangered smalltooth saw!sh. Global Ecology and Conservation, 3, 764-775. Poulakis, G. R., & Seitz, J. C. (2004). Recent occurrence of the smalltooth saw!sh, Pristis pectinata (Elasmobranchiomorphi: Pristidae), in Florida Bay and the Florida Keys, with comments on saw!sh ecology. Florida Scientist, 27-35. Poulakis, G. R., Stevens, P. W., Timmers, A. A., Wiley, T. R., & Simpfendorfer, C. A. (2011). Abiotic a%nities and spatiotemporal distribution of the endangered smalltooth saw!sh, Pristis pectinata, in a south-western Florida nursery. Marine and Freshwater Research, 62(10), 1165- 1177. Poulakis, G. R., Urakawa, H., Stevens, P.W., DeAngelo, J.A. and others (2017). Sympatric elasmobranchs and fecal samples provide insight into the trophic ecology of the smalltooth saw!sh. Endangered Species Research, 32, 491- 506 Simpfendorfer, C. A. (2001). Essential habitat of smalltooth saw!sh (Pristis pectinata). Mote Marine Laboratory. Wearmouth, V. J., & Sims, D. W. (2010). Sexual segregation in elasmobranchs. Biologia Marina Mediterranea, 17, 236-239. Whitty, J. M., Morgan, D. L., Peverell, S. C., #orburn, D. C., & Beatty, S. J. (2009) Ontogenetic depth partitioning by juvenile freshwater saw!sh (Pristis microdon: Pristidae) in a riverine environment. Marine and Freshwater Research, 60, 306-316. Wiley, T. R., & Simpfendorfer, C. A. (2010). Using public encounter data to direct recovery e$orts for the endangered smalltooth saw!sh Pristis pectinata. Endangered Species Research, 12(3), 179-191. 7 Sex and length as predictors of vertical movements in smalltooth saw!sh (Pristis pectinata) Aisthesis Volume 11, 2020 Figures and Tables 8 Sex and length as predictors of vertical movements in smalltooth saw!sh (Pristis pectinata) Aisthesis Volume 11, 20209 Sex and length as predictors of vertical movements in smalltooth saw!sh (Pristis pectinata) Aisthesis Volume 11, 202010