SH II A Fish U.S. Department of Commerce Volume 111 Number 2 April 2013 Fishery Bulletin U.S. Department of Commerce Rebecca M. Blank Acting Secretary of Commerce National Oceanic and Atmospheric Administration Kathryn D. Sullivan Acting Administrator National Marine Fisheries Service Samuel D. Rauch III Acting Assistant Administrator for Fisheries ^T°Pc0 ★ Scientific Editor Bruce C. Mundy Associate Editor Kathryn Dennis National Marine Fisheries Service Pacific Islands Fisheries Science Center Aiea FHeights Research Facility 99-193 Aiea Heights Drive, Suite 417 Aiea, Hawaii 96701-3911 Managing Editor Sharyn Matriotti National Marine Fisheries Service Scientific Publications Office 7600 Sand Point Way NE Seattle, Washington 981 15-0070 The Fishery Bulletin (ISSN 0090-0656) is published quarterly by the Scientific Publications Office, National Marine Fish- eries Service, NOAA, 7600 Sand Point Way NE, Seattle, WA 98115-0070. Although the contents of this publica- tion have not been copyrighted and may be reprinted entirely, reference to source is appreciated. The Secretary of Commerce has deter- mined that the publication of this peri- odical is necessary according to law for the transaction of public business of this Department. Use of funds for printing of this periodical has been approved by the Director of the Office of Management and Budget. For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, DC 20402. Subscrip- tion price per year: $32.00 domestic and $44.80 foreign. Cost per single issue: $19.00 domestic and $26.60 foreign. See back for order form. Editorial Committee Richard Brodeur John Carlson Kevin Craig Jeff Leis Rich McBride Rick Methot Adam Moles Frank Parrish Dave Somerton Ed Trippel Mary Yoklavich National Marine Fisheries Service, Newport, Oregon National Marine Fisheries Service, Panama City, Florida National Marine Fisheries Service, Beaufort, North Carolina Australian Museum, Sydney, New South Wales, Australia National Marine Fisheries Service, Woods Hole, Massachusetts National Marine Fisheries Service, Seattle, Washington National Marine Fisheries Service, Auke Bay, Alaska National Marine Fisheries Service, Honolulu, Hawaii National Marine Fisheries Service, Seattle, Washington Department of Fisheries and Oceans, St. Andrews, New Brunswick, Canada National Marine Fisheries Service, Santa Cruz, California Fishery Bulletin web site: www.fisherybulletin.noaa.gov The Fishery Bulletin carries original research reports and technical notes on investigations in fishery science, engineering, and economics. It began as the Bulletin of the United States Fish Commission in 1881; it became the Bulletin of the Bureau of Fisheries in 1904 and the Fishery Bulletin of the Fish and Wildlife Service in 1941. Separates were issued as documents through volume 46; the last document was No. 1103. Beginning with volume 47 in 1931 and continuing through volume 62 in 1963, each separate appeared as a numbered bulletin. A new system began in 1963 with volume 63 in which papers are bound together in a single issue of the bulletin. Beginning with volume 70, number 1, January 1972, the Fishery Bulletin became a periodical, issued quarterly. In this form, it is available by subscription from the Superintendent of Documents, U.S. Government Printing Office, Washington, DC 20402. It is also available free in limited numbers to libraries, research institutions, State and Federal agencies, and in exchange for other scientific publications. U.S. Department of Commerce Seattle, Washington Volume 111 Number 2 April 2013 Fishery Bulletin Contents Articles 0 N //] ^ APR S 0 2013 A® RAR't§. 111-121 Fulford, Richard S. and Kevin Dillon Quantifying intrapopulation variability in stable isotope data for Spotted Seatrout ( Cynoscon nebulosus) The National Marine Fisheries Service (NMFS) does not approve, recommend, or endorse any proprie- tary product or proprietary material mentioned in this publication. No reference shall be made to NMFS, or to this publication furnished by NMFS, in any advertising or sales promotion which would indicate or imply that NMFS approves, rec- ommends, or endorses any propri- etary product or proprietary mate- rial mentioned herein, or which has as its purpose an intent to cause directly or indirectly the advertised product to be used or purchased be- cause of this NMFS publication. The NMFS Scientific Publications Office is not responsible for the contents of the articles or for the standard of English used in them. 122-146 Frable, Benjamin W., Carole C. Baldwin, Brendan M. Luther, and Lee A.Weigt A new species of western Atlantic lizardfish (Teleostei: Synodontidae: Synodus) and resurrection of Synodus bond i Fowler, 1939, as a valid species from the Caribbean with redescriptions of S. bond/, S. foetens (Linnaeus, 1766), and S. mtermedius (Agassiz, 1829) 147-160 Wells, R. J. David, Susan E. Smith, Suzanne Kohin, Eilen Freund, Natalie Spear, and Darlene A. Ramon Age validation of |uvenile Shortfin Mako (./sums oxynnchus ) tagged and marked with oxytetracycline off southern California 161-174 McBride, Richard S., Tiffany E. Vidal, and Steven X. Cadrin Changes in size and age at maturity of the northern stock of Tilefish ( Lopholatilus chamaeleonticeps ) after a period of overfishing 175-188 Syamsuddin, Mega L., Sei-lchi Saitoh, Tom Hirawake, Samsul Bachri, and Agung B. Harto Effects of El Niho-Southern Oscillation events on catches of Bigeye Tuna (.Thunnus obesus ) in the eastern Indian Ocean off Java 189-201 Capossela, Karen M., Mary C. Fabrizio, and Richard W. Brill Migratory and within-estuary behaviors of adult Summer Flounder ( Paralichthys dentatus ) in a lagoon system of the southern mid-Atlantic Bight 202-204 Guidelines for authors Abstract — Stable isotope (SI) values of carbon (513C) and nitrogen (515N) are useful for determining the tro- phic connectivity between species within an ecosystem, but interpreta- tion of these data involves important assumptions about sources of intra- population variability. We compared intrapopulation variability in 513C and 815N for an estuarine omnivore, Spotted Seatrout ( Cy noscion nebulo- sus), to test assumptions and assess the utility of SI analysis for delinea- tion of the connectivity of this spe- cies with other species in estuarine food webs. Both 813C and §15N val- ues showed patterns of enrichment in fish caught from coastal to off- shore sites and as a function of fish size. Results for §13C were consistent in liver and muscle tissue, but liver 815N showed a negative bias when compared with muscle that increased with absolute 515N value. Natural variability in both isotopes was 5-10 times higher than that observed in laboratory populations, indicating that environmentally driven intra- population variability is detectable particularly after individual bias is removed through sample pooling. These results corroborate the utility of SI analysis for examination of the position of Spotted Seatrout in an estuarine food web. On the basis of these results, we conclude that inter- pretation of SI data in fishes should account for measurable and ecologi- cally relevant intrapopulation vari- ability for each species and system on a case by case basis. Manuscript submitted 17 January 2012. Manuscript accepted 4 January 2013. Fish. Bull. 111:111-121 (2013). doi:10.7755/FB.l 11.2.1 The views and opinions expressed or implied in this article are those of the author (or authors) and do not necessar- ily reflect the position of the National Marine Fisheries Service, NOAA. Quantifying intrapopulation variability in stable isotope data for Spotted Seatrout ( Cy noscion nebulosus ) Richard S. Fulford (contact author) Kevin Dillon Email address for contact author: fulford.richard@epa.gov Department of Coastal Sciences University of Southern Mississippi Gulf Coast Research Laboratory 703 East Beach Road Ocean Springs, Mississippi 39564 The mandate for management of fisheries has shifted toward a focus on ecosystem-based fisheries man- agement (EFM) (Brodziak and Link, 2002). The EFM approach differs from historical single-species man- agement in that EFM acknowledges the linkages between ecosystem com- ponents as well as the spatial and temporal variability in those linkag- es (Arkema et ah, 2006; Christensen et al., 1996; NRC 1999; Thomas and Huke, 1996). The EFM paradigm has great potential to improve the abil- ity of managers to accurately pre- dict and appropriately respond to the impacts of multiple stressors on exploited fish populations. However, the data needs of EFM are high, and the acquisition of these data has generally lagged behind development of EFM theory (Dame and Christian, 2006). Quantification of trophic connec- tions among species is an important precursor for the shift to EFM be- cause such data support the devel- opment of food-web network models that allow fishing pressure to be an- alyzed in an ecological context with other sources of mortality (Chris- tensen and Pauly, 2004). Quantita- tive techniques, like stable isotope (SI) analysis, have been widely ap- plied to fish populations as a method for quantification of the trophic links between predator and prey (Power et ah, 2002; Rooker et ah, 2006). Sta- ble isotope data are well suited to the development of network models of food webs because they are time- integrated and provide data on the relative importance of entire trophic pathways down to sources of primary production (Peterson and Fry, 1987). Yet, the interpretation of stable iso- tope data requires several important assumptions, and these assumptions should be carefully analyzed if re- sults may affect management deci- sion making. Key assumptions made in inter- preting stable isotope data involve the sources of variability in isoto- pic values for a given population (Peterson and Fry, 1987). Popula- tion variability integrates several sources of variability, including en- vironmental influences (Barnes et al., 2008), differences in individual behavior (Sweeting et al., 2005), ontogeny, growth, and tissue turn- over rates (Herzka, 2005; Perga and Gerdeaux, 2005), and prey availabil- ity and feeding success (Sweeting et al., 2005). Partitioning out these sources of variability, particularly for an omnivorous consumer in an estuarine ecosystem, can be com- plex. In particular, quantification of intrapopulation variability in isotope data (e.g., variability due to differ- ences in individual behavior or tis- sue turnover rates) is important for separating this variability from changes that are ecologically impor- tant (e.g., changes due to ontogeny or the environment). 112 Fishery Bulletin 111(2) Intrapopulation variability in SI compositions can be particularly difficult to interpret in species and eco- systems with multiple overlapping trophic pathways. Omnivorous fish species, such as Eurasian Perch {Per- ea fluviatilis ) and Spotted Seatrout ( Cynoscion nebulo- sus), hold important positions in aquatic food webs as trophic mediators of primary production, but they have complex stable isotope signatures that require a close examination of assumptions regarding intrapopulation variability (Quan et ah, 2007; Quevedo et ah, 2009). The estuarine habitat of Spotted Seatrout, in particu- lar, is highly complex; it contains both potential aquat- ic sources (e.g., phytoplankton, microphytobenthos, submerged aquatic vegetation, marsh macrophytes) and terrestrial sources of primary production that can vary in importance for estuarine food webs, both tem- porally and spatially. One of the key questions with the food-web approach to EFM that can be addressed with stable isotope data is an estimation of the relative importance of different sources of organic matter for the production of estuarine consumers. Yet, individual fishes exposed to the same suite of prey resources may have very different isotopic values, and that variation may overemphasize the importance of omnivory and confound detection of any population-level differences in isotope ratios (Zanden et ah, 2010). The utility of stable isotope data for analysis of estuarine ecosystems will be greatly enhanced when sources of uncertainty have been well-quantified for important populations. The goal of this analysis was to measure intrapopu- lation variability in carbon (S13C) and nitrogen (515N) SI values in a representative estuarine omnivore (Spot- ted Seatrout) to improve understanding and interpre- tation of SI data in an estuarine ecosystem in the northern Gulf of Mexico. The study objectives were 1) to quantify individual variation in carbon and nitrogen isotopic values of Spotted Seatrout in the laboratory and in wild populations; 2) to examine spatial and sea- sonal changes in the isotopic value of various tissue types with different turnover rates; and 3) to quantify intrapopulation variability for carbon and nitrogen iso- topes in the laboratory and compare these estimates with field-based and literature reported values for iso- tope variability. A characterization of these sources of variability will allow for a clearer interpretation of how opportunistic changes in fish diet may be reflected in seasonal or spatial variances in isotope values. Methods Spotted Seatrout were collected from regular surveys conducted monthly by the University of Southern Mis- sissippi Center for Fisheries Research and Develop- ment at 8 sites in coastal Mississippi (Fig. 1). Addition- al fish were also captured at 2 additional sites as part of a second survey of composition of fish communities associated with oyster reefs in the western Mississippi Sound (Fig. 1). Both surveys were conducted from 2007 to 2009. All fish were captured during 1-h sets of an experimental gillnet with 5 panels (30.48x1.83 m with mesh sizes of 50.8, 63.5, 76.2, 88.9, and 101.6 mm). All Spotted Seatrout collected were identified based on taxonomic descriptions of Hoese and Moore (1977), measured to the nearest 1 mm (total length [TL] ) and weighed to the nearest 1 g (wet weight), returned to the laboratory on ice, and immediately frozen at -20° C. In the laboratory, fish were thawed and tissue samples were collected: 2 samples of white muscle were collect- ed from the left and right side of the fish dorsal to the midline, and 2 samples were collected of liver tissue. Both muscle and liver samples were freeze-dried to a constant weight, ground to a fine power, homogenized, and stored in a desiccator for stable isotope analysis. Stable isotope ratios for 813C and §15N were em- ployed to delineate spatial and temporal changes in the SI values of individual fish within and between lo- cal populations. The high lipid content of liver tissue necessitated the need for a correction factor because lipids are typically depleted in 13C relative to carbohy- drates and protein (Deniro and Epstein, 1977). Ten liv- er samples were split into 2 parts; lipid was extracted from half the sample and the other half was unaltered. Each lipid subset of liver tissue samples was extract- ed sequentially with 2:1 chloroform: methanol and 1:2 chloroforrmmethanol mixtures, dried, and homogenized before analysis (Bligh and Dyer, 1959; Ruiz et ah, 2007). The 813C results of the lipid-extracted samples were plotted against the results of the nonextracted samples, and the resultant linear equation was used to correct the liver 813C values for lipid content: 5,3C „r„cted = (0.9975 x 5«C + 2.0578,(1) where the coefficient of determination [r2] = 0.93. Samples were analyzed with a DELTA V Advantage1 stable isotope ratio mass spectrometer (Thermo Scien- tific, Waltham, MA) coupled to an ECS 4010 elemental combustion system (Costech Analytical Technologies, Inc., Valencia, CA). All samples were analyzed in du- plicate and referenced to known isotopic standards and are reported in per mil notation {%o). Nitrogen is ref- erenced to atmospheric nitrogen and carbon to the Pee Dee Belemnite (PDB) standard. The 815N value was used to calculate mean trophic level by site and month based on a general formula reported by Rooker et al. (2006): Trophic level = 1.0 + (815N — 6.0)/3.2, (2) where 3.2 is the reported mean fractionation factor for nitrogen in fish (Peterson and Fry, 1987; Post, 2002) and 6.0 is the mean 815N value of primary producers in Gulf of Mexico coastal estuaries on the basis of both analysis of samples collected as part of separate proj- ects in Mississippi Sound (senior author, unpubl. data) 1 Mention of trade names or commercial companies is for identification purposes only and does not imply endorsement by the National Marine Fisheries Service, NOAA. Fulford and Dillon: Intrapopulation variability in stable isotope data for Cynosaon nebulosus 113 -97 10' -92 45' -88 20' -83 55' -79 30' Figure 1 Map of 10 collection sites for Spotted Seatrout (Cynoscion nebulosus) captured in coastal Mississippi and the western Mississippi Sound during 2007-09 for this study of intrapopulation variability in stable isotope values for this species. Sites are labeled as Bayou Caddy (BC), Dupont (DP), Bayou Portage (BP), oyster reefs (OR), Back Bay (BB, 2 locations), Fort Bayou (FB), Davis Bayou (DB), Graveline Bayou (GB), Marsh Lake (ML), and Barrier Islands (BI). Fish were collected at random at the Barrier Island site, and therefore fish collected at the Barrier Islands may have been collected adjacent to either of the 2 largest of the Barrier Islands. and similar values reported in the literature for the coastal Gulf of Mexico (Winemiller et ah, 2007). Pre- viously reported 815N values for primary producers — needlegrass rush ( Juncus roemerianus Scheele), smooth cordgrass ( Spartina alterni flora), and edaphic algae — in a salt marsh within Graveline Bayou ranged from 5.2% Barrier Islands -18 -17 16.5 16 0 B 15.5 - 15.0 - 14.5 - 2 14.0 - 13.5 13.0 12.5 12.0 - 11.5 — -25 1 1 1 1 1 1 1 -24 -23 -22 -21 -20 -19 -18 Figure 2 Bivariate plot showing mean (±standard deviation (SD|) carbon ( 5 1 3 C ) and nitrogen (515N) stable isotope values per mil for Spotted Seatrout ( Cynoscion nebulosus) by (A) collection site and (B) collection month. All samples were collected in 2007, 2008, or 2009 off the coast of Mississippi. Symbols in the top panel indicate the 10 sites where fish were collected (see Fig. 1 for locations); black=bayou sites, white=open-bay sites, and gray=the offshore site. The asterisk (*) indicates a significantly different group (a=0.05). ticeable bias. The relationship between lipid-corrected 513C data from liver samples and data for muscle dis- played a relationship significantly different from 1:1 U-test, slope=0.25, PcO.QOl; Fig. 4A) with a shift from positive to negative bias in liver 813C values as muscle 513C values increased. However, although 515N values also showed a linear trend between liver and muscle samples with a slope that differed significantly from one (515N slope=0.50, PcO.OOl), liver tissue showed more consistent depletion bias than muscle tissue (Fig. 4B). This bias increased with overall enrichment of 515N in muscle tissue. The resampling procedure was performed on data from 3 sites: Bayou Portage, Bayou Caddy, and the Bar- rier Islands. These sites had the largest sample siz- es and showed some contrast between nearshore and open-water sites. Intersample variability was related to pooled resample size ( n ; Fig. 5). A high mean absolute residual value ( >0 . 1 ) was observed for n< 3 but dropped rapidly to a value <0.05 for an n> 5. This pattern was consistent across sites despite a large difference in re- sidual values at smaller pooled resample sizes. In gen- eral, bias in mean isotopic ratio was minimal at any pooled resample size greater than n= 5. 116 Fishery Bulletin 111(2) Discussion Effects of tissue type, fractionation, and fish size Both 513C and 515N SI values can differ between tis- sue types (Pinnegar and Polunin, 1999). Differences between the isotopic values of fish liver and muscle tissue have been ascribed to differences in tissue turn- over rates of liver and muscle (Sweeting et al., 2005), and to seasonal differences in growth rates, feeding behavior, and prey availability (Perga and Gerdeaux, 2005). However, in subtropical waters, such as Missis- sippi Sound, fish growth likely occurs throughout the year (Vetter, 1982), and, in this study, 513C values of liver and muscle were generally very similar, indicat- ing that the pattern reported by Perga and Gerdaeux (2005) at more northern latitudes was not observed in this study. The isotopic values for 515N did show a negative bias in liver compared with muscle. Because all sampling occurred during periods of probable high growth (summer and fall), this result may be related to differences in isotopic fractionation and turnover rates between tissue types. The increase in both 813C and 5I5N with fish size Fulford and Dillon Intrapopulation variability in stable isotope data for Cynoscion nebulosus 117 was small but significant and generally supports an in- crease in trophic level as fish grow and move offshore. The increase in 515N observed over the size range of fish in our study was close to 3%c, nearly a full tro- phic level. The size range of fish in this study was relatively wide but probably includes only fish beyond their first year — an assumption made on the basis of published growth curves (Fulford and Hendon, 2010). The probable age range of these fish suggests that an ontogenetic diet shift associated with maturation is dif- ficult to extrapolate from these data because the trend of increasing heavy isotope values is confounded by changes in prey availability, by increases in fish prey capture efficiency with seatrout size, and by the poten- tial variability of carbon and nitrogen sources within the Mississippi Sound and its subestuaries. The largest Spotted Seatrout were caught in open water, mainly offshore near the Barrier Islands, where prey fishes of higher trophic level, such as anchovies (e.g., Striped Anchovy [Anchoa hepsetus ]) that have a mean trophic 118 Fishery Bulletin 111(2) level similar to that of Spotted Seatrout captured at inshore sites, may be more prevalent. Spatial and temporal differences in SI Spatial and temporal patterns in SI composition were interrelated and consistent with a seasonal offshore movement of individual fish (Hendon et ah, 2002). Overall, across all sites, mean 513C for Spotted Seatrout in coastal Mississippi suggests a stronger influence of pelagic primary production than of terrestrial sources on the diet of Spotted Seatrout. Yet, there was a detect- able trend: the least-enriched samples were collected in nearshore waters in small embayments, and the most- enriched samples were collected farther offshore near the Barrier Islands. A clear inshore-to-offshore pattern of enrichments was also displayed by 515N SI values, a finding that is consistent with increases in trophic Fulford and Dillon: Intrapopulation variability in stable isotope data for Cynoscion nebulosus 119 level as fish begin to migrate offshore, although spatial variability in 815N may also contribute to the observed differences. There was much overlap between sites, but the separation of small embayments from the Barrier Islands was detectable in both carbon and nitrogen, and therefore the seasonal-spatial pattern seems robust for within-site variability and occurs at a temporal scale larger than that for the tissue isotopic turnover rate. These results also are consistent with isotopic val- ues of Spotted Seatrout in Graveline Bayou reported by Sullivan and Moncreiff (1990), who found that Spotted Seatrout had a mean 813C value of -20.7 and a mean 515N ratio of 11.8 — results that are very similar to the values reported here for Spotted Seatrout caught in Graveline Bayou. Our data, however, also indicate that these values are the middle of the overall range for 813C and near the bottom of the range for 815N across all sites in Mississippi. A more comprehensive analysis of Spotted Seatrout position in the coastal food web will require sampling over a broad spatial and tempo- ral scale. Still, ecological patterns, such as spatial and temporal differences in SI values, are detectable, and therefore stable isotope data can be used for under- standing trophic relationships in this species. Individual variability in SI composition within a population can confound ecological patterns particu- larly for omnivores (Quevedo et ah, 2009; Sweeting et ah, 2005). Population variability can be caused by individual physiological differences (i.e., differences in isotopic fractionation or growth and tissue turnover rates); individual differences in behavior, prey avail- ability or prey preference; and changes in environmen- tal conditions that may influence individual movement, metabolism, growth, and diet (Barnes et ah, 2008). Our analysis shows that individual variability measured in the laboratory with a diet of a single commercial feed is quite low (standard deviation=0.24) in comparison with that derived from field studies (median standard deviation=1.5), showing that ecologically relevant iso- topic variability in Spotted Seatrout is detectable in field populations. Detection of population-level differences through space and time can be confounded by this individual variability within population-level samples. Pooling of individuals into a common sample for analysis is, therefore, an often used method for reduction of indi- vidual variability to increase the focus on population- level questions. How much pooling is enough to reduce individual variation? Our resampling experiment in- dicates that the effectiveness of pooling is maximized for a subsample size of 5-6 individual fish. Smaller subsamples introduced potential bias due to individual variation, and larger subsamples reduced sampling ef- ficiency. In general, we advocate pooling only for inves- tigations of interpopulation differences or for changes in a single population through time. Individual vari- ability has been found to be important in detection of behavior patterns (Sweeting et ah, 2005) and in com- munity-level studies (Bolnick et ah, 2003; Matthews and Mazumder, 2004). Conclusions Understanding the trophic role of omnivores, such as Spotted Seatrout, is a critical part of estuarine eco- system-based management, and SI analysis is a valu- able tool for trophic studies of this species. Diet data are needed for identification of important prey species, but such data can be biased by small sample sizes and do not provide information on differences in total as- similation across prey species. Stable isotope analysis is an important complementary tool that allows the characterization of trophic pathways at broader spatial and temporal scales. However, variability in SI values within a population can be caused by factors unrelated to diet shifts, and these mitigating factors must be ac- counted for and incorporated into trophic studies to achieve meaningful results. Our findings substantiate the utility of stable iso- tope analysis for quantification of the trophic role of Spotted Seatrout in an estuarine system and demon- strate that this trophic role changes spatially and tem- porally in predictable ways that can be incorporated into ecosystem studies. The trophic role of Spotted Seatrout is not static, it is dependent on factors, such as habitat, life stage, and fish size that are not inde- pendent of one another. Spotted Seatrout are known to display minimal movement before sexual maturation (Hendon et al., 2002), and this behavior indicates that the observed patterns in trophic role are related more to alterations in migration patterns as fish grow rather than to opportunistic movement not related to ontog- eny. Such distinctions are important for understanding the role of a fish population within an ecosystem, and this research will directly benefit the more comprehen- sive study that is needed to understand the impact of habitat on Spotted Seatrout production. More broadly, patterns in both intra- and interpopulation variability in SI values need to be well understood for SI analysis to provide an unbiased measure of trophic connectivity within estuarine ecosystems. Studies such as this one that explicitly measure these variables in a target eco- system are extremely valuable for the application of SI data to ecosystem management. Acknowledgments This study was made possible with the time and efforts of personnel at the University of Southern Mississippi Center for Fisheries Research and Development. We thank W. Dempster and J. Dietrich for their help in col- lecting fish and E. Myers and C. Messer for their help in processing samples for analysis. This project was funded by a grant from the Mississippi-Alabama Sea Grant Consortium and the Northern Gulf Institute. 120 Fishery Bulletin 111(2) The research described in this article was developed by the author, an employee of the U.S. Environmen- tal Protection Agency (EPA), on his own time. This re- search was conducted independent of EPA employment, and has not been subjected to the Agency’s peer and administrative review. Therefore, the conclusions and opinions drawn are solely those of the author and are not necessarily the views of the EPA. Literature cited Arkema, K. K., S. C. Abramson, and B. M. Dewsbury. 2006. Marine ecosystem-based management: from char- acterization to implementation. Front. Ecol. Environ. 10:525-532. Barnes, C., S. Jennings, N. V. C. Polunin, and J. E. Lancaster. 2008. The importance of quantifying inherent variability when interpreting stable isotope field data. Oecologia 155:227-235. Bligh, E. G., and W. J. Dyer. 1959. A rapid method of total lipid extraction and puri- fication. Can. J. Biochem. Physiol. 37:911-917. Bolnick, D. I., R. Svanback, J. A. Fordyce, L. H. Yang, J. M. Davis, C. D. Hulsey, and M. L. Forister. 2003. The ecology of individuals: incidence and implica- tions of individual specialization. Am. Nat. 161:1-28. Brodziak, J., and J. Link. 2002. Ecosystem-based fishery management: what is it and how can we do it? Bull. Mar. Sci. 70:589-611. Christensen, N. L., A. M. Bartuska, J. H. Brown, S. R. Carpen- ter, C. D'Antonio, R. C. Francis, J. F. Franklin, J. A. MacMa- hon, R. F. Noss, D. J. Parsons, C. H. Peterson, M. G. Turner, and R. G. Woodmansee. 1996. The report of the Ecological Society of America committee on the scientific basis for ecosystem manage- ment. Ecol. Appl. 6:665-691. Christensen, V., and D. Pauly. 2004. Placing fisheries in their ecosystem context, an in- troduction. Ecol. Model. 172:103-107. Dame, J. K., and R. R. Christian. 2006. Uncertainty and the use of network analysis for ecosystem-based management. Fisheries 31:331-341. Deniro, M. J., and S. Epstein. 1977. Mechanism for carbon isotope fractionation associ- ated with lipid synthesis. Science 197:261-263. Fulford, R. S., and J. R. Hendon. 2010. Evaluating management actions for spotted seatrout, Cynoscion nebulosus, in Mississippi with an age-structured projection model. Gulf Caribb. Res. 22:51-61. Hendon, J. R., J. R. Warren, J. S. Franks, and M. V. Buchanan. 2002. Movements of spotted seatrout ( Cynoscion nebulo- sus) in Mississippi coastal waters based on tag-recap- ture. Gulf Mex. Sci. 20:91-97. Herzka, S. Z. 2005. Assessing connectivity of estuarine fishes based on stable isotope ratio analysis. Estuar. Coast. Shelf Sci. 64:58-69. Hoese, H. D., and R. H. Moore. 1977. Fishes of the Gulf of Mexico: Texas, Louisiana, and adjacent waters, 327 p. Texas A&M Univ. Press, Col- lege Station, TX. Matthews, B., and A. Mazumder. 2004. A critical evaluation of intrapopulation variation of 513C and isotopic evidence of individual specializa- tion. Oecologia 140:361-371. Moncreiff, C. A., and M. J. Sullivan. 2001. Trophic importance of epiphytic algae in subtropi- cal seagrass beds: evidence from multiple stable isotope analyses. Mar. Ecol. Prog. Ser. 215:93-106. NRC (National Research Council). 1999. Sustaining marine fisheries, 184 p. National Academy Press, Washington, D.C. Perga, M. E., and D. Gerdeaux. 2005. 'Are fish what they eat’ all year round? Oecologia 144:598-606. Peterson, B. J., and B. Fry. 1987. Stable isotopes in ecosystem studies. Annu. Rev. Ecol. Sys. 18:293-320. Pinnegar, J. K., and N. V. C. Polunin. 1999. Differential fractionation of 813C and 515N among fish tissues: implications for the study of trophic inter- actions. Funct. Ecol. 13:225-231. Post, D. M. 2002. Using stable isotopes to estimate trophic posi- tion: models, methods, and assumptions. Ecology 83:703-718. Power, M., G. Power, F. Caron, R. R. Doucett, and K. R. A. Guiguer. 2002. Growth and dietary niche in Saluelinus alpinus and Saluelinus fontmalis as revealed by stable isotope analysis. Environ. Biol. Fishes 64:75-85. Quan, W., C. Fu, B. Jin, Y. Luo, B. Li, J. Chen, and J. Wu. 2007. Tidal marshes as energy sources for commercially important nektonic organisms: stable isotope analysis. Mar. Ecol. Prog. Ser. 352:89-99. Quevedo, M., R. Svanback, and P. Ekloav. 2009. Intrapopulation niche partitioning in a gener- alist predator limits food web connectivity. Ecology 90:2263-2274. Rooker, J. R., J. P. Turner, and S. A. Holt. 2006. Trophic ecology of Sargassum -associated fishes in the Gulf of Mexico determined from stable isotope analysis. Mar. Ecol. Prog. Ser. 313:249-259. Ruiz, N., N. Dubois, G. Wielgosz-Collin, T. R. duPont, J. P. Berge, Y. F. Pouchus, and G. Barnathan. 2007. Lipid content and fatty acid composition of a marine-derived Trichoderma longibrachiatum strain cultured by agar surface and submerged fermentation. Process Biochem. 42:676-680. Sullivan, M. J., and C. A. Moncreiff. 1990. Edaphic algae are an important component of salt marsh food-webs: evidence from a multiple stable iso- tope analysis. Mar. Ecol. Prog. Ser. 62:149-159. Sweeting, C. J., S. Jennings, and N. V. C. Polunin. 2005. Variance in isotopic signatures as a descriptor of tissue turnover and degree of omnivory. Funct. Ecol. 19:777-784. Thomas, J. W., and S. Huke. 1996. The Forest Service approach to healthy ecosys- tems. J. Forest. 94( 8 ): 14— 18. Vetter, R. D. 1982. Seasonal metabolic compensation in sympatric seatrout: adaptation to the estuary. Trans. Am. Fish. Soc. 111:193-198. Fulford and Dillon Intrapopulatson variability in stable isotope data for Cynoscon nebulosus 121 Winemiller, K. O., S. Akin, and S. C. Zeug. 2007. Production sources and food web structure of a temperate tidal estuary: integration of dietary and sta- ble isotope data. Mar. Ecol. Prog. Ser. 343:63-76. Zanden, H. B. V., K. A. Bjorndal, K. J. Reich, and A. B. Bolten. 2010. Individual specialists in a generalist population: results from a long-term stable isotope series. Biol. Lett. 6:711-714. 122 A new species of western Atlantic lizardfish (Teleostei: Synodontidae: Sy nodus) and resurrection of Sy nodus bondi Fowler, 1939, as a valid species from the Caribbean with redescriptions of S. bondi, S . foetens (Linnaeus, 1766), and S. intermedius (Agassiz, 1829) Benjamin W. Frable (contact author)1- 2 Carole C. Baldwin1 Brendan M. Luther1 Lee A. Weigt1 Email address for contact author: frableb@si.edu 1 National Museum of Natural History Smithsonian Institution P.O. Box 37012 Washington, D C 20013-7012 2 Department of Fisheries and Wildlife Oregon State University 104 Nash Hall Corvallis, Oregon 97331 Abstract— Western Atlantic synodon- tid species were studied as part of an ongoing effort to reanalyze Ca- ribbean shorefish diversity. A neigh- bor-joining tree constructed from cytochrome c oxidase I (COI) data revealed 2 highly divergent genetic lineages within both Synodus inter- medius (Agassiz, 1829) (Sand Diver) and S. foetens (Linnaeus, 1766) (In- shore Lizardfish). A new species, Synodus macrostigmus, is described for one of the S. intermedius lin- eages. Synodus macrostigmus and S. intermedius differ in number of lateral-line scales, caudal pigmen- tation, size of the scapular blotch, and shape of the anterior-nostril flap. Synodus macrostigmus and S. intermedius have overlapping geo- graphic and depth distributions, but S. macrostigmus generally inhabits deeper water (>28 m) than does S. intermedius and is known only from coastal waters of the southeastern United States and the Gulf of Mex- ico, in contrast to those areas and the Caribbean for S. intermedius. Synodus bondi Fowler, 1939, is res- urrected from the synonymy of S. foetens for one of the S. foetens ge- netic lineages. The 2 species differ in length and shape of the snout, num- ber of anal-fin rays, and shape of the anterior-nostril flap. Synodus bondi and S. foetens co-occur in the central Caribbean, but S. bondi otherwise has a more southerly distribution than does S. foetens. Redescriptions are provided for S. intermedius, S. foetens , and S. bondi. Neotypes are designated for S. intermedius and S. foetens. A revised key to Synodus species in the western Atlantic is presented. Manuscript submitted 11 February 2012. Manuscript accepted 8 February 2013. Fish. Bull. 111:122-146 (2013). Publication date: 28 March 2013. doi 10.7755/FB.111.2.2 The views and opinions expressed or implied in this article are those of the author (or authors) and do not necessar- ily reflect the position of the National Marine Fisheries Service, NOAA. The aulopiform lizardfish family Synodontidae is represented in the western Atlantic by 3 genera: Syno- dus (5 species recognized before this study), Saurida (4), and Trachino- cephalus (1). Lizardfishes are ben- thic predators in numerous ecosys- tems, including coral reefs, estuaries, and reef structure or sandy bottom areas on continental shelves (Ander- son et al., 1966; Cressey, 1981; Ran- dall, 2009). Once thought to occupy a mid-trophic position and employ a sit-and-wait predation strategy, adult lizardfishes now are known to occupy a high trophic position — apex in some systems — as active hunters feeding primarily on other predatory fishes (Cruz-Escalona et al., 2005). Despite having no commercial value as food fishes, Synodus spp. and Saurida brasiliensis (Largescale Lizardfish) are caught as bycatch in shrimp-trawl fisheries in the west- ern Atlantic, accounting for 1.5% and 1.8% of total-catch biomass in the Gulf of Mexico and North Carolina,1 1 Brown, K. 2009. Interstate fisheries management program implementation respectively (Jeffers et al., 2008; Manjarres et al., 2008). Populations of Synodus foetens (Linnaeus, 1766) (Inshore Lizardfish) are estimated to be at fully exploited levels as bycatch in Gulf of Mexico shrimp-trawl fish- eries (Garcia-Abad et al., 1999; Wells, 2007; Jeffers et al., 2008). Proper management and ecological inves- tigation of commercially fished spe- cies require an accurate understand- ing of species diversity. For example, Collette et al. (1978) discovered that the Caribbean and Brazilian popula- tions of the commercially important Scomberomorus maculatus (Spanish Mackerel) constitute a distinct spe- cies, which they named S. brasilien- sis (Serra). Without this systematic study, populations of S. brasiliensis would still be managed under the same plan as S. maculatus. for North Carolina. Job 3: character- ization of the near-shore commercial shrimp trawl fishery from Carteret County to Brunswick County, North Carolina. Southeast Fisheries Science Center, Miami, FL, 29 p. [Available on- line from http://www.sefsc.noaa.gov/se- dar/download/SEDAR20-ASMFC-DW09. pdf?id=DOCUMENT] Frable et al.: Description of a new species of Synodus in the western Atlantic Ocean 123 Scientists have periodically investigated the system- atics of synodontids since Linnaeus (1758) described Esox synodus in the mid-18th century. Throughout the 19th century, researchers described 17 Synodus species in the western Atlantic (Anderson et al., 1966; Meek, 1884; Norman, 1935). In the most recent comprehensive treatments of western Atlantic lizardfishes (Anderson et al., 1966; Russell, 2003), 5 species of Synodus have been recognized. Ongoing research to evaluate diver- sity of the Caribbean ichthyofauna with DNA barcod- ing (Hebert et al., 2003) and traditional morphological investigation have led to the recent discovery of many new cryptic fish species and the resurrection of several formerly synonymized ones in what were thought to be well-studied taxa (Baldwin and Weigt, 2012; Baldwin et al., 2009, 2011; Tornabene et al., 2010; Victor, 2007, 2010). In the course of the present work, discrepancies were revealed between barcode data and the currently accepted species classification of western Atlantic Syn- odus lizardfishes. Specifically, each of 2 Synodus spe- cies, S. intermedius (Agassiz, 1829) (Sand Diver) and S. foetens comprise 2 distinct cytochrome c oxidase I (COI) lineages. The purpose of this study was to reconcile genetic lineages with the nominal species of western Atlantic Synodus. Through comparative morphological study, our first goal was to determine if the “extra” genetic lineages correspond with morphologically distinct spe- cies and, if so, to assess whether they represent pre- viously synonymized or undescribed species. Herein, we resurrect and redescribe Synodus bondi Fowler, 1939, from the synonymy of S. foetens and describe S. macrostigmus as a new species distinct from S. inter- medius. We establish neotypes for S. intermedius and S. foetens and redescribe both species. We discuss pre- liminary evidence of population structure within S. foe- tens, S. synodus (Red Lizardfish), Saurida brasiliensis , and Trachinoeephalus myops (Snakefish) and species- level genetic structure within Synodus poeyi (Offshore Lizardfish). Finally, we provide a revised key for the 7 species of Synodus found in the western Atlantic. Materials and methods Specimens for genetic analysis were collected in Tobago (Trinidad and Tobago), Turks and Caicos Islands, the Bahamas, Curasao, Belize, North Carolina, South Caro- lina, and Florida. Type material and additional speci- mens from other localities were examined from ANSP, MCZ, UF, FSBC, KU, and USNM (institutional abbre- viations are listed at http://www.asih.org/node/204, ac- cessed February 2012). Specimens examined are listed in appropriate species sections or in the Appendix. Specimens were collected with the use of quinaldine sulfate, rotenone, or a pole spear during snorkeling or scuba diving, as well as by bottom trawling and hook and line fishing. For most specimens, field protocol in- volved taking digital photographs of fresh specimens to document living color patterns, and subsequently a small sample of tissue from the trunk musculature for genetic analysis. Voucher specimens were fixed in 10% formalin and ultimately preserved in 75% ethanol for archival storage. Measurements were taken to the nearest 0.1 mm with Mitutoyo digital calipers (Mitutoyo Corp., Ja- pan2). Measurements and counts follow Hubbs and La- gler (1964) and Randall (2009), except as noted below. Length of the anterior-nostril flap was measured from the posterior tip of the anterior nostril to the distal end of the flap when depressed. For the S. intermedius group only, the length of the scapular blotch was mea- sured on an anterior-posterior axis at its greatest ex- panse. For the S. foetens group only, the width of the adipose lid was measured as the maximum distance between the bony orbit and distal edge of the lid. Numbers of vertebrae and dorsal-, anal-, pectoral-, and caudal-fin rays were counted from digital radio- graphs or preserved specimens. Scales flanking the dor- sal- and anal-fin bases are half the size of other trunk scales and are reported as half-scales. Tissue samples were stored in saturated salt buffer (Seutin et al., 1990). DNA was extracted from up to approximately 20 mg minced, preserved tissue through an automated phenol: chloroform extraction on an Au- togenprep965 DNA extraction system (Autogen, Hoi- liston, MA) using the mouse tail tissue protocol to a final elution volume of 50 pL. In the polymerase chain reaction (PCR), 1 pL of extracted DNA was used in a 10 pL reaction with 0.5 U BioLine (Bioline USA, Inc., Boston, MA) Taq polymerase, 0.4 pL 50 mM MgCl2, 1 pL 10x buffer, 0.5 pL 10 mM deoxyribonucleotide tri- phosphate, and 0.3 pL 10 pM each primer FISH-BCL ( 5'TCAAC YAATC AYAAAGATATYGGC AC ) and FISH- BCH (5'-TAAACTTCAGGGTGACCAAAAAATCA). The PCR theromcycle protocol was: 1 cycle of 5 min at 95°C; 35 cycles of 30 s at 95°C, 30 s at 52°C, and 45 s at 72°C; 1 cycle of 5 min at 72°C; and a hold at 10°C. PCR prod- ucts were purified with ExoSAP-IT (Affymetrix, Santa Clara, CA) with 2 pL 0.2x enzyme and incubated for 30 min at 37°C. The reaction was then inactivated for 20 min at 80°C. Sequencing reactions were performed with 1 pL of this purified PCR product in a 10 pL reaction that contained 0.5 pL primer, 1.75 pL BigDye reaction buffer (Life Technologies Corp., Carlsbad, CA), and 0.5 pL BigDye in the thermal cycler for 30 cycles of 30 s at 95°C, 30 s at 50°C, 4 min at 60°C, and then were held at 10°C after completion of cycles. These sequenc- ing reactions were purified with MultiScreen-HV plates (MAHVN4550; EMD Millipore Corp., Billerica, MA) ac- cording to the manufacturer’s instructions and stored dry until analyzed. Sequencing reactions were analyzed on an Applied Biosystems 3730XL automated DNA se- quencer (Life Technologies Corp.), and sequence trace 2 Mention of trade names or commercial companies is for iden- tification purposes only and does not imply endorsement by the National Marine Fisheries Service, NOAA. 124 Fishery Bulletin 111(2) files were exported into Sequencher, vers. 4.7 (Gene Codes Corp., Ann Arbor, MI). With the Sequencher pro- gram, ends were trimmed from the raw sequences until the first and last 10 bases contained fewer than 5 base calls with a confidence score (phred score) lower than 30. After trimming, forward and reverse sequences for each specimen were assembled. Each assembled pair was examined and edited by hand, and each sequence was checked for stop codons. Finally, the consensus se- quence (655 bp) from each contig was aligned and ex- ported in a NEXUS file format with PAUP* software, vers. 4.0 beta 10 (sensu Swofford, 2003). MEGA software, vers. 5.05, (Tamura et ah, 2011) was used to generate a distance matrix of Kimura 2-param- eter distances (Kimura, 1980) of genetic sequence simi- larity, from which a neighbor-joining tree (Saitou and Nei, 1987) was constructed. The neighbor-joining tree is not intended to reflect phylogenetic relationships. The label for each entry on the tree is our DNA number, and we include that number in the material examined sec- tions and figure captions. Abbreviations used in DNA numbers reflect geographical location, expedition name, or institutions that provided COI sequences or speci- mens for genetic analysis: BAH=Bahamas; BLZ=Belize; CUR=Cura SL (5.5%) and 16.7-24.7% HL (20.8%), orbit with bony ridge extending over an- terodorsal margin; interorbital width 2. 8-4. 3% SL (3.6%-) and 9.5-15.3% HL (13.8%-). Predorsal length 37.8-43.6% SL (40.4%); prepelvic length 32.2—36.5% SL (34.3%); preanal length 67.6- 72.5% SL (70.4%); preadipose length 75.7-82.1% SL (79.3%); dorsal-fin base 15.0-18.2% SL (16.5%), longest 128 Fishery Bulletin 111(2) 20 mm 20 mm 20 mm 20 mm Figure 3 Comparison of coloration: (A) Synodus macrostigmus , holotype, UF 182810, coloration of live specimen; (I ) S. macrostigmus, new species, holotype, UF 182810, 189 mm SL, coloration of preserved specimen (C ) S. inte me- dius , neotype, USNM 398292, 285 mm SL, coloration of live specimen, photo by C. Boucher, NOAA; and (I ) S. intermedius, neotype, USNM 398292, 285 mm SL, coloration of preserved specimen. Frable et al.: Description of a new species of Synodus in the western Atlantic Ocean 129 North America X ’’ ' W.m Guifof \ ^ Mexico .*▼/ ▼▼ T ▼ mi I “wim» * * = Synodus intermedius ▼ = Synodus macrostigmus Atlantic Ocean Q 1 0°N \ X X N 0° Figure 4 Distribution map Synodus intermedius and S. macrostigmus, new species. Symbols indicate sampling localities, not individual specimens. Dashed-line ovals highlight area of overlap in distribution of these species. Map by Robert Myers (Coral Graphics), reprinted with permission. dorsal-fin ray 15.2-17.0% SL (16.1%); anal-fin base 13.0-15.5% SL (14.3%), longest anal-fin ray 6. 9-9. 4% SL (7.7%); pectoral-fin length 12.5-14.7% SL (13.5%); pelvic-fin length 21.6-25.2% SL (23.3%); caudal-pe- duncle length 9.9-16.7% SL (13.4%), caudal-fin length 16.5-21.5% SL (19.1%), ventral caudal lobe slightly longer than dorsal lobe. Upper jaw slightly shorter than lower jaw, mouth oblique, forming an 8° angle with the horizontal body axis; upper-jaw length 13.7-17.6% SL (15.9%) and 55.8-63.8% HL (60.3%). Teeth thin, needlelike; upper jaw with 2 rows of teeth, teeth in inner row longer than teeth in outer row; lower jaw with 3 rows of teeth, size of teeth increasing medially, outer row covered by lips; posterior teeth in both jaws directed slightly an- teriorly. Palatine with 3-4 rows of depressible teeth, teeth directed posteromedially. Tongue with 5 rows of posteriorly directed teeth, largest near anterior tip. An- terior nostril on level with median axis of orbit; poste- rior nostril same size and situated slightly dorsal to anterior nostril; anterior nostril with large, leaf-shaped flap that tapers distaily, length of flap when depressed 0.8-1. 2% SL (1.0%) and 2. 7-4.9% HL (3.8%); anterior portion of posterior nostril occluded by basal rim of flap (Fig. 2A); posterior nostril without flap. Cheek with 4-6 rows of large scales, modally 6; opercle with vertical row of 5-6 scales on anterior margin and several scales posteriorly; predorsal scales extending anteriorly to a position approximately one orbit diameter posterior to eye; interorbital region na- ked. Dorsal and anal fins each flanked basally by 10-12 half scales, remainder of dorsal and anal fins without scales; large, diamond-shaped scales covering pelvic-fin base, remainder of pelvic fin and pectoral fin without scales; 6 rows of body scales extending onto base of caudal fin; upper and lower caudal lobes each with a single, enlarged, horizontally elongate scale. Coloration Before preservation (Fig. 3A) Head and dorsal half of trunk gray to greenish brown, ventral half of trunk pale; 6-8 dark brown markings along length of trunk laterally, at least some of them resembling wide crosses that are widest along lateral midline; markings on left and right portions of trunk connected to one another across dorsal midline to form saddles; trunk also with several thin orange-yellow stripes dorsally, stripes thicker or broken into orange blotches ventrally; groove above maxilla with stripe of black pigment from below anterior edge of orbit to posterior end of jaw; posterior edge of branchiostegal membrane pale yellow; scapu- lar region with large black blotch of pigment partially hidden by operculum, length 3.8-4.9%< SL (4.3%) and 130 Fishery Bulletin 111(2) Frable et at: Description of a new species of Synodus in the western Atlantic Ocean 131 Table 3 Frequency distributions of counts of pored species (w=51), and S. intermedius (n= 52). lateral-line scales in Synodus macrostigmus. new Pored lateral-line scales 45 46 47 48 49 50 51 52 Synodus macrostigmus 4 12 Synodus intermedius 27 8 7 16 23 6 14.2-19.0% HL (16.4%); dorsal fin not erect in avail- able photographs (Fig. 3A), but depressed fin pale yel- low with oblique rows of dark pigment; adipose fin pale with at least one dark marking dorsally; pectoral fin pale with approximately 6 wavy, irregular bars of dark brown pigment; pelvic fins bright yellow; anal fin yel- low; center of caudal fin (at anterior end of fork) and posterior portion of ventral caudal lobe dark gray to brown; dorsal caudal lobe pale with light, dusky pig- ment on posterior margin; dorsal caudal lobe with 2-3 small brown bars on procurrent rays. In preservative (Fig. 3B) Trunk tan to light brown dor- sally, pale ventrally; dark markings along trunk pres- ent but faded in some specimens; black pigmentation on scapular region, head, and caudal fin (described in the previous paragraph) easily visible; barring on pec- toral fin indistinct. Etymology Synodus macrostigmus is named for its large, black scapular blotch. The species name is a Latinized con- junction of the Greek macros, meaning large or long, and stigma, meaning brand or mark. Distribution The holotype and paratypes are from the northeastern Gulf of Mexico off the coast of Florida from depths of 71-75 m. Additional specimens of this species, previ- ously identified as S. intermedius in museum collec- tions, are from the northern Gulf of Mexico off Ala- bama and Florida, Yucatan Mexico, and Atlantic coast off Georgia and South Carolina. All specimens were collected at depths below 28 m. The specimens from the deepest known collection, UF46974, are from the Gulf Coast of Florida at 194 m. Synodus intermedius (Agassiz, 1829) English common name: Sand Diver Figures 1, 2B, 3 (C and D), and 4, Tables 1-3 Saurus intermedius Agassiz in Spix and Agassiz, 1829:81, pi. 44. Type locality: “Brazil;” described from specimen, 216 mm in total length in the Museo Monacensi, current whereabouts unknown; neotype herein designated, USNM 398292. Saurus anolis Valenciennes in Cuvier and Valenci- ennes, 1850:483. Type localities: Martinique, Guade- loupe, and Bahia, Brazil; 2 types — MNHN A. 8611, 390 mm, and MNHN B.1022, 400 mm. Synodus cubanus Poey, 1876:143. Type locality: Cuba; one type — 375 mm, current whereabouts unknown. Holotype No types known. Designation of neotype Agassiz (in Spix and Agassiz, 1829) described Saurus intermedius from a single specimen collected off Brazil (Kottelat, 1988). The original description is brief and provides little information to separate this species from a general Synodus body plan with the exception of the mention of transverse barring on the trunk, 55 mucous canal scales in the lateral line, a projected snout, and 12 pectoral- and anal-fin rays. The number of pored lat- eral-line scales is higher than the numbers observed in our study (49-52 scales); however, specimens of S. inter- medius are known to have up to 55 lateral-line scales (Anderson et al., 1966; Anderson and Gehringer, 1975). The number of anal-fin rays (12) is also greater than the number we observed (11). This difference might be explained by a deep bifurcation in the last anal-fin ray that makes it appear to be 2 separate rays. The deep split is noted to result in miscounting (Anderson et al., 1966). There is no mention of caudal pigmentation or a scapular blotch. Subsequent descriptions of S. intermedius by Gun- ther (1864), Poey (1868), and Meek (1884) all includ- ed diagnostic features, such as a scapular blotch and caudal barring. Gunther (1864) expressed frustration over the lack of diagnostic information in the original description. Kottelat (1988) was unable to locate the original type of Saurus intermedius. It may have been stored at the Zoologische Staatssammlung Munchen (ZSM), but the entire Spix and Agassiz collection housed there was destroyed in World War II (Kottelat, 132 Fishery Bulletin 111(2) 1988). Or, it may be housed in the Museum d’Histoire Naturelle de Neuchatel (MHNN), in Switzerland, where it has not been located. A color illustration of the ho- lotype was published in Spix and Agassiz (1829, plate 44). The body shape, pigmentation, and barring on the caudal fin depicted in this illustration match those characters in later descriptions of Saurus intermedius and Synodus intermedius and in specimens recognized herein as S. intermedius. In light of our discovery of the similar S. macrostigmus , designation of a neotype is appropriate to clarify the taxonomic status of S. in- termedius. We designate USNM 398292 from Brazil as a topologically equivalent neotype. We examined speci- mens of Synodus intermedius from Brazil and numer- ous localities in the Caribbean, and we found no sig- nificant morphological differences. Note on authorship Although Synodus intermedius was formally described in Spix and Agassiz (1829), Kottelat (1988) determined that Agassiz was the sole author of this name and the description was subsequently incorporated into this text. According to priority, the authorship of S. in- termedius has been changed from (Spix and Agassiz, 1829) to (Agassiz, 1829) following the adopted format for other species described by Agassiz in the same work, including Saurus longirostris (Reis et al, 2003). Neotype USNM 398292, 285.0 mm SL, off northeastern Brazil, 0°13 48.00 N, 44°49 47.99 W, 62-64 m, sta. 67, Oregon II, 16 May 1975, field number BBC 1631, Coll: B. B. Collette. Additional material (DNA numbers and GenBank accession numbers for specimens that are vouchers are given in parentheses following catalog numbers). Florida: UF 146453, 3; UF 174147, 2; UF 234684, 1; UF 111225, 2; UF 116648, 5; UF 176286, 1; UF 152825, 1; UF 154795, 3; UF 152128, 3; UF 176286,1; UF 29818, 4; USNM 38711, 1; USNM 35045, 1; Puerto Rico: UF 234199, 2; Bahamas Islands: USNM 405474, 1; USNM 405461 (BAH 9029: JX519387), 1; USNM 405462 (BAH 10043: JX519402), 1; USNM 405463 (BAH 10180: JX519386), 1; USNM 405464 (BAH 10181: JX519366), 1; Belize: USNM 327555, 1; USNM 404211 (BLZ 10211: JX519367), 1; USNM 405465 (BLZ 7151: JQ841412), 1; USNM 405468 (BLZ 8163: JQ841835), 1; USNM 405469 (BLZ 8006: JQ841836), 1; Bermuda: USNM 368585, 1; USNM 385983, 1; Brazil: USNM 398292, 1; Colombia: UF 123373, 1; UF 137038, 3; USNM 384339, 1; Curacao: USNM 405470 (CUR 8386: JQ842339), 1; French Gui- ana: UF 211584, 1; Guyana: UF 137049, 3; Honduras: UF 136971, 1; UF 123383, 1; Jamaica: UF 123615, 1; UF 231377, 5; Saba, Leeward Islands: UF 207329, 2; To- bago, Trinidad and Tobago: USNM 405471 (TOB 9160: JQ843084), 1; USNM 405472 (TOB 9090: JQ843082), 1; UF 123378, 1; Turks and Caicos: USNM 405473, 1; Venezuela: UF 224467, 4; UF 123377, 1. Diagnosis A species of Synodus differentiated from its congeners by the following combination of characters: dorsal-fin rays 11-12; anal-fin rays 11; total caudal rays 40-42; dorsal segmented caudal rays 9; ventral segmented caudal rays 9-10; dorsal procurrent caudal rays 11-12; ventral procurrent caudal rays 9-10; pelvic-fin rays 8; pectoral-fin rays 11-13; total vertebrae 49-52, modally 50; pored lateral-line scales 49-52, modally 51; median predorsal scalesl6 or 17; scale rows on cheek 6-8; HL 15.2-28.4% SL (26.6%); orbit diameter 3. 6-6. 4% SL (4.7%) and 13.4—22.5% HL (17.2%); interorbital width 2. 8-5. 3% SL (4.3%) and 10.1-19.4% HL (15.9%); flap on anterior nostril small and broad, length 1.5-3. 6% HL (2.4%); snout triangular in dorsal view; pelvic-fin length 23.1-26.7% SL (24.9%). In preservative, 3 to 6 dark bars on caudal fin spanning upper and lower lobes; a small, rectangular-shaped, black scapular blotch (length 5.1-12.2% HL, mean 8.4%); 9 to 13 dark bars on trunk; in life, adults tan to olive and with un- evenly distributed yellow stripes along body. Description Description based on 51 specimens, 59.2-285.0 mm SL. Counts and measurements of neotype given in Table 2. Frequency distribution of pored lateral-line scales given in Table 3. Dorsal-fin rays 11-12; anal-fin rays 11; total caudal- fin rays 40-42; dorsal segmented caudal rays 9; ventral segmented caudal rays 9-10; dorsal procurrent caudal rays 11-12; ventral procurrent caudal rays 9-10; pel- vic-fin rays 8; pectoral-fin rays 11-13; total vertebrae 49-52, modally 50; predorsal vertebrae 12 or 13; scales ctenoid; pored lateral-line scales, 49-52, modally 51, although counts of up to 55 have been reported (An- derson et ah, 1966; Anderson and Gehringer, 1975); me- dian predorsal sca!esl5-17; scales above lateral line to dorsal-fin base 3.5; scales below lateral line to anal-fin base 4.5. Body cylindrical; depth at pelvic-fin origin 12.4- 16.3% SL (14.1%); depth at anal-fin origin 8.7-11.7% SL (10.2%); caudal-peduncle depth 5. 1-6.2% SL (5.7%); body width at dorsal-fin origin 9.7-16.2% SL (14.2%); HL 24.5-28.4% SL (26.6%); snout length 6. 0-7. 3% SL (6.7%) and 22.0-27.5% HL (24.7%), snout triangular in dorsal view; orbit diameter 3. 6-6. 4% SL (4.7%) and 13.4-22.5% HL (17.2%), orbit with bony ridge extend- ing over anterodorsal margin; interorbital width 2.8- 5.3% SL (4.3%) and 10.1-19.4% HL (15.9%>). Predorsal length 38.0—43.5% SL (41.0%); prepelvic length 33.4-37.9% SL (35.0%); preanal length 69.2- Frable et al Description of a new species of Synodus in the western Atlantic Ocean 133 76.4% SL (72.2%); preadipose length 75.5-84.5% SL (80.2%); dorsal-fin base 13.9-19.4% SL (16.8%), longest dorsal-fin ray, usually third, 14.5-20.0% SL (17.0%); anal-fin base 12.7-15.9% SL (14.2%), longest anal-fin ray, usually second, 7.7-10.7% SL (9.1%); pectoral-fin length 12.1-14.0% SL (13.1%); pelvic-fin length 23.1- 26.7% SL (24.9%); caudal-peduncle length 8.5-19.4% SL (14.9%), caudal-fin length (only measured in 19 specimens because many specimens had damaged cau- dal fins) 16.1-22.5% SL (19.4%), ventral caudal lobe slightly longer than dorsal lobe. Mouth terminal, upper jaw slightly oblique, form- ing an 8° angle with the horizontal body axis; upper- jaw length 14.7-18.1% SL (16.4%) and 53.9- 64.2% HL (60.4%). Teeth needlelike; upper jaw with 2 rows of teeth, teeth in inner row longer than teeth in outer row; lower jaw with 3 rows of teeth, size of teeth in- creasing medially, outer row covered by lips; posterior teeth in both jaws directed slightly anteriorly. Palatine with 3-4 rows of depressible teeth, teeth directed pos- teromedially. Tongue with 5 rows of posteriorly direct- ed teeth, largest near anterior tip. Anterior nostril on level with median axis of orbit; posterior nostril same size and situated slightly dorsal to anterior nostril; an- terior nostril with small, spade-shaped flap, length of flap when depressed 0. 4-1.0% SL (0.6%) and 1.5-3. 6% HL (2.4%); anterior nostril flap does not reach posterior nostril (Fig. 2B); posterior nostril without flap. Cheek with 6-8 rows of large scales, modally 7; opercle with vertical row of 6-8 scales on anterior margin and several scales posteriorly; predorsal scales extending anteriorly to a position approximately one orbit diameter posterior to eye; interorbital region na- ked. Dorsal and anal fins each flanked basally by 10-12 half scales, remainder of dorsal and anal fins without scales; large, diamond-shaped scales covering pelvic-fin base, remainder of pelvic fin and pectoral fin without scales; 5-7 rows of body scales extending onto base of caudal fin; upper and lower caudal lobes each with a single, enlarged, horizontally elongate scale. Coloration Before preservation (Fig. 30 Head and dorsal half of trunk gray-green to brown, ventral half of trunk pale; specimens more than 200 mm SL with 9-13 dark brown bars along length of trunk laterally, bars more faint in smaller specimens and resemble wide crosses that are widest along lateral midline; markings on left and right portions of trunk connected to one another across dorsal midline to form saddles; trunk with sev- eral thin, golden-yellow stripes dorsally, with stripes becoming disrupted near lateral midline and forming distinct stripes again ventrally; groove above max- illa with stripe of black pigment from below anterior edge of orbit to posterior end of jaw; posterior edge of branchiostegal membrane yellow; scapular region with small rectangular black blotch of pigment partially hid- den by operculum, length 1.4-3. 3% SL (2.3%) and 5.1- 12.2%- HL (8.4%); dorsal fin yellow-brown with multiple oblique rows of dark pigment; adipose fin brown; pec- toral fin translucent with 4-5 diagonal bars of dark brown pigment; pelvic fins golden-yellow with darker pigment between rays; anal fin yellow; caudal fin light brown with 3-5 rows of dark pigmentation spanning both lobes, posterior fringes of fin dark brown. In preservative (Fig. 3D) Trunk tan to light brown dor- sally, pale ventrally; dark markings along trunk pres- ent but faded in some specimens; black pigmentation on scapular region and jaw easily visible; barring on caudal fin less distinct in older specimens; barring on dorsal fin faded and pigmentation on pectoral and pel- vic fins indistinct. Distribution Synodus intermedins was described originally from “the inlets and river outflows of Brazil” (Agassiz in Spix and Agassiz, 1829). The species also occurs widely through- out the western Atlantic. Specimens examined in this study are from Bermuda, the Gulf of Mexico, and the eastern coasts of the United States, Puerto Rico, Saba, Leeward Islands, Belize, Venezuela, Colombia, French Guiana, Guyana, Brazil, and Trinidad and Tobago (Fig. 4). Specimens are known from depths up to 183 m, but most specimens have been collected at depths <60 m. Comparisons of Synodus macrostigmus, 5. intermedius, and congeners The anterior-nostril flap in S. macrostigmus is signifi- cantly larger (2.7- 4.9% HL, mean 3.8%) than that flap of S. intermedius (1. 5-3.6%- HL, mean 2.4%) and tapers distally (ends more abruptly in S. intermedius — Fig. 2, A and B). Synodus macrostigmus on average possesses a larger orbit diameter (20.8% versus 17.2% HL) and more narrow interorbital (width 13.8% versus 15.9% HL). It possesses a more blunt snout than does S. inter- medius (Fig. 3, A-D) and has fewer lateral-line scales (45-48, modally 47, versus 49-52, modally 51, in S. in- termedius) (Table 3). Dark pigment on the caudal fin in S. macrostigmus is restricted primarily to the cen- ter of the fin and ventral lobe, whereas S. intermedius has a distinctive banded pattern with 3 to 6 bars. The scapular blotch in S. macrostigmus is much larger and more ovoid than the miniscule, rectangular marking in S. intermedius, mean length of blotch 16.4%- HL in S. macrostigmus and 8.4% HL in S. intermedius. Additional fresh specimens of the new species are needed to verify differences in color patterns, but, in available material, S. macrostigmus has orange lateral stripes and markings on the trunk and S. intermedi- us has yellow stripes (Fig. 3, A and C). Specimens of S. intermedius more than 200 mm SL also appear to have more vertical bars along the trunk (9-13) than do large specimens of S. macrostigmus (6-8). Finally, S. macrostigmus has not been collected at depths shal- 134 Fishery Bulletin 111(2) lower than 28 m, whereas nearly half of the specimens of S. intermedius examined are from those depths. Syn- odus macrostigmus generally inhabits deeper waters (mean depth 96.5 m) than does S. intermedius (mean depth 49.3 m), but they co-occur at many depths. For example, 4 specimens of S. intermedius and 1 speci- men of S. macrostigmus (UF 29818) were collected in a single trawl off Florida. From their congeners, S. macrostigmus and S. in- termedius can be differentiated from S. poeyi by an- terior dorsal-fin rays that do not extend beyond the distal tips of succeeding rays when the fin is depressed and from S. saurus (Bluestripe Lizardfish), S. bondi, S. foetens, and S. synodus in having fewer lateral-line scales (45-48 in S. macrostigmus, 55 or more in the other species). Remarks Two previously described species are currently rec- ognized as synonyms of Synodus intermedius (Agas- siz, 1829): Saurus anolis (Valenciennes in Cuvier and Valenciennes, 1850) and Synodus cubanus Poey, 1876. Valenciennes described Saurus anolis in 2 brief paragraphs in Cuvier and Valenciennes (1850), but he did not mention a scapular blotch and compared the specimens only with S. synodus. Type material was deposited in the Museum National d’Histoire Naturel- le (MNHN) in Paris; 2 syntypes exist in this mate- rial, one dried (MNFIN A-8611) and one in alcohol (B- 1022) (Bertin and Esteve, 1950). The wet specimen ap- pears to be S. intermedius on the basis of caudal bar- ring and remnants of a miniscule scapular blotch. Diagnostic features are not discernible on the dried syntype. Albert K. L. G. Gunther purportedly synonymized Saurus anolis with Synodus intermedius before 1868 (Poey, 1868); however, no published reference has been located. Meek (1884) recognized Saurus anolis as valid in his systematic review of Synodontidae in the west- ern Atlantic, but he stated the original description is “so insufficient that no certain identification can be made” (Meek, 1884: 134). Anderson et al. (1966) listed Saurus anolis as a synonym of Synodus intermedius without providing discussion. We concur with Meek’s assessment of the original description but recognize Saurus anolis as a synonym of Synodus intermedius on the basis of features of one of the MNHN Saurus anolis syntypes. The second synonym, Synodus cubanus, was de- scribed as having caudal barring, yellow body stripes, short pelvic fins, and a scapular spot (Poey, 1876). The description was based on a 375-mm-SL specimen from Cuba, but the whereabouts of this specimen are unknown. Jordan (1884) identified a specimen from Florida (USNM 35045) as Poey’s S. cubanus, but he noted that it matches the description of S. interme- dius by previous authors (e.g., Gunther, 1864). Our examination of USNM 35045 revealed a small scapu- lar marking typical of S. intermedius and diagnostic morphological features, such as 51 lateral-line scales. We agree that S. cubanus Poey, 1876, is a synonym of S. intermedius. Synodus foetens (Linnaeus, 1 766) English common name: Inshore Lizardfish Figures 1, 2C, 5 (A and B), and 6, Tables I, 4, and 5 Salmo foetens Linnaeus, 1766: 513 (12th ed.). Descrip- tion based on Catesby, 1743; specimens sent by Dr. Alexander Garden of Charleston, South Carolina; no types designated. Osmerus albidus Lacepede, 1803: 229. Name given in list and based on descriptions by Catesby and Linnaeus. Type locality: South Carolina; no types designated. Coregonus ruber Lacepede, 1803: 243. Name given in list and based on description by Plumier. Type local- ity: Martinique; no types designated. Esox salmoneus Mitchill, 1815: 442. Type locality: New York Bay; described from specimens, 203-229 mm in total length; no types designated. Saurus longirostris Agassiz in Spix and Agassiz, 1829: 80, pi. 43. Type locality: Brazil; described from speci- mens, 178-216 mm in total length; current where- abouts unknown. Saurus mexicanus Cuvier, 1829: 314. Type locality: Gulf of Mexico; no types designated. Saurus spixianus Poey, 1860: 304. Type locality: Cuba; one type — MCZ 6884, 330 mm. Holotype No types known. Designation of neotype Linnaeus (1766) described Salmo foetens from material collected off the coast of South Carolina. Type material is not present in known collections of Linnaean speci- mens (Wheeler, 1985; 1991), and, although the origi- nal description is vague, it indicates the presence of 12 anal-fin rays. We believe that the genetic lineage of “ Synodus foetens” in our data set that comprises speci- mens with 12 anal-fin rays is Synodus foetens (Linnae- us, 1766); the type locality of Salmo foetens provides corroborative evidence, because several specimens in our Synodus foetens lineage are from South Carolina. To stabilize the taxonomic status of Synodus foetens and distinguish it from S. bondi, formerly considered a synonym of S. foetens but resurrected herein (see entry for S. bondi below), we establish a neotype for S. foe- tens. We have selected USNM 405413 as the neotype on the basis of its collection off the type locality, South Carolina. Frable et al Description of a new species of Synodus in the western Atlantic Ocean 135 Neotype USNM 405413, 205 mm SL, South Carolina, 32°47'34.80'N, 79°39'46.80'W, 10 m, Sta. 20110526, 17 October 2011, Coll: P. Webster. Neotype COI sequence GenBank sequence JX519368, USNM 405413 (neotype), DNA number SC 11001 Additional material (DNA numbers for specimens that are vouchers are given in parentheses following catalog numbers). Ala- bama: USNM 398343 (FDA 170: JX5 19375), 1; USNM 358612, 6; Georgia: UF 143583, 1; Florida: KU 29686 (KUIT 3949: JX519400, KUIT 3950: JX519401), 2; UF 177024, 1; UF 177327, 3; UF 177249, 1; USNM 160462, 3; USNM 57123, 1; USNM 38710, 1; USNM 405452, 15; USNM 405450 (SMS 7502: JQ842743), 1; USNM 40545KFCC 8070: JQ841978), 1; Louisiana: UF 99242, 2; USNM 185713, 2; Mississippi: UF 137002, 3; Mary- land: USNM 125789, 1; New Jersey: USNM 399108, 1; USNM 395747, 1; North Carolina: KU 27009 (KUIT 1127: JX519409, KUIT 1128: JX519410), 2; UF 5239, 3; UF 77354, 2; UF 178337, 9; South Carolina: UF 44476, 2; UF 39667, 1; UF 39650, 6; USNM 25998, 2; USNM 405414 (SC11002: JX519369, SC11003: JX519370), 2; USNM 405475 (SC11004: JX519371), 1; USNM 405449, 2; Texas: KU 30177 (KUIT 5069: JX519411), 1; UF 54575, 10; Bahamas Islands: UF 200425, 1; Belize: BLZ 6429: JQ841024 (tissue only), 1; Bermuda: USNM 337726, 3; Cuba: MCZ 6884, 1; USNM 331820, 1; Ja- maica: UF 123664, 1; St. Martin, Leeward Islands: UF 205900, 1; Mexico: UF 125765, 1; UF 123640, 1; UF 7108, 1; UF 123660, 1; Puerto Rico: UF 137023, 1. Diagnosis A species of Synodus distinguished from all congeners by the following combination of characters: dorsal-fin rays 10-12, rarely 10; anal-fin rays 11-13, rarely 11; pectoral-fin rays 11-13; total caudal-fin rays 40-41; dorsal segmented caudal rays 9-10; ventral segmented caudal rays 9; dorsal procurrent caudal rays 11-12; ventral procurrent caudal rays 11; vertebrae 59-62; lateral-line scales 59-63, modally 60; predorsal scales 20-30; scale rows above the lateral line to dorsal-fin base 5. 5-6. 5; HL 23.5-28.0% SL; snout long and trian- gular (length 6. 0-8. 3% SL, mean 6.8%, and 24.5-31.9% HL, mean 27.1%), tip slightly rounded; anterior-nostril flap broad and triangular, length 1.7-3. 9% HL (3.0%); orbit diameter 3. 3-5. 3% SL (3.9%) and 11.6-20.1% HL (15.6%); interorbital area wide (width 9.2-18.5% HL, mean 14.6%); adipose lid around orbit narrow (width 1.9-5. 2% HL, mean 3.5%); dorsal-fin base 9.2-12.7% SL, mean 11.3%, equal in length to or shorter than anal-fin base (9.6-14.7% SL, mean 12.1%); posterior tip of pectoral fin not extending to pelvic-fin origin; color in preservation: trunk dark tan to brown dorsally, paler ventrally; 6-8 dark, vertical cross-shaped mark- ings along trunk in specimens <100 mm SL, crosses disappearing or becoming indistinct dark patches in larger specimens; dark pigment present along posterior margin of ventral portion of upper caudal-fin lobe and along entire posterior margin of lower lobe, sometimes extending onto main portion of ventral lobe; dorsal half of adipose fin dark; in life, adults olive to tan. Description Description based on 52 specimens, 58.7-380.0 mm SL. Counts and measurements of neotype given in Table 4. Frequency distribution of anal-fin rays given in Table 5. Dorsal-fin rays 10-12, rarely 10; anal-fin rays 11-13, rarely 11; total caudal-fin rays 40-41; dorsal segment- ed caudal rays 9-10; ventral segmented caudal rays 9; dorsal procurrent caudal rays 11-12; ventral procur- rent caudal rays 11; pectoral-fin rays 11-13, pelvic- fin rays 8; total vertebrae 59-62; predorsal vertebrae 17-19; scales ctenoid; pored lateral-line scales 59-63, modally 60; median predorsal scales 20-30; scale rows above the lateral line to dorsal-fin base 5. 5-6. 5; scales below lateral line to anal-fin base 6.5. Body cylindrical and elongate; depth at pelvic-fin origin 7.5-14.2% SL (11.8%); depth at anal-fin origin 6.2-10.5% SL (8.4%); caudal-peduncle depth 4. 5-5. 7% SL (5.1%); body width at dorsal-fin origin 8.7-14.2% SL (11.3%); HL 23.5-28.0%- SL (25.2%); snout length 6.0- 8.3% SL (6.8%) and 24.5-31.9%. HL (27.1%), snout long and pointed, tip rounded in dorsal view; orbit diam- eter 3. 3-5. 3% SL ( 3.9 %) and 11.6-20. 1% HL (15.6%), orbit with bony ridge extending over anterodorsal mar- gin and with narrow adipose lid on its posterior edge, adipose lid width 0.5-1. 3% SL (0.9%) and 1.9-5. 2%. HL (3.5%); interorbital region wide, width 2. 3-5. 2% SL (3.7%) and 9.2-18.5% HL (14.6%). Predorsal length 41.2-48.7% SL (44.6%); prepelvic length 32.1-40.3% SL (37.1%); preanal length 71.9- 80.1% SL (76.0%); preadipose length 77.3-86.1% SL (81.5%); dorsal-fin base 9.2-12.7%- SL (11.3%), longest dorsal-fin ray, usually third, 13.0-16.5%- SL (15.0%); anal-fin base 9.6-14.7% SL (12.1%), longest anal-fin ray, usually third, 6. 7-9. 9% SL (8.3%); pectoral-fin length 10.9-13.7% SL (12.0%); pelvic-fin length 17.7— 22.6% SL (20.1%); caudal-peduncle length 8.8-14.0% SL (10.8%), caudal-fin length (only measured in 27 specimens because many specimens had damaged cau- dal fins) 14.1-19.5% SL (17.1%), caudal lobes generally same length. Mouth terminal, upper jaw slightly oblique, form- ing a 10° angle with the horizontal body axis, upper jaw extending anterior to lower jaw; upper-jaw length 13.3-17.2% SL (15.5%)) and 55.0-65.2%- HL (61.1%). Teeth needlelike; upper jaw with 2 rows of teeth, teeth in inner row longer than teeth in outer row; lower jaw with 3 rows of teeth, size of teeth increasing medially, 136 Fishery Bulletin 111(2) Table 4 Counts and proportional measurements of the holotype and paratype of Syno- dus bondi and the neotype of S. foetens. Standard length presented in millime- ters; other measurement values in percentages of standard length; values in parentheses are percentages of head length. Synodus bondi Holotype ANSP 68634 Synodus bondi Paratype ANSP 68635 Synodus foetens Neotype USNM 408413 Standard length 220.0 155.9 205.0 Vertebrae 60 - 62 Dorsal-fin rays 13 12 14 Anal-fin rays 10 10 12 Pectoral-fin rays 13-12 14-13 14-14 Pelvic-fin rays 8-8 8-8 8-8 Caudal-fin rays 42 42 42 Pored lateral-line scales 60 59 62 Predorsal scales 24 23 28 Cheek scales 6 6 7 Body depth (pelvic origin) 12.0 10.5 13.2 Body depth (anal origin) 10.8 8.8 8.5 Caudal-peduncle depth 6.0 5.9 5.4 Body width 13.2 12.1 11.8 Head length 25.7 24.7 25.3 Snout length 7.7 (30.0) 6.5 (26.3) 7.3 (28.9) Nostril-flap length 0.8 (3.1) 0.7 (2.7) 0.6 (2.5) Orbit diameter 3.2 (12.4) 3.4 (13.6) 3.7 (14.5) Adipose-lid width 1.2 (4.7) 1.1 (4.6) 0.8 (3.1) Interorbital width 4.3 (16.6) 3.8 (15.5) 3.7 (14.5) Upper-jaw length 16.2 (62.9) 15.3 (62.0) 15.5 (61.4) Predorsal length 44.4 44.1 43.6 Preanal length 77.3 76.4 74.7 Preadipose length 81.8 81.4 79.0 Prepelvic length 36.6 37.5 36.2 Dorsal-fin base 12.8 11.9 11.4 Longest dorsal ray 16.2 16.1 15.0 Anal-fin base 11.0 11.0 12.2 Longest anal ray 9.5 8.5 8.2 Pectoral-fin length 11.8 11.6 12.3 Pelvic-fin length 18.4 18.7 20.5 Caudal-peduncle length 10.7 13.9 9.4 Caudal-fin length 18.0 20.3 16.7 outer row covered by lips; posterior teeth in both jaws directed slightly anteriorly. Palatine with 3-4 rows of depressible teeth, teeth directed posteromedially. Tongue with 5 rows of posteriorly directed teeth, larg- est near anterior tip. Anterior nostril on level with me- dian axis of orbit; posterior nostril same size and situ- ated slightly dorsal to anterior nostril, nostrils almost on dorsal surface of snout; anterior nostril with small, triangular flap, length of flap when depressed 0. 4-1.0% SL (0.8%) and 1. 7-4.0% HL (3.0%); anterior nostril flap reaches center of posterior nostril when depressed (Fig. 2C); posterior nostril without flap. Cheek with 7 rows of large scales, opercle with 3-5 vertical rows of 9-10 scales on anterior margin; pre- dorsal scales extending anteriorly above anterior mar- gin of preopercle; interorbital re- gion naked. Dorsal and anal fins each flanked basally by 8-10 half scales, remainder of dorsal and anal fins without scales; 3 large, diamond-shaped scales covering pelvic-fin base, pelvic-fins flanked by a row of 4-6 half-scales lateral- ly, remainder of pelvic fin and pec- toral fin without scales; 7-9 rows of body scales extending onto base of caudal fin; upper and lower cau- dal lobes each with a single, en- larged, horizontally elongate scale. Coloration Before preservation (Fig. 5A) Head and dorsal half of trunk olive to brown, ventral half of trunk pale; 6—8 dark, vertical, cross-shaped markings along trunk in speci- mens <100 mm SL, with crosses disappearing or becoming indis- tinct dark patches in larger speci- mens; rows of light spots extend- ing laterally over trunk; groove above maxilla with stripe of black pigment from anterior edge of orbit to posterior end of jaw; posterior edge of branchiostegal membrane pale yellow; dorsal fin brown with translucent membrane between rays; adipose fin tan with dark dorsal edge; pectoral fin translucent with dark pigment on dorsal edge; pelvic fin translucent to pale yellow; anal fin pale; dark pigment present along posterior margin of ventral portion of upper caudal-fin lobe and along entire posterior margin of lower lobe, sometimes extending onto main portion of ventral lobe. In preservative (Fig. 5B.) Trunk dark tan to brown dor- sally, paler ventrally; dark markings in small speci- Table 5 Frequency distributions of counts of anal-fin rays in Synodus bondi in= 46) and S. foetens (n= 52). Anal-fin rays 10 11 12 13 Synodus bondi 12 32 2 - Synodus foetens - 2 35 15 Frable et al : Description of a new species of Synodus in the western Atlantic Ocean 137 20 mm Figure 5 Comparison of coloration: (A) Synodus foetens, USNM 405414, 130 mm SL, coloration of live specimen; (B) S. foe- tens, neotype, USNM 405413, 205 mm SL, coloration of preserved specimen, photo by S. Raredon (USNM); (C) S. bondi, AMNH 1-245350, 139 mm SL, coloration of live specimen, photo by D. R. Roberston (Smithsonian Institution) and J. Van Tassell (AMNH); and (D) S. bondi , holotype, ANSP 68634, 220 mm SL, coloration of preserved specimen. 20 mm 20 mm 138 Fishery Bulletin 111(2) North America ' ▼ ■IP ♦ ♦ = Synodus foetens • = Synodus bondi ♦♦ ♦ ♦ ♦ ^ ♦ Gulf of Mexico »♦ 1 Atlantic Ocean ce 280 mm SL). Distribution Synodus foetens was originally described from South Carolina (Linnaeus, 1766). Additional specimens ex- amined span the mid- and south-Atlantic coast of the United States: New Jersey, Maryland, North Carolina, South Carolina, Georgia, and Florida; the Gulf of Mexico: Alabama, Mississippi, Louisiana, Texas, and Mexico; Bermuda; Bahamas; Cuba; Ja- maica; Puerto Rico; and St. Martin in the Leeward Islands. One larval specimen was collected in Belize (Fig. 6). Synodus bondi Fowler, 1939 Proposed English common name: Sharpnose Lizardfish Figures 1, 2D, 5 (C and D), and 6, Tables 1, 4, and 5 Holotype ANSP 68634, 220 mm SL, Kingston, Jamaica, field number JB35-K-1, Jan 1935, Coll: J. Bond. Paratype ANSP 68635, 156 mm SL, Kingston, Jamaica, field number JB35-K-1, Jan 1935, Coll: J. Bond. Additional material (DNA numbers for specimens that are vouchers are given in parentheses following catalog numbers). Be- lize: KU 34275 (KUIT 5804: JX519392), 1; UF 137041, 8; USNM 300454, 1; USNM 328256, 1; Brazil: UF 213980, 1; UF 123639, 6; French Guiana: UF 211687, 1; Guyana: UF 123665, 1; Haiti: USNM 133666, 1; Hondu- ras: UF 123637, 1; Jamaica: UF 5191, 1; USNM 38538, 1; USNM 32076, 1; Panama: UF 75450, 1; UF 75707, 1; USNM 390121, 3; USNM 389847, 2; USNM 389982, 7; Tobago, Trinidad and Tobago: UF 123636, 3; Venezuela: UF 123656, 1; UF 123659, 1; USNM 405447, 1. Diagnosis A species of Synodus distinguished by the following combination of characters: dorsal-fin rays 11-12; anal- fin rays 10-12, rarely 12; pectoral-fin rays 12-14; total caudal-fin rays 42; configuration of caudal-fin rays al- ways 12 dorsal procurrent caudal rays + 10 segmented dorsal rays + 9 segmented ventral rays + 11 ventral Frable et al : Description of a new species of Synodus in the western Atlantic Ocean 139 procurrent rays; total vertebrae 56-60, modally 59; lat- eral-line scales 57-60, modally 60; predorsal scales 23- 25; scales above the lateral-line to dorsal-fin base 5.5 or 6.5; HL 22.4-26.7% SL (25.2%); snout very long and triangular (length 6.3-9. 1% SL, mean 7.3%, and 26.3- 37.2% HL, mean 29.0%), tip sharply pointed; anterior nostril with long, narrow flap that tapers to a filament distally, length 2. 6-5. 3% HL (3.4%); orbit diameter 2.8- 4.9% SL (3.5%) and 11.1-19.0% HL (13.8%); interorbit- al width 10.4-16.9% HL (15.5%); thick adipose lids on anterior and ventral margins of orbit (width 3. 4-7. 4% HL, mean 4.8%); dorsal-fin base 10.5-13.0% SL, mean 11.9%, usually longer than anal-fin base (9.6-12.7% SL, mean 11.0%). In preservative, body dark gray to reddish brown above the lateral axis, pale below; a few lateral stripes darker than background color sometimes present along length of body; no vertical bars on trunk; dark pigmentation present on margins of caudal fin, usually extending onto main portion of ventral lobe; dorsal half of adipose fin dark. Body pale green to tan in life. Description Description based on 45 specimens, 55.3-279.0 mm SL. Counts and measurements of type specimens given in Table 4. Frequency distribution of anal-fin rays given in Table 5. Dorsal-fin rays 11-12; anal-fin rays 10-12, rarely 12; total caudal-fin rays 42; configuration of caudal-fin rays in all specimens examined: 12 dorsal procurrent caudal rays + 10 segmented dorsal rays + 9 segmented ventral rays +11 ventral procurrent rays; pectoral-fin rays 12-14; pelvic-fin rays 8; total vertebrae 56-60, modally 59; predorsal vertebrae 15-18; scales ctenoid, lateral-line scales 57-60, modally 60; predorsal scales 23-25; scales above the lateral-line to dorsal-fin base 5.5 or 6.5; scales below lateral line to anal-fin base 6.5. Body cylindrical and elongate; depth at pelvic-fin or- igin 8.8—15.9% SL (11.8%); depth at anal-fin origin 5.7— 10.8% SL (8.9%); caudal-peduncle depth 3. 5-6. 5% SL (5.6%); body width at dorsal-fin origin 10.5-14.2% SL (12.7%); HL 22.4-26.7% SL (25.2%); snout length 6.3- 9.1% SL (7.3%) and 26.3-37.2% HL (29.0%), snout long and triangular, tip pointed in dorsal view; orbit diame- ter 2. 7-4.9% SL (3.5%) and 11.1-19.0% HL (13.8%), or- bit with bony ridge extending over anterodorsal margin and with wide adipose lid on its posterior edge, adipose lid width 0.8-1. 9% SL (1.2%) and 3. 4-7. 4% HL (4.8%); interorbital width 2. 7-4. 5% SL (3.9%) and 10.4-16.9% HL (15.5%). Predorsal length 40.9-45.8% SL (43.9%); prepelvic length 35.0-41.5% SL (38.6%); preanal length 75.0- 81.3% SL (77.7%); preadipose length 80.0-86.4% SL (82.4%); dorsal-fin base 10.5-13.0%> SL (11.9%), longest dorsal-fin ray, usually third, 14.0-17.2% SL (15.5%); anal-fin base 9.6-12.7% SL (11.0%), longest anal-fin ray, usually third, 7. 0-9. 6% SL (8.5%); pectoral-fin length 11.0-13.2% SL (11.9%); pelvic-fin length 15.9- 21.2% SL (18.8%); caudal-peduncle length 7.9-14.4% SL (12.1%), caudal-fin length (only measured in 31 specimens because many specimens had damaged cau- dal fins) 15.7-20.3% SL (17.6%), caudal lobes generally same length. Mouth terminal, upper jaw slightly oblique, form- ing an 8° angle with the horizontal body axis, upper jaw extending anterior to lower jaw; upper-jaw length 14.6-17.0% SL (15.9%) and 58.0-70.1% HL (62.9%), lower jaw with fleshy nub at symphysis. Teeth needle- like; upper jaw with 2 rows of teeth, teeth in inner row longer than teeth in outer row; lower jaw with 3 rows of teeth, size of teeth increasing medially, outer row covered by lips; posterior teeth in both jaws directed slightly anteriorly. Palatine with 3-4 rows of depress- ible teeth, teeth directed posteromedially. Tongue with 4-5 rows of posteriorly directed teeth, largest near an- terior tip. Anterior nostril slightly ventral to median axis of orbit; posterior nostril same size and situated slightly dorsal to anterior nostril; anterior nostril with long, narrow flap that tapers to a filament distally, length when depressed 0.7-1. 4% SL (0.9%) and 2.6- 5.3%- HL (3.4%); anterior nostril flap extending beyond posterior nostril when depressed (Fig. 2D); posterior nostril without flap. Cheek with 6 rows of large scales, opercle with 2 vertical rows of 7-9 scales on anterior margin and several scales posteriorly; predorsal scales extending anteriorly above anterior margin of preopercle; in- terorbital region naked. Dorsal fin flanked basally by 10-12 half-scales, anal fin flanked by 8-10 half-scales, remainder of dorsal and anal fins without scales; 3 large, diamond-shaped scales covering pelvic-fin bases, pelvic fins flanked laterally by 6 half-scales, remainder of pelvic fin and pectoral fin without scales; 7 rows of body scales extending onto base of caudal fin; upper and lower caudal lobes each with a single, enlarged, horizontally elongate scale. Coloration Before preservation (Fig. 50 Head and dorsal half of trunk olive to golden-brown, ventral half of trunk pale; cross-shaped markings dispersed along trunk in speci- mens <100 mm SL, those markings not present in larg- er specimens; rows of light spots present on central re- gion of trunk, spots darker in some specimens; groove above maxilla with stripe of black pigment from pos- terior edge of orbit to posterior end of jaw; dorsal-fin rays brown with translucent membrane between rays; adipose fin opaque, dorsal half dark; pectoral fin trans- lucent with a few darker spots; pelvic fins translucent to pale yellow; anal fin pale; dark pigmentation present on margins of caudal fin, usually extending onto main portion of ventral lobe. 140 Fishery Bulletin 111(2) In preservative (Fig. 5D) Trunk dark orange-tan to brown dorsally, paler ventrally; dark markings in small specimens faded; lines on trunk faded or no longer vis- ible; dark pigmentation on jaw and adipose fin visible; dark caudal-fin pigmentation less distinct in larger specimens (>280 mm SL). Distribution The holotype and paratype of S. bondi were collected off Kingston, Jamaica. Additional specimens of S. bondi, previously identified as S. foetens, are known from Ja- maica, Haiti, Belize, Honduras, Panama, Trinidad and Tobago, Brazil, French Guiana, Guyana, and Venezuela. On the basis of the material examined, the distribution of this species is concentrated in the southern portions of the Caribbean (Fig. 6). No specimens currently are known from the United States, Mexico, Bahamas, or Bermuda. Comparisons of Synodus foetens, S. bondi, and congeners Synodus bondi was synonymized with S. foetens under the reasoning that “comparison of the type and para- type with specimens of S. foetens removes all doubt that S. bondi is identical to S. foetens ” (Anderson et al., 1966: 73). The S. foetens discussed by Anderson et a!. (1966) is undoubtedly the lineage we have identified as S. foetens because a majority of the specimens they examined possess 12-13 anal-fin rays and are from the United States and northern Caribbean. Anderson et al. (1966) noted that the apparent differences in counts between S. foetens and S. bondi for lateral-line scales and scale rows above the lateral line recorded by Fowler (1939) are incorrect. Their re-examination of the paratype of S. bondi revealed the presence of 60 (as opposed to Fowler’s 54) lateral-line scales and 5 (instead of Fowler’s 6) rows of scales above the lat- eral line. However, those counts were non-diagnostic in Fowler’s species description, and additional reasoning for the synonymy was not provided. Synodus bondi is, in fact, morphologically distinct from S. foetens. The snout of S. bondi ends in a sharper point than does the snout of S. foetens (Fig. 5) and is significantly longer (mean 29.0% HL in S. bondi, compared with 27.1% HL in S. foetens ). The anterior- nostril flap in S. bondi, on average, is slightly longer than that flap in S. foetens (3.4% and 3.0% HL, re- spectively); however, the flap in S. bondi is narrow and tapers to a filament distally but in S. foetens is broad and triangular (Fig. 2, C and D). The upper jaw is longer in S. bondi than in S. foetens, 58.0-70.0% HL (63.0%) versus 55.0-65.2% HL (61.1%). The adi- pose lids surrounding the dorsal and ventral margins of the orbit are wider in S. bondi (mean 4.8% HL) than in S. foetens (mean 3.5% HL), making the orbit look superficially smaller in S. bondi. Synodus bondi usu- ally possesses fewer anal-fin rays than does S. foetens (10-12, usually 10 or 11, versus 12 or 13); as a result, the dorsal-fin base is usually longer than the anal-fin base in S. bondi (the opposite is true in S. foetens). In addition, the configuration of caudal-fin rays in S. bon- di is consistently 12+10+9+11 in specimens examined, but it is more variable in S. foetens and other Syno- dus species. Finally, although the geographic ranges of these 2 species overlap in a swath across the central Caribbean, S. bondi otherwise has a more southern Caribbean distribution relative to S. foetens, which oc- curs northward to New York (Fig. 6). Synodus bondi and S. foetens are distinguished from S. synodus by having a sharply pointed snout, pectoral fins that do not extend beyond the base of the pelvic fins, higher numbers of predorsal scales (20-30 versus 15-18), and no dark spot on the upper jaw. Synodus bondi and S. foetens differ from S. saurus by having 5 or 6 complete scale rows above the lateral line versus 3 and a snout that is longer than the diameter of the orbit. Finally, S. bondi and S. foetens are differentiated from S. poeyi, S. intermedius, and S. macrostigmus by having more than 55 lateral-line scales. In the original description of S. bondi, Fowler (1939) noted that the type specimens have dark pigment on the isthmus and posterior margins of the branchioste- gals. We did not observe this pigment in the ANSP types or additional material, but it may have faded in preservative. We have not examined fresh material beyond photographs. Remarks Although numerous synonyms exist for Synodus foe- tens, S. bondi is the only name that can be defini- tively associated with the additional genetic lineage of “S. foetens ” in our material. Two nominal species, Osmerus albidus (Lacepede, 1803) and Coregonus ruber (Lacepede, 1803), were described only in brief paragraphs and lack known type material. The only diagnostic feature given in the original description for C. ruber is that it has a rounded snout. Based on the snout shape and lack of information of type ma- terial, neither name is applicable to specimens rec- ognized herein as S. bondi. Mitchill (1815) described Esox salmoneus from New York without designating any type material. On the basis of the type locality, it is clear that this name also is not applicable to our S. bondi material. Cuvier (1829) described Saurus mexicanus from small, transparent specimens from the Gulf of Mexico without designating type materi- al. These specimens were most likely larvae or small juveniles, but there is not enough detail in the origi- nal description to identify them to species. Agassiz (in Spix and Agassiz 1829) briefly described Saurus longirostris on the basis of 2 specimens from Brazil (deposited at ZSM) that have a short anal fin that comprises 12 rays and a very pointed snout. Kottelat (1988) was unable to find the type material of Saurus longirostris at ZSM or MHNN, and their whereabouts Frable et al Description of a new species of Synodus in the western Atlantic Ocean 141 Key to the western Atlantic species of Synodus This key is modified from a provisional key for Synodontidae constructed by Russell (2003). la Scales in lateral line 43 to 52 2 lb Scales in lateral line 54 to 65 4 2a Dorsal fin with anterior rays extending to, or usually beyond, tips of succeeding rays when depressed; lower jaw ending in fleshy knob; no black scapular blotch on shoulder under gill cover Synodus poeyi 2b Dorsal fin with anterior rays not extending beyond, but occasionally extending to, tips of succeeding rays when depressed; lower jaw rounded anteriorly, without fleshy knob; black scapular blotch present on shoulder under gill cover 3 3a Caudal fin with 3-5 dark bars spanning both lobes; pored scales in lateral line 49-52; scapular blotch small and rectangular (length <12% HL); anterior-nostril flap broad and short, not tapering significantly posteriorly and not extending beyond posterior nostril when depressed Synodus intermedius 3b Caudal fin without prominent dark bars but with dark pigment on lower lobe; pored scales in lateral line 45-48; scapular blotch large and ovoid (length >14% HL); anterior- nostril flap large, tapering posteriorly and extending beyond posterior nostril when depressed Synodus macrostigmus new sp. 4a Three rows of complete scales between lateral line and base of dorsal fin Synodus saurus 4b Four to 6 rows of complete scales between lateral line and base of dorsal fin 5 5a Snout rounded and blunt, its length less than diameter of eye; anal-fin base much shorter than dorsal-fin base; tip of pectoral fin extending well beyond base of pelvic fin; dark spot present on tip of upper jaw; predorsal scales 15 to 18 Synodus synodus 5b Snout triangular and pointed, its length greater than diameter of eye; anal-fin base slightly shorter to longer than dorsal-fin base; tip of pectoral fin falling short of or just reaching pelvic-fin base; no dark spot on tip of upper jaw; predorsal scales 20 to 30 6 6a Anal-fin rays usually 10 or 11 (rarely 12); dorsal-fin base as long as or longer than anal-fin base; adipose lids around orbit thick; tip of snout sharply pointed; anterior-nostril flap narrow and tapering to filament distally; species currently known from off Central and South America, Jamaica and Haiti Synodus bondi 6b Anal-fin rays usually 12 or 13 (rarely 11); dorsal-fin base usually shorter than anal-fin base (rarely same length); adipose lids around orbit narrow; tip of snout not sharply pointed, slightly rounded; anterior-nostril flap broad and triangular, not tapering to filament; species currently known from New York south to the Leeward Islands, the Gulf of Mexico, and Belize Synodus foetens are unknown. Kottelat (1988) did find a specimen at MHNN collected off Brazil about the same time as Spix and Agassiz’s ( 1829) Saurus longirostris (MHNN 793), which may be a syntype. However, as noted by Kottelat (1988), the length of that specimen does not match the lengths of S. longirostris recorded in the original description or in Agassiz’s notes. Although this specimen could be the same one described by Fowler (1939) as S. bondi on the basis of its pointed snout, short anal Fin, and collection locality (as noted previously, S. bondi is also known from Brazil), it would be imprudent to resurrect this name based on a specimen that may or may not be a primary type. Finally, we examined the holotype of another syn- onym, Saurus spixianus Poey, 1860 (MCZ 6884) and found that it has 13 anal-fin rays, a rounded snout, and morphometries similar to those of S. foetens. Our observations, therefore, corroborate the synonymy of Saurus spixianus with S. foetens. Discussion Molecular variation Our DNA barcoding analysis of Synodontidae in the At- lantic (Fig. 1) revealed 13 highly divergent lineages (av- erage divergence in COI among lineages: 20.2%, range: 9.0-30.7%). Morphological examination of voucher specimens indicated that 10 of the lineages correspond to the following previously described species: Saurida brasiliensis, S. caribbaea (Smallscale Lizardfish), S. normani (Shortjaw Lizardfish), S. suspicio (Suspicious Lizardfish), Synodus foetens, S. intermedius, S. poeyi, S. saurus, S. synodus, and Trachinocephalus myops. Of the additional 3 lineages, 2 represent morphologically dis- tinct species originally identified as S. intermedius and S. foetens. We describe S. macrostigmus for the former and resurrect S. bondi for the latter. The thirteenth lineage in the data set represents specimens from deep 142 Fishery Bulletin 111(2) water off Central America originally identified as S. po- eyi (Sy nodus poeyi lineage 2 in Fig. 1). That lineage is 17.9% divergent from S. poeyi specimens from the Gulf of Mexico (Sy nodus poeyi lineage 1 in Fig. 1). A comprehensive study of S. poeyi from throughout its range, similar to this study conducted for S. inter- medius and S. foetens, should help clarify the taxonomy of that species and likely will result in the recogni- tion of a new species or one resurrected from synony- my. Preliminary investigation of Synodus synodus and Saurida brasiliensis from Cape Verde (not included in Fig. 1) reveal approximately 5-6% divergences in COI from western Atlantic Synodus synodus and Saurida brasiliensis, indicating deep population structure or potential cryptic species in those lizardfish lineages. Initial investigation of some Pacific lizardfish species (genetic data not included in this article) indicates that Trachinocephalus my ops from the Philippines is highly divergent in COI from western Atlantic T. myops — a finding that sheds doubt on the current circumtropical distribution of T. myops and the monotypy of Trachi- nocephalus. Further investigation of this genus world- wide is needed to evaluate species diversity. The intrageneric genetic variation in COI in western Atlantic Synodus and Saurida is high compared to the variation observed for other marine fishes that have been analyzed. Ward et al. (2005) found the average in- trageneric variation in 207 species of Australian fishes to be 9.93%, and Hubert et al. (2008) found an average 8.30% intrageneric distance for 193 species of Cana- dian freshwater fishes. Average intrageneric divergence in western Atlantic synodontids (20.5%) exceeds even the average intergeneric distance of 16.6% calculated by Kartavtsev (2011) for animal species in general. Interspecific and intergeneric distances are similar in western Atlantic synodontids (Table 1). Large diver- gences likely reflect older speciation events, but the factors that drive lizardfish evolution are unknown. Geographic and population variation Although a species phylogeny for synodontids is need- ed to hypothesize sister-group relationships and ex- amine patterns of speciation in this family, we note that morphologically and genetically similar species, such as S. foetens and S. bondi or S. intermedius and S. macrostigmus, exhibit different geographical and, sometimes, depth distributions. Synodus foetens and S. bondi have nearly distinct geographical distributions (Fig. 6). Synodus foetens occurs off the East Coast of the United States, in the Gulf of Mexico, and in the central Caribbean; S. bondi occurs in coastal Central and South America northward to Haiti. The distribu- tion for these species overlaps in Belize and eastward to Jamaica and Haiti (Fig. 6). Having largely disjunct distributions that overlap in the central Caribbean is a pattern congruent with distributions observed in other predatory fish genera, such as Scomberomorus (mackerels) and Rhizoprion- odon (sharpnose sharks). In Scomberomorus, S. macu- latus occurs off the East Coast of the United States, in the Gulf of Mexico, and in the northern and north- western Caribbean, whereas S. brasiliensis occurs in the southern and central Caribbean (Collette et al., 1978; Banford et al., 1999). These 2 species overlap off northern Central America and potentially in the south- ern Gulf of Mexico. Similarly, Rhizoprionodon terrae- novae (Atlantic Sharpnose Shark) inhabits the Gulf of Mexico, northern Caribbean, and East Coast of the United States, whereas the closely related species, R. porosus (Caribbean Sharpnose Shark), occurs off South and Central America. These 2 species may overlap in the central Caribbean (Springer, 1964; Compagno et al., 2005). Synodus foetens and S. bondi do not appear to have different depth preferences because both spe- cies inhabit depths between the surface and 95 m. The geographic distributions of Synodus macrostig- mus and S. intermedius overlap (Fig. 4). S. macrostig- mus is known from the eastern and southern Gulf of Mexico and East Coast of the United States, whereas S. intermedius inhabits the eastern Gulf of Mexico, East Coast of the United States, Bermuda, Bahamas, and the Caribbean. These overlapping distributions are similar to the distributions observed in 2 genetic lineages of the goby, Bathygobius soporator (Frillfin Goby), by Tornabene et al. (2010). Those authors did not describe the lineages as separate species because no morphological differences were found to corroborate the genetic data, but Tornabene and Pezold (2011) not- ed that the B. soporator lineages could represent re- cent divergence and ongoing speciation in the western Atlantic. However, because the genetic divergence was observed in mitochondrial DNA alone, they could not rule out the possibility of deep coalescence. Unlike the gobies, S. intermedius and S. macrostig- mus are morphologically distinct, and they exhibit dif- ferent depth preferences. Although S. macrostigmus primarily inhabits depths below 28 m (mean 96.5 m), S. intermedius is found typically at shallower depths (mean 49.3 m). However, S. intermedius has a broad depth distribution and has been collected in deep water along with S. macrostigmus (i.e., UF29818). Further in- vestigation is needed to ascertain possible reproductive barriers in the evolutionary history of these species and whether or not ecological speciation (e.g., Rocha et al., 2005) could have played a role. The neighbor-joining tree (Fig. 1) also reveals evi- dence of population structure in some species. For ex- ample, S. foetens specimens from the Gulf of Mexico differ in COI by 1.2% from a specimen in Belize, and Caribbean specimens of Trachinocephalus myops differ genetically by 1.6% from 2 deepwater Gulf of Mexico specimens. Finally, within Saurida normani, specimens from the Gulf of Mexico differ from Central American specimens by 1.7%. Additional material and genetic analyses are needed to describe the population struc- ture of western Atlantic lizardfishes. Frable et al.: Description of a new species of Synodus in the western Atlantic Ocean 143 Fisheries concerns Globally, some synodontids are commercially important as food fishes. Trachinocephalus rnyops, Synodus syno- dus, Saurida tumbil (Greater Lizardfish), and Harpa- don nehereus (Bombay Duck) are targeted throughout the Mediterranean, Indian Ocean, and Southeast Asia (Raje et al., 2004; Ghosh et al., 2009; Xu et al., 2011). In the western Atlantic, synodontid species are encoun- tered almost exclusively as bycatch in recreational and commercial fisheries. As a result, they have received little attention from fisheries researchers and manag- ers despite their role as significant fish predators in various ecosystems (Sweatman, 1984; Cruz-Escolano et al., 2005). As mentioned previously, S. foetens makes up 1. 5-2.0% of shrimp-trawl biomass and is usually among the 10 most prevalent finfish encountered by shrimp trawlers in the Gulf of Mexico, Caribbean, and off the southeastern United States.1 (Gutherz, 1987; Jeffers et al., 2008; Manjarres et al., 2008). These stud- ies also report collection of S. intermedins and S. poeyi in much smaller amounts. Jeffers et al. (2008) found that S. foetens in the northern Gulf of Mexico has a very high instantaneous annual mortality related to bycatch in fisheries ( Fh ), ranging from 0.4-0.6/year, and estimated exploitation ratios ( E ) of 0.43-0.55. Exploitation ratios of more than 0.5 generally indicate heavy fishing pressure and po- tential for overexploitation (Gulland, 1977). These data indicate that, although S. foetens is not commercially targeted, it is still fully exploited as bycatch. To date, only commercially targeted fish species have been evaluated quantitatively as bycatch in the north- ern Gulf of Mexico (Jeffers et al., 2008), and little is known about bycatch rates and their effects on popula- tions for noncommercial species. Broad ecosystem-scale studies of the effects of bycatch are needed that include nontargeted but exploited species, such as lizardfishes. For such studies, as well as any management plans that might result from them, an accurate understand- ing of species diversity and distribution is essential. This study provides new insights into the systematics of lizardfishes in the western Atlantic that should be incorporated into future management plans and fisher- ies research programs. Conclusions The description of Synodus macrostigmus as a new species and the recognition of S. bondi as a valid spe- cies bring the total number of valid western Atlantic Synodus species to 7. The integration of molecular and morphological data greatly facilitated identification of the new species and recognition of S. bondi as valid. A thorough systematic revision of western Atlantic Syn- odontidae that incorporates both molecular and mor- phological data is needed, and it seems likely that ad- ditional new species may exist. Geographic variation and population structure within this group also war- rant further study. Acknowledgments We would like to thank the many people who facili- tated this study: A. Driskell, A. Ormos, D. Smith, R. Myers, A. Bentley, P. Webster, E. Wiley, R. Robins, M. Sabaj Perez, J. Lundberg, B. Nagareda, R. Causse, C. Ferrara, D. Johnson, C. Boucher, D. R. Robertson, J. Van Tassell, C. Castillo, S. Raredon, B. Collette, and Z. Baldwin. Research in Florida was conducted pursuant to Special Activities License no. 07SR-1024B to the sec- ond author. This article is contribution number 935 of the Caribbean Coral Reef Ecosystems Program, Smith- sonian Institution, supported in part by the Hunter- don Oceanographic Research Fund and Smithsonian Marine Station at Fort Pierce contribution number 905. Literature cited Anderson, W. W., and J. W. Gehringer. 1975. The correlation between numbers of vertebrae and lateral-line scales in western Atlantic lizardfishes (Syn- odontidae). Fish. Bull. 73:202-206. Anderson, W. W., J. W. Gehringer, and F. H. Berry. 1966. Family Synodontidae: lizardfishes. In Fishes of the western North Atlantic. Part five: Lizardfishes, other Iniomi, deepsea gulpers (Y. H. Olsen and J. W. Atz, eds.), p. 30-102. Memoir 1, Sears Foundation for Marine Research, Yale Univ., New Haven, CT. Baldwin, C. C., C. I. Castillo, L. A. Weigt, and B. C. Victor. 2011. Seven new species within western Atlantic Stark- sia atlantica, S. lepicoelia, and S. sluiten (Teleostei: Labrisomidae), with comments on congruence of DNA barcodes and species. ZooKeys 79:21-72. Baldwin, C. C., and L. A. Weigt. 2012. A new species of soapfish (Teleostei: Serranidae: Rypticus), with redescription of R. subbifrenatus and comments on the use of DNA barcoding in systematic studies. Copeia 2012:23-36. Baldwin, C. C., L. A. Weigt, D. G. Smith, and J. H. Mounts. 2009. Reconciling genetic lineages with species in west- ern Atlantic Coryphopterus (Teleostei: Gobiidae). In Proceedings of the Smithsonian Marine Science Net- work Symposium; 15-16 November 2007. Smithson. Contrib. Mar. Sci. 38 (M. A. Lang, I. G. Macintyre, K. Rutzler, eds.), p. 113-140. Smithson. Inst. Scholarly Press, Washington, D.C. Banford, H. M., E. Bermingham, B. B. Collette, and S. S. McCafferty. 1999. Phylogenetic systematics of the Scomberomorus regalis (Teleostei: Scombridae) species group: molecules, morphology and biogeography of Spanish mackerels. Copeia 1999:596-613. Bertin, L., and R. Esteve. 1950. Catalogue des types de poissons du Museum Na- tional d’Histoire Naturelle, 6e partie: haplomes, hete- romes, catosteomes, 60 p. Imp. Nationale, Paris. 144 Fishery Bulletin 111(2) Catesby, M. 1743. The natural history of Carolina, Florida and the Bahama Islands, vol. 2, 100 p. Royal Soc. House, Lon- don. [Available at http://www.biodiversitylibrary.org/ bibliography/620 15#. I Collette, B. B., J. L. Russo, and L. A. Zavala-Camin. 1978. Scomberomorus brasiliensis, a new species of Spanish mackerel from the western Atlantic. Fish. Bull. 76:273-280. Compagno, L., M. Dando, and S. Fowler. 2005. Sharks of the world, 1st ed., 368 p. Princeton Univ. Press, Princeton, NJ. Cressey, R. 1981. Revision of the Indo-West Pacific lizardfishes of the genus Synodus (Pisces: Synodontidae). Smithson. Contrib. Zool. 342, 53 p. Cruz-Escalona, V. H., M. S. Peterson, L. Campos-Davilla, and M. Zetina-Regon. 2005. Feeding habits and trophic morphology of inshore lizardfish ( Synodus foetens ) on the central continental shelf off Veracruz, Gulf of Mexico. J. Appl. Ichthyol. 21:525-530. Cuvier, G. 1829. Le regne animal distribue d’apres son organisa- tion, pour servir de base a l’histoire naturelle des ani- maux et d’introduction a l’anatomie comparee, vol. 2, 2nd ed, 405 p. Chez Deterville, Paris. [Available at http://archive.Org/details/lergneanimaldi03cuvie.l Cuvier, G., and A. Valenciennes. 1850. Histoire naturelle des poissons, vol. 22, 532 p. Berg- er-Levrault, Strasbourg, France. [Available at http:// a rch ive.org/stream/histoi re nature 112 2 cu vie# page/ 1 1/ mode/lup.] Fowler, H. W. 1939. Notes on fishes from Jamaica with descriptions of three new species. Notulae Naturae 35, 16 p. Acad. Sci., Philadelphia, PA. Garcia-Abad, M. C., P. Sanchez-Gil, and M. Tapia-Garcia. 1999. Distribution, abundancia y reproduction de Syno- dus foetens (Linnaeus, 1766) (Pisces: Synodontidae) en la plataforma continental del sur del Golfo de Mexico. Thalassas 15:9-18. Ghosh, S„ N. G. K. Pillai, and H. K. Dhokia. 2009. Fishery and population dynamics of Harpadon ne- hereus (Ham.) off the Saurashtra coast. Indian J. Fish. 56:13-19. Gunther, A. 1864. Catalogue of the fishes in the British Museum, vol. 5, 455 p. British Mus., London. [Available at http:// archive. org/stream/catalogueoffishe05brit#page/n7/ mode/2up.| Gutherz, E. J. 1987. MEXUS-Gulf demersal fish research, 1978-1985. Mar. Fish. Rev. 49( 1 ):3 1— 35. Gulland, J. A. 1977. Fish population dynamics, 2nd ed., 384 p. John Wiley and Sons, New York. Hebert, P. D. N., A. Cywinska, S. L. Ball, and J. R. DeWaard. 2003. Biological identifications through DNA barcodes. Proc. R. Soc. Lond., Ser. B: Biol. Sci. 270:313-321. Hubbs, C. L., and K. F. Lagler. 1964. Fishes of the Great Lakes region, 4th ed., 213 p. Univ. Michigan Press, Ann Arbor, MI. Hubert, N., R. Hanner, E. Holm, N. E. Mandrak, E. Taylor, M. Burridge, D. Watkinson, P. Dumont, A. Curry, P. Bentzen, J. Zhang, J. April, and L. Bernatchez. 2008. Identifying Canadian freshwater fishes through DNA barcodes. PLoS ONE. 3(6):e2490. Jeffers, S. A., W. F. Patterson, III, and J. H. Cowan Jr. 2008. Habitat and bycatch effects on population param- eters of inshore lizardfish (Synodus foetens) in the north central Gulf of Mexico. Fish. Bull. 106:417-426. Jordan, D. S. 1884. List of Fishes collected at Key West, Florida, with notes and descriptions. Proc. U.S. Natl. Mus. 7:103-150. Kartavtsev, Y. P. 2011. Sequence divergence at mitochondrial genes in animals: applicability of DNA data in genetics of spe- ciation and molecular phylogenetics. Mar. Genomics 4:71-81. Kimura, M. 1980. A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J. Mol. Evol. 16:111-120. Kottelat, M. 1988. Authorship, dates of publication, status and types of Spix and Agassiz’s Brazilian fishes. Spixiana 11:69-93. Lacepede, B. G. E. 1803. Histoire naturelle des poissons, vol. 5, 803 p. Chez Plassan, Paris. Linnaeus, C. 1758. Systema naturae per regna tria naturae, secun- dum classes, ordines, genera, species, cum characteri- bus, differentiis, synonymis, locis, vol. 1, 10th ed., 824 p. Laurentii Salvii, Holmiae. [Available at http://archive. org/stream/mobot31753000798865#page/nl/mode/2up.[ 1766. Systema naturae sive regna tria naturae, secun- dum classes, ordines, genera, species, cum characteri- bus, differentiis, synonymis, locis, vol. 1, 12th ed., 532 p. Laurentii Salvii, Holmiae. Manjarres, L., L. O. Durante, J. Altamar, F. Escobar, C. Garcia, and F. Cuello. 2008. Effects of using bycatch reduction devices on the Colombian Caribbean Sea shrimp fishery. Cienc. Mar. 34:223-238. Meek, S. E. 1884. A review of the American species of the genus Synodus. Proc. Acad. Nat. Sci. Phila. 36:130-136. Mitchill, S. L. 1815. The fishes of New York described and arranged. New York, Trans. Lit. Phil. Soc. 1:355-492. Norman, J. R. 1935. A revision of the lizardfishes of the genera Syno- dus, Trachinocephalus, and Saurida. Proc. Zool. Soc. Lond. 105:99-136. Poey, F. 1860. Memorias sobre la historia natural de la Isla de Cuba, acompanadas de sumarios Latinos y extractos en Frances, vol. 2, p. 97-336. Imprente de la Viuda de Barcina, Habana, Cuba. 1868. Synopsis piscium cubensium. Repertorio fisico- natural de la isla Cuba, vol. 2, p. 413-415. Imprente de la viuda de Barcina y Comp., Habana, Cuba. 1876. Enumeratio piscium cubensium. Parte segunda. Anales de la Sociedad Espanola de Historia Natural 5:131-218. Frable et al Description of a new species of Synodus in the western Atlantic Ocean 145 Raje, S. G., V. D. Deshmukh, and T. Das. 2004. Observations on the lizard fish fishery and some aspects of biology of Saurida tumbil (Bloch) off Mum- bai. Indian J. Fish. 51:199-207. Reis, R. E., S. O. Kullander, and C. J. Ferraris. Check list of the freshwater fishes of South and Central America. Edi- pucrs, 2003. Randall, J. E. 2009. Five new Indo-Pacific lizardfishes of the genus Synodus ( Aulopiformes: Synodontidae). Zool. Stud. 48:402-417. Rocha, L. A., D. R. Robertson, J. Roman, and B. W. Bowen. 2005. Ecological speciation in tropical reef fishes. Proc. R. Soc. Lond., Ser. B: Biol. Sci. 272:573-579. Russell, B. C. 2003. Synodontidae In The living marine resources of the Western Central Atlantic. Vol. 2: Bony fishes; part 1 ( Acipenseridae to Grammatidae). FAO species identifi- cation guide for fishery purposes and American Society of Icthyologists and Herpetologists Special Publication No. 5 (K. E. Carpenter, ed.), p. 923-930. FAO, Rome, Saitou, N., and M. Nei. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406-425. Seutin, G., P. Bagley, and B. N. White. 1990. Preservation of avian blood and tissue samples for DNA analyses. Can. J. Zool. 69:82-90. Spix, J. B. von, and L. Agassiz. 1829. Selecta genera et species piscium quos in itinere per Brasiliam annos MDCCCXVII-MDCCCXX jussu et auspiciis Maximiliani Josephi I . . . collegit et pingendso curavit Dr J. B. de Spix... Part 1, 82 p. Typis C. Wolf, Monachii. [Available at http://archive.org/stream/ selectageneraets00spix#page/n5/mode/2up.] Springer, V. G. 1964. A revision of the carcharhinid shark genera Sco- liodon, Loxodon, and Rhizoprionodon. Proc. U.S. Natl. Mus. 115:559-632. Sweatman, H. P. A. 1984. A field study of the predatory behavior and feed- ing rate of a piscivorous coral reef fish, the lizardfish Synodus englemani. Copeia 1984:187-194. Swofford, D. 2003. PAUP*: Phylogenetic analysis using parsimony Appendix: Comparative material of western Atlantic Synodontidae examined in this study In the lists below, the following sequence is observed: first, catalog number; second, DNA number; and third, GenBank accession number (the latter two numbers are shown within parentheses). Synodus synodus USNM 405415 (BAH 8058: JQ839916), USNM 405416 (BAH 8059: JQ839915), USNM 405417 (BAH 9028: JX519403), USNM 405418 (BAH 9030: JX519388), USNM 405419 (BAH 10008: JX519382), USNM 405420 (BAH 10009: JX519383), USNM 405421 (BAH 10010: JX519384), USNM 405422 (and other methods), vers. 4.0 beta 10. Sinauer Asso- ciates, Sunderland, MA. Tamura, K., D. Peterson, N. Peterson, G. Stecher, M. Nei, and S. Kumar. 2011. MEGA5: molecular evolutionary genetics analy- sis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28:2731-2739. Tornabene, L., and F. Pezold. 2011. Phylogenetic analysis of western Atlantic Bathy- gobius (Teleostei: Gobiidae). Zootaxa 3042:27-36. Tornabene, L., C. C. Baldwin, L. A. Weigt, and F. Pezold. 2010. Exploring the diversity of western Atlantic Bathygobius (Teleostei: Gobiidae) using mitochondrial cytochrome c oxidase-I, with descriptions of two new species. Aqua J. Icthyol. Aquat. Biol. 16:141-170. Victor, B. C. 2007. Coryphopterus kuna , a new goby (Perciformes: Gobiidae: Gobiinae) from the western Caribbean, with the identification of the late larval stage and an estimate of the pelagic larval duration. Zootaxa 1526:51-61. 2010. The redcheek paradox: the mismatch between ge- netic and phenotypic divergence among deeply-divided mtDNA lineages in a coral-reef goby, with the descrip- tion of two new cryptic species from the Caribbean Sea. J. Ocean Sci. Found. 3:1-16. Ward, R. D., T. S. Zemlak, B H. Innes, P. R. Last, and P. D. N. Hebert. 2005. DNA barcoding Australia’s fish species. Philos. Trans. R. Soc. Lond., Ser. B: Biol. Sci. 360:1847-1857. Wells, R. J. D. 2007. The effects of trawling and habitat use on red snapper and the associated community. Ph.D. diss., 180 p. Louisiana State LTniv., Baton Rouge, LA. Wheeler, A. 1985. The Linnaean fish collection in the Linnaean Soci- ety of London. Zool. J. Linn. Soc. 84:1-76. 1991. The Linnaean fish collection in the Zoological Mu- seum of the University of Uppsala. Zool. J. Linn. Soc. 103:145-195. Xu, T J . D. Q. Sun, H. Y. Li, and R. X. Wang. 2011. Development and characterization of microsatel- lite markers for the lizardfish known as the Bombay duck, Harpadon nehereus (Synodontidae). Genet. Mol. Res. 10:1701-1706. (BLZ 4026: JQ840361), USNM 405423 (BLZ 4027: JQ840360), USNM 405424 (BLZ 4555: JQ840362), USNM 405425 (BLZ 5094: JQ840726), USNM 405426 (BLZ 5133: JQ840727), USNM 405427 (BLZ 5159: JQ840729), USNM 405428 (BLZ 5240: JQ840728), USNM 405429 (BLZ 5214: JQ840730), USNM 405430 (BLZ 5455: JQ840725), USNM 405431 (BLZ 6297: JQ841025), USNM 405432 (TCI 9275: JX519404), USNM 405433 (TCI 9276: JX519405), USNM 405434 (TCI 9277: JX519389), USNM 405435 (TCI 9278: JX519390), USNM 405436 (TCI 9279 JX519406), USNM 405437 (TCI 9576: JX519408), USNM 405438 (TCI 9645: JX519407), USNM 405439 (TCI 9646: JX519391), USNM 405440 (TOB 9199: JQ843085). 146 Fishery Bulletin 111(2) Synodus poeyi FSBC 020586 (FWRI 20586a: JX519378, FWRI 20586b: JX519379), FSBC 020708 (FWRI 20708: JX519381), USNM 405441 (MOC 11014: JX519397), USNM 405442 (MOC 11437: JX519398), USNM 405443 (MOC 11784: JX519399). Synodus saurus USNM 405444 (BAH 10071: JX5 19385). Saurida brasiliensis FSBC 020698 (FWRI 20698). Saurida caribbaea USNM 405445 (MOC 11159: JX519372), USNM 405446 (MOC 11471: JX519373). Saurida normani FSBC 020653 (FWRI 20653a: JX519393), USNM 405453 (MOC 11434: JX519374), USNM 405454 (MOC 11438: JX519365). Saurida suspicio USNM 405455 (BLZ 8329 JQ841833), USNM 405456 (BLZ 8393: JQ841834), USNM 405457 (BLZ 8397: JQ841786), USNM 405458 (BLZ 8398: JQ841832). Trachinocephalus myops FSBC 020509 (FWRI 20509: JX519396); FSBC 020698 (FWRI 20698a: JX519394); USNM 405460 (BLZ 6415: JQ841030); USNM 405459 (TOB 9157: JQ843093); USNM 405459. 147 Abstract— The purpose of this study was to validate aging results of ju- venile Shortfin Mako ( Isurus oxy- rinchus) by vertebral band counts. Vertebrae of 29 juvenile Shortfin Mako marked with oxytetracycline (OTC) were obtained from tag-re- capture activities to determine cen- trum growth-band deposition. Tag- ging occurred off southern Califor- nia from 1996 to 2010, and time at liberty of the 29 sharks ranged from 4 months to 4.4 years (mean=1.3 years). Growth information also was obtained from length-frequency modal analyses (MULTIFAN and MIXDIST) by using a 29-year data set of commercial and research catch data, in addition to a tag-recapture growth model (e.g, the GROTAG model). For vertebrae samples used for age validation, shark size at time of release ranged from 79 to 142 cm fork length (FL) and from 98 to 200 cm FL at recapture. Results from band counts of vertebrae distal to OTC marks indicate 2 band pairs (2 translucent and 2 opaque) are formed each year for Shortfin Mako of the size range examined. Length- frequency analyses identified 3 age- class modes. Growth rate estimates from 26.5 to 35.5 cm/year were cal- culated for the first age-class mode (85 cm FL) and from 22.4 to 28.6 cm/year for the second age-class mode (130 cm FL). Results from the tag-recapture growth model revealed fast growth during time at liberty for tagged fish of the 2 youngest age classes. Collectively, these methods suggest rapid growth of juvenile Shortfin Mako in the southern Cali- fornia study area and indicate bian- nual deposition of growth bands in vertebrae for the first 5 years. Manuscript submitted 23 August 2012. Manuscript accepted 20 February 2013. Fish. Bull 111:147-160 (2013). doi 10.7755/FB.111.2.3 The views and opinions expressed or implied in this article are those of the author (or authors) and do not necessar- ily reflect the position of the National Marine Fisheries Service, NOAA. Age validation of juvenile Shortfin Mako Usurus oxyrinchus ) fagged and marked with oxytetracycline off southern California R. J. David Wells (contact author)1-2 Susan E. Smith1 Suzanne Kohin1 Ellen Freund1 Natalie Spear1 Darlene A. Ramon1 Email address for contact author wellsr@tamug edu ’ Fisheries Resources Division Southwest Fisheries Science Center National Marine Fisheries Service, NOAA 8901 La Jolla Shores Drive La Jolla, California 92037 Present address for contact author: Department of Marine Biology Texas A&M University at Galveston 1001 Texas Clipper Rd Galveston, Texas 77553 2 Department of Marine Biology Texas A&M University at Galveston 1001 Texas Clipper Rd Galveston, Texas 77553 For almost 3 decades, researchers in various parts of the world have fo- cused on the problem of accurately interpreting age and growth in the Shortfin Mako ( Isurus oxyrinchus) (e.g., Cailliet and Bedford, 1983; Cailliet et al., 1983; Pratt and Casey, 1983; Casey and Kohler, 1992; Cam- pana et al., 2002; Bishop et al.1; Ri- bot-Carballal et al., 2005; Bishop et al., 2006; Natanson et al., 2006; Ar- dizzone et al., 2006; Maia et al., 2007; Cerna and Licandeo, 2009; Okamura and Semba, 2009; Semba et al., 2009). Driving these efforts is the need to better assess the vulnerability of Shortfin Mako to harvest in commer- cial and recreational fisheries and as bycatch in longline and driftnet fish- eries from high-seas fleets (Stevens, 2008). Studies of the demographic 1 Bishop, S. D. H., M. P. Francis, and C. Duffy. 2004. Age, growth, maturity, longevity and natural mortality of the Shortfin Mako shark ( Isurus oxyrinchus ) in New Zealand waters. Working Paper SCTB17, BIO-4, National Institute of Water and Atmospheric Research, New Zealand, Ministry of Fisheries, NZ. 34 p. Presented at the 17th Meeting of the Standing Committee on Tuna and Bill- fish (SCTB), Majuro, Marshall Islands, 9-18 August 2004. dynamics of sharks have shown that average age at first maturity and the relative rate of growth that deter- mines this parameter is one of the leading factors that affect the abil- ity of most sharks to rebound from harvest pressures (Smith et al., 1998; Cortes, 2002; Garcia et al., 2008). Ac- curate age determinations also are necessary for calculations of growth and mortality rates, age at recruit- ment, and longevity. The Shortfin Mako is an epipe- lagic species distributed in temper- ate and tropical seas worldwide (Compagno, 2001) and seldom found in water temperatures lower than 13-17°C (Casey and Kohler, 1992; Stevens, 2008). Shortfin Mako occur off the U.S. West Coast principally off California and Oregon, and catches are associated primarily with warm sea-surface temperatures from 15° to 25°C (PFMC, 2011). Most Shortfin Mako off the U.S. West Coast are sex- ually immature, and high recapture rates for tagged juveniles show that young individuals remain for about 2 years in nearshore California waters (Taylor and Bedford, 2001), where they are more frequently taken in the summer months and are only 148 Fishery Bulletin 111(2) rarely found below the thermocline (Holts and Bed- ford, 1993). Large fish approaching 200 cm fork length (FL) disappear from the catch and presumably move offshore to a more oceanic and highly migratory exis- tence, where most males are mature but females are not (Mollet et al., 2000; Joung and Hsu, 2005; Semba et ah, 2011). Researchers who study the age and growth of this species have used pairs of alternating bands of dif- ferent mineralization in vertebral centra to estimate age. These estimates differ depending on assumptions about the number of band pairs that represent a single year of growth. Bishop et ah (2006) noted that despite different conclusions in these studies, all of them ap- pear to have produced relatively similar growth curves if the same rate of band deposition was assumed. This finding indicates that techniques used by different research groups to identify countable band pairs are generally alike, with the greatest difference being as- sumptions about rates of band deposition. Smaller dis- parities may be due to ontogeny, geographic variability, divergent methods of preparing and interpreting verte- bral banding patterns, difficulties in interpreting band- ing patterns, differences in sample sizes, or limitations of the growth models used (Bishop et ah, 2006). Although considerable advances have been made in determination of age and growth of Shortfin Mako, questions remain as to why current length-at-age models underestimate growth of young fish, especial- ly fish <200 cm FL (Pratt and Casey, 1983; Bishop et ah1; Maia et ah, 2007). Bishop et ah (2006) suggested that this discrepancy may be due to the inability of commonly used growth models to reconcile such rap- id juvenile growth with the slow growth predicted for subadults and adult sharks. Direct age-validation techniques, such as oxytetracycline (OTC) marking and recapture of an adequate sample size of Shortfin Mako <200 cm FL, therefore, are needed to help resolve the issue of band-deposition rates in juvenile Shortfin Mako vertebrae. In this study, we examined OTC-marked vertebrae of 29 juvenile Shortfin Mako (<200 cm FL) released off California and later recaptured from 2000 to 2010 (at liberty from 145 days to 4.4 years). OTC is deposited at sites of active calcification, such as vertebral centra, and is known to remain distinct for at least 20 years in certain finfish, such as the Sablefish ( Anoplopoma fim- bria) (Beamish and McFarlane, 2000), and sharks, such as the Leopard Shark ( Triakis semifasciata) (Smith et al., 2003). The combination of tag-recapture and chemi- cal marking is thought to be the most robust method of age validation (Campana, 2001; Goldman, 2005). These methods test the accuracy of the counts of vertebral band pairs as annual indicators through observation of the banding pattern deposited distally to the OTC mark during the known time at liberty. To supplement this information, we also conducted analyses of at-lib- erty growth of tagged-recaptured sharks and length- frequency data from 3 decades of length data from commercial and research sources. Materials and methods Tagging methods Sharks for tagging and OTC injection were captured in the Southern California Bight (SCB) (Fig. 1) with baited pelagic longlines and identified as Shortfin Mako through the method described by Compagno (2001). Leaders were unsnapped from the main line, and sharks were guided into a semisubmerged metal tagging cradle at the stern of the vessel. The cradle was then raised to facilitate tagging, measuring, and OTC injection, while the eyes of the shark were cov- ered with a wet chamois cloth and a saltwater ven- tilation hose continuously ran water over the shark’s gills. Each shark was tagged on the dorsal fin with a plastic Rototag2 3 (Dalton ID, Henley-on-Thames, UK) la- beled with contact and reward information in English and Spanish and with instructions to measure the fish and save the vertebrae. Most sharks also were double- tagged with a spaghetti tag placed in the dorsal mus- culature beneath the first dorsal fin. At tagging, the sex of each shark was determined and each shark was measured (straight line FL or to- tal length [TL] ) to the nearest centimeter with a sta- tionary meter stick fitted to the shark tagging cradle. Sharks were given an intraperitoneal injection of OTC at a dose rate of 25 mg/kg of body weight; the dose was estimated with a length-weight dose table developed from length-weight measurements of Shortfin Mako measured by NOAA scientific observers for the Califor- nia drift gillnet fishery (Rasmussen3). Laboratory methods for processing vertebrae Past studies have shown that band counts are consis- tent throughout the vertebral column in the Shortfin Mako, indicating that vertebrae from any region can be used in age analysis, although the larger central ver- tebrae are easier to read because of wider band spac- ing (Bishop et al.1; Bishop et al., 2006; Natanson et al., 2006). OTC-marked vertebrae were obtained from Shortfin Mako recaptured on research cruises and com- mercial and recreational fishing vessels between 2000 and 2010. The widest diameter vertebral centra in a given sample were chosen for sectioning. We used only OTC-marked vertebrae from tag recaptures at liberty longer than 124 days (or -4 months) to incorporate one full season into band-pair counts. 2 Mention of trade names or commercial companies is for iden- tification purposes only and does not imply endorsement by the National Marine Fisheries Service, NOAA. 3 Rasmussen, R. 1995. Unpubl. data. Southwest Fisheries Science Center, NOAA, La Jolla, CA. Wells et al. Age validation of |uvenile Isurus oxyrmchus tagged off southern California 149 160°E 180° 160°W 140°W 120°W 100°W 130°W 128°W 126°W 124°W 122°W 120°W 118°W 116°W 114°W Figure 1 Map of study area in the Southern California Bight with tag and recapture locations (triangles) for (A) all recap- tured Shortfin Mako ( Isurus oxyrinchus) (OTC-marked and unmarked ( n =3 1 7 ) and (B) recaptured OTC-injected juvenile Shortfin Mako (n= 29) during the study period from 1996 to 2010. In panel B, black and gray arrows represent males and females, respectively, and the inset shows the more distant recapture locations. To elucidate the vertebral bands, we chose to du- plicate the “hard” (high-frequency) X-radiography technique of Cailliet and Bedford (1983) and Cail- liet et al. (1983), which was used in the only cen- trum aging study of Shortfin Mako off California to date. This type of high-resolution radiography, with industrial, fine-grain, high-contrast film for sharpness and sensitivity, operates in the range of 150 Fishery Bulletin 111(2) 10-110 kV at 3 mA and penetrates the highly calci- fied vertebrae. Samples were stored frozen until processed and kept from light and excessive UV exposure to preserve the OTC time mark. Whole centra were separated, cleaned of excess tissue, rinsed and air-dried. Once dry, 2 centra from each sample were chosen, and 2 types of sections were prepared with a low-speed circular saw (IsoMet, Buehler, Lake Bluff, IL). To duplicate the Cailliet et al. (1983) method of X-raying the whole centrum face and to avoid the double image of the anterior and poste- rior face superimposed in X-ray images, a transverse cut was made through the centrum to create 2 equal halves of the centrum face. Additionally, longitudinal (sagittal) sections, 0. 3-1.0 mm thick, were taken from the center of the second vertebrae to further elucidate banding patterns, especially growth zones along the centrum edge that are sometimes difficult to read on centrum face X-rays (Cailliet et al., 1983). Sections were then mounted and examined under a dissecting microscope at 7. 5-10. Ox magnification, with reflected long-wave UV light (365 nm) to illuminate the OTC mark. A metal pin was glued into place to visualize the position of the brightest, most distal edge of the OTC mark before it was X-rayed. Samples were X-rayed with a General Electric (Fairfield, CT) Mobile 100-15 X-ray unit for exposures between 15 and 45 s at 5 mA and 20—40 kV by using Kodak Industrex M100 film (Readypack II; Eastman Kodak Co., Rochester, NY). X- radiographs were photographed with a Leica Z16 APO dissecting microscope with substage illumination and a Leica DFC420 digital camera (Leica Microsystems, Wetzlar, Germany). Standardization of band-reading techniques and terminology We examined data collected by commercial drift gill- net observers and from research longline surveys and determined the peak in abundance of postpartum-size Shortfin Mako <70 cm FL to identify peak parturition time off southern California. Most individuals (90%) were collected between August and November, provid- ing a tentative parturition time in our study region. Size at birth has been estimated at approximately 63 cm FL (70 cm TL), with parturition occurring year- round according to Mollet et al. (2000); parturition may occur primarily in summer in the South Pacific (Duffy and Francis, 2001) and eastern North Atlantic (Maia et al., 2007). The birth band for counting purposes was identified as the most pronounced calcified first band distal to the centrum focus and indicated by a change in the angle of the centra (Bishop et al., 2006). Seasons were defined according to solstice and equinox periods in the Northern Hemisphere. Band-pair counts Band pairs were counted from digital images of X-ray photographs on a computer screen. We referred to the original X-rays if more detail was desired. As in Bishop et al.1 and Bishop et al. (2006), counts excluded the birth band, which represents age 0. Alternating pairs of translucent bands (hypomineralized; appearing dark in X-ray) and opaque bands (hypermineralized; appearing light in X-ray) were assumed to represent one complete band pair. Two separate band counts were made: 1) to- tal band pairs, or bands distal to the presumed birth band, and 2) band pairs distal to the OTC mark. Band- pair counting began for the former at the distal edge of the first translucent zone beyond the birth band and for the latter at the distal edge of the first translucent zone beyond the OTC mark. In many cases, the OTC mark was directly on a translucent zone, but this zone was not included in the distal-to-OTC counts because only partial growth occurred during this period. This method results in counts that are conservative; counts are lower than they would be if this zone were includ- ed. If a count ended with a partial band pair (observed when the centrum edge was opaque), a plus sign was appended to the count number. For statistical analy- ses, the plus sign was converted to an arbitrary par- tial band count of 0.5. The corpus calcareum (centrum “arm”) was used as a primary counting surface, and bands in the intermedialia were used for confirmation of a band pair, although we were not always successful in acquisition of sections with the fragile intermedialia intact (Goldman, 2005; Branstetter and Musick, 1994). Each sample was examined and counted indepen- dently by 3 readers. Bands were “blind “counted with- out knowledge of Shortfin Mako length, sex, or time at liberty. Readers consulted with each other on cri- teria for counts before readings. Samples for which there was disagreement were counted a second time with X-ray images from which corresponding sample numbers had been removed and placed in a random order. Counts with similar readings among readers were deemed final; however, several samples did not have similar counts. For those samples without similar counts, the average number of band pairs was reported because differences were minor and no readings were deemed irreconcilable. Differences among final readings of each of the 3 in- dependent readers were examined through an analysis of variance (ANOVA) with readers as the dependent variable. A least-squares linear regression analysis was performed, and the null hypothesis that the slope ( b ) of the relationship between the number of band pairs and time was 1:1 (a situation that occurred if one opaque and one translucent band were deposited each year) was tested with a two-tailed f-test (Kusher et al., 1992). Age bias was investigated with age-bias plots and chi-square tests of symmetry by using the contingency table methods of Bowker (1948) and Hoe- nig et al. (1995). Differences in band-pair counts among readers were evaluated by the average percent error (APE) (Beamish and Fournier, 1981) and coefficient of variation (CV) (Chang, 1982) for readings distal to both the OTC mark and birth band. Wells et al Age validation of |uvemle Isurus oxynnchus tagged off southern California 151 Length-frequency analysis Length-frequency distributions of juvenile Shortfin Mako were analyzed to estimate size at age of the first 3 age classes and to compare to vertebrae readings. The MULTIFAN model (Fournier et al., 1990) was one of 2 techniques used to analyze length-frequency data on the basis of size distributions. This model simultane- ously analyzes multiple length-frequency distributions using a maximum likelihood method to estimate the number of age classes and proportions of fish at age. Tests for significance (best-fit growth parameters) were made through the use of likelihood ratio tests. First, a systematic search was performed to estimate the num- ber of age classes and to hold the standard deviations of length constant across all age classes. Next, stan- dard deviations of length were allowed to vary across age classes. The MIXDIST package (MacDonald and Pitcher, 1979) in R, vers. 2.8.0 (R Development Core Team, 2008) was the second length-frequency analysis used. This analysis uses a maximum likelihood method to estimate proportions of fish at age with the added benefit of fitting non-normally distributed data. Data for both techniques were analyzed annually and came from 2 sources: 1) fishery-dependent data from the California drift gillnet fishery (1981-2009), which operates between May and January, and 2) fish- ery-independent data from longline research surveys of juvenile Shortfin Mako conducted by the NOAA South- west Fisheries Science Center (1993-2009) between June and August of each year. A mixture of length measurements taken across study years: TL, FL and alternate length (AL, straight line distance between the origins of the first and second dorsal fins) were standardized into FL for this study to allow for com- parison of our results with the results of other studies. The following length conversions were obtained from the 2 sources and used to standardize data for subse- quent length-frequency analyses: FL = 0.913 x (TL) - 0.397, coefficient of determination (r2) = 0.986 (zz =2 177) FL = 2.402 x (AL) + 9.996, r2= 0.957 (n=3250). Size data were combined between sexes because no significant difference existed (P=0.769), and size bins of 5 cm, ranging from 55 to 265 cm FL, were used. Growth of tagged and recaptured mako sharks Growth rates were calculated for 1) recaptured, OTC-marked sharks and 2) angler- and research-re- leased, unlabeled sharks for which reliable length es- timates were available (n=6 2 of 317 returns) (Fig. 1A). Growth was estimated with the tag-recapture growth model GROTAG (Francis 1988a, 1988b) on the basis of length and time-at-liberty. This model was chosen be- cause age-based and length-based growth models often differ, and it is a useful alternative for comparison of growth rates at particular sizes (Francis, 1988a; Na- tanson et al., 2006; Claisse et al., 2009). We used a maximum likelihood approach with this model to esti- mate growth rates (ga and g^) at 2 selected lengths (a and /?), a CV of growth variability, mean measurement error and standard deviation, and outlier probability. Therefore, estimated growth of a Shortfin Mako, i, was calculated with the following equation: AL, = uPea - agp) / Pa - gp ) - l,) / (i - a + (ga -gp) / a -pFTl), where L( = length at release; and A T, = the tag deployment time. Results Tagging and recapturing oxytetracycline-marked sharks Off southern California from 1996 to 2010, 940 OTC- injected Shortfin Mako were tagged and released (Fig. 1, A and B). Of the subset of released sharks for which sexes were determined, 67% were males and 33% were females. Average size at release was 110 cm FL (±0.80 SE). Of the released OTC-marked sharks, 35 were re- captured from 2000 to 2010. Of these 35 sharks, 29 fish were selected for this study because they had been at liberty for >4 months and OTC marks in the vertebrae fluoresced (Table 1, Fig. IB). Five samples were ex- cluded from analysis because of a short time-at-liberty (within a range from 24 to 60 days), provided no infor- mation on band-pair progression, and in the case of 1 shark, the vertebrae did not fluoresce. For the 29 fish used in this study, average time-at- liberty was 522 days (±71.0 SE), within a range from 145 to 1594 days, and average size was 109 cm FL (±2.9 SE) at tagging and 148 cm FL (±5.0 SE) at re- capture (Table 1). The average displacement distance (great-circle distance from tagging to recapture loca- tion) for the 29 OTC-marked Shortfin Mako was 902 km (±291 km SE), within a range from 2 to 5295 km (Fig. IB). According to results from analyses with lin- ear regressions, no significant effects of time-at-liberty or size-at-tagging existed in relation to total displace- ment distance (P>0.05). Age validation results Results from readings of the OTC-marked vertebrae indicated that 2 band pairs are deposited each year in samples analyzed in this study. The observed slope of the relationship between the number of band pairs each year and years-at-liberty significantly differed from the 1:1 relationship (P<0.01). Specifically, the av- erage number of band pairs predicted each year was modeled with the following linear regression: average number of band pairs = 1.988 x (number of years-at- large) - 0.136, r2=0.942 (P<0.05) (Fig. 2). 152 Fishery Bulletin 111(2) Table 1 Summary table of OTC-marked vertebrae samples from juvenile Shortfin Mako ( Isurus oxyrinchus) recaptured from 2000 to 2010 in the Southern California Bight for this study. Samples are sorted by time at liberty, and information includes tag and recapture dates and fish lengths and sex. The average number of band pairs (from 3 independent readers) is provided for after the OTC mark and birth band mark (±1 standard error [SE] ). NL=either no length estimate or an unreliable one. *=fork length (FL) was converted from total length or alternate length (first dorsal to second dorsal fin). Fish ID Time at liberty (days) Tagging date Recapture date Tagging length (cm FL) Recapture length (cm FL) Sex Average number of band pairs after OTC mark (±1SE) Average number of band pairs after birth band mark (±1SE) A037789 1594 7/11/2000 11/21/2004 108 200 F 7.3 (0.9) 9.3 (0.7) A037559 1512 6/21/2001 8/11/2005 110* 190* M 8.7 (0.3) 11.7 (0.3) A038423 1198 7/9/2005 10/19/2008 91 157* M 7.8 (0.6) 11.5 (0.3) A037655 1133 7/06/2001 8/12/2004 00 CD * 157* F 5.3 (0.2) 8.2 (0.2) A038734 836 7/5/2004 10/19/2006 106* 172* M 5.0 (0.0) 9.0 (0.0) A038611 733 6/19/2004 6/22/2006 111* 152 F 4.0 (0.0) 6.0 (0.0) A039992 730 7/28/2007 7/27/2009 92 146 M 4.0 (0.0) 6.0 (0.0) A039946 555 8/1/2007 2/6/2009 109 150 M 2.5 (0.0) 4.5 (0.0) A038623 531 6/22/2004 12/05/2005 130* 154 M 2.3 (0.2) 7.3 (0.2) A058923 463 6/30/2002 10/06/2003 o OO -X- 139* M 2.5 (0.0) 5.2 (0.7) A058969 446 6/23/2003 9/11/2004 * GO 125* M 2.3 (0.2) 4.7 (0.4) A040354 407 8/13/2009 9/24/2010 98 137* M 2.3 (0.2) 3.8 (0.2) A039341 400 7/24/2007 8/27/2008 131 160* M 1.5 (0.0) 5.5 (0.0) A058916 382 7/01/2002 7/18/2003 137* 162* F 1.5 (0.0) 4.5 (0.0) A038924 373 7/21/2005 7/29/2006 129 163* M 2.0 (0.0) 5.5 (0.3) A039374 364 7/15/2007 7/13/2008 129 154 M 2.0 (0.0) 7.0 (0.0) A038091 350 6/24/2005 6/9/2006 128 135 M 2.0 (0.0) 7.5 (0.3) A038404 335 7/8/2005 6/8/2006 117 NL M 1.5 (0.0) 5.0 (0.3) A040327 323 8/11/2009 6/30/2010 96 114* M 2.0 (0.0) 5.0 (0.6) A038589 313 8/14/2004 6/23/2005 124 141* M 1.5 (0.0) 7.5 (0.6) A040251 298 8/4/2009 5/29/2010 105 141* F 1.0 (0.0) 3.3 (0.3) A040374 298 8/15/2009 6/9/2010 100 NL F 1.2 (0.2) 3.2 (0.2) A040302 277 8/8/2009 5/12/2010 89 103 M 1.2 (0.2) 3.5 (0.3) A039912 270 8/2/2007 4/28/2008 79 98 F 2.3 (0.2) 3.3 (0.2) A040788 253 8/23/2009 5/3/2010 124 164* M 0.8 (0.2) 5.3 (0.2) A038854 211 7/29/2009 2/25/2010 108 NL F 0.8 (0.2) 2.7 (0.2) A039669 206 6/14/2008 1/6/2009 92 NL M 1.0 (0.0) 3.0 (0.0) A040798 193 8/24/2009 3/5/2010 122 NL F 1.0 (0.0) 4.3 (0.2) A038445 145 7/10/2005 12/2/2005 88 NL M 1.0 (0.0) 3.3 (0.3) Results of the vertebral centrum analysis are sum- distal to the OTC mark and 3.5, 3.5, and 3 distal to marized in Table 1, including time at liberty, average the birth band. number of band pairs observed distal to the OTC mark. All tagging activities occurred during summer and average number of band pairs observed distal to the birth band. In addition, Figure 3 shows banding patterns and band-pair counts for 5 vertebrae from Shortfin Mako at liberty from 206 to 1512 days. Visual band-pair counts (identified by arrows in Fig. 3) were agreed upon post hoc by all 3 readers, and 3 samples had full agreement among readers dur- ing independent readings; however, 2 samples did not have complete agreement during independent readings. Band-pair counts distal to the OTC mark for sample A037559 were 9, 9, and 8, and distal counts to the birth band were 12, 12, and 11. Sam- ple A039912 had band-pair counts of 2.5, 2.5, and 2 months when translucent zones appeared in the verte- brae (Fig. 3). In contrast, both translucent and opaque zones were found at the outer edge of vertebrae for samples collected throughout the year without any ap- parent seasonal patterns. There were no significant differences among band- pair counts distal to the OTC mark (ANOVA; P= 0.978) or among total counts distal to the birth band (ANOVA; P= 0.955). Age bias was negligible: age-bias plots (Fig. 4) and chi-square tests of symmetry showed no system- atic bias (P>0.05), confirming that differences were due to random error. Variability among reader counts was low with an APE of 4.36% and CV of 5.71% for counts Wells et al.: Age validation of juvenile Isurus oxyrinchus tagged off southern California 153 12 - 0 365 730 1095 1460 1825 Days at liberty Figure 2 Average number of band-pairs (translucent and opaque) for the number of days at liberty, determined from the oxytetracycline (OTC) mark to the outer centrum edge on vertebrae of juvenile Shortfin Mako ( Isurus oxyrinchus) (n= 29) tagged from 2000 to 2010 in the Southern California Bight for this study. Readings were based on 3 independent readers (±1 standard error (SE1). The solid line shows the linear regression of number of band pairs relative to days at liberty, and the lines with short and long dashes show predicted number of band pairs for 1 band pair/year and 2 band pairs/year, respectively. distal to the OTC mark and with an APE of 5.65% and CV of 7.73% for counts distal to the birth band. Among readers, 93% (27 of 29) of the final band-pair estimates distal to the OTC were within 1 band pair of each other, and 86% (25 of 29) of the estimates distal to the birth band were within 1 band pair. The readings distal to the OTC that differed by more than one band pair were from sharks that had been at liberty for the longest time, in- dicating that variability in band-pair counts in- creases with age, likely because of disagreement among readers caused by structural artifacts. For example, sample A037789 was from a shark that was at liberty for 1594 days and band-pair estimates distal to the OTC varied from 6 to 9 among the 3 readers for this shark (Table 1). Length-frequency analysis A total of 14,720 individual Shortfin Mako were used for the length-frequency analysis that was completed with fishery-dependent data from the California drift gillnet fishery (1981-2009) and fishery-independent data from juvenile Shortfin Mako surveys (1993-2009) (Fig. 5). No differ- ences in length-frequency modes were detected when analyzed by season or sex; therefore, pa- rameter estimates were generated by grouping across factors (season and sex) with age-depen- dent standard deviation incorporated into the model. Average size of Shortfin Mako used in growth modeling was 121.3 cm FL (±0.25 SE). The majority of fish (85%) ranged in size from 80 to 160 cm FL with 3 identifiable modes con- sistently present in annual length frequencies regardless of survey type or sex. Results from MULTIFAN showed these 3 modes at 86, 112, and 134 cm FL, and average modal sizes from MIXDIST were 83, 118, and 147 cm FL. Tak- ing the difference between modal sizes from MULTI- FAN analysis provided average annual growth rates of 27 and 23 cm FL for the period from the first to sec- ond mode and for the period from the second to third modes, respectively (Table 2). Similarly, annual growth rates from MIXDIST averaged 36 and 29 cm FL for the same 2 periods (Table 2). Growth of tagged and recaptured sharks No difference was observed between growth rates cal- culated from at-liberty OTC-injected sharks (28 and 21 cm per year at 85 and 130 cm FL, respectively) and tag-recaptured sharks not injected (29 and 19 cm per year at 85 and 130 cm FL, respectively); there- fore, these data were pooled. The growth rates that resulted from GROTAG analysis of lengths of tagged and recaptured sharks were similar to the growth rates from length-frequency calculations, averaging 29 and 20 cm per year at 85 and 130 cm FL, respectively (Table 2). Discussion Results of this study indicate that 2 band pairs are deposited each year in juvenile Shortfin Mako tagged and released in southern California. The fast growth rates collectively obtained through the use of OTC- marked vertebrae, MULTIFAN and MIXDIST length- frequency analyses, and tag-recapture growth models were consistent, providing strength to our results be- yond the usefulness possible with the employment of a single method alone. Further, a sample size of 29 OTC-marked vertebrae of Shortfin Mako at liberty from 4 months to more than 4 years is the most com- prehensive OTC tag-return data set reported for this species. We also compared our total band-pair counts at length with those of Cailliet et al (1983), the only other Shortfin Mako age and growth study completed 154 Fishery Bulletin 111(2) Figure 3 X-ray images of vertebrae sections that show band-pair progression of 5 OTC-marked juvenile Shortfin Mako ( Isurus oxyrinchus) recaptured during this study off southern California. The small inset in each image shows the OTC fluorescence under UV light. Translucent or hyaline (rapid growth) bands or darker areas of vertebrae show periods of rapid growth, and opaque bands or lighter areas of vertebrae show periods of slow growth. Arrows indi- cate band-pair counts. Wells et al: Age validation of |uvemle Isurus oxynnchus tagged off southern California 155 Counts after the OTC mark <1) ■o 200 cm FL at tagging or recapture, but the 3 largest sharks at recapture (172, 190, and 200 cm FL) exhibited a pattern consistent with biannual deposition. Two of these sharks (200 and 190 cm FL) were at liberty for more than 4 years and averaged 7.3 and 8.7 band pairs distal to the OTC mark, respectively, indicating that the biannual pattern continues in fish of this size. Both had similar total band-pair counts, averaging 9.3 and 11.7, respectively. If one assumes a biannual pattern, as indicated by our OTC marking results, these fish would range from 4 Pauly, D. 1978. A critique of some literature data on the growth, reproduction and mortality of the lamnid shark, Ce- torhinus maximus (Gunnerus). ICES Coucil Meeting Doc. 1978/H: 17, 10 p. 158 Fishery Bulletin 111(2) 4.5 to 6 years of age when collected. Examination of vertebral centra from some of our larger, non-OTC- tagged, recaptured Shortfin Mako reveals that as fish approach maturity, banding patterns appear to become more distinct, with both fast and slow growth zones becoming more regular and evenly spaced. Therefore, it is possible that these larger fish, on entering a more oceanic realm, change the periodicity of their band- ing pattern. On the basis of these results, it appears that Shortfin Mako <200 cm FL and found off Cali- fornia grow much faster than previously thought, with observed rapid growth (biannual band deposition) for approximately the first 6 years of life. Slower growth (annual deposition) thereafter may occur but cannot be confirmed by this study. Mechanisms that drive the observed biannual band- pair deposition may be linked to seasonal migration patterns and subsequent food availability. Through the use of similar OTC tag-recapture methods, Murphy et al. (1998) determined that Black Drum ( Pogonias crornis ) begin to deposit biannual band pairs after 4 years of age because of a shift in migration patterns. Similarly, Shortfin Mako tagged in southern California appear to exhibit a biannual growth cycle, with fast- growth periods (wide, translucent band formation) in summer and winter and slow-growth periods (narrow, opaque band formation) in spring and fall. One possible explanation is that, as coastal waters warm and cool seasonally, juvenile Shortfin Mako move between rich feeding grounds off California in the summer and off Mexico in the winter and spend spring and fall migrat- ing between these feeding areas. Of the 15 OTC-marked sharks recaptured in Baja Mexico waters (Fig. IB), only 2 sharks were recaptured during summer months. In contrast, 70% of the returns of OTC-marked sharks in California coastal waters oc- curred during summer and there were no returns in the winter. In addition, on the basis of data from all tag recaptures (OTC and non-OTC-marked sharks, «=317, senior author, unpubl. data), 40% of all shark recaptures in Mexico (south of 30°N) occurred dur- ing winter, but only 7% of sharks were recaptured in Mexico during summer. Likewise, 58% and 7% of all recaptures in California coastal waters occurred during summer and winter seasons, respectively. Preliminary analysis of movements of juvenile (<150 cm FL) Short- fin Mako tracked with satellite tags (2003-10, n= 26) also shows high relative densities of Shortfin Mako at reported locations off Baja California, Mexico, in the winter and off California in the summer and more dis- persive offshore movements in the spring and fall (S. Kohin, unpubl. data). Tag-recapture techniques with OTC are among the most powerful methods for validation of age and growth patterns in fishes, but there are difficulties with this method. One disadvantage in the use of chemical marking and recaptures is that the number of band pairs formed for short times between tagging and recapture is often low, resulting in a potentially large relative error if one of the band pairs (such as on the growing edge) is misinterpreted (Campana, 2001). For example, misinterpretation of a single growth zone in a fish at liberty for 2 years would result in a 50% error, but the same misinterpretation in a fish at liberty for 10 years would produce an error of only 10%. Fortunately, the relatively large sample size of 15 sharks at liberty for more than 1 year and 4 sharks at liberty for more than 3 years (2 of them for more than 4 years) allowed us to confirm the consistent banding patterns between both short- and long-term deployments. An additional problem with the use of chemical marking techniques, such as with OTC, is that growth may be inhibited (Pfizer,5 1975; Monaghan, 1993); how- ever, other researchers that have worked with elasmo- branchs have shown OTC to have little adverse effects on growth (Tanaka [1990] for the Japanese Wobbegong [Orectolobus japonicus] and Gelsleichter et al. [1998] for the Nurse Shark [ Ginglymostoma cirratum]). Further, this problem does not appear relevant in this study be- cause fast (i.e., not inhibited) growth was observed with our OTC-marked juveniles. Conclusions Direct and indirect methods used in this study indi- cate rapid growth of juvenile Shortfin Mako in south- ern California and the OTC-marked vertebrae support a pattern of biannual deposition for the first 5 years of life. Accurate age and growth parameters in stock assessment models are important for fisheries man- agement and are necessary for calculations of growth and mortality rates, age at recruitment, and longevity. This article highlights the need to continue life history studies of elasmobranch species throughout different regions and ocean basins. Acknowledgments We thank personnel of the Southwest Fisheries Sci- ence Center for assistance in tagging and logistical operations to make this study possible: specifically, D. Holts, R. Rasmussen, R. Vetter, and J. Wraith. We also thank NOAA Fisheries observers for their diligent work in collecting specimens and length data and to the many anglers and commercial operators for return- ing OTC-marked vertebrae. We especially thank O. So- sa-Nishizaki for help with collection and shipment of samples from Mexico, G. Morse for sharing microscope and imaging equipment, and both A. Andrews and M. Francis for valuable suggestions and comments on ear- lier drafts. 5 Pfizer, Inc. 1975. Oxytetracycline intramuscular solution leaflet 60-1051-00-8, 2 p. Pfizer Lab. Div., New York. Wells et at: Age validation of juvenile Isurus oxyrmchus tagged off southern California 159 Literature cited Anislado-Tolentino, V., M. Gallardo-Cabello, F. Amezcua-Lin- ares, and C. Robinson-Mendoza. 2008. Age and growth of the scalloped hammerhead shark, Sphyrna lewini (Griffith & Smith, 1834) from the Southern coast of Sinaloa, Mexico. Hidrobiologica 18:31-40. Ardizzone, D., G. M. Cailliet, L. J. Natanson, A. H. Andrews, L. A. Kerr, and T. A. Brown. 2006. Application of bomb radiocarbon chronologies to shortfin mako ( Isurus oxyrinchus) age validation. En- viron. Biol. Fishes 77:355-366. Beamish, R. J., and D. A. Fournier. 1981. A method for comparing the precision of a set of age determinations. Can. J. Fish. Aquat. Sci. 38:982-983. Beamish, R. J., and G. A. McFarlane. 2000. Reevaluation of the interpretation of annuli from otoliths of a long-lived fish , Anoplopoma fimbria. Fish. Res. 46:105-111. Bishop, S. D. H., M. P. Francis, C. Duffy, and J. C. Montgomery. 2006. Age, growth, maturity, longevity and natural mor- tality of the shortfin mako shark (Isurus oxyrinchus) in New Zealand waters. Mar. Freshw. Res. 57:143-154. Bowker, A. H. 1948. A test for symmetry in contingency tables. J. Am. Stat. Assoc. 43:572-574. Branstetter, S., and J. A. Musick. 1994. Age and growth estimates for the sand tiger in the northwestern Atlantic Ocean. Trans. Am. Fish. Soc. 123:242-254. Brown, P., C. Green, K. P. Sivakumaran, D. Stoessel, and A. Giles. 2004. Validating otolith annuli for annual age de- termination of common carp. Trans. Am. Fish. Soc. 133:133-196. Cailliet, G. M., and D. S. Bedford. 1983. The biology of three pelagic sharks from Cali- fornia waters, and their emerging fisheries: a review. CalCOFI Rep. 24:57-69. Cailliet, G. M., L. K. Martin, J. T. Harvey, D. Kusher, and B. A. Welden. 1983. Preliminary studies on the age and growth of blue, Prionace glauca, common thresher, Alopias vulpi- nus, and shortfin mako, Isurus oxyrinchus, sharks from California waters. In Proceedings of the international workshop on age determination of oceanic pelagic fish- es: tunas, biilfishes, and sharks; 15-18 February 1982 (E. D. Prince and L. M. Pulos, eds.), p. 179-188. NOAA Tech. Rep. NMFS 8. Campana, S. E. 2001. Accuracy, precision and quality control in age de- termination, including a review of the use and abuse of age validation methods. J. Fish Biol. 59:197-242. Campana, S. E., L. J. Natanson, and S. Myklevoll. 2002. Bomb dating and age determination of the large pelagic sharks. Can. J. Fish. Aquat. Sci. 59:450-455. Casey, J. G., and N. E. Kohler. 1992. Tagging studies on the shortfin mako shark (Isu- rus oxyrinchus) in the western North Atlantic. Aust. J. Mar. Freshw. Res. 43:45-60. Cerna, F., and R. Licandeo. 2009. Age and growth of the shortfin mako ( Isurus oxy- rinchus) in the south-eastern Pacific off Chile. Mar. Freshw. Res. 60:394-403. Chang, W. Y. B. 1982. A statistical method for evaluating the reproduc- ibility of age determination. Can. J. Fish. Aquat. Sci. 39:1208-1210. Chen, C. T., T. C. Leu, S. J. Joung, and N. C. H. Lo. 1990. Age and growth of the scalloped hammerhead, Sphyrna lewini, in northeastern Taiwan waters. Pac. Sci. 44:156-170. Claisse, J. T., M. Kienzle, M. E. Bushnell, D. J. Schafer, and J. D. Parrish. 2009. Habitat- and sex-specific life history patterns of yellow tang Zebrasoma flavescens in Hawaii, USA. Mar. Ecol. Prog. Ser. 389:245-255. Compagno, L. J. V. 2001. Sharks of the world. An annotated and illustrated catalogue of shark species known to date. Vol. 2: Bull- head, mackerel and carpet sharks (Heterodontiformes, Lamniformes and Orectolobiformes). FAO Species Cat- alogue for Fishery Purposes 1, 269 p. FAO, Rome. Cortes, E. 2002. Incorporating uncertainty into demographic mod- eling: application to shark populations and their conser- vation. Conserv. Biol. 16:1048-1062. Duffy, C., and M. P Francis. 2001. Evidence for summer parturition in shortfin mako (Isurus oxyrinchus) sharks from New Zealand waters. Z. J. Mar. Freshw. Res. 35:319-324. Fournier, D. A., J. R. Sibert, J. Majkowski, and J. Hampton. 1990. MULTIFAN: a likelihood-based method for esti- mating growth parameters and age composition from multiple length frequency data sets illustrated us- ing data for southern bluefin tuna (Thunnus mac- coyii). Can. J. Fish. Aquat. Sci. 47:301-317. Francis, R. I. C. C. 1988a. Are growth parameters estimated from tagging and age-at-length data comparable? Can. J. Fish. Aquat. Sci. 45:936-942. 1988b. Maximum likelihood estimation of growth and growth variability from tagging data. N. Z. J. Mar. Freshw. Res 22:43-51. Garcia, V. B., L. O. Lucifora, and R. A. Myers. 2008. The importance of habitat and life history to ex- tinction risk in sharks, skates, rays and chimeras. Proc. R. Soc. Lond., Ser. B: Biol. Sci. 275:83-89. Geisleichter, J., E. Cortes, C. A. Manire, R. E. Heuter, and J. A. Musick. 1998. Evaluation of oxytetracycline on growth of captive nurse sharks, Ginglymostoma cirratum. Fish. Bull. 96:624-627. Goldman, K. J 2005. Age and growth of elasmobranch fishes. In Man- agement techniques for elasmobranch fisheries (J. A. Musick and R. Bonfil, eds.), p. 76-102. FAO Fish. Tech. Pap. 474. FAO, Rome. Hoenig, J. M., J. M. Morgan, and C. A. Brown. 1995. Analyzing differences between two age determi- nation methods by tests of symmetry. Can. J. Fish. Aquat. Sci. 52:364-368. 160 Fishery Bulletin 111(2) Holts, D. B., and D. W. Bedford. 1993. Horizontal and vertical movements of the shortfin mako shark, Isurus oxyrinchus, in the southern Califor- nia bight. Aust. J. Mar. Freshw. Res. 44:901-909. Joung, S. J., and H. H Hsu. 2005. Reproduction and embryonic development of the shortfin mako, Isurus oxyrinchus Rafinesque, 1810, in the northwestern Pacific. Zool. Stud. 44:487-496. Kusher, D. I., S. E. Smith, and G. M. Cailliet. 1992. Validated age and growth of the leopard shark, Triakis semifasciata, with comments on reproduc- tion. Environ. Biol. Fishes 35:187-203. MacDonald, P. D. M., and T. J. Pitcher. 1979. Age-groups from size-frequency data: a versatile efficient method of analyzing distribution mixtures. J. Fish. Res. Board Can. 36:987-1001. Maia, A., N. Queiroz, H. N. Cabral, A. M. Santos, and J. P. Correia. 2007. Reproductive biology and population dynamics of the shortfin mako, Isurus oxyrinchus Rafinesque, 1810, off the southwest Portuguese coast, eastern North At- lantic. J. Appl. Ichthyol. 23:246-251. Mollet, H. F., G. Cliff, H. L. Pratt, and J. D. Stevens. 2000. Reproductive biology of the female shortfin mako, Isurus oxyrinchus Rafinesque, 1810, with comments on the embryonic development of lamnoids. Fish. Bull. 98:299-318. Monaghan, J. P. 1993. Comparison of calcein and tetracycline as chemi- cal markers in summer flounder. Trans. Am. Fish. Soc. 122:298-301. Murphy, M. D., D. H. Adams, D. M. Tremain, and B. L. Winner. 1998. Direct validation of ages determined for adult black drum, Pogonias cromis , in east-central Flori- da, with notes on black drum migration. Fish. Bull. 96:382-387. Natanson, L. -J., N. E. Kohler, D. Ardizzone, G. M. Cailliet, S. P Wintner, and H. F. Mollet. 2006. Validated age and growth estimates for the short- fin mako, Isurus oxyrinchus , in the North Atlantic Ocean. Environ. Biol. Fishes 77:367-383. Natanson, L. J., S. P. Wintner, F. Johansson, A. Piercy, P. Campbell, A. D Maddalena, S. J. B. Gulak, B. Human, F. C. Fulgosi, D. A. Ebert, F. Hemida, F. H. Mollen, S. Vanni, G. H. Burgess, L. J. V. Compagno, and A. Wedderburn-Maxwell. 2008. Ontogenetic vertebral growth patterns in the basking shark Cetorhinus maximus. Mar. Ecol. Prog. Ser. 361:267-278. Okamura, H., and Y. Semba. 2009. A novel statistical method for validating the peri- odicity of vertebral growth band formation in elasmo- branch fishes. Can. J. Fish. Aquat. Sci. 66:771-780. PFMC (Pacific Fishery Management Council). 2011. Fishery management plan for U.S. West Coast fisheries for highly migratory species. As amended through Amendment 2, 106 p. Pacific Fishery Man- agement Council, Portland, OR. [Available from http://www.pcouncil.org/highly- migratory -species/ fishery-management-plan-and-amendments/.] Parker, H. W., and F. C. Stott. 1965. Age, size and vertebral calcification in the basking shark, Cetorhinus maximus (Gunnerus). Zool. Meded. 40:305-319. Pauly, D. 2002. Growth and mortality of the basking shark Ceto- rhinus maximus and their implications for management of whale sharks Rhincodon typus. In Elasmobranch biodiversity, conservation and management. Proceed- ings of the international seminar and workshop; Sabah, Malaysia, July 1997 (S. L. Fowler, T. Reid, and F. A. Dip- per, eds.), p. 199-208. Occasional Papers of the IUCN Survival Commission, no. 25. Int. Union Conserv. Na- ture, Gland, Switzerland. Pratt, H. L., Jr., and J. G. Casey. 1983. Age and growth of shortfin mako, Isurus oxyrin- chusi, using four methods. Can. J. Fish. Aquat. Sci. 40:1944-1957. R Development Core Team. 2008. R: a language and environment for statistical computing. R Foundation for Statistical Computing. [Available from: http://R-project.org, accessed November 2008.] Ribot-Carballal, M. C., F. Galvan-Magana, and C. Quinonez- Velazquez. 2005. Age and growth of the shortfin mako shark, Isurus oxyrinchus , from the western coast of Baja California Sur, Mexico. Fish. Res. 75:14-21. Semba, Y., H. Nakano, and I. Aoki. 2009. Age and growth analysis of the shortfin mako shark, Isurus oxyrinchus , in the western and central North Pacific Ocean. Environ. Biol. Fishes 84:377-391. Semba, Y. I. Aoki, and K. Yokawa. 2011. Size at maturity and reproductive traits of short- fin mako, Isurus oxyrinchus , in the western and central North Pacific. Mar. Freshw. Res. 62:20-29. Smith, S. E., D. W. Au, and C. Show. 1998. Intrinsic rebound potentials of 26 species of Pacific sharks. Mar. Freshw. Res. 49:663-678. Smith, S. E., R. A. Mitchell, and D. Fuller. 2003. Age-validation of a leopard shark recaptured after 20 years. Fish. Bull. 101:194-198. Stevens, J. D. 2008. The biology and ecology of the shortfin mako shark, Isurus oxyrinchus. In Sharks of the open ocean: biology, fisheries and conservation (M. D. Camhi, E. K. Pikitch, and E. A. Babcock, eds), p. 87-94. Blackwell Publ., Ltd., Oxford. Tanaka, S. 1990. Age and growth studies on the calcified struc- tures of newborn sharks in the laboratory aquaria us- ing tetracycline. In Elasmobranchs as living resources: advances in the biology, ecology, systematics, and the status of the fisheries (H. L. Pratt Jr., S. H. Gruber, and T. Taniuchi, eds.), p. 189-202. NOAA Tech. Rep. NMFS 90. Taylor, V. B., and D. W. Bedford. 2001. Shortfin mako shark. In California living marine resources: a status report (W. S. Leet, C. M. DeWees, R. Klingbeil, and E. J. Larson, eds), p. 336-337. Publica- tion SG01-11, California Dep. Fish Game, Sacramento, CA. Winker, H., O. L. F. Weyl, A. J. Booth, and B. R. Ellender. 2010. Validating and corroborating the deposition of two annual growth zones in asteriscus otoliths of common carp Cyprinus carpio from South Africa’s largest im- poundment. J. Fish Biol. 77:2210-2228. 161 Changes in size and age at maturity of the northern stock of Tilefish ( Lopholatilus chamaeleonticeps) after a period of overfishing Email address for contact author: richard.mcbride@noaa gov Abstract— The modern fishery for Tilefish (Lopholatilus chamaeleon- ticeps) developed during the 1970s, offshore of southern New England, in the western North Atlantic Ocean. The population quickly became over exploited, with documented declines in catch rates and changes in demo- graphic traits. In an earlier study, median size at maturity (L50) of males declined from 62.6 to 38.6 cm fork length (FL) and median age at maturity (A50) of males declined from 7.1 to 4.6 years between 1978 and 1982. As part of a cooperative research effort to improve the da- ta-limited Tilefish assessment, we updated maturity parameter esti- mates through the use of an otolith aging method and macroscopic and microscopic evaluations of gonads. The vital rates for this species have continued to change, particularly for males. By 2008, male L50 and A50 had largely rebounded, to 54.1 cm FL and 5.9 years. Changes in female reproductive schedules were less variable among years, but the smallest L50 and youngest A50 were recorded in 2008. Tilefish are di- morphic, where the largest fish are male, and male spawning success is postulated to be socially mediated. These traits may explain the initial rapid decline and the subsequent re- bound in male L50 and A50 and less dramatic effects on females. Other factors that likely contribute to the dynamics of maturity parameter es- timates are the relatively short pe- riod of overfishing and the amount of time since efforts to rebuild this fishery began, as measured in num- bers of generations. This study also confirms the gonochoristic sexual pattern of the northern stock, and it reveals evidence of age trunca- tion and relatively high proportions of immature Tilefish in the recent catch. Manuscript submitted 11 July 2012. Manuscript accepted 21 February 2013. Fish. Bull. 111:161-174 (2013). doi 10.7755/FB.111.2.4 The views and opinions expressed or implied in this article are those of the author (or authors) and do not necessar- ily reflect the position of the National Marine Fisheries Service, NOAA. Richard S. McBride (contact author)1 Tiffany E. Vidal1 2 Steven X. Cadrin2 1 Northeast Fisheries Science Center National Marine Fisheries Service, NOAA 166 Water Street Woods Hole, Massachusetts 02543 2 University of Massachusetts Dartmouth School for Marine Science & Technology 200 Mill Road, Suite 325 Fairhaven, Massachusetts 02719 As the largest malacanthid, growing to more than 1 m and 25 kg, the Tile- fish ( Lopholatilus chamaeleonticeps ) is a valuable fishery species, often marketed as “golden tilefish.” The Tilefish ranges from New England to the Gulf of Mexico and into the Caribbean Sea (Freeman and Turn- er1; Dooley, 1978), where 2 stocks have been identified, north or south of the Virginia and North Carolina border (Kitts et al., 2007). North- ern Tilefish are morphologically and genetically distinct from southern Tilefish (Katz et al., 1983). Although individuals can range as far north as Nova Scotia, Tilefish are gener- ally in low abundance in the Gulf of Maine (Able, 2002). Fishing on the northern stock is concentrated from Veatch Canyon, on the southern flank of Georges Bank off Massa- chusetts, to the Hudson Canyon off the coast of New Jersey (Grimes et al., 1980; Grimes and Turner, 1999; Kitts et ah, 2007). Recent (2007-11) Tilefish landings north of the Caroli- 1 Freeman, B. L., and S. C. Turner. 1977. Biological and fisheries data on tilefish, Lopholatilus chamaeleonticeps Goode and Bean. NOAA Fisheries, Sandy Hook Lab. Tech. Ser. Rep. no. 5, 41 p. [Avail- able from http://www.nefsc.noaa.gov/ publications/series/shtsr/shlts r5.pdf, ac- cessed October 2012. nas were valued at $4. 2-5. 6 million annually.2 Unlike some historic fisheries of New England (e.g., Atlantic Cod [Ga- el us morhua ], American Shad [Alosa sapidissima]-, Lear, 1998), the Tile- fish fishery developed only recently, and it had a most inauspicious start. The Tilefish was described in 1879 from fishery catches off New Eng- land (Goode and Bean, 1879). Oc- casional catches followed, but, in 1882, Tilefish became widely known because they constituted the largest single kill of vertebrates ever record- ed. Tilefish are stenothermal, occur- ring along a narrow band of warm water, 9-14°C, at the continental shelf-slope break (Able et al., 1982; Grimes et al., 1986; Grimes and Turner, 1999); Marsh et al. (1999) as- sembled the evidence that this mass mortality was caused by intrusion of the Labrador Current into these outer shelf habitats. After a decade of no reported landings and specu- lation that this species had become extirpated in northern waters, land- ings resumed in the 1890s. Specific 2 NOAA Fisheries, Annual Commer- cial Landings Statistics. http://www. s t. nmfs.noaa.gov/st 1/com mercia 1/1 and - ings/annual_landings.html, accessed De- cember 2012. 162 Fishery Bulletin 111(2) efforts to popularize Tilefish as a food fish resulted in record-high landings (4500 metric tons [t] ) in 1916 (Freeman and Turner, 1977; Grimes and Turner, 1999). These efforts had only modest market success, and, except when prices were high, as in the 1920s and 1950s, landings rarely exceeded 1000 t until the 1970s (Fig. 1). Events in the 1970s proved that persistent annual landings that exceeded 1000 t were unsustainable. Be- ginning in 1971, landings rose rapidly from <100 t to nearly 4000 t within a decade (Fig. 1). Landings re- mained high in the 1980s but were accompanied by ev- idence of overexploitation: decreased fish density, lower catch rates, smaller maximum size, and higher mortal- ity (Grimes et ah, 1980; Turner et ah, 1983; Grimes et al., 1988; Grimes and Turner, 1999). By the 1990s, only a subset of the fleet remained dedicated to fishing for Tilefish in the northeastern United States, and a 2001 fishery management plan capped annual landings at 905 t (Kitts et ah, 2007). The northern Tilefish stock is now considered largely rebuilt but uncertainty in the stock assessment ham- pers confidence in stock status and projections (NEF- SC3). There is, for example, no fishery-independent index of abundance, and monitoring of biological data has been infrequent. Comparisons of assessment model results indicate that the presence of large Tilefish, and the biomass estimate in general, is dependent on peri- odically strong year classes, such as the 1970 and 1973 year classes and most recently the 1993 and 1999 year classes (Turner, 1986; NEFSC3). High levels of exploita- tion during the 1970s and 1980s also may have altered the demographics of the population. Vidal (2010) reports a maximum age of Tilefish of 25 years, much younger than the maximum age of 46 years reported by Turner (1986), indicating that the population has not recovered from age truncation that occurred during the period of high exploitation. This study updates several aspects of Tilefish life history from samples collected in cooperation with the commercial fishery. We began by revisiting the question of whether Tilefish are gonochoristie at the northern ex- tent of their range (Dooley, 1978; Grimes et al., 1988). It has been proposed but not proven that Tilefish are functional hermaphrodites (Sadovy de Mitcheson and Liu, 2008); therefore, we examined and clarified the gonochoristie sexual pattern of the northern stock with gonad histology. Ages were estimated with an otolith method, and age and size at maturity were calculated for both sexes to re- examine sexual dimorphism and temporal dynamics of 3 NEFSC (Northeast Fisheries Science Center). 2009. As- sessment of golden tilefish, Lopholatilus chamaeleonticeps, in the Middle Atlantic-Southern New England region. In 48th northeast regional stock assessment workshop (48th SAW) assessment report, p. 11-180. Northeast Fish. Sci. Cent. Ref. Doc. 09-15 [Available from http://www.nefsc.noaa.gov/ publications/crd/crd0915/pdfs/tilefish.pdf, accessed February 2013.] Year Figure 1 Tilefish (Lopholatilus chamaeleonticeps) landings from Virginia to New England for the period of 1915-2011 in thousands of metric tons (t). Landings from 1915 to 2008 are reported in NEFSC.3 Landing data for 2009- 11 are from a NOAA Fisheries database (http://www. st. nmfs.noaa.gov/st 1/commercial/land ings/annual_land- ings.html, accessed December 2012). maturity ogives. Male Tilefish grow faster and achieve a larger maximum size than females (Turner et al., 1983). Sexual dimorphism also is observed with respect to ma- turity: males develop larger predorsal adipose flaps than females at maturity, and males mature at older ages and larger sizes than do females (Grimes et al., 1988). Grimes et al. (1988) made 2 other important conclusions with respect to measurement of maturity: 1) males show evidence of spermiation, as detected by gonad histology, 1-2 years earlier than macroscopic ripening of the go- nad, indicating that males delay spawning for a couple of years after this initial sign of maturity, and 2), from 1978 to 1982, male age at spawning declined about 2-3 years in association with high rates of exploitation and reduced population density. The effect was so extreme that, by 1982, males matured at a younger age than fe- males (Fig. 2). The topic of dramatic shifts in size and age at ma- turity was still controversial in the 1980s (Beacham, 1987) , but such dynamic metrics have now been asso- ciated with overexploitation in Tilefish (Grimes et al., 1988) and other fish stocks (Trippel, 1995; Wright et al., 2011). To continue this line of inquiry, we compared our results with the benchmark values reported by Grimes et al. (1988). Although rapid responses by maturity traits to changes in mortality can be adaptive at the McBride et al : Changes in size and age at maturity of the northern stock of Lopholatilus chamae/eonticeps 163 Age (years) Figure 2 Maturity ogives for (A) female and ( B ) male Tilefish (Lopholatilus chamae/eonticeps) in 1978 (solid line) and 1982 (dashed line), at the height of the expansion of the modern fishery. Maturity was determined by mac- roscopic appearance of the gonad. Raw data were ex- tracted from Grimes et al. (1988: tables 5 and 6). For model fitting, a generalized linear model and the logit link function in R software were used. The predicted curves together with the median age at maturity (A50) are depicted by sex and year. individual level, such responses can signal a decline in fishery yields and reproductive potential at the popu- lation level (Law, 2000; Fitzhugh et al., 2012; Cooper et al., 2013). Therefore, such data can be important to monitor and include in stock assessments (Caselle et al., 2011; Collins and McBride, 2011). In particular, the pos- sibility that fishing selects for a certain genotype and may thereby cause fishing-induced evolution can be a grave concern in terms of rebuilding fisheries to be sus- tainable (Conover, 2000; Heino and Dieckmann, 2008; Enberg et al., 2011). Materials and methods Field collections During 2 trips by commercial fishing vessels targeting Tilefish in 2008, 688 Tilefish were sampled on 16 dif- ferent days of normal longline operations. The first trip occurred in June, offshore of southern New England, where fish were collected between 70° and 72°W at depths of 104-280 m (Fig. 3). The second trip occurred in July, offshore of southern New England and farther south, where fish were collected between 70° and 74°W at depths of 119-283 m (Fig. 3). This geographic cover- age overlapped all the major fishing areas by the Tile- fish fishery north of the Carolinas (Turner et al., 1983; Kitts et al., 2007). Fish were identified on the basis of taxonomic char- acters summarized by Able (2002). Fork length (FL) was measured to the nearest centimeter, and sex and maturity were determined macroscopically for 421 males and 267 females. Macroscopic determination of maturity followed Idelberger (1985; Table 1), which conforms to the standard maturity classifications used in the region (Burnett et al., 1989). To reduce cluster sampling in high-density areas, especially where fish from the same longline set may have had similar age or reproductive status, at least one fish of each sex was sampled for each 1-cm interval (Wigley et al., 1999; Helle and Pennington, 2004). This sampling strategy resulted in a broad range of fish sizes that was similar to the size composition in the landings, and, if any- thing, this strategy increased the number of larger, older fish to aid in fitting the maturity data to a model (Fig. 4). Gonad histology To confirm macroscopic evaluations of sex and matu- rity, gonad tissue was taken from 157 males and 67 females and fixed in 10% buffered formalin (Fig. 4B). Fixed tissue was prepared according to standard par- affin embedding techniques, stained with hematoxylin, and counterstained with eosin. Histological sections collected from 3 locations (anterior, medial and poste- rior) within the ovary lobe for 15 females were initially examined, but there was no effect of location on the most advanced stage of oocyte development, as also re- ported by Erickson et al. (1985); therefore, no further attention was given to the intragonad location. The sexual pattern, meaning the functional expres- sion of sexuality by individuals, was characterized on the basis of gonad histology. Morphological features noted were the presence of a remnant ovarian lumen 164 Fishery Bulletin 111(2) 74°0'0"W 72°0'0"W 70°0'0"W Figure 3 Map of areas where Tilefish ( Lopholatilus chamaeleonticeps ) were collected between the 100 and 300 m isobaths off southern New England in 2008 for this study to update the maturity schedules of this species. Fish were sampled during 2 trips by commercial fishing vessels targeting Tilefish: in June to the east (right polygon) and in July throughout the region (both polygons). The exact locations are not plotted to maintain confidentiality for commercial fishing operations. in testes or a mix of ovarian and seminiferous tissue in a single gonad, the latter of which was reported by Grimes et al. (1988) and Erickson and Grossman (1986) for functional males. Although we examined morphol- ogy, our interpretation of sexual pattern was made on the basis of functionality, specifically whether gameto- genesis was complete for both ovarian and seminifer- ous tissue during an individual’s lifetime (Sadovy de Mitcheson and Liu, 2008). The existence of gonads that contained nonfunctional tissue of the opposite sex (i.e., intersex) and evidence that individuals matured and spawned as only one sex were considered to be a gonochoristic rather than a hermaphroditic trait. The infrequent presence of isolated oocytes in seminiferous tissue was not considered a bisexual condition because such oocytes also did not confer any function as a fe- male (Sadovy and Domeier, 2005). Because sampling occurred during the spawning season, the histological criterion for female maturity was the presence of secondary oocytes as the most ad- vanced oocyte stage. Secondary oocytes were defined as germ cells that showed evidence of vitellogenin uptake and transformation of lipoprotein yolk in the cytoplasm (Grier et ah, 2009). Cortical alveolar-stage oocytes as the most advance stage were uncommon, and we com- ment on their presence and interpretation in the Re- sults section. Male maturity was marked by the pres- ence of spermatozoa in the spermatogenic lobules. Age determination Sagittal otoliths were extracted at sea and stored dry. These otoliths were thin-sectioned through the core according to standard methods with a low-speed, di- amond-blade saw. Marginal increment analysis indi- cated that annuli are laid down by June of each year (Turner et al., 1983); therefore, given the timing of our collections, the number of complete bands equaled the age of the fish, in years. Otoliths from 100 Tilefish used in Steve Turner’s aging study (hereafter, called the reference collection ) were used to train and calibrate the primary age read- er (T. Vidal) in relation to the age-assignment practic- es reported in Turner et al. (1983) and Turner (1986). McBride et al.: Changes in size and age at maturity of the northern stock of Lopho/atilus chamae/eonticeps 165 Table 1 Macroscopic criteria for classifying maturity of Tilefish ( Lopholatilus chamaeleonticeps ), modified from Idelberger (1985: table 1), with references to new microscopic observations from gonad histology of Tilefish sampled in 2008 off southern New England for this study. Maturity class Description of ovary Description of testes Immature Ovaries are small and transparent, becoming increasingly yellowish, rounded, and veined at the surface as fish nears maturity. Gonads compose <0.5% of body weight. Testes consist of very narrow, transparent bands of tissue, composing <0.05% of body weight. Histological sections reveal isolated oocytes in a low percentage of young males. Developing Ovaries are firm, bulbous, yellow to light orange in color, and 0. 5-2.0% of body weight. Vitellogenic (yolked) oocytes (0.3-0. 7 mm in diameter) are visible through the gonad wall. Gonads become opaque white, increasing modestly in size (0.03-0.12% of body weight). Ripe Enlarged gonads (1. 0-5.0% of body weight) become lobular and have a speckled appearance. A homogenous mixture of vitellogenic and mature (hydrated; 0. 7-1.0 mm) oocytes are evident through the gonad wall. Further enlarged (0. 1-0.2% of body weight) although still relatively small organs; long, flattened, and opaque milky white with phosphorescent sheen. Running ripe Ovaries are turgid and compose 5.0-10.0% of body weight. Gonads have a granular yellow appearance from vitellogenic oocytes in the lamellae and a transparent lumen containing hydrated oocytes, visible ventrally. Eggs (-1.2 mm) flow freely from the vent without any or only light pressure to the abdomen. Sperm released with light abdominal pressure was diagnostic of this maturity class, although it also was observed rarely, even among the largest fish. Spent Gonads are reddish-orange, flaccid, and reduced to 0. 5-1.0% of body weight. Slightly flaccid and reduced to 0.04-0.07% of body weight. Resting Ovaries are uniformly yellow in color, becoming firm, composing 0. 5-1.0% of body weight. Indistinguishable from developing testes. Training included testing for aging precision (i.e. , re- peatability of age assignment by different readers to the same otolith). Precision was first measured by per- cent agreement, A Percen t agreement = 100 x — , ( 1 ) N where A = the number of replicate ages in agreement (of 2); and N = the total number of fish aged. Precision also was evaluated with Chang’s coefficient of variation (CV; Chang, 1982): CV = 100 x 1 % \ N “ 7=1 2 (X y '7 ,=i R- 1 X, XjY (2) where N = the total number of fish aged; Xt/ = the /th age determination (i.e., of 2) of the jth fish; and X = the mean age estimate of the yth fish. m- 1 m (/I - n )z y y y V Ji ’ (=i j=i+ 1 n . + n (3) where m = the maximum age in the data set; and n = the number of fish in the /th row and yth column, etc. These precision tests also were used to evaluate repeatability of multiple readings by the primary age reader. In terms of final age assignment, when the first 2 readings from each otolith collected in 2008 dis- agreed, a third reading was performed. The value that occurred twice was used as the final age (i.e., there were no situations in which the third reading was dif- ferent from both of the first 2 readings). Models and analysis Generalized linear models were programmed with R, vers. 2.15.24 (R Core Team, 2012), to estimate param- eters of maturity ogives. A full range of immature and mature fish, by size and age, was collected, making pa- rameter estimation straightforward (Trippel and Har- Bowker’s test was used to detect departures from sym- metry between the new reader and the reference col- lection, with the formulation of Hoenig et al. (1995): 4 Mention of trade names of commercial companies is for iden- tification purposes only and does not imply endorsement by the National Marine Fisheries Service, NOAA 166 Fishery Bulletin 111(2) o 30 40 50 60 70 80 90 110 Fork length (cm) Figure 4 Fork lengths of Tilefish ( Lopholatilus chamaele- onticeps) sampled in 2008 from (A) ports of the northeastern United States ( /? =5 1 10) and (B) trips on cooperating commercial Fishing vessels during June and July 2008. In panel B, plotted histograms are overlapping (not stacked) and depict the numbers of individuals examined by macroscopic characters only (gray), histological characters only (black), and both methods (di- agonal lines). vey, 1991). An information-theoretic approach was used to select among full and reduced models; the second- order Akaike’s information criterion (AICc) was used to account for sample sizes in all comparisons. Model selection began with evaluation of logit, pro- bit, and complementary log-log model fits to FL or age, and histological maturity data, by sex. The logit model was selected because it consistently had a lower or a tied score (<2 AAICc value) compared with the score of the probit model and a much lower score than that of the complementary log-log model. The logit-link function was used for selection of full or reduced models (i.e., by sex [logit {mature) ~ FL + Sex + FL:Sex], method [macroscopic, histological], or year [1978, 1982, 2008]). Model results from 2008, as presented in Tables 2 and 3, are from analyses of in- dividuals with both size and age data. Additional fish were measured for FL only, but inclusion of these fish in the analyses did not substantially change any re- sults. Historic data were extracted from Grimes et al. (1988). Because FLs in Grimes et al. (1988: tables 3 and 4) were pooled by 5-cm bins, when entering their data for analysis, we assigned FL values for each fish as a midpoint value. Grimes et al. (1988: tables 5 and 6) reported older fish as a plus group (i.e., 15+); there- fore, we also grouped all fish >15 years old together as a plus group to be consistent between studies, but the use of this plus group did not alter any result. Results Gonad histology No seminiferous tissue was evident in any functioning females, but morphologically intersex males, occurring in 2 morphs, were observed (Fig. 5). In the first morph, a lumen was evident but seminiferous tissue arose di- rectly along the gonad wall and no oocytes were pres- ent (Fig. 5A). Males with a lumen were common, but their frequency was not quantified because many histo- logical sections were incomplete across the transverse plane and the lumen was relatively small and difficult to recognize in larger testes. The other morph of inter- sex males appeared as rare, isolated, primary growth oocytes interspersed in continuous seminiferous tissue (Fig. 5B). This morph was observed in 4 young fish, ages 3 or 5, and at least some of the embedded oocytes were visibly degrading. In other males, remnant gaps in seminiferous tissue were evident, indicating that isolated oocytes had been present but were now fully degraded (Fig. 5C). The immature testis was initially dominated by spermatogonia surrounded by connective tissue, with limited spermatogenesis in the form of spermatocytes and spermatids in crypts surrounded by germinal epi- thelia (Fig. 5C). As spermatogenesis proceeded, sper- matozoa were released into lobules lined with a discon- tinuous germinal epithelia (i.e., spermiation), signaling physiological maturity (Fig. 5D). The ovary of immature females was dominated by oocytes at a perinucleolar stage (Fig. 5E). In many females classified as immature, the appearance of or- ganelles (i.e., a Balbiani body) and early formation of cortical alveoli were common in the cytoplasm of the largest primary growth oocytes (Fig. 5F); we consid- ered these traits characteristic of a maturing state (rather than a mature one), where such individuals were only preparing to mature and would not spawn until the following year. Only a single individual had advanced cortical alveoli as the most advanced oocyte stage (Fig. 5G), and 2 other females had only begun McBride et al.: Changes in size and age at maturity of the northern stock of Lopholatilus chamaeleonticeps 167 Figure 5 Microphotographs of gonadal tissue of Tilefish (Lopholatilus chamaeleonticeps) functional males (left) and females (right): (A) a functional male with seminiferous tissue developing directly along a lumen (arrow); (B) isolated oocytes, one partially degraded (left) and one in- tact (right), embedded in seminiferous tissue of a functional male; (C) seminiferous tissue of an immature male (ct=connective tissue, sg=spermatogonia, sc=spermatocytes, st=spermatids, do=degraded oocyte jfully degraded, no cell remaining]); (D) tissue of a mature (developing) male (sz=spermatozoa); (E) ovarian tissue of an immature female ( pn=perinucleolar oocyte, triangle=oogonial nest); (F) primary growth oocytes, marking the Balbiani body (black arrow) and other inclusions that appear to be precursors to cortical alveoli (white arrow); (G) an oocyte with advanced cortical alveoli throughout the cytoplasm (arrow); and (H) ovarian tissue of a mature (ripe) female (ev=early vitellogenic oocyte [not fully with yolk], mn=migrating nucleus, ho=hydrated oocyte, pof=postovulatory follicle). Scale bars vary with image: 50 pm (E, F, G, B), 100 pm (C, D), 250 pm (A), and 500 pm (H). 168 Fishery Bulletin 111(2) vitellogenesis (not shown; i.e., the most advanced oo- cyte stage was only partially yolked, where the yolk inclusions did not extend from the nucleus to the cho- rion). These 3 fish were young, age 5 or 6, and, if they had been capable of spawning imminently, they evi- dently would have started spawning later than other conspecifics. In comparison, 92% of mature females were already actively spawning, with oocytes that either exhibited migrating nuclei or were in various stages of hydration; postovulatory follicles were ob- served as well (Fig. 5H). Age determination Tilefish otoliths are difficult to age, but good results were obtained after training the primary age reader with the reference collection. There was 62% percent agreement with the reference collection (85% and 89% agreement within 1 or 2 years, respectively), with a Chang’s CV of 5.1. There was a tendency to underage fish approximately 15 years old and older; however, Bowker’s test indicated this departure was not signifi- cant (%2= 19.5; P=0.42). Ages of Tilefish collected in 2008 ranged from 3 to 25 years for females and 3 to 23 years for males. Nearly all fish (98%; n = 180) were <15 years old (Fig. 6). Precision, based on re-reading the 2008 otoliths, was good. Percent agreement between the first 2 read- ings was 79.6% (97% and 98% within 1 or 2 years, re- spectively), with Chang’s CV of 2.2. Again, the Bowk- er’s test of symmetry was not significant (%2=17.4; P=0. 18). McBride et al.: Changes in size and age at maturity of the northern stock of Lopholatilus chamae/eonticeps 169 Table 2 Median fork length (L50, cm) and age (A50, years) at maturity, standard error (SE), range, and sample size (n) of Tilefish ( Lopholatilus chamaeleonticeps) collected in 2008 off southern New England — by year, sex, and method used to evaluate maturity. The methods used were macroscopic evaluation of the whole gonad and microscopic evaluation of gonad histology. Raw data from 1978 and 1982 are extracted from Grimes et al. (1988: tables 3-6; see text for details). Predicted ages at maturity from tabulated data in Grimes et al. (1988) also are plotted in Figure 2. Year Sex Method ^50 SE Range n "^50 SE Range' n 2008 Male Macroscopic 54.1 1.4 32-100 99 5.9 0.2 3-23 99 2008 Male Histological 46.8 1.5 32-100 99 4.9 0.2 3-16 99 2008 Female Macroscopic 44.1 1.1 32-90 58 4.9 0.2 3-25 58 2008 Female Histological 46.3 1.2 32-90 58 5.2 0.2 3-21 58 1982 Male Macroscopic 38.6 4.6 41-95 241 4.6 0.8 4-12 88 1978 Male Macroscopic 62.6 1.0 31-115 384 7.1 0.2 4-15 246 1982 Female Macroscopic 49.8 0.4 26-100 360 5.5 0.2 4-15 121 1978 Female Macroscopic 45.4 1.2 31-95 393 5.2 0.1 4-15 267 'Ages >15 years were grouped because Grimes et al. ( 1988) had grouped ages at this value in their data tables. Maturity methods compared Before we compared maturity schedules between years, we evaluated a potentially confounding effect on esti- mation of maturity of Tilefish: the effect of method. There was a general agreement in maturity classifi- cation (i.e., immature versus mature) between mac- roscopic and microscopic (histological) methods. When both methods were used on the same fish, the agree- ment was higher for females (94%) than for males (84%). Mismatched females (n= 4) were immature ac- cording to gonad histology but mature macroscopically. Table 3 Comparisons of different data aggregates in testing for the effect of fish size (fork length [FL], cm) or age (years) on maturity of Tilefish ( Lopholatilus chamaeleonticeps). The first comparison tests the effect of (A) method, macroscopic (macro) versus histological (histo), in evaluation of maturity status of fish collected in 2008 off southern New England for this study (controlling for factors of sex, M=male; F= female). The next comparison tests for (B) sexual dimorphism (by method with data from examination of fish collected in 2008). The last 2 comparisons test whether macroscopic estimates of maturity (used in all years) were different in 2008 than they were in (C) 1982 and (D) 1978, by using historic data for 1978 and 1982 from Grimes et al. (1988). See Table 2 for fitted parameter values by year, sex, and method. “Units” are modeled as a covariate, either as a main effect ( + ) or an interaction (*). Model sets are evaluated row-wise, with the second-order Akaike’s information criterion (AICc) value. The lowest AICc value, indicating the least uncertainty, is underlined. If AAICc values are <2, indicating the effects are indistinguishable, both or all cells are underlined. AICc values of full and reduced models Units compared (covariates) Other Length models Age models factors FL* units FL+ units FL Age* units Age+ units Age A Method (macro/histo) M, 2008 86.8 85.0 97.1 94.9 93.4 104.2 Method (macro/histo) F, 2008 55.8 53.9 53.8 68.5 66.5 66.0 B Sex ( male/female) Macro, 2008 74.7 74.9 95.8 87.3 85.9 93.7 Sex (male/female) Histo, 2008 67.9 66.1 64.2 76.1 74.1 73.8 C Year (1982/2008) M, Macro 305.3 323.9 329.4 143.4 153.1 151.8 Year (1982/2008) D Year ( 1978/2008) F, Macro 189.8 187.8 208.4 78.7 77.2 78.8 M, Macro 353.5 356.6 369.7 248.9 249.6 261.5 Year (1978/2008) F, Macro 344.8 353.8 352.8 230.0 229.8 229.7 170 Fishery Bulletin 111(2) Mismatched males (n= 25) were mature according to histology but immature macroscopically. Disagreements were for fish within the size range of transition from immature to mature (females: 42-49 cm FL; males: 44-71 cm FL). Mismatches among fe- male classifications resulted in values of median size and age at maturity (L50 and A50, respectively) that were 2.2 cm FL larger and 0.3 years older for histology- based results than for macroscopic classifications (mac- roscopic: L50=44.1 cm FL, A50=4.9 years; histological: L50=46.3 cm FL, A50=5.2 years) (Table 2). Mismatches among male classifications resulted in the L50 and A50 values that were 7.3 cm FL smaller and 1.0 year younger for histological examination than for macro- scopic results (macroscopic: L50= 54.1 cm FL, A50=5.9 years; histological: L50=46.8 cm FL, A50=4.9 years) (Ta- ble 2). Therefore, a histological method not only shifted the median parameter estimates in opposite directions for each sex, but the magnitude of uncertainty due to method was much greater for males than for females (Table 3A). Maturity and spawning The fishery harvests immature fish of both sexes. Mac- roscopic collections indicated that 14% of females (32 — 49 cm FL, 3-6 years old) and 38% of males (32-71 cm FL, 3-9 years old) caught on the 2008 sampled trips were immature. Histological classifications did not support differenc- es in L50 or A50 between sexes; however, macroscopic observations supported sexual dimorphism in the Lr>0 and A50 parameters (Tables 2, 3B). These results likely mean that gonad histology detects hormonal matu- ration, a physiological state that occurs at a similar size and age for each sex but that may not be an ac- curate predictor of spawning activity for males. If so, then spawning activity, which is more closely aligned to measuring spawning stock biomass, occurred when males were 10 cm FL larger and 1 year older than fe- males, on average, in 2008 (Table 2; Fig. 7). Discussion This study confirms that the northern stock of Tilefish is functionally gonochoristic. Grimes et al. (1988) also concluded that this stock is gonochoristic, noting the presence of isolated oocytes in 2 of 50 testes. They did not report finding a lumen in testes, but they may have overlooked it, stating they were unsure about the sexu- ality of small fish with a lumen. We observed that the lumen was not always obvious in large fish, even when looking for it. The term “prematurational sex change” — where individuals express themselves as a female first but do not mature as a female before they switch to a male — does not seem to apply here. Instead, we believe that a testis containing a lumen is a common feature in Tilefish, as occurs for other fishes (e.g., Pomacentri- 1.0 - ii in (i i! iii — ~ n i rr A :'r 0.8- 1 / 0.6- // 0.4 - , I , /495=6.1 yr / /; ^5o=4-9 y 0.2- / / 1' A5=3.6 yr o.o- U U_J i i i o 0 5 10 15 Q. 1.0 - ii ii mi ii in n r B / 0.8- ;7 / 0.6- ,7/ 0.4- : : ^=7.7 yr / ^so^-9 yr 0.2 - / / cn II yr 0.0- 1L B L 1 1 0 5 10 15 Age (years) Figure 7 Maturity schedules of Tilefish ( Lopholatilus chamaeleonticeps) (A) females and (B) males col- lected in 2008 off southern New England for this study. Plotted are the predicted ogive (solid line), the 95% confidence limits (dashed lines), and individual data (internal tick marks, staggered relative to each other to reveal sample size). The median age at maturity (A50) is listed, along with the age at which 5% (A5) and 95% (A95) of the individuals were mature. Ages >15 years were grouped. dae and Serranidae; Sadovy and Domeier, 2005), but is unrelated to function. We also do not categorize Tilefish as bisexual — a term that does not apply with the ap- pearance of a lumen or the presence of isolated oocytes as described here (Sadovy and Domeier, 2005). Our conclusion about gonochorism emphasizes func- tion; in other words, all individuals reproduce exclu- sively as either male or female during their lives (Sa- dovy de Mitcheson and Liu, 2008). Because we sam- pled fish during the spawning period, and functional, McBride et al : Changes in size and age at maturity of the northern stock of Lopholatilus chamaeleonticeps 171 simultaneous hermaphroditism was not observed, this type of hermaphroditism is unlikely. We did not sam- ple in winter to test for sequential sex change during the nonspawning period, but the histological evidence of similar A50 for both functional sexes makes this change unlikely. We predict that intersex fish would be no more common during the nonspawning season than they are reported herein for the spawning season. Also, we predict that, if collections of younger (<3 years old) fish were possible, isolated oocytes would be more commonly seen because isolated oocytes were observed to be degrading in our collections of testes. If these predictions about morphology are correct, they would confirm our conclusions that Tilefish are functionally gonochoristic. Erickson and Grossman (1986) investi- gated the sexual pattern of Tilefish farther south, in Atlantic waters of the Georgia Bight, and also conclud- ed that Tilefish are functionally gonochoristic. In con- trast, Lombardi-Carlson (2012) reported higher rates of intersex Tilefish in the Gulf of Mexico, evident for both functional males and females; therefore, there appears to be geographic variation in the morphological expres- sion of intersex fish and possibly the sexual pattern by Tilefish. Our study is the first attempt to age the north- ern stock of Tilefish in nearly 30 years. Turner et al. (1983) reported ages of fish collected in the longline and recreational fisheries in 1978, and Turner (1986) reported ages from the longline fisheries in 1979, 1980, and 1982. Females older than 31 years were collected in each of these sampling years, and males older than 31 years were collected in half of these years (Turner, 1986: appendix 1, A-H). The oldest fish observed was 46 years old (Turner, 1986), nearly twice as old as the oldest fish observed in 2008 in our study. Age structure during 1978-82 also appeared to be dominated by the 1970 and 1973 year classes, but dominant year classes (the most recent one being 1999; NEFSC3) were not ob- vious from the age structure measured in 2008 (Fig. 6). The reduced numbers of older fish today indicate that age truncation still exists, a finding that should not be surprising because landings >1000 t persisted well into the 1990s. We predict that fish older than 30 years will return to the population in the next decade. The effect of method (macroscopic versus histologi- cal) in determination of maturity was more pronounced for males than for females (Tables 2, 3A). Differences in the 2008 female L50 and A50 attributable to method were minor (2.2 cm FL, 0.3 year). Our confidence in macroscopic evaluation of maturity was good at this time of year, when the main histological criterion, vi- tellogenic oocytes, were large enough (0.3-0. 7 mm) to be seen macroscopically; hydrated oocytes were even more readily visible: 0. 7-1.0 mm (as measured from histological slides by T. Vidal, unpubl. data). Data for females at other times of the year, especially during the nonspawning season, are likely to be less precise or accurate (Vitale et al., 2006; McBride et al., 2013). That the observed differences in L50 and A50 attrib- uted to method were larger for males (7.3 cm FL, 1.0 year) than for females was not unexpected. Grimes et al. (1988) also observed larger and older male L50 and A50 with a macroscopic method versus a histological method. We agree with Grimes et al. (1988): these dif- ferences in male maturity are not merely a method- ological artifact but are of biological significance — like- ly the result of a physiological lag in gonad growth and the time that exists between spermiation (an indica- tion of hormonal activity) and full ripening of the tes- tes that precedes functional spawning by males. Such a lag may also be associated with behavioral differences. Grimes and Turner (1999) postulated that males first mature in a subordinate role and become dominate within 1-2 years. Although such hypotheses demand further study, it is obvious that the method to determine maturity can matter in comparative analyses. The macroscopic method is likely aligned with functional spawning, and functional spawning more accurately defines spawn- ing stock biomass. Therefore, it is the more appropri- ate method to use in routine measures to characterize this reference point for males. Grimes et al.’s (1988) approach emphasized the macroscopic method; there- fore, our comparisons with relatively large sample sizes should be robust between all years (i.e. , 1978, 1982, 2008), The large percentage of immature Tilefish in the catch (14-38%, by sex) appears to point to violation of the principle to let fish reproduce at least once before they are harvested (Sissenwine and Shepherd, 1987). Although it is once again a topic of debate (Garcia et al., 2012), this principle prompts a re-evaluation of the effect of hook size on the proportion of immature fish landed. Female size at maturity differed between 2008 and earlier years, but female age at maturity did not differ strongly between years (Tables 2, 3[C and D|). Female L50 was smaller in 2008 (44.1 cm FL) than in 1978 (45.4 cm FL) and 1982 (49.8 cm FL). The difference in A50 between all years was <1 year, and the low AICc score (i.e., <2) indicated that these differences in fe- male A50 over time were similar. Nonetheless, female fitness is related to size and age (Green, 2008). In our study, the youngest A50 was measured in 2008. Other studies have shown that such shifts in maturity sched- ules are associated with reduced yield, survival, and fecundity (Law, 2000; de Roos et al., 2006; Conover et al., 2009). In particular, the mature female Tilefish that showed no immediate evidence of spawning were young (5-6 years old), indicating that newly matured females have lower spawning frequency and, therefore, a lower reproductive potential than older females. Age-specific effects on spawning frequency and batch fecundity are commonly observed in fishes and can alter stock as- sessment outcomes (Fitzhugh et al., 2012), and there- fore continued research is warranted to clarify such 172 Fishery Bulletin 111(2) additional effects on reproductive potential of female Tilefish. Male size and age at maturity differed between years much more dramatically than did female size and age at maturity (Tables 2, 3[C and D] ). In 2008, male L50 was 54.1 cm FL, smaller than in 1978 (62.6 cm FL) but larger than in 1982 (38.6 cm FL). The same rebounding pattern was evident for male A50, but, again, the differ- ence in A50 between 1978 and 1982 (2.5 years) was not completely regained by 2008. Grimes et al. (1988) re- ported the initial trend, when they concluded that high fishing pressure was associated with and presumably induced a reduced size at maturity for males between 1978 and 1982. Our results indicate that this smaller size and younger age at maturity observed in 1982 did not become fixed. Because male fitness can be related strongly to size and age (Trippel, 2003), it is likely that male reproductive success is still hampered by reduced maturity parameters relative to that observed in 1978. The interpretation and predictability of these results, however, are hindered by the limited amount of data available to determine the stability of sex-specific ma- turity schedules between years. Early reports that maturity schedules were flexible and could be dynamic in response to rates of fishing were treated with skepticism (Beacham, 1987), but they are becoming increasingly common and well sup- ported (de Roos et ah, 2006; Conover et ah, 2009; Chu- wen et ah, 2011). If rates of maturation are heritable and survival rates of reproducing individuals are low, fishing will select individuals that reproduce at smaller sizes and younger ages (Reznick et ah, 1990; Hutchings, 1993). If this selective pressure were to be eliminated, maturation rates would still likely require several gen- erations to rebound (Conover et ah, 2009). Tilefish have a minimum generation time of 5-7 years, according to the generation time estimated from the values of A50 reported here. Therefore, several generations actually have passed between 1982 and 2008. Also, the period of high exploitation (1977-87) did not extend beyond 2-3 Tilefish generations, unlike the several decades- long, chronic effects of overfishing observed in some other fisheries (Worm et ah, 2009). Finally, Tilefish maturation evidently is not entirely under genetic con- trol because males are presumed to use proximate sex and size cues to determine their reproductive potential (Grimes et ah, 1988). In summary, we cannot rule out that genetic selection caused by increased fishing rates occurred 30 years ago for Tilefish, but we can point to these other factors as likely contributors to the ap- pearance of a rebounding maturity rate among males following heavy exploitation in the 1970s and 1980s. Conclusions Intersex males exist, but the northern stock of Tilefish is functionally gonochoristic. Current demographics in- dicate age truncation and the lack of any strong year classes in the rebuilt fishery of 2008. A macroscopic method for assessment of sex-specific maturity dur- ing the spawning season was verified as a reliable and cost-effective approach to monitoring trends in Tilefish maturation. Previously published data (Grimes et ah, 1988) were reanalyzed, and an information-theoretic approach revealed differences in estimates of maturity ogives attributed to methods, sexes, and years. Once method and sex were accounted for, it was evi- dent that male maturation rates have rebounded from an earlier decline associated with a period of overex- ploitation. This rebound probably occurred because the period of overexploitation did not last long, several generations have passed during a period of improved conditions for the fishery, and male maturation is so- cially mediated. At present, only 3 years of age and maturity data exist, but these data were available for the 2009 Tilefish assessment, and they show the value of continued cooperative biological monitoring in this data-limited fishery. Acknowledgments The fishermen and vessel owners of the FV Sea Cap- ture and the FV Kimberly were cooperating partners, together with the assistance of the Northeast Fisher- ies Science Center (NEFSC) Study Fleet program: J. Hoey, M. Palmer, J. Moser, K. Anderson, M. Ball, and I. Conboy. J. Burnett, J. Brenton, S. Correia, B. Jackson, P. Nitschke, K. Oliveira, N. Perry, G. Shepherd, and K. Stokesbury also assisted materially. Funding for this project was provided through NOAA Fisheries Coop- erative Marine Education and Research program, the University of Massachusetts, the NEFSC Cooperative Research Program, and the Massachusetts Fisher- ies Institute. C. Grimes, R. Muller, and 3 anonymous reviewers provided constructive criticism on earlier drafts. We thank all of those involved. Literature cited Able, K. W. 2002. Tilefishes. Family Malacanthidae. In Bigelow and Schroeder’s fishes of the Gulf of Maine, 3rd ed. (B. B. Collette and G. Klein-MacPhee, eds.), p. 397-400. Smithsonian Inst. Press, Washington, D.C. Able, K. W., C. B. Grimes, R. A. Cooper, and J. R. Uzmann. 1982. Burrow construction and behavior of tilefish, Lopholatilus chamaeleonticeps, in Hudson Submarine Canyon. Environ. Biol. Fishes 7:199-205. Beacham, T. D. 1987. Variation in length and age at sexual maturity of Atlantic groundfish: a reply. Environ. Biol. Fishes 19:149-153. Burnett, J., L. O’Brien, R. K. Mayo, J. A. Darde, and M. Bohan. 1989. Finfish maturity sampling and classification schemes used during Northeast Fisheries Center bottom McBride et al.: Changes in size and age at maturity of the northern stock of Lopholatilus chamaeleonticeps 173 trawl surveys, 1963-89. NOAA Tech. Memo. NMFS-F/ NEC-76, 14 p. Caselle, J. E., S. L. Hamilton, D. M. Schroeder, M. S. Love, J. D. Standish, J. A. Rosales-Casian, and O. Sosa-Nishizaki. 2011. Geographic variation in density, demography, and life history traits of a harvested, sex-changing, temper- ate reef fish. Can. J. Fish. Aquat. Sci. 68:288-303. Chang, W. Y. B. 1982. A statistical method for evaluating the reproduc- ibility of age determination. Can. J. Fish. Aquat. Sci. 39:1208-1210. Chuwen, B. M., I. C. Potter, N. G. Hall, S. D. Hoeksema, and L. J. B. Laurenson. 2011. Changes in catch rates and length and age at ma- turity, but not growth, of an estuarine plotosid ( Cmdo - glanis macrocephalus) after heavy fishing. Fish. Bull. 109:247-260. Collins, A. B., and R. S. McBride. 2011. Demographics by depth: spatially explicit life-his- tory dynamics of a protogynous reef fish. Fish. Bull. 109:232-242. Conover, D. O. 2000. Darwinian fishery science. Mar. Ecol. Prog. Ser. 208:303-307. Conover, D. O., S. B. Munch, and S. A. Arnott. 2009. Reversal of evolutionary downsizing caused by se- lective harvest of large fish. Proc. R. Soc. Lond., Ser. B: Biol. Sci. 276:2015-2020. Cooper, W. T., L. R. Barbieri, M. D. Murphy, and S. K. Lowerre-Barbieri. 2013. Assessing stock reproductive potential in species with indeterminate fecundity: effects of age truncation and size-dependent reproductive timing. Fish. Res. 138:31-41. de Roos, A. M., D. S. Boukal, and L. Persson. 2006. Evolutionary regime shifts in age and size at mat- uration of exploited fish stocks. Proc. R. Soc. Lond., Ser. B: Biol. Sci. 273:1873-1880. Dooley, J. K. 1978. Systematics and biology of the tilefishes (Perci- formes: Branchiostegidae and Malacanthidae), with de- scriptions of two new species. NOAA Tech. Rep. NMFS Circ. 411, 78 p. Enberg, K., C. Jorgensen, E. S. Dunlop, 0. Varpe, D. S. Boukal, L. Baulier, S. Eliassen, and M. Heino. 2011. Fishing-induced evolution of growth: concepts, mechanisms and the empirical evidence. Mar. Ecol. 33:1-25. Erickson, D. L., and G. D. Grossman. 1986. Reproductive demography of tilefish from the South Atlantic Bight with a test for the presence of protogynous hermaphroditism. Trans. Am. Fish. Soc. 115:279-285. Erickson, D. L., M. J. Harris, and G. D. Grossman. 1985. Ovarian cycling of tilefish, Lopholatilus chamae- leonticeps Goode and Bean, from the South Atlantic Bight, U.S.A. J. Fish Biol. 27:131-146. Fitzhugh, G. R., K. W. Shertzer, G. T. Kellison, and D. M. Wyanski. 2012. Review of size- and age-dependence in batch spawning: implications for stock assessment of fish spe- cies exhibiting indeterminate fecundity. Fish. Bull. 110:413-425. Garcia, S. M., J. Kolding, J. Rice, M. J. Rochet, S. Zhou, T. Ari- moto, J. E. Beyer, L. Borges, A. Bundy, D. Dunn, E. A. Ful- ton, M. Hall, M. Heino, R. Law, M. Makino, A. D. Rijnsdorp, F. Simard, and A. D. M. Smith. 2012. Reconsidering the consequences of selective fisher- ies. Science 335:1045-1047. Goode, G. B., and T. H. Bean. 1879. Description of a new genus and species of fish, Lopholatilus chamaeleonticeps , from the south coast of New England. Proc. U.S. Natl. Mus. 2:205-209. Green, B. S. 2008. Maternal effects in fish populations. Adv. Mar. Biol. 54:1-105. Grier, H. J., M. C. U. Aranzabal, and R. Patino. 2009. The ovary, folliculogenesis, and oogenesis in tele- osts. In Reproductive biology and phylogeny of fishes (agnathans and bony fishes): phylogeny, reproductive system, viviparity, spermatozoa. Vol. 8 A: Reproductive biology and phylogeny (B. G. M. Jamieson, ed.), p. 25- 84. Science Pubis., Enfield, NH. Grimes, C. B., K. W. Able, and R. S. Jones. 1986. Tilefish, Lopholatilus chamaeleonticeps , habitat, behavior and community structure in mid-Atlantic and southern New England waters. Environ. Biol. Fishes 15:273-292. Grimes, C. B., K. W. Able, and S. C. Turner. 1980. A preliminary analysis of the tilefish, Lopholati- lus chamaeleonticeps , fishery in the Mid-Atlantic Bight. Mar. Fish. Rev. 42( 1 1 ): 13-18. Grimes, C. B., C. F. Idelberger, K. W. Able, and S. C. Turner. 1988. The reproductive biology of tilefish, Lopholatilus chamaeleonticeps Goode and Bean, from the United States Mid-Atlantic Bight, and the effects of fishing on the breeding system. Fish. Bull. 86:745-762. Grimes, C. B., and S. C. Turner. 1999. The complex life history of tilefish, Lopholatilus chamaeleonticeps, and vulnerability to exploitation. In Am. Fish. Soc. Symp. 23, p. 17-26. Am. Fish. Soc., Bethesday, MD. Heino, M., and U. Dieckmann. 2008. Detecting fisheries-induced life-history evolution: an overview of the reaction-norm approach. Bull. Mar. Sci. 83:69-93. Helle, K., and M. Pennington. 2004. Survey design considerations for estimating the length composition of the commercial catch of some deep-water species in the northeast Atlantic. Fish. Res. 70:55-60. Hoenig, J. M., M. J. Morgan, and C. A. Brown. 1995. Analysing differences between two age determi- nation methods by tests of symmetry. Can. J. Fish. Aquat. Sci. 52:364-368. Hutchings, J. A. 1993. Adaptive life histories effected by age-specific sur- vival and growth rate. Ecology 74:673-684. Idelberger, C. F. 1985. Reproductive biology of the tilefish, Lopholatilus chamaeleonticeps, in the Middle Atlantic Bight. M.S. thesis, 51 p. Rutgers, New Brunswick, NJ. Katz, C. H., C. B. Grimes, and K. W. Able. 1983. Delineation of tilefish, Lopholatilus chamaeleonti- ceps, stocks along the United States east coast and in the Gulf of Mexico. Fish. Bull. 81:41-50. Kitts, A., P. Pinto da Silva, and B. Rountree. 2007. The evolution of collaborative management in the Northeast USA tilefish fishery. Mar. Policy 31:192-200. 174 Fishery Bulletin 111(2) Law, R. 2000. Fishing, selection, and phenotypic evolution. ICES J. Mar. Sci. 57:659-668. Lear, W. H. 1998. History of fisheries in the northwest Atlantic: the 500-year perspective. J. Northwest Atl. Fish. Sci. 23:41-73. Lombardi-Carlson, L. A. 2012. Life history, population dynamics, and fish- ery management of the golden tilefish, Lopholatilus chamaeleonticeps , from the southeast Atlantic and Gulf of Mexico. Ph.D. diss., 150 p. Univ. Florida, Gaines- ville, FL. Marsh, R., B. Petrie, C. R. Weidman, R. R. Dickson, J. W. Loder, C. G. Hannah, K. Frank, and K. E. N. Drinkwater. 1999. The 1882 tilefish kill — a cold event in shelf waters off the north-eastern United States? Fish. Oceanogr. 8:39-49. McBride, R. S., M. J. Wuenschel, P. Nitschke, G. Thornton, and J. R. King. 2013. Latitudinal and stock-specific variation in size- and age-at-maturity of female winter flounder, Pseu- dopleuronectes americanus, as determined with gonad histology. J. Sea Res. 75:41-51. R Core Team. 2012. R: a language and environment for statistical computing. R Foundation for Statistical Computing. Vienna, Austria. (Available from http://www.R-project. org, accessed October 2012.1 Reznick, D. N., H. Bryga, and J. A. Endler. 1990. Experimentally induced life-history evolution in a natural population. Nature 346:357-359. Sadovy de Mitcheson, Y., and M. Liu. 2008. Functional hermaphroditism in teleosts. Fish Fish. 9:1-43. Sadovy, Y., and M. L. Domeier. 2005. Perplexing problems of sexual patterns in the fish genus Paralabrax (Serranidae, Serraninae). J. Zool. 267:121-133. Sissenwine, M. P, and J. G. Shepherd. 1987. An alternative perspective on recruitment over- fishing and biological reference points. Can. J. Fish. Aquat. Sci. 44:913-918. Trippel, E. A. 1995. Age at maturity as a stress indicator in fisheries. Bioscience 45:759-771. 2003. Estimation of male reproductive success of marine fishes. J. Northwest Atl. Fish. Sci. 33:81-113. Trippel, E. A., and H. H. Harvey. 1991. Comparison of methods used to estimate age and length of fishes at sexual maturity using populations of white sucker (Catostomus commersoni). Can. J. Fish. Aquat. Sci. 48:1446-1459. Turner, S. C. 1986. Population dynamics of and the impact of fishing on tilefish, Lopholatilus chamaeleonticeps , in the Mid- dle Atlantic-Southern New England regions during the 1970s and early 1980s. Ph.D. diss., 289 p. Rutgers, New Brunswick, NJ. Turner, S. C., C. B. Grimes, and K. W. Able. 1983. Growth, mortality, and age/size structure of the fisheries for tilefish, Lopholatilus chamaelonticeps, in the Middle Atlantic-southern New England region. Fish. Bull. 81:751-763. Vidal, T. E. 2010. Re-evaluation of age and maturity of golden tile- fish, Lopholatilus chamaeleonticeps , in the Mid-Atlantic Bight. M.S. thesis, 75 p. Univ. Massachusetts, North Dartmouth, MA. Vitale, F., H. Svedang, and M. Cardinale. 2006. Histological analysis invalidates macroscopically determined maturity ogives of the Kattegat cod ( Ga - dus morhua) and suggests new proxies for estimating maturity status of individual fish. ICES J. Mar. Sci. 63:485-492. Wigley, S. E., J. M. Burnett, and P. J. Rago. 1999. An evaluation of maturity estimates derived from two different sampling schemes: are the observed changes fact or artifact? J. Northwest Atl. Fish. Sci. 25:133-140. Worm, B., R. Hilborn, J. K. Baum, T. A. Branch, J. S. Collie, C. Costello, M. J. Fogarty, E. A. Fulton, J. A. Hutchings, S. Jennings, O. P. Jensen, H. K. Lotze, P. M. Mace, T. R. Mc- Clanahan, C. Minto, S. R. Palumbi, A. M. Parma, D. Ricard, A. A. Rosenberg, R. Watson, and D. Zeller. 2009. Rebuilding global fisheries. Science 325:578-585. Wright, P. J., C. P. Millar, and F. M. Gibb. 2011. Intrastock differences in maturation schedules of Atlantic cod, Gadus morhua. ICES J. Mar. Sci. 68:1918-1927. 175 Effects of El Nino-Southern Oscillation events on catches of Bigeye Tuna ( Thunnus obesus ) in the eastern Indian Ocean off Java Email address for contact author: vegha16@gmail.com 1 Laboratory of Marine Bioresource and Environment Sensing Faculty of Fisheries Sciences Hokkaido University Minato-cho 3-1-1, Hakodate Hokkaido 041-8611, Japan 3 Study Program of Geodetic and Geomatics Engineering Faculty of Earth Sciences and Technology Bandung Institute of Technology Jalan Ganesha 10 Bandung 40132, Indonesia Abstract— The effects of El Nino- Southern Oscillation events on catches of Bigeye Tuna (Thun- nus obesus) in the eastern Indian Ocean (EIO) off Java were evaluated through the use of remotely sensed environmental data (sea-surface- height anomaly [SSHAj, sea-surface temperature [SST], and chlorophyll- a concentration), and Bigeye Tuna catch data. Analyses were conducted for the period of 1997-2000, which included the 1997-98 El Nino and 1999-2000 La Nina events. The em- pirical orthogonal function (EOF) was applied to examine oceano- graphic parameters quantitatively. The relationship of those parameters to variations in catch distribution of Bigeye Tuna was explored with a generalized additive model (GAM). The mean hook rate was 0.67 dur- ing El Nino and 0.44 during La Nina, and catches were high where SSHA ranged from -21 to 5 cm, SST ranged from 24°C to 27.5°C, and chlorophyll-a concentrations ranged from 0.04 to 0.16 mg rcr3. The EOF analysis confirmed that the 1997- 98 El Nino affected oceanographic conditions in the EIO off Java. The GAM results indicated that SST was better than the other environmen- tal factors (SSHA and chlorophyll-a concentration) as an oceanographic predictor of Bigeye Tuna catches in the region. According to the GAM predictions, the highest probabilities (70-80%) for Bigeye Tuna catch in 1997-2000 occurred during oceano- graphic conditions during the 1997- 98 El Nino event. Manuscript submitted 4 June 2012. Manuscript accepted 1 March 2013. Fish. Bull 111:175-188 (2013) doi 10.7755/FB. 11 1.2.5 The views and opinions expressed or implied in this article are those of the author (or authors) and do not necessar- ily reflect the position of the National Marine Fisheries Service, NOAA. Mega L. Syamsuddin (contact author)' 2 Sei-lchi Saitoh1 Toru Hirawake' Samsul Bachri3 Agung B. Hart©3 2 Faculty of Fisheries and Marine Sciences Pad|ad|aran University Jalan Raya Bandung-Sumedang KM 21 Jatinangor, Bandung 40600, Indonesia The El Nino-Southern Oscillation (ENSO) is a large-scale pattern of climate fluctuation that strongly in- fluences much of the globe. In the Pacific, the ENSO cycle causes warm phases (El Nino) and cool phases (La Nina) that have been shown to affect catches in tuna fisheries (Lehodey et al., 1997; Lehodey, 2001). The ap- proximate onset of the 1997-98 El Nino event occurred during March- April 1997, and the mature phase in November-December 1997 (Enfield, 2001). The El Nino event ended in May 1998 and a cold La Nina was es- tablished in the eastern Pacific. The 1997-98 El Nino was the strongest on record and affected the climate in many parts of the world (McPhaden, 1999). Catches of tunas around the world are affected by ENSO events (Howell and Kobayashi, 2006; Le- hodey et al., 2010). Therefore, for sustainable management of Bigeye Tuna ( Thunnus obesus ) resources in the eastern Indian Ocean (EIO) off Java, one of the main islands in In- donesia, understanding the effects of ocean climate variability on catch distribution is essential. The Bigeye Tuna is a productive tropical species that accounts for more than 10% of the total catch of market tuna species worldwide (Miyake et al., 2010). Bigeye Tuna is a commer- cially targeted species and represents one of the most valuable species of longlme fisheries in the EIO off Java (ISSF1). It is a highly migratory spe- cies that is distributed between 40°N and 40°S in all 3 major oceans, except in the southwestern sector of the At- lantic (Hanamoto, 1987). Bigeye Tuna generally favor water temperatures between 17°C and 22°C. They prefer to stay near, and usually below, the 1 ISSF (International Seafood Sustain- ability Foundation). 2012. ISSF stock status ratings — 2012: status of the world fisheries for tuna. ISSF Tech. Rep. 2012-04, 88 p. [Available from http://iss-foundation.org/resources/ downloads/?did=328, accessed November 2012.] 176 Fishery Bulletin 111(2) thermocline and come to the surface periodically (Pep- pered, 2010). The main depth range of fishing for Big- eye Tuna in the Indian Ocean is 161—280 m (Mohri and Nishida, 1999), although they can inhabit the depth range of 0-100 m during the night (Howell et al., 2010). Sea-surface-height can be used to infer oceanic fea- tures such as current dynamics, fronts, eddies, and con- vergences (Polovina and Howell, 2005), and sea-surface temperature (SST) has been used to investigate pro- ductive frontal zones (Zainnudin et ah, 2004), both of which can be used to indicate potential tuna fishing grounds. Thermal (or color) gradients in satellite im- ages that arise from the circulation of water masses often indicate areas of high productivity (Saitoh et al., 2009). Chlorophyll-a data can also be used as a valu- able indicator of water mass boundaries and may iden- tify upwelling that can influence tuna distribution in a region. The effects of ENSO on oceanographic conditions and tuna catches in the Pacific have been reported widely (Lehodey et al., 1997; Torres-Orozco et al., 2006; Briand et al., 2011). High catch rates of Albacore ( Thunnus alalunga ) in the southwest Pacific Ocean were found to correspond with high negative Southern Oscillation Index values during strong El Nino events (Briand et al., 2011). The effects of ENSO events on Bigeye Tuna catches have been well studied in the western Pacific Ocean (Miller, 2007) but less studied in the Indian Ocean. Most Indian Ocean studies have focused on the relationship between oceanographic parameters and the distribution of Bigeye Tuna (Mohri and Nishida, 1999; Song et al., 2009, Song and Zhou, 2010), the cor- relation of a single oceanographic factor with ENSO (Yoder and Kennely, 2003), or oceanographic variabil- ity in the interior Indonesian seas (Zhou et al., 2008; Sprintall et al., 2009). Here, we focus on the ways in which climate vari- ability affects oceanographic conditions and catch rates of Bigeye Tuna in the EIO off Java. To obtain a more detailed description of the spatiotemporal charac- teristics of those oceanographic param- eters, we applied the empirical orthogo- nal function (EOF). Further analysis was undertaken with a generalized additive model (GAM) to examine the relationship between oceanographic conditions and catch rates of Bigeye Tuna. The ultimate goal of this study was to understand how catch rates of Bigeye Tuna in the EIO off Java are affected by ENSO events. 90°E 1 00°E 1 1 0°E 1 20°E 130°E 14Q°E 150°E 0 220 440 km Figure 1 (A) Map of the Indonesian seas, with the inset box representing the study area. (B) Map of the study area in the eastern Indian Ocean (EIO) off Java for our analyses of how E! Nino-Southern Oscillation events may affect catch rates of Bigeye Tuna ( Thunnus obesus). In panel B, the wave and current systems in the EIO off Java are indicated by the dotted line for the South Java Current (SJC), solid lines for the Indonesian Throughflow (ITF), the line with dashes and 2 dots for the Indian Ocean Kelvin Waves (IOKWs), the line with dashes and 1 dot for the Rossby Waves (RWs), and the dashed line for the Indian Ocean South Equatorial Current (SEC). Materials and methods Study area The study area was located in the EIO, south of Java, spanning be- tween 6-16°S and 104— 126°E (Fig. IB). This area has complex dynamic currents and wave systems. The dominant current and wave features include 1) the Indo- nesian Throughflow (ITF), outflow wa- ter from the Pacific to the Indian Ocean (Molcard et al., 2001; Gordon et al., 2010); 2) the seasonally reversing South Java Current (SJC) along the southern coast of the Indonesian Sea (Sprintall et al., 2010); 3) the Indian Ocean South Equatorial Current (SEC) that flows from the southern Indian Ocean to an area off southern Java (Zhou et al., 2008); 4) downwelling Indian Ocean Kelvin Waves (IOKWs) that propagate to the east along the coasts of west Sumatra, Java, and the lesser Sunda islands (Syamsudin et al., Syamsuddin et al.: Effects of El Nino- Southern Oscillation on catches of Thunnus obesus in the eastern Indian Ocean 177 2004); and 5) westward Rossby Waves propagation at 12-15°S (Gordon, 2005; Sprintall et al., 2009). Besides these current and wave systems, winds over the Indonesian maritime continent and the position of the Intertropical Convergence Zone are dominant fea- tures of strong monsoon signatures. During the south- east monsoon (May to October), southeasterly winds from Australia generate upwelling along the southern coasts of Java and Bali. These conditions are reversed during the northwest monsoon (November to April) (Gordon, 2005). Data For our study, we used fishery catch data and satel- lite remotely sensed data. Bigeye Tuna catch data and remotely sensed environmental data for the period of 1997-2000 were analyzed. These data included the ENSO components of an El Nino event (April 1997- May 1998) and a La Nina event (July 1998-June 2000). Fisheries data sets Catch data for Bigeye Tuna were obtained from longline fishing logbooks provided by PT Perikanan Nusantara,2 an incorporated company of the Indonesian government, at Benoa, Bali. Data included fishing position (latitude and longitude), operational days, fish weight (in kilograms), vessel number, num- ber of hooks, and the number of fish caught per month during 1997-2000. The fishing locations recorded in the logbook were only the fishing positions where Bigeye Tuna were caught (there were no data for the locations where no fish were caught). These data were compiled into grids of 1° latitudexl0 longitude because catch data for Bigeye Tuna were available only at a resolu- tion of 1°. From this data set, the catch rate of Bigeye Tuna was expressed as a percentage of hook rate (HR). The HR was calculated as the number of fish caught (individuals/month) per 100 hooks. The HR, therefore, shows how many tuna were hooked per unit of 100 longline hooks, and the HR can be referred to as catch per unit of effort. The majority of fishing operations were conducted by medium-size vessels (100 gross tonnage). The num- ber of vessels in operation was 19-20 per month, and vessels used the same fishing gear (longline sets) and similar fishing techniques. The number of fishing sets was within the range of 910-1607 per year. The long- line sets were specifically designed and constructed to reach the swimming depths of the Bigeye Tuna. PT Perikanan Nusantara3 used 2 types of longline sets constrained by the fishing depth during operation: 1) a shallow set (depth <100 m; consisting of 4-6 branch lines between floats) and 2) a deep set (depth 100-300 2 Mention of trade names or commercial companies is for iden- tification purposes only and does not imply endorsement by the National Marine Fisheries Service, NOAA. 3 Perikanan Nusantara, Inc. 2001. Vessels operation data of Incorporated Company of Perikanan Nusantara, 78 p. Peri- kanan Nusantara, Jakarta, Indonesia. m; consisting of 10-14 branch lines between floats). The deep set was used to catch Bigeye Tuna, and the shallow set was better suited to catchYellowfin Tuna ( Thunnus albacares). The longline fishery targeted Big- eye Tuna at operational depths of 109—288 m during night sets. Each deep set consisted of 10-14 branch lines between floats and 800-1600 hooks per set. The fishing ground covered an area located around 10-16°S and 108-120°E; fishing operations were limited to 15 days per trip because of fuel costs and the need to keep caught fish fresh (Perikanan Nusantara3). Bigeye and Yellowfin Tunas were distinguished by various characteristics, including the following fea- tures: the Bigeye Tuna is longer, has a large head, large eyes, a dusky-colored tail, yellowish finlets edged in black, and a tail with a flat, trailing edge. The Yel- lowfin Tuna is shorter, has a smaller head, round and small eyes, a narrow body, a yellowish tail, and a notch in the center of its tail (Itano4). Remotely sensed data Remotely derived environmen- tal variables included the sea-surface-height anomaly (SSHA), SST, and chlorophyll-a concentration. The SSHA data with a spatial resolution of 1/3°, derived from the TOPEX/Poseidon and ERS-1/2 altimeter mea- surements, were produced and distributed by Archiving, Validation and Interpretation of Satellite Oceano- graphic Data (AVISO, http://www.aviso.oceanobs.com). We obtained 7-day composite cycles of SSHA products to calculate the monthly mean SSHA. SST data were derived from the Advanced Very High Resolution Radi- ometer sensor on board NOAA satellites. This data set is distributed by the Physical Oceanography Distrib- uted Active Archive Center (http://podaac.jpl.nasa.gov) of the Jet Propulsion Laboratory of the National Aero- nautics and Space Administration (NASA). We used the monthly mean SST data set at a pixel resolution of 4x4 km. Chlorophyll-a data were derived from images obtained from the Sea-viewing Wide Field-of-view Sensor (SeaWiFS) Project (level 3) and were of a spatial resolution of 9x9 km for the period from September 1997 to December 2000 (monthly composite data were downloaded from http://oceancolor.gsfc.nasa.gov). These data were pro- cessed with the SeaWiFS Data Analysis System vir- tual appliance (SeaDAS VA, vers. 6.1) of NASA (http:// seadas.gsfc.nasa.gov/seadasva.html). SSHA and SST images were matched with the 9-km- resolution spatial scale for chlorophyll-a concentra- tions. The 9-km-resolution data were used to capture dynamic features of the oceanographic conditions that represented El Nino and La Nina events and to show spatial patterns for the EOF analysis. However, for the 4 Itano, D. G. 2005. A handbook for identification of yel- lowfin tuna and bigeye tuna in fresh condition, vers. 2, 27 p. Pelagic Fisheries Research Program, Univ. Hawaii, Ho- nolulu. lAvailable from ftp://ftp.soest.hawaii.edu/PFRP/ itano/1 _ BE -YF%20ID%20Fresh_ ENGL ISH_v2 Iogo.pdf, ac- cessed November 2012.] 178 Fishery Bulletin 111(2) GAM input, SSHA, SST, and chlorophyll-a data were calculated on a spatial grid of l°xl° to match with the spatial resolution of the fisheries data. We resampled the remotely sensed data to resolutions of 9 km and 1° through the use of geographic information system tools, including Generic Mapping Tools (GMT, vers. 4.5.7 [Wessel and Smith, 1998]), with the nearest-neighbor technique. Nearest-neighbor assignment can be applied with the resample function as a preprocessing step be- fore combination of raster data of different resolutions. This assignment does not change any of the values of cells from the input raster data sets; the cell center from the input raster that is closest to the cell center for the output processing is used. Nino 3.4 index The Nino 3.4 index was used as a cli- matic index of ENSO indicators based on SST. The in- dex is the average SST anomaly in the region bounded by 5°N to 5°S and 120-170°W. The Nino 3.4 index was downloaded from the NOAA Climate Prediction Cen- ter (http://www.cpc.ncep.noaa.gov). El Nino and La Nina events were identified if the 5-month running average of the Nino 3.4 index exceeded +0.5°C for El Nino or -0.5°C for La Nina for at least 5 consecutive months (this index is shown as the dashed line in Fig. 2A). Catchability coefficient Catchability was defined as the proportion of available fish in the population that would be caught by a unit of fishing effort. Catchability depends on the distribu- tion of fishing effort in relation to the distribution of the target species (Ellis and Wang, 2007). The catch- ability coefficient is defined as the proportion of the total stock taken by one unit of effort (Haddon, 2011) and is expressed as C / (q E) = B, (1) where C = catch; q = the catchability coefficient; E = the amount of fishing effort; and B - stock biomass. The unit of effort used for calculation of longline catch rates is the number of hooks, as a result of changes in gear practices (Ward and Meyers, 2004). We computed the catchability coefficient for all months during 1997-2000. Empirical orthogonal function We applied the EOF as a statistical method to quan- titatively examine oceanographic parameters. EOF analysis is a useful technique for decomposition of a time series of geophysical data into temporal and spa- tial variability in terms of orthogonal functions or sta- tistical modes. EOF analysis has been used commonly to describe spatiotemporal ocean variability (Yoder and Kennely, 2003; Otero and Siegel, 2004; Radiarta and Saitoh, 2008; Tolan and Fisher, 2009). EOF analysis was applied to the raw weekly data set of SSHA and SST and monthly data of chiorophyll- a concentrations because of a lack of data in many of the weekly images. Here, we examined only the first and second dominant modes, which were statistically independent and significant. A more comprehensive ex- planation of the concept of EOF analysis has been pro- vided by Bjornsson and Venegas (1997). We constructed the EOF analysis using Matlab, vers. 7.1, software (The MathWorks, Inc., Natick, MA). On the basis of the re- sults of the EOF, we generated maps of patterns of spa- tial and temporal ocean variability in the EIO off Java with ArcGIS tools (Esri, Redlands, CA). Generalized additive model The GAM was first proposed by Hastie and Tibshirani (1990). The advantage of this model is that the predic- tor variables have nonlinear effects upon the response variable. We applied the binomial GAM to analyze positive catches, interpreted as presence (1), and null catches, interpreted as absence (0) (Fraile et al., 2010), of Bigeye Tuna and to determine the catch probability of Bigeye Tuna in the EIO off Java. The Bigeye Tuna catch data presented here are based only on the grids fished because, as mentioned previously, catch infor- mation for Bigeye Tuna was not available for unfished grids; therefore, we ignored the unfished strata (Wal- ters, 2003). Some sampling bias may have affected our results regarding catch variability. However, the catch data are nevertheless useful for examination of Big- eye Tuna variability in the study area, and the HR data based on those catch data may be indicative of relative changes in availability. We analyzed the pres- ence of Bigeye Tuna through exploration of the spatial trends in their catch distribution that were influenced by SSHA, SST, and chlorophyll-a concentrations. All explanatory model terms were treated as continuous variables and the spline smoothers were fitted initially to each term in the model (Zuur et al., 2009). A step- wise GAM was performed to determine the best-fitting model before application of the final GAM to the entire data set. Akaike’s information criteria (AIC) were used to determine the optimal set of explanatory variables. The model with the smallest AIC can be selected as the optimal model. GAMs were constructed in R software (vers. 2.14.0; R Development Core Team, 2011) with the gam function of the mgcv package (Wood, 2006). The GAMs were fitted in the form g(ut) = a0 + SjUjj) + s2(x2i) + s3(x3i) + sn0eni), (2) where g = the link function; ut = the expected value of the dependent variable (Bigeye Tuna catch); a0 = the model constant; and Syamsuddin et al Effects of El Nino- Southern Oscillation on catches of Thunnus obesus in the eastern Indian Ocean 179 sn = a smoothing function for each of the model covari- ates xn. The constructed GAM could then be used to predict the Bigeye Tuna catch probability with the predict, gam function in the mgcv package in R (Wood, 2006). The Bigeye Tuna catch probability is the relative prob- ability of catching one or more Bigeye Tuna at a location given the oceano- graphic conditions in the area (Teo and Block, 2010). The prediction maps were produced with the best model selected from a set of 7 models. The distribution of the GAM-predictions of Bigeye Tuna catch was compared with monthly fishery data. With GMT tools, we made monthly predictions of Big- eye Tuna catch probabilities. Results Catch rates of Bigeye Tuna During the El Nino event in 1997-98, catch rates of Bigeye Tuna peaked in May-July 1997 (0.87-0.94 HR) and were still high in February^July 1998 (0.69-0.90 HR), followed by a continu- ous decline in the later months of 1998 (Fig. 2A). During the La Nina event in 1999-2000, catch rates were reduced throughout the year (HR <0.67). The average HR was higher during the 1997-98 El Nino event (0.67) than during the 1999-2000 La Nina event (0.44). The number of hooks deployed ranged from 14,392 to 178,109 per month (Fig. 2B), and more hooks were used during the 1997-98 El Nino event (147,676 hooks) than during the 1999-2000 La Nina event (144,283 hooks). In June 1997, 178, 109 hooks (maximum number) were deployed and in November 2000, 14,392 hooks (minimum number) were deployed. The catchability coefficient (q) varied with the number of hooks (Fig. 2B). Catchability was high in May 1997 (1.56xl0-7), June 1998 (1.47xl()-7), and November 2000 (2.73xl0~7). Catch rates of Bigeye Tuna showed seasonal variations and were higher during the southeast monsoon (May- October; SE) than during the north- El Nino (>0.5°C) La Nina (>-0.5°C) -1 -2 Jan97 Jan98 0.8 0 6 0.4 0.2 0.0 I I Southeast (May-Oct) □ Northwest (Nov-Apr) 1997 1998 1999 2000 Year Figure 2 (A) Variability in catch rates of Bigeye Tuna ( Thunnus obesus) in the east- ern Indian Ocean off Java by hook rate (HR) percentage (solid line) and SST anomalies from the Nino 3.4 index during 1997-2000 (dashed line). El Nino and La Nina events were identified when the average of the Nino 3.4 index exceeded +0.5°C for El Nino or -0.5°C for La Nina for at least 5 consecutive months. The El Nino event is indicated by a black bar on the top of this graph, and the La Nina event is indicated by a gray bar on the top. (B) The total number of hooks deployed (gray bars) and time series variation of the catchabillity coefficient (solid line) during 1997-2000. (C) Seasonal variations in Bigeye Tuna HR in 1997-2000. The gray bar repre- sents the southeast monsoon ( May-October), and the white bar represents the northwest monsoon (November-April). 180 Fishery Bulletin 111(2) -10°S -15°S -10°S -1 5°S -10°S -1 5°S 1 05°E 1 1 0°E 1 15°E 1 20°E 125°E 105°E 110°E 115°E 120°E 125°E Figure 3 On the left, images of mean values for 3 months obtained from remotely sensed data of the eastern Indian Ocean off Java during El Nino event (September, October, and November 1997) for 3 parame- ters— (A) sea-surface-height anomaly (SSHA), (B) sea-surface temperature (SST), and (C) chlorophyil- o (chl-a) concentration. On the right, images of mean values for 3 months obtained during La Nina event (February, March, and April 1999) for the same 3 parameters: (D) SSHA, (E) SST, and (F) chl-a concentrations. Scale units are in centimeters, degrees Celcius, and milligrams per cubic meter for SSHA, SST, and chl-a data, respectively. west monsoon (November-April; NW) for all years ex- cept 1999 (Fig. 2C). High HRs (average 0.55 HR) were recorded during the southeast monsoon, and lower HR (average 0.45 HR) occurred during the northwest monsoon. Features of oceanographic conditions The following results are presented to show how the different phases of ENSO affected oceanographic condi- tions in the EIO off Java. Figure 3 shows 6 snapshots of SSHA, SST, and chlorophyli-a conditions, each show- ing the means of values from 3 different months dur- ing different phases of El Nino (mean of September, October, and November 1997) and La Nina (mean of February, March, and April 1999) in the EIO off Java. The months that represent the El Nino and La Nina events were chosen on the basis of the 3 consecutive months with the highest individual values in the Nino 3.4 index during El Nino and La Nina during 1997- 2000. There were contrasting conditions of SSHA and SST along the coasts and in offshore areas during the El Nino and La Nina events. During the El Nino event, SSHA showed negative values (from -1 to -20 cm) along coastal regions of the EIO off Java (Fig. 3A). In contrast, offshore regions in- fluenced by frontal areas between 10-12°S and 15°S exhibited positive SSHA (4 cm). Lower SST (24-27°C) occurred in the coastal to offshore regions of the EIO off Java (Fig. 3B), and warmer waters (29°C) appeared in the eastern part of the EIO off Java. Conversely, during the La Nina event, positive SSHA (1-13 cm) occurred along the coasts, and negative SSHA (—5 to -3 cm) were found offshore between 12° and 16°S (Fig. 3D). SST during the La Nina event both near the coasts and in offshore areas of the EIO off Java (27-29°C) were higher than values during the El Nino event in this region (Fig. 3E). Chlorophyll-a concentrations were higher during the El Nino event (0. 2-2.0 mg nr3; mean of September, October, and November 1997; Fig. 30 than they were during the La Nina event (0.05-0.10 mg nr3; mean of February, March, and April 1999; Fig. 3F). The highest concentration of 2 mg m-3 was detected during the El Nino event along the coast of the EIO off Java, and values decreased toward offshore (0.01-0.20 mg nr3) around the frontal area at 10-12°S (Fig. 30. This vari- ation in chlorophyll-a values coincides well with the occurrence of lower SST in the same areas during the El Nino event (Fig. 3B). Syamsuddm et al Effects of El Nino-Southern Oscillation on catches of Thunnus obesus in the eastern Indian Ocean 181 co CO T. > cT 3 CO CO H o 1 05°E 1 1 0°E 1 1 5°E 120°E 125°E 105°E 110°E 1 1 5°E 120°E 125°E Figure 4 Spatial patterns of the empirical orthogonal function modes in the eastern Indian Ocean off Java from September 1997 to December 2000: (A) first mode of sea-surface-height anomaly (SSHA), (B) second mode of SSHA, (C) first mode of sea-surface temperature (SST), (D) second mode of SST, (E) first mode of chlorophyll-a (chl-a) concentrations, and (F) second mode of chl-a concentrations. Scale units are in centimeters, degrees Celcius, and milligrams per cubic meter for SSHA, SST, and chl-a data, respectively. Spatial and temporal modes of ocean variability The remote forcing in the timescale of interannual and seasonal variations due to ENSO and monsoons are represented by the first and second EOF modes of SSHA, SST, and chlorophyll-a concentrations, with the total energy variance from these 2 modes about 78.60% for chlorophyll-a values, 71.39% for SSHA, and 34.53% for SST (Table 1). The first spatial mode of SSHA con- tributed 48.82% of the total variance, indicating waters with relatively low temperatures and with negative SSHA concentrated along the southern coast of Java (Fig. 4A). The second spatial mode of SSHA contrib- uted 22.57% of the total variance and showed that wa- ters with relatively low temperatures were distributed along the southern coast of the Indonesian archipelago (Fig. 4B). The first spatial mode of SSHA corresponded with interannual variability, as shown by the first tem- poral mode (Fig. 5A). A map of the second spatial mode of SSHA clearly shows that this mode corresponded with seasonal variability, as indicated by the second temporal mode (Fig. 5B). The SSHA results were associated with the first and second spatial modes of SST, with total variances of 25.93% and 8.60%, respectively. The first spatial mode of SST showed that waters with relatively low tem- peratures were concentrated along the southern coast of Java, extending to the offshore area (7-13°S) (Fig. Table 1 Summary of the amplitude function (percentage of the total variability) of the first and second empirical orthogonal function modes of the sea-surface-height anomaly (SSHA), sea-surface temperature (SST), and chlorophyll-a (chl-a) concentrations in the eastern In- dian Ocean off Java from September 1997 to December 2000. Total modes Parameter Mode 1 (%) Mode 2 (%) 1 and 2 (%) SSHA 48.82 22.57 71.39 SST 25.93 8.60 34.53 Chl-a 63.39 15.21 78.60 4C). The second spatial mode of SST showed that those waters were spread along the western coast of Java (7- 12°S), while waters with relatively high temperatures covered the eastern part of the EIO off Java (Fig. 4D). The amplitude function of the first mode of SST corre- sponded with interannual variability (Fig. 50, and the second mode corresponded with the seasonal cycle in which the maximum and minimum SST occurred during the northwest monsoon (November-April) and southeast monsoon (May— October), respectively (Fig. 5D). 182 Fishery Bulletin 111(2) 20 -20 50 -50 150 -150 75 A Mode 1 SSHA 1 1 i i i Q. E < -75 100 E 6 r y — \ Mode 2 SSHA C._J, 1 Mode 1 SST D 1 1 Mode 2 SST Mode 1 Chl-a -100 40 -40 Mode 2 Chl-a Jan98 Jul98 Jan99 Jul99 JanOO JulOO Year Figure 5 Amplitude function from the temporal mode of the analysis for the study area in the eastern Indian Ocean off Java from Sep- tember 1997 to December 2000: (A) first mode of sea-surface- height anomaly (SSHA) (interannual signal), (B) second mode of SSHA (seasonal signal), (C) first mode of sea-surface tempera- ture (SST) (interannual signal), (D) second mode of SST (season- al signal), (E) first mode of chlorophyll-a (chl-a) concentrations (interannual signal), and (F) second mode of chl-a concentrations (annual signal). The x-axis represents the year, and the y-axis shows the amplitude function for each mode ( nondimensional ). The first EOF mode of SSHA and SST showed an inverse relationship with the first EOF mode of chlorophyll-a, with a negative SSHA value and rela- tively low SST followed by higher chlorophyll-a con- centrations along the southern coast of Java. The first mode of chlorophyll-a contained 63.39% of the energy variance, and the second mode contributed 15.21% of the energy variance, with notably higher chlorophyll-a levels concentrated along the southern coast of Java (7-9°S). The ampli- tude function of the first mode of chlorophyll- a corresponded with interannual variability, and the second mode corresponded with the annual cycle. Positive values (chlorophyll-a concentrations greatly elevated above values seen in other periods) occurred during Sep- tember-November 1997 (Fig. 5, E-F). Generalized additive models The results of the GAMs are presented as 1-parameter, 2-parameter, and 3-parameter models (Table 2). All of the variables used were statistically highly significant (PcO.OOOl) for SSHA, SST, and chlorophyll-a concentrations. The addition of predictor variables at different levels resulted in an increase in the deviance in catch rates explained. In the 1-parameter models, SST explained the highest deviance (6.48%) and chlorophyll-a concentrations ex- plained the lowest deviance (2.04%). The 3-pa- rameter combination models explained the highest deviance (16.30%) and had the lowest AIC values. GAM plots can be interpreted as the indi- vidual effects of predictor variables associated with SSHA, SST, and chlorophyll-a concentra- tions on Bigeye Tuna catch (Fig. 6, A-C). High probabilities of Bigeye Tuna presence were observed for SSHA ranging from -21 to 5 cm, for SST ranging from 24° to 27.5°C, and for chlorophyll-a levels ranging from 0.04 to 0.16 mg m~3. Negative effects on Bigeye Tuna were observed for SSHA >5 cm, SST values >27.5°C, and chlorophyll-a values of 0.01-0.03 mg mr3 and >0.16 mg m 3. Spatial predictions for catch distribution of Bigeye Tuna were compared with the actual monthly fishery data collected during the El Nino (September and October 1997) and La Nina (March and April 1999) events. The predicted catch distribution of Bigeye Tuna in September 1997 during the El Nino event indicated a po- tential area with higher catch probability of ap- proximately 70-80% at 10-16°S and 104-122°E, and the actual Bigeye Tuna fishing locations (with a HR of 0.41) occurred in the area between 12-16°S and 110-115°E (Fig. 7A). In October 1997, spatial predictions for catch distribution of Bigeye Tuna indicated locations with higher catch probability (60-70%) in the west at 7-12°S, 104-108°E and 14-16°S, 109-1 14°E, but the actual Bigeye Tuna fishing locations were located at 12-15°S, 110-116°E, where there was a predicted catch probability of around 20-40% and actual HR of only 0.20 (Fig. 7B). The Syamsuddin et at: Effects of El Nino- Southern Oscillation on catches of Thunnus obesus in the eastern Indian Ocean 183 Table 2 Results from general additive models (GAMs) derived from catch rates of Big- eye Tuna (Thunnus obesus) in the eastern Indian Ocean off Java in 1997- 2000 as a function of the oceanographic parameters (Af=2843 samples). The best model was selected on the basis of the significance of predictor terms, reduction of Akaike’s information criterion (AIC), and increase in cumulative deviance explained (CDE). SST=sea-surface temperature; SSHA=sea-surface- height anomaly; chl-a=chlorophyll-a concentrations estimated from SeaWiFS level-3 images (http://oceancolor.gsfc.nasa.gov). Model Variable P value AIC CDE (%) SST SST <2.00 x 10-16*** 2117.04 6.48 SSHA SSHA <2.00 x 10-16*** 2153.88 4.95 Chl-a Chl-a <2.00 x IQ-16*** 2219.38 2.04 SSHA + SSHA <2.00 x 10-16*** 2110.55 7.18 chl-a Chl-a <2.00 x p)-16*** SST + SST <2.00 x IQ-16*** 2065.54 9.20 chl-a Chl-a <2.00 x 10-16*** SSHA + SSHA <2.00 x 10-16*** 1973.96 13.30 SST SST <2.00 x 10-16*** SSHA + SSHA <2.00 x 10-16*** 1912.79 16.30 SST + SST <2.00 x 10-16*** chl-a Chl-a <2.00 x 10-ie*** ***indicates statistical significance at the 0.001 level. suspected shift in the distribution of Bigeye Tuna away from the fishing grounds caused the drop in the HR from 0.41 in September 1997 to 0.20 in October 1997. During La Nina in 1999, spatial predictions indi- cated Bigeye Tuna catches with lower probabilities of 20-40% occurring in the offshore area of the western part of the EIO off Java. In March 1999, the spatial prediction of Bigeye Tuna catch (20-30%) was located around 12-16°S, 104-115°E, but actual Bigeye Tuna fishing locations were at 13-15° S, 106-107° E; 12— 15°S, 109-112°E; and 12-13°S, 114-115°E (Fig. 70. The spatial prediction of Bigeye Tuna catch (20-40%) moved to the west at 11-16° S and 104-1 11°E in April 1999 (Fig. 7D), but the actual Bigeye Tuna catch areas were located at 13-14°S and 11-15°S in the longitude range of 108-1 18°E. Discussion Catch rates of Bigeye Tuna varied over a range of time scales and apparently in relation to environmental changes. Changes in oceanographic conditions during ENSO events resulted in perceivable variations in Big- eye Tuna catches, with an average HR of 0.67 during the 1997-98 El Nino event. The 1999-2000 La Nina event, with an average HR of 0.44, was less favorable for catches. The spatial patterns of the first and second EOF modes for SSHA, SST, and chlorophyll-a concentra- tion gave typical negative SSHA, low SST, and high chlorophyll-a concentration along the southern coast of the Indonesian archipelago, and changes in these pat- terns could be exposed by the temporal mode as inter- annual variation related to the forcing of the 1997-98 El Nino event and upwelling evidence. Those typical spatial patterns are consistent with the oceanograph- ic conditions during September-November 1997 (Fig. 3, A— C). The first and second modes of chlorophyll-a showed the characteristics of an upwelling pattern, with chlorophyll-a concentrations higher along the southern coast of Java than in other areas. Although it normally ends in October, upwelling was observed into November during the southeast monsoon in the El Nino event in 1997. Our results are consistent with the findings of Sprintall et al. (1999) and Ffield et al. (2000), who reported that the ITF brings colder and warmer wa- ters to the Indian Ocean during El Nino and La Nina events, respectively, and the local, alongshore winds south of Java are favorable for upwelling through De- cember. They also confirmed that the upwelling signal could account for the reduced downwelling signal from the November Kelvin waves. In November 1997, Kelvin waves were not generated in the region and this condi- tion caused the persistence of colder water along the southern coast of Java. Chlorophyll-a concentrations contributed the high- est energy variance, indicating that chlorophyll-a con- centration was the main indicator of the forcing mecha- nisms responsible for the 1997-98 El Nino event. Our 184 Fishery Bulletin 111(2) -20 -10 0 10 20 24 25 26 27 28 29 SST 0.00 0.10 0.20 0.30 Chi-a Figure 6 Effect of the 3 oceanographic variables — (A) sea- surface-height anomaly (SSHA), (B) sea-surface temperature (SST), and (C) chlorophyll-a (chl-a) concentrations — on Bigeye Tuna ( Thunnus obe- sus) catches derived from generalized additive model (GAM) of presence-absence data. The r-axis shows the values of the explanatory variables, and the y-axis shows the results of smoothing the fit- ted values. The tick marks on the horizontal axis represent the values of the observed data points; the solid line indicates the fitted function. Dashed lines represent 95% confidence intervals. The hori- zontal line at zero indicates no effect. The percent frequency of occurrence was higher for all values for which the fitted GAM function was above the zero axis and lower for values <0. results are consistent with the work of Murtugudde et al. (1999), who showed that, in the Indian Ocean, intense El Nino events, such as the one in 1997-98, have direct effects on primary production and cause anomalous high values of chlorophyll-a concentration ob- served in the EIO. Upwelling areas are potential conver- gence zones for plankton aggregation, attracting larger predators, such as tunas (Lehodey et ah, 1997). Such concentrations of chlorophyll-a may cause the increased catches during El Nino event (Polovina et ah, 2001; Le- hodey et ah, 2003; Polovina et ah, 2004; Miller, 2007). We used a binomial GAM to investigate the effects of environmental variables that affect the catchability of Bigeye Tuna. The effects of oceanographic conditions inferred from the GAM indicated that oceanographic fac- tors strongly influence the catchability of Bigeye Tuna. SST was a more important oceanographic predictor of Bigeye Tuna catches than were the other environmental variables (SSHA and chlorophyll-a) in this region. Fur- thermore, this result from GAM analyses of SST indicates that remote forcing from the Pacific Ocean has a large ef- fect on HR during an El Nino because of the reduction in heat transported from the Pacific to the Indian Ocean by the Indonesian Throughflow during El Nino events (Ffield et ah, 2000; Gordon et al., 2010). Bigeye Tuna are very sensitive to changes in SST (Holland et al., 1992; Brill et ah, 2005). Our results indicate that Bigeye Tuna catches increased in areas with relatively low temperatures (24- 27.5°C) and decreased at temperatures >27.5°C (Fig. 6B). Our results are supported by previous research from the North Pacific Ocean, in which SST had the greatest effect on Bigeye Tuna at temperatures of 23-26. 5°C (Howell et al., 2010). SSHA was the second-most significant oceanographic predictor of Bigeye Tuna catch distribution in the EIO off Java. We used SSHA to understand oceanic variabil- ity, such as current dynamics, eddies, convergences, and divergences, which could be used as proxies for the poten- tial location of tuna catches (Polovina and Howell, 2005). Our study showed that the preferred habitat for Bigeye Tuna was in the range of SSHA values of -21 to 5 cm (Fig. 6A). This finding indicates that Bigeye Tuna forage in areas of low and negative SSHA values in contrast to divergences in SSHA values. Howell and Kobayashi (2006) also found the presence of a strong gradient of sea- surface height in the region of Palmyra Atoll during the 1997-98 El Nino, coinciding with an increase in the geo- strophic (subsurface) flow that may increase shoaling of longline sets. Negative SSHA would push the thermocline upward, nearer the surface, and the elevation of the ther- mocline would allow Bigeye Tuna from below to become more accessible to longline gear. This preferred condition may enhance the potential Bigeye Tuna habitat, as it ap- parently did during the 1997-98 El Nino, when increased Bigeye Tuna catches occurred. Our findings seem to agree with the results of Holland et ai. (1992) and Brill (1994), who reported that Bigeye Tuna move toward cooler habi- tats to prevent overheating, with negative values of SSHA indicating that Bigeye Tuna are attracted only to shallow water when the thermocline is closer to the surface (Ar- rizabalaga et ah, 2008). Among the 3 environmental predictors assimilated in the model, chlorophyll-a concentrations exhibited the Syamsuddm et at: Effects of El Nino-Southern Oscillation on catches of Thunnus obesus in the eastern Indian Ocean 185 105°E 1 1 0°E 1 1 5°E 1 20°E 125°E 1 05°E 1 1 0°E 1 1 5°E 120°E 125°E % Probability 0 20 40 60 80 100 Figure 7 Spatial prediction for catch probabilities of Bigeye Tuna ( Thunnus obesus) overlaid with actual fishing locations of Bigeye Tuna catch in the eastern Indian Ocean off Java for the El Nino event in (A) Sep- tember 1997 and (B) October 1997 and for the La Nina event in (C) March 1999 and (D) April 1999. Color bars show the level of predicted catch probability of one or more Bigeye Tuna catches (0-100%); blue indicates the lowest catch probability (0), and red indicates the highest catch probability (100%). Circles outlined in black show the actual fishing locations for Bigeye Tuna catch (1° intervals). The original data set was gridded in 1° increments and smoothed for the purposes of better visualization. lowest contribution to the model prediction. However, the derived relationship between this parameter and catches of Bigeye Tuna was statistically significant (PcO.OOOl). Bigeye Tuna fishing sets were located in waters with relatively low-to-moderate chlorophyll- a values. Chlorophyll-a data is a valuable proxy for water mass boundaries and upwelling events. Overall, the GAM results showed that the distribution of Big- eye Tuna catch in the study area was influenced pri- marily by SST and SSHA. The lag time in food chain processes may explain the rather weak effect of chlo- rophyll-a concentration on HR. Bigeye Tuna catchability could be influenced by many factors, in addition to oceanographic parame- ters, such as the depth range of longline sets, dura- tion of longline operations, competition among gears, number of hooks, and experience level of the fisher- men (Polacheck, 1991; Ward, 2008). In this study, the fishermen used the same fishing gear with similar fishing techniques. Therefore, we assumed that differ- ences in fishing gear did not affect the catchability of Bigeye Tuna and we considered the number of hooks and environmental conditions to explain the catch- ability of Bigeye Tuna. Catchability fluctuated within and between years in relation to the number of hooks. During El Nino, high catchability coefficients occurred in May 1997 (1.56xl(D7) and June 1998 ( 1.47x HD7), coinciding with high HR of 0.94 and 0.83, respectively. This high coefficient number of catchability could be due to the higher number of hooks and catch of Bigeye Tuna during El Nino events related to oceanographic conditions favorable to Bigeye Tuna. The favorable oceanographic conditions were indicated by negative SSHA and by a colder SST than the normal condi- tion of around 28-29°C. Marsac and Blanc (1999) re- ported that the anomalous upwelling conditions that occurred in the EIO, providing biological enrichment and a shallower thermocline, should have favored catchability in the purse-seine fishery for Bigeye Tuna during the 1997-98 El Nino. The catchability coeffi- cient for November 2000 in our study appeared as an outlier. In that month, adverse weather conditions re- sulted in a decreased number of hooks and a lower HR; the increased catchability may have been due to the reduced fishing competition among longliners. The favorable oceanographic conditions for Bigeye Tuna catches during El Nino resulted in increasing predicted catch probabilities for Bigeye Tuna. The predicted distribution of Bigeye Tuna catch during El Nino showed a potential area with higher catch prob- ability compared with catch probability for La Nina events. Between the El Nino and La Nina events, fish- ing effort within the fishing grounds did not shift as much as did the predicted tuna habitat. Most catches 186 Fishery Bulletin 111(2) were made in the area of lower catch probabilities; however, some were from the boundary regions of high probability in the offshore waters of Java. Predictions of higher catch probabilities of Bigeye Tuna catches appeared to be associated with frontal areas in 10- 12°S, a region that seemed to reveal the importance of the confluence on the eastward IOKW and SJC that met with the outflow of the ITF and SEC in the off- shore area of the EIO off southern Java. There were many potential fishing locations that were not used optimally, and this inefficient use re- duced the total catch to much less than the level that was expected during the El Nino 1997-98 year. None- theless, the catch remained significantly higher during this El Nino event than during the 1999-2000 La Nina event. The mismatch between optimal fishing locations and actual fishing locations can be attributed to one or more of the reasons outlined below: The fishermen did not have the capability to deter- mine potential Bigeye Tuna habitats on the basis of large-scale regional shifts in oceanographic and climate regimes. They still used traditional methods, such as targeting locations similar to the ones where they had previously found Bigeye Tuna. The cost of fuel limited how far the fishermen could travel in search of Bigeye Tuna. Their fishing ground covered the area around 10-16°S and 108-120°E, the same region used in our data analysis. The fishermen did not target other species when fewer Bigeye Tuna were caught. They caught other species, such as small pelagic fishes ( Sardinella sp. or Euthynnus sp.), only for their own consumption during fishing trips (not as main targets). Therefore, we as- sumed that other catches did not influence the fishing effort. Political or management boundaries were not ma- jor problems facing this traditional fishing ground. Instead, the main constraints were the financial lim- its on long or distant fishing operations and the rapid fluctuations in fuel costs from week to week. The price of Bigeye Tuna depends on these factors: the location in which it is caught (fish caught farther from market are more expensive), the season (during the northwest monsoon, when fish abundance decreases, the price in- creases), and climate variability that affects environ- mental conditions (during El Nino event, fish generally are abundant and the price drops, and vice versa dur- ing La Nina event). Conclusions This study has shown the effects of ENSO-induced oceanographic conditions on catch rates of Bigeye Tuna in the EIO off Java. Spatiotemporal patterns in oceano- graphic conditions shown by the EOF, combined with the results of the GAMs, indicate that the 1997-98 El Nino event had a positive effect on catch rates of Big- eye Tuna in the EIO off Java. The EOF modes further highlight that interannual and seasonal time scales are the main factors that af- fect ocean current variability in the study area. The EOF analysis also provides evidence for the effects of the 1997-98 El Nino event in the EIO off the southern coast of Java — the dominant features being negative SSHA, cold SST, and high chlorophyll-a concentrations. In terms of these environmental variables, the binomi- al GAM confirmed that SST was the major factor that influenced Bigeye Tuna catches. These results indicate that the use of a GAM with 3 predictor variables may facilitate the identification of areas with potentially high Bigeye Tuna catch in the EIO off Java. Our results show significant effects of ENSO on Big- eye Tuna catches. For example, favorable oceanograph- ic conditions corresponded with the El Nino event, as indicated by the EOF and GAM analyses. We did not consider depth range data for Bigeye Tuna catches. Further investigations into prediction of fishing ground locations through the use of long-term, historical time series of environmental conditions and fishing ef- forts— fishery data sets of greater spatial and temporal resolutions than the data sets used in ours study — are needed to better understand the effects of climate vari- ability and fishing effort on changes in Bigeye Tuna catches in the EIO off Java. Acknowledgments The authors would like to thank the Directorate Gen- eral of Higher Education of the Republic of Indonesia and the Japan Science Society, under the Sasagawa Scientific Research Grant, for their support in funding this research. We thank the 3 anonymous reviewers for their valuable comments. We appreciate J. R. Bower for reading and improving this article. We thank the God- dard Space Flight Center/NASA and Physical Ocean- ography Distributed Active Archive Center for the production of chlorophyll-a and SST data, AVISO for the distribution of SSHA data, and the NOAA Climate Prediction Center for the use of its Nino 3.4 index. We also thank the incorporated company of Perikanan Nu- santara, Indonesia, which provided the fishery data. Literature cited Arrizabalaga, H., J. G. Pereira, F. Royer, B. Galuardi, N. Goni, I. Artetxe, I. Arregi, and M. Lutcavage. 2008. Bigeye tuna ( Thunnus obesus) vertical movements in the Azores Islands determined with pop-up satellite archival tags. Fish. Oceanogr. 17:74-83. Bjornsson, H., and S.A. Venegas. 1997. A manual for EOF and SVD analyses of climatic data. McGill Univ. C2GCR Rep. 97-1, 52 p. Briand, K., M. Brett, and L. Patrick. 2011. A study on the variability of albacore ( Thunnus alalunga ) longline catch rates in the southwest Pacific Ocean. Fish. Oceanogr. 20:517-529. Syamsuddin et at Effects of El Nino- Southern Oscillation on catches of Thunnus obesus in the eastern Indian Ocean 187 Brill, R. W. 1994. A review of temperature and oxygen tolerance studies of tunas pertinent to fisheries oceanography, movement models and stock assessments. Fish. Ocean- ogr. 3:204-216. Brill, R. W., K. A. Bigelow, M. K. Musyl, K. A. Fritsches, and E. J. Warrant. 2005. Bigeye tuna (Thunnus obesus) behavior and physi- ology and their relevance to stock assessments and fish- ery biology. Col. Vol. Sci. Pap. ICCAT 57(2): 142— 161. Ellis, N., and Y. G. Wang. 2007. Effects of fish density distribution and effort dis- tribution on catchability. ICES J. Mar. Sci. 64:178-191. Enfield, D.B. 2001. Evolution and historical perspective of the 1997- 1998 El Nino-Southern Oscillation event. Bull. Mar. Sci. 69:7-25. Ffield, A., K. Vranes, A. L. Gordon, R. D. Susanto, and S. L. Garzoli. 2000. Temperature variability within Makassar Strait. Geophys. Res. Lett. 27:237-240. Fraile, I., H. Murua, N. Goni, and A. Caballero. 2010. Effects of environmental factors on catch rates of FAD-associated yellowfin ( Thunnus alhacares) and skip- jack ( Katsuwonus pelamis ) tunas in the western Indian Ocean. Indian Ocean Tuna Commission, IOTC-2010- WPTT-46, 22 p. Gordon, A. L. 2005. Oceanography of the Indonesian seas and their throughflow. Oceanography 18(41:13-26. Gordon, A., J. Sprintall, V. H. M. Aken, R. D. Susanto, S. Wi- jffels, R. Molcard, A. Ffield, W. Pranowo, and S. Wirasantosa. 2010. The Indonesian throughflow during 2004-2006 as observed by the INSTANT program. Dyn. Atmos. Oceans 50:115-128. Haddon, M. 2011. Modelling and qualitative methods in fisheries, 2nd ed., 449 p. CRC Press, Boca Raton, FL. Hanamoto, E. 1987. Effect of oceanographic environment on bigeye tuna distribution. Bull. Jap. Soc. Sci. Fish. Oceanogr. 3:203-216. Hastie, T. , and R. Tibshirani. 1990. Generalized additive models, 335 p. Chapman and Hall, New York. Holland, K. N., R. W. Brill, R. K. C. Chang, J. R. Sibert, and D.A. Fournier. 1992. Physiological and behavioural thermoregulation in bigeye tuna (Thunnus obesus). Nature 358:410-412. Howell, E. A., and D. R. Kobayashi. 2006. El Nino effects in the Palmyra Atoll region: ocean- ographic changes and bigeye tuna (Thunnus obesus) catch rate variability. Fish. Oceanogr. 15:477-489. Howell, E. A., D. R. Hawn, and J. J. Polovina. 2010. Spatiotemporal variability in bigeye tuna (Thun- nus obesus ) dive behavior in the central North Pacific Ocean. Prog. Oceanogr. 86:81-93. Lehodey, P. 2001. The pelagic ecosystem of the tropical Pacific: dy- namic spatial modelling and biological consequences of ENSO. Prog. Oceanogr. 49:439-468. Lehodey, P, M. Bertignac, J. Hampton, A. Lewis, and J. Picaut. 1997. El Nino Southern Oscillation and tuna in the western Pacific. Nature 389:715-718. Lehodey, P, F. Chai, and J. Hampton. 2003. Modelling climate-related variability of tuna pop- ulations from a coupled — ocean biogeochemical-popula- tions dynamics model. Fish. Oceanogr. 12:483-494. Lehodey, P., I. Senina, J. Sibert, L. Bopp, B. Calmettes, J. Hampton, and R. Murtugudde. 2010. Preliminary forecast of Pacific bigeye tuna popula- tion trends under the A2 IPCC scenario. Prog. Ocean- ogr. 86:302-315. Marsac, F. and J. L. Blanc. 1999. Oceanographic changes during the 1997-1998 El Nino in the Indian Ocean and their impact on the purse seine fishery. lOTC Proceedings 2:147-157. McPhaden, M. J. 1999. Genesis and evolution of the 1997-1998 El Nino. Science 283:950-954. Miller, K. A. 2007. Climate variability and tropical tuna: manage- ment challenges for highly migratory fish stocks. Mar. Policy 31:56-70. Miyake, M., P. Guillotreau, C-H. Sun, and G. Ishimura. 2010. Recent developments in the tuna industry: stocks, fisheries, management, processing, trade and markets. FAO Fisheries and Aquaculture Tech. Paper No. 543, 125 p. Mohri M., and T. Nishida. 1999. Distribution of bigeye tuna (Thunnus obesus) and its relationship to the environmental conditions in the Indian Ocean based on the Japanese longline fisheries information. IOTC Proceedings 2:221-230. Molcard, R., M. Fieux, and F. Syamsudin. 2001. The throughflow within Ombai Strait. Deep-Sea Res. (I Oceanogr. Res. Pap.) 48:1237-1253. Murtugudde, R. G., S. R. Signorini, J. R. Christian, A. J. Busa- lacchi, C. R. McClain, and J. Picaut. 1999. Ocean color variability of the tropical Indo-Pacific basin observed by SeaWiFS during 1997-1998. J. Geo- phys. Res. (C Oceans) 104:18351-18366. Otero, M. P, and D. A. Siegel. 2004. Spatial and temporal characteristics of sediment plumes and phytoplankton blooms in the Santa Bar- bara channel. Deep-Sea Res. (II Top. Stud. Oceanogr.) 51:1129-1149. Pepperell, J. 2010. Pishes of the open ocean: a natural history and illustrated guide, 272 p. Univ. Chicago Press, Chicago. Polacheck, T. 1991. Measures of effort in tuna longline fisheries: changes at the operational level. Fish. Res. 12:75-87. Polovina, J. J., G. H. Balazs, E. A. Howell. D. M. Parker, M. P. Seki, and PH. Dutton. 2004. Forage and migration habitat of loggerhead (Caretta caretta) and olive ridley ( Lepidochelys olivacea) sea turtles in the central North Pacific Ocean. Fish. Oceanogr. 13:36-51. Polovina, J. J., and E. A. Howell. 2005. Ecosystem indicators derived from satellite re- motely sensed oceanographic data for the North Pacific. ICES J. Mar. Sci. 62:319-327. Polovina, J. J., E. Howell, D. R. Kobayashi, and M. P. Seki. 2001. The transition zone chlorophyll front, a dynamic global feature defining migration and forage habitat for marine resources. Prog. Oceanogr. 49:469-483. R Development Core Team. 2011. R: a language and environment for statistical 188 Fishery Bulletin 111(2) computing. R Foundation for Statistical Computing, Vienna, Austria. [Available from http://www.r-project. org, accessed February 2012.1 Radiarta, I. N., and S. Saitoh. 2008. Satellite-derived measurements of spatial and temporal chlorophyll-a variability in Funka Bay, south- western Hokkaido, Japan. Estuar. Coast. Shelf Sci. 79:400-408. Saitoh, S., E. Chassot, R. Dwivedi, A. Fonteneau, H. Kiyofuji, B. Kumari, M. Kuno, S. Matsumura, T. Platt, M. Raman, S. Sathyendranath, H. Solanki, and F. Takahashi. 2009. Remote sensing applications to fish harvest- ing. In Remote sensing in fisheries and aquaculture. Reports of the International Ocean-Colour Coordinating Group (IOCCG), No. 8 (M.-H. Forget, V. Stuart, and T. Platt, eds.), p.57-76. IOCCG, Dartmouth, Canada. Song, L., and Y. Zhou. 2010. Developing an integrated habitat index for bigeye tuna ( Thunnus obesus ) in the Indian Ocean based on longline fisheries data. Fish. Res. 105:63-74. Song, L., J. Zhou, Y. Zhou, T. Nishida, W. Jiang, and J. Wang. 2009. Environmental preferences of bigeye tuna, Thun- nus obesus , in the Indian Ocean: an application to a longline fishery. Environ. Biol. Fishes 85:153-171. Sprintall, J., J. C. Chong, F. Syamsudin, W. Morawit, S. Hau- tala, N. Bray, and S. E. Wijffels. 1999. Dynamics of the South Java Current in the Indo- Australian Basin. Geophys. Res. Lett. 26:2493-2496. Sprintall, J., S. E. Wijffels, R. Molcard, and I. Jaya. 2009. Direct estimates of the Indonesian Throughflow entering the Indian Ocean: 2004-2006. J. Geophys. Res. (C Oceans) 114:1-19. 2010. Direct evidence of the South Java Current system in Ombai Strait. Dyn. Atmos. Oceans 50:140-156. Syamsudin, F., A. Kaneko, and D. B. Haidvogel. 2004. Numerical and observational estimates of In- dian Ocean Kelvin wave intrusion into Lombok Strait. Geophys. Res. Lett. 31:L24307. doi: 10. 1029/2004GL02 1227. Teo, S. L. H, and B. A. Block. 2010. Comparative influence of ocean conditions on yellowfin and Atlantic bluefin tuna catch from long- lines in the Gulf of Mexico. PLOS ONE 5(5):el0756. doi: 10. 137 1/journal, pone. 00 10756. Toian, J. M, and M. Fisher. 2009. Biological response to changes in climate patterns: population increases of gray snapper ( Lutjanus griseus) in Texas bays and estuaries. Fish. Bull. 107:36-44. Torres-Orozco, E., A. Muhlia-Melo, A. Trasvina, and S. Ortega-Garcia. 2006. Variation in yellowfin tuna (Thunnus albacares) catches related to El Nino-Southern Oscillation events at the entrance to the Gulf of California. Fish. Bull. 104: 197-203. Walters, C. 2003. Folly and fantasy in the analysis of spatial catch rate data. Can. J. Fish. Aquat. Sci. 60:1433-1436. Ward, P. 2008. Empirical estimates of historical variations in the catchability and fishing power of pelagic longline fishing gear. Rev. Fish Biol. Fish. 18:409-426. Ward, P, and R. A. Myers. 2004. Inferring the depth distribution of catchability for pelagic fishes and correcting for variations in the depth of longline fishing gear. Can. J. Fish Aquat. Sci. 62:1130-1142. Wessel, P, and W. H. F. Smith. 1998. New, improved version of Generic Mapping Tools released. EOS Trans Am Geophys 79:579. [Available from http://gmt.soest.hawaii.edu, accessed February 2013.1 Wood, S. N. 2006. Generalized additive models: an introduction with R, 392 p. Chapman & Hall/CRC, Boca Raton, FL. Yoder, J. A, and M. A. Kennely. 2003. Seasonal and ENSO variability in global ocean phytoplankton chlorophyll derived from 4 years of SeaWiFS measurements. Global Biogeochem. Cycles 17(4): 1 1 12. doi: 10. 1029/2002GB001942. Zainuddin, M., S. Saitoh, and K. Saitoh. 2004. Detection of potential fishing ground for al- bacore tuna using synoptic measurements of ocean color and thermal remote sensing in the northwest- ern North Pacific. Geophys. Res. Lett. 3LL20311. doi:10.1029/2004GL021000. Zhou, L., R. Murtugudde, and M. Jochum. 2008. Dynamics of the intraseasonal oscillations in the Indian Ocean South Equatorial Current. J. Phys. Oceanogr. 38:121-132. Zuur, A. F., E. N. leno, N. J. Walker, A. A. Saveliev, and G. M. Smith. 2009. Mixed effects models and extensions in ecology with R, 574 p. Springer, New York. 189 Migratory and within-estuary behaviors of adult Summer Flounder {.Paralichthys den tat us) in a lagoon system of the southern mid-Atlantic Bight Email address for contact author: kcapossela@dnr.state.md.us Virginia Institute of Marine Science Department of Fisheries Science College of William 8. Mary 1208 Greate Road, PO Box 1346 Gloucester Point, Virginia 23062 Present address for contact author: Maryland Department of Natural Resources 301 Marine Academy Drive Stevensville, Maryland 21666 James J Howard Marine Sciences Laboratory Northeast Fisheries Science Center, National Marine Fisheries Service, NOAA 74 Magruder Road, Sandy Hook Highlands, New Jersey 07732 Abstract— We monitored the move- ments of 45 adult Summer Floun- der ( Paralichthys dentatus) between June 2007 and July 2008 through the use of passive acoustic telemetry to elucidate migratory and within- estuary behaviors in a lagoon system of the southern mid-Atlantic Bight. Between 8 June and 10 October 2007, fish resided primarily in the deeper (>3 m) regions of the system and exhibited low levels of large- scale (100s of meters) activity. Mean residence time within this estuarine lagoon system was conservatively estimated to be 130 days (range: 18- 223 days), which is 1.5 times longer than the residence time previously reported for Summer Flounder in a similar estuarine habitat -250 km to the north. The majority of fish remained within the lagoon system until mid-October, although some fish dispersed earlier and some of them appeared to disperse tempo- rarily (i.e., exited the system for at least 14 consecutive days before returning). Larger fish were more likely to disperse before mid-Octo- ber than smaller fish and may have moved to other estuaries or the in- ner continental shelf. Fish that dis- persed after mid-October were more iikely to return to the lagoon system the following spring than were fish that dispersed before mid-October. In 2008, fish returned to the system between 7 February and 7 April. Dispersals and returns most closely followed seasonal changes in mean water temperature, but photoperiod and other factors also may have played a role in large-scale move- ments of Summer Flounder. Manuscript submitted 24 May 2012. Manuscript accepted 5 March 2013. Fish. Bull. 111:189-201 (2013). doi 10.7755/FB.111.2.6 The views and opinions expressed or implied in this article are those of the author (or authors) and do not necessar- ily reflect the position of the National Marine Fisheries Service, NOAA. Karen M. Capossela (contact author)' Mary C. Fabrizio' Richard W. Brill2 The continued degradation of estua- rine environments associated with eutrophication, shoreline develop- ment, and global climate change ne- cessitates a better understanding of how seasonal residents, like Summer Flounder ( Paralichthys dentatus ), use mid-Atlantic estuaries (Gibson, 1994; Beck et al., 2001). Estuaries provide juvenile and adult Summer Flounder with the water tempera- tures, food resources, and protection from predation that are necessary for their growth and survival (Stierhoff et al., 2006). Summer Flounder mi- grate offshore in the fall and winter to spawn over the outer continental shelf before they migrate back in- shore the following spring, often re- turning to the same estuary in sub- sequent years (Sackett et al., 2007). As a result, stock abundance is influ- enced by local estuarine conditions (Ray, 2005). The use of estuaries as nursery habitat by Summer Flounder and the responses of juvenile Sum- mer Flounder to estuarine conditions have been extensively examined (e.g., Malloy and Targett, 1994; Tyler, 2004; Necaise et al., 2005; Stierhoff et al., 2006, 2009). Only recently, however, have migratory and within-estuary behaviors of adult Summer Flounder been examined (Sackett et al., 2007, 2008; Henderson, 2012). Migration timing traditionally has been determined through assessment of the abundance of fishes in an es- tuary over time with standard fish- eries methods, such as bottom trawl surveys. However, population-level monitoring is insufficient to under- stand the dynamics of emigration or the variation in individual responses (DeCelles and Cadrin, 2010). In re- cent years, acoustic telemetry has been established as a powerful tool for observation of individual variabil- ity in behaviors (Heupel et al., 2006; DeCelles and Cadrin, 2010). A study of acoustically monitored adult Sum- mer Flounder in the Mullica River- Great Bay estuary in New Jersey (located in the northern mid-Atlantic Bight [MAB]) indicated that a large number of fish departed the estuary in July, but the precise timing varied between years (Sackett et al., 2007). 190 Fishery Bulletin 111(2) Therefore, variation in emigration timing may exist not only by latitude but also among individual fish within a system. Some adult Summer Flounder have been known to return to the same estuary in subsequent years (Poole, 1962; Sackett et al., 2007; Henderson, 2012), but factors that influence site fidelity in Sum- mer Flounder are not well understood. Acoustic telemetry has also been used to identify variations in Summer Flounder within-estuary activity. For example, Summer Flounder in the Mullica River- Great Bay estuary primarily used the lower bay (near the inlet), but some fish resided in other areas (Sack- ett et al., 2008). Likewise, most fish remained in the Mullica River-Great Bay estuary until emigration to the outer shelf, but several adults exited and entered the system multiple times (i.e., exhibited temporary emigration). Similar patterns were observed in the Chesapeake Bay (southern MAB), where some adults remained sedentary and resided at structured sites for long periods of time, and others were more active and traveled long distances (Henderson, 2012). The continuation of acoustic studies is necessary to identify similarities and differences in behavioral pat- terns between regions and to investigate the drivers behind these patterns. Our objectives for this study were to use acoustic telemetry to describe the migra- tory and within-estuary behaviors of adult Summer Flounder from a previously unstudied lagoon system in the southern portion of the MAB. The lagoon systems off Virginia’s Eastern Shore are subject to large fluc- tuations in temperature typical of most MAB systems (0-30°C), but they differ from larger estuaries in that they are shallow (mean depth <3 m), well-mixed, and polyhaline areas. These lagoon systems are a nursery ground for juvenile Summer Flounder (Schwartz, 1964; Norcross and Wyanski, 1994; Desfosse, 1995; Kraus and Musick, 2001), but they also support a large num- ber of adults and an active recreational fishery (Rich- ards and Castagna, 1970). Previous descriptions of the use of our chosen lagoon system by Summer Flounder have been limited to descriptions of juvenile habitat preferences (Wyanski, 1990; Norcross and Wyanski, 1994) and adult migration patterns determined by tra- ditional mark-recapture methods (Kraus and Musick, 2001; Desfosse, 1995). We used the data from our acoustic telemetry study to determine 1) dispersal and return rates, 2) duration of residency, 3) spatiotemporal distribution, and 4) ac- tivity of fish within the system. Because tidal stage, time of day, and temperature all have been associated with flatfish activity (Olla et al., 1972; Casterlin and Reynolds, 1982; Wirjoatmodjo and Pitcher, 1984; Malloy and Targett, 1991; Szedlmayer and Able, 1993; Hender- son, 2012), these factors were considered in our exami- nation of within-estuary activity. We also analyzed the effects of seasonal temperature, photoperiod, and fish size on dispersal, returns, and residency times (Smith, 1973; Able and Kaiser, 1994). Materials and methods Study site The estuarine lagoon system near Wachapreague, Vir- ginia, resides behind a series of low barrier islands and primarily connects with the Atlantic Ocean through Wachapreague Inlet (Fig. 1). The 2 main channels leading from Wachapreague Inlet divide into smaller channels that cut through marsh areas (dominated by smooth cordgrass [Spartina alterniflora]) before they open into large, shallow tidal flats. Channels were identified as areas -3-12 m deep, and tidal flats were identified as areas <3 m deep. As with most seaside lagoon systems in Virginia, the system near Wachapre- ague is characterized by restricted access to the ocean, minimal freshwater input, and a moderate tidal range (1.2-1. 4 m; NOAA Center for Operational Oceano- graphic Products and Services, http://tidesandcurrents. noaa.gov/tides07/tab2ec2b. html#44). Strong currents are typical because of natural constrictions at the in- let and in the channels (Conrath, 2005), although cur- rents generally dissipate with distance from the inlet. Sediment type follows the energy gradient, with coarse sand within and near the inlet, and progressively finer (muddy) sediments at greater distances from the inlet (Wyanski, 1990). We divided our study area into 4 regions (Fig. 1): 1 Wachapreague Inlet — the primary point of ingress and egress of fish characterized by depths of 6-15 m and strong currents; the inlet is about 625 m wide. 2 Upper channels — the channel leading north from Wachapreague Inlet and its divergent channels. 3 Lower channels — the channel leading south from Wachapreague Inlet and its divergent channels. 4 Tidal flats (also known locally as bays ) — the shal- lowest bodies of water included in our study. Al- though several tidal flats are present in this area, only Swash Bay was included in our study area be- cause we could monitor the movements of Summer Flounder into and out of this area. We recorded environmental conditions in the in- let, channels, and tidal flat from 8 June 2007 to 29 July 2008 with 3 YSI 6920-O1 multiparameter water- quality sondes (YSI, Inc., Yellow Springs, OH; Fig.l), which recorded temperatures and dissolved oxygen con- centrations once per hour. Sondes were replaced with calibrated units every 1-2 weeks in the summer and (as fouling diminished) every 2-4 weeks thereafter. Errone- ous recordings due to membrane fouling, battery failure, and calibration drift were removed from the data set. Data from the water-quality sondes confirmed that dis- solved oxygen concentrations generally remained above the critical oxygen level (27.2%, 2.0 mg O2 L_1) for adult Summer Flounder at typical summer bottom-water 1 Mention of trade names or commercial companies is for iden- tification purposes only and does not imply endorsement by the National Marine Fisheries Service, NOAA. Capossela et al.: Migratory and within-estuary behaviors of adult Paralichthys dentatus of the southern mid-Atlantic Bight 191 75°41'W 75°40'W 75“39’W 75°38'W 75"37'W 75°36’W A :: Wachapreague S12+ JL Chesapeako Bay q V ^5 #14 'bi/l 23 • 0 £28 29 • - ■ -fl ■ o w+ 24 • 26 27 • 25* ^ ’ i ■ • 20 19 • • WQ2 se®" A lit .'ir-v ■ 3 5 t- A 21 ■ ■ fi 18 Area of Interest Wachapreague Inlet ■9 - - 13 WOI 1Q * © ■ v v 4. t 4 Swash Bay Virginia Atlantic Ocean 0 0.5 1 1.5 2 k 1. 1 1 1 1 1 I 1 I # 12 ; 0 Upper channel receivers A Wachapreague Inlet receivers 1 Lower channel receivers # Tidal flat receivers 4* Supplemental receivers O Water-quality sonde Figure 1 (A) The distribution of acoustic receivers and water-quality sondes installed in the Wachapreague lagoon system for this study of Summer Flounder ( Paralichthys dentatus) behaviors between June 2007 and July 2008. The location of the Wachapreague lagoon system in the southern mid-Atlantic Bight is shown by the square in the smaller map. Regions specified in the text are Wacha- preague Inlet, upper channels, lower channels, and tidal flat. Receivers S3, S5-S9, Sll, and S12 provided supplemental data on the activity in regions outside of our study area. (B) In this map of acoustic receivers, each circle represents the approximate detection range (radius=35Q m) of the receivers deployed in the Wachapreague lagoon system. temperatures (Capossela et al., 2012). Photoperiod (i.e., day length) was acquired from tide prediction software (Jtides, vers. 4.9; http://www.arachnoid.com/JTides). Telemetry and tagging On 22 May 2007, 50 Summer Flounder (261—558 mm total length [TL] ) were captured at the study area by hook and line, identified from Murdy et al (1997), and immediately anesthetized with 60 mg L_1 AQUI- S (AQUI-S New Zealand Ltd., Lower Hutt, New Zea- land) to allow surgical implantation of individually coded 69-kHz transmitters (V9-2L-R64K; VEMCO Di- vision, AMIRIX Systems, Inc., Bedford, Canada) by using established procedures (Fabrizio and Pessutti, 2007). Transmitters were 30 mm long and 9 mm in diameter and had a delay time of 60-180 s and a pro- jected 14-month battery life. All fish were tagged and released in the upper channels with the exception of a single fish that was captured, tagged, and released on the tidal flat. Before release, all fish were allowed to fully recover in an onboard aquarium that accom- modated total length, and externally tagged with an individually numbered T-bar anchor tag inserted near the caudal peduncle to alert anglers to report recap- tures. We considered all fish to be adults because Sum- mer Flounder can reach maturity at 240-300 mm TL (Morse, 1981). Summer Flounder migratory and within-estuary be- haviors were examined from 8 June 2007 until the last fish departed on 17 January 2008. We chose the start date (8 June 2007), which was approximately 2 weeks after the release of tagged fish, to limit the influence of any atypical activity patterns due to recovery from capture and surgery (Knights and Lasse, 1996; Rogers and White, 2007). We recorded fish locations with 31 192 Fishery Bulletin 111(2) receivers (VR2, VR2W; VEMCO) deployed throughout the study area (Fig. 1A). Most receivers were deployed by 8 June 2007 (receivers numbered 1-27), but 4 re- ceivers were deployed on 26 June (receivers numbered 28-30) and 16 July (receiver numbered 31) to provide additional coverage. Receivers were attached to an an- chored line fitted with a buoy and positioned near the bottom of the water column (<1 m from the bottom of the ocean floor) with the hydrophone oriented down- ward. Range tests conducted throughout the study area indicated an approximate detection range of 350 m. We placed as many receivers as possible -700 m apart in the upper and lower channels to be able to monitor fish movements on the scale of 100s of me- ters (Fig. IB). Currents and boat traffic limited the placement of receivers in certain locations in the up- per channels and prevented the use of a directional gate (Heupel et al., 2006) at Wachapreague Inlet. In these cases, receivers were placed in the next suitable location. The tidal flat was too shallow for extensive receiver coverage; instead, we used receivers to moni- tor fish as they entered and exited the tidal flat. Most receivers were retrieved on 31 January 2008, but 12 receivers (receivers numbered 5-7, 16-18, and 22-27; Fig. 1A) were left in the system to detect fish return- ing to the Wachapreague system the following year. To prevent receiver loss, we began retrieval of the receiv- ers that were farthest from the inlet in mid-April. All receivers were retrieved by 29 July 2008. A separate acoustic telemetry study conducted by re- searchers to examine movements of Cownose Ray ( Rh - inoptera bonasus) in the Wachapreague system over- lapped with the timing of our Summer Flounder study. Receivers from the Cownose Ray study were placed mostly in small channels far from Wachapreague Inlet in an area not covered by our receivers. Receivers for that study were deployed on 26 June 2007 (receiver labeled S3) and 26 July 2007 (receivers labeled S5-S9, S11-S12; Fig. 1A) and retrieved on 17 November 2007. The receivers in the Cownose Ray study were spaced too far apart to meet the specific objectives of our study and detections from these receivers were not used in our analyses. We did note, however, the extent to which Summer Flounder were detected in these small back channels and henceforth refer to these receivers as supplemental receivers. Migratory behaviors Data were examined over weekly intervals to examine patterns of seasonal migration. We considered a fish to have dispersed on the last day it was detected at or near Wachapreague Inlet (receivers 17-22, 31; Fig 1A). Likewise, we considered a fish to have returned when it was first redetected at Wachapreague Inlet or with- in the lagoon system. Weekly probabilities of disper- sal and return were calculated with the Kaplan-Meier estimator, a nonparametric approach that requires no assumptions about the underlying hazard function and accommodates censored fish (Pollock et al., 1989; Bennetts et al., 2001). Fish were censored from (i.e., not included in) this analysis if they were no longer detected but did not depart from the system through Wachapreague Inlet; the fate of such fish could not be conclusively determined. Censored fish may have re- sided in the system undetected, been removed by fish- ermen or predators, or have left the system through another route. We used a piecewise linear regression model to identify when dispersal rates changed (i.e., the change- point), and we fitted the model to the data with non- linear least-squares estimation (the NLIN procedure in SAS, vers. 9.2, SAS Institute, Inc., Cary, NC; e.g., Ryan et al., 2007). The time before dispersal rates changed was considered the residency period, a time during which most fish were found within the lagoon system. The time after dispersal rates changed was considered the emigration period, during which most fish were observed finally to have dispersed. We classified fish according to observed migratory behaviors: those fish that dispersed early (during the residency period) and those fish that dispersed late (during the emigration period). An odds ratio (Agresti, 2007) was used to test the association between the timing of dispersal (i.e., residency period vs. emigration period) and the like- lihood that a fish would return to the Wachapreague system the following year. Some fish were detected at or near the inlet (receiv- ers 17-22, 31) but were subsequently undetected for 14 or more consecutive days before redetection. These fish were classified as temporary emigrants because they were presumed to have exited and re-entered the lagoon system. Such behaviors were consistent with activity reported in a previous study (Sackett et al., 2007). Tagged fish, including temporary emigrants, were considered residents until final dispersal out of the inlet, and residence time was defined as the total number of days from the start of our study (8 June 2007) until the last detection at or near the inlet be- fore final dispersal. The residence time of uncensored fish was used to calculate a mean residence time for Summer Flounder in the Wachapreague system. The mean residence time reported throughout this article is, therefore, an estimate of least (minimum) residence time because we do not know how long tagged fish were present in the lagoon system before the start of our study. Mean residence time and other mean values are reported as mean ±1 standard error. The effects of mean monthly temperature and mean monthly photoperiod on the percentage of fish that fi- nally dispersed in a given month (log-transformed to improve homogeneity of variance) were examined with a multiple linear regression (general linear model [GLM] procedure in SAS). We also examined the effect of fish size on the probability of final dispersal before and after dispersal rates changed with logistic regres- sion (LOGISTIC procedure in SAS). Goodness-of-fit sta- Capossela et al Migratory and within-estuary behaviors of adult Paralichthys dentatus of the southern mid-Atlantic Bight 193 Table 1 The total detectable area (km2) that we monitored for the presence of Summer Flounder and the percentage of the total detectable area in each defined region (upper channels, lower channels, tidal flat, and inlet) of the Wachapreague lagoon system, on the basis of a 350-m detection range, for this study of Summer Flounder ( Paralichthys dentatus ) be- haviors. Also included are the proportions of time Summer Flounder spent in each region and the proportions of fish found in each region over the residency and emigration periods (8 June 2007-17 January 2008). The sum of the proportion of fish that used each region exceeds 1 because a single fish could occupy more than 1 region over the study period. Region Detectable area (km2) Percentage of total area (%) Total mean proportion of time (%) Total proportion of fish (%) Upper channels 2.75 39.2 78.1 97.8 Lower channels 2.21 31.5 19.4 28.9 Tidal flat 1.29 18.5 0.4 4.4 Inlet 0.76 10.8 2.1 67.7 Total area 7.01 tistics were calculated to assess the fit of the model through the use of the LOGISTIC procedure in SAS. Within-estuary behaviors We ascertained the temporal and spatial distributions of Summer Flounder in the upper channels, lower channels, tidal flat, and Wachapreague Inlet by ex- amination of monthly distributions of Summer Floun- der until all fish finally dispersed. Because the total detectable area that we monitored for the presence of Summer Flounder varied between regions (Table 1), our assessment of fish activity by region did not rely on continuous detection. We calculated the proportion of fish in each region by month as the number of fish detected in a region divided by the total number of fish present in the system that month. We also calculated the proportion of time the average fish resided within a region each month as the total time a fish spent in a region divided by the total time spent in all regions that month. Time in a region was defined as the total time between the first and last detection before de- tection in another region; receivers provided sufficient coverage to monitor fish movement into and out of the 4 regions (Fig. IB). For fish that moved between re- gions, we did not use the length of time between the last region-specific detection and the next region-spe- cific detection because we could not objectively assign fish location during that interval to a specific region. Because not all fish could be assigned objectively to a region each month, the sum of the proportions of fish using each region could be <1 for a given month. Con- versely, the sum of the proportions of fish using each region could be >1 because a single fish could occupy more than one region in any given month. In addition to monthly analyses, we calculated the proportions of fish present and time spent in each region for the residency and emigration periods. The 2 statistic was used to test for differences in the mean proportions between the residency and emigration periods (Fleiss, 1981). All proportions were expressed as percentages. We used movement between receivers to calculate the activity index, which we defined as the total num- ber of times an individual moved between receivers during nonconsecutive 6-h periods. We limited the data to fish in the upper channels during the residency pe- riod because the sample size was highest in this loca- tion and during this time (8 June 2007 to 10 October 2007; see the Results section). For each 6-h period, we assigned an activity index value of zero when a fish did not move between receivers, and a value of 1 for each arrival at a different receiver (adjacent or non- adjacent). The activity index was weighted to account for variation in distances between receivers (rounded to the nearest integer) and summarized weekly for individual fish by tidal stage (ebb, slack before ebb, flood, and slack before flood) within each time-of-day interval (day or night). Day (10:00-16:00) and night (22:00-4:00) were restricted to these nonconsecutive 6-h periods to minimize autocorrelations associated with successive observations on the same fish during day and night periods (Rogers and White, 2007). We also computed mean temperature for each period (tidal stage, time of day, and week combination). We examined the relationship between activity indi- ces and week, time-of-day, tidal stage, and temperature with a generalized repeated measures model (GEN- MOD procedure in SAS). This equation represents the statistical model fitted to the data: logUijk) = n + «i + C Vem1gration=46%). Dispersal followed the steep decline in temperature more closely than it did the gradual shift in day length, which (in contrast to changes in water temperature) was smooth and al- most constant over time (CVresicjency=7.7%, CVernigratjon=5.8%; Figs. 4, 5). The multiple linear regression that included both tem- perature and photoperiod as predictors of dispersal was significant (P=20.3, P<0.05) and explained 89% of the variation in monthly dispersals. Temperature was a sig- nificant predictor of mean percent disper- sal (F=6.39, P= 0.05), but photoperiod was not (F=0.94, P=0.38). The length of time over which we observed returning fish (3 months) was in- adequate to statistically examine the effects of mean monthly temperature and photoperiod on the timing of return. The mean sizes at tagging for fish that dis- persed during the residency and emigration pe- riods were 437 ±21 mm TL and 367 ±13 mm TL, respectively. We found that the timing of dis- persal was inversely related to fish size at the time of tagging (%2=8.45, P<0.05). Larger fish were more likely to leave the system during the residency period (before October 11) than were smaller fish. Conversely, smaller fish were more likely to disperse during the emigration period. The goodness-of-fit of this model indicated that predicted and observed frequencies were not sig- nificantly different (%2=3.86, P- 0.80), indicating the adequacy of the logistic regression model as a descriptor of these data. Within-estuary behaviors Summer Flounder primarily used the upper channels during the residency period, although fish were detected in all habitats (Fig. 6, A and B). Fish occupied the upper and lower channels for 78% and 19%, respectively, of the total time that fish were detected (Table 1). With the exception of the single fish released in the tidal fiat, all fish were 7/1/2007 Figure 4 The proportion of Summer Flounder (Paralichthys dentatus ) that dispersed from (^) and returned to (■) the Wachapreague lagoon system from 8 June 2007 to 7 April 2008 (when the last fish was detected returning), on the basis of the Kaplan-Meier estimator. Mean daily temperature (°C; gray line) is also plotted. Confidence intervals have been omitted for clarity. 196 Fishery Bulletin 111(2) r 14.28 48 14 27 30 14.21 54 14 20 24 7/1/2007 3/1/2008 Figure 5 The proportion of Summer Flounder (Paralichthys dentatus ) that dis- persed from (^) and returned to (■) the Wachapreague lagoon system from 8 June 2007 to 7 April 2008 (when the last fish was detected return- ing), on the basis of the Kaplan-Meier estimator. Photoperiod (day length in hours; gray line) is also plotted. Confidence intervals have been omit- ted for clarity. detected in the upper channels, but only 27% (12 in- dividuals) of the fish that were detected in the upper channels were also detected in the lower channels. This finding indicates that the majority of fish released in the upper channels (73%, 32 individuals) remained near the release site in the upper channels until dispersal. The proportion of time and the proportion of fish in the upper channels were significantly greater during the residency period than during the emigration period (Table 2; ztlme=17.0, P<0.05; zfish=4.2, P<0.05). Use of the lower channels was greatest during the emigration period, both in terms of proportions of time spent in these habitats and the number of fish detected (Ta- ble 2; ztime=14.6, P<0.05; zfish=2.6, P<0.05). Most fish (85%) detected in the lower channels occupied the up- per channels for a mean of 132 ±14 days before they were detected in the lower channels. Fish detected in both the upper and lower channels had a later mean emigration date (15 November 2007) than that of fish that did not use the lower channels (24 August 2007). Only 4% (2 individuals) of Summer Flounder briefly occupied the tidal flat between October and December 2007 (6 ±5 days, range: 1-11 days). Summer Flounder did not appear to regularly occupy the additional por- tions of the Wachapreague system monitored by the supplementary receivers. Only 7% (3 individuals) of Summer Flounder were detected by these receivers, and the mean residency was 6 ±4 days (range: 0.2-13 days). Fish presence was, however, likely underesti- mated because of the limited coverage and the shorter period of receiver deployment. Although the inlet region was fre- quented by Summer Flounder over the course of our study (Fig. 6B), fish spent a smaller proportion of time at the in- let (2%) than in the upper and lower channels (97%; Table 1). The mean time at the inlet was 2 ±0.6 days. Not surprisingly, both the proportions of time and fish at the inlet were greatest during the emigration period (Table 2; Ztime=4.9, P< 0.05; zfish=3.0, P<0.05). Only 5 Summer Flounder in the up- per channels moved between adjacent or nonadjacent receivers more than 10 times during the residency period. The mean observed activity did not vary significantly by week (x2=19.06, P=0.33), but it did vary significantly with time of day; the mean activity index was significantly greater dur- ing night than during day (x2=6.13, P<0.05). Individuals appeared most ac- tive during the flood tide or during the slack tide before ebb, but differences in mean activity among tidal stages were not statistically significant (x2=6.97, P=0.07). Activity also was not affected by differences in mean temperature for a given tidal stage (x2=0.46, P= 0.55). Discussion Migratory behaviors The observed timing of Summer Flounder dispersal from the Wachapreague system (October though Janu- ary) is consistent with the established seasonal pro- gression of spawning migration from north to south (Smith, 1973; Morse, 1981; Kraus and Musick, 2001; Sackett et al., 2007). It most closely matches the re- ported timing of emigration for Summer Flounder in the nearby Chesapeake Bay. Summer Flounder pri- marily emigrate from Chesapeake Bay from October through December, and some fish emigrate as late as February (Desfosse, 1995; Henderson, 2012). In New Jersey’s Mullica River-Great Bay estuary (-250 km to the north), acoustically tagged fish generally emi- grated earlier — between August and December (Able et al., 1990; Roundtree and Able, 1992b; Szedlmayer and Able, 1993). By mid-September, 75% of tagged Summer Flounder had dispersed from a study site on the inner shelf near New Jersey (Fabrizio et al.3). In contrast, 3 Fabrizio, M. C., J. P. Pessutti, J. P. Manderson, A. F. Drohan, and B. A. Phelan. 2005. Use of the historic area remedia- tion site by black sea bass and summer flounder. Northeast Fish. Sci. Cent Ref. Doc. 05-06, 95 p. Capossela et al.: Migratory and within-estuary behaviors of adult Paralichthys dentatus of the southern mid-Atlantic Bight 197 10 i MB Upper channel l l Lower channel Inlet 1 0 ■o 2 08 06 S 0 4 - o CL O 0 2 - 0 0 m Jun Jul Aug Sep Oct Month Nov Dec Jan Figure 6 (A) The monthly mean proportion of time that Summer Floun- der ( Paralichthys dentatus) occupied the upper channels, lower channels, and Wachapreague Inlet in the Wachapreague lagoon system from June 2007 to January 2008. For a given month, the proportion of time that individual fish occupied each re- gion was calculated as the ratio of the amount of time that a fish resided in a region in relation to the total time it was detected that month, with proportions for a month adding to 1. iB) The monthly proportion of individual Summer Flounder detected in the upper channels, lower channels, and Wacha- preague Inlet. The proportion of fish that occupied a region was determined as the ratio of the number of individual fish identified in that region to the number of fish detected in the system that month. The sum of the proportions of fish in each region could be <1 if not all fish could be assigned objectively to a region in any given month. The sum of the proportions of fish that used each region could be >1 if a single fish occupied more than 1 region in any given month. 75% of tagged fish in the Wachapreague lagoon system did not disperse until early December, and mean residence time was 1.5 times longer (130 days, June-January) than the time previ- ously reported for the Mullica River— Great Bay estuary (86 days, May-December; Sackett et ah, 2008). Seasonal changes in temperature strong- ly influenced residence time, as indicated by the increase in dispersal rates with the seasonal decline in temperature. A similar relationship between water temperature and seasonal mi- gration was observed in winter flounder through the use of passive acoustic telemetry (DeCelles and Cadrin, 2010). On the basis of the life history of Summer Flounder, fish that dispersed from the Wacha- preague lagoon system during the emigration period (after 11 October 2007) were most likely moving offshore to spawn. Our study revealed that smaller fish were more likely than larger fish to leave during the emigration period, con- firming previous reports that larger Summer Flounder commence spawning migrations ear- lier than smaller fish (Smith, 1973). Summer Flounder that dispersed from the Wachapreague system during the emigration period had signifi- cantly greater odds of returning to the system the following year than did those fish that dis- persed during the residency period. The percentage of fish returning to the Wachapreague lagoon system (36%) was similar to the percentage reported for more northern estuaries (25-35% and 39% in New York and New Jersey, respectively; Poole, 1962; Sackett et ah, 2007). Unlike returns in a previous mark- recapture study (Desfosse, 1995), returns to the Wachapreague lagoon system were not detected after April, although the expected battery life of our transmitters would have permitted detec- tion through July 2008. Summer Flounder did return to the Wachapreague lagoon system as early as February, indicating that some fish may actually remain in this system for upwards of 10 months (i.e., from February to the following December). Acoustic telemetry permitted the identifi- cation of early and temporary emigrants from estuaries in this and a previous study (Sackett et ah, 2007). It is possible that early emigrants migrate to the outer continental shelf to spawn, but the timing of these events is much earlier (typically in the early summer) than the tim- ing reported for the spawning migration of this species. Fish that disperse early or temporarily may instead occupy habitats on the inner continental shelf or in other estuaries before final emigration to the outer continental shelf to spawn. On the basis of the confirmed observation of a single fish that was sub- sequently detected in Delaware Bay approximately 2 weeks after tagging, there is at least some movement of Summer Flounder between coastal estuarine sys- tems within the same summer. Previous mark-recap- ture studies have also indicated that Summer Flounder move from Virginia to more northern MAB estuaries 198 Fishery Bulletin 111(2) Table 2 Mean proportions of time Summer Flounder (Paralichthys dentatus ) spent in each region of the Wachapreague lagoon system (upper channels, lower channels, and inlet) by month and period (resi- dency and emigration); proportions of Summer Flounder found in each area by month and period; and numbers of fish present in the system (AO by month and period. The residency period was from 8 June to 10 October 2007, and the emigration period was from 11 October 2007 to 17 January 2008. Activity in tidal flats and supplementary channels was not included because of the low numbers of detections in these areas. Mean proportion of time (%) Proportion of fish (%) Upper Lower Upper Lower N channels channels Inlet channels channels Inlet Month 2007 June 45 97.3 2.0 0.6 95.6 2.3 17.8 July 31 98.9 1.1 0.1 80.7 6.5 6.5 August 30 83.6 15.0 1.4 74.2 16.1 32.3 September 25 61.1 34.4 4.6 44.0 28.0 12.0 October 23 53.7 43.7 2.3 43.5 30.4 21.7 November 16 49.4 46.1 1.6 43.8 50.0 50.0 December 10 49.0 29.1 17.7 30.0 60.0 90.0 2008 January Period 2 32.7 42.7 24.6 50.0 50.0 50.0 Residency 45 86.3 11.5 1.5 97.8 17.8 40.0 Emigration 16 48.9 14.6 5.6 50.0 56.3 87.5 (Lucy and Gillingham4), although not within the same year (as observed in our study). Early or temporary emigration from estuaries may occur in response to en- vironmental cues not monitored in this study, such as barometric pressure or rainfall, or may simply reflect variation in migratory behavior among fish (Sackett et ah, 2007; Henderson, 2012). Future research is needed to investigate the drivers of early and temporary emi- gration and the destination of fish that engage in these behaviors. Within-estuary behaviors The distribution of Summer Flounder in the Wachap- reague lagoon system was comparable to that observed in the Mullica River-Great Bay estuary (Sackett et ah, 2008); in both studies, adult Summer Flounder were primarily detected in the lower bay near the inlet. In our 1-year study, nearly all tagged fish were released in the upper channels (where most fish remained). It is possible that tagged fish released in other regions exhibit fidelity to those regions and that the distribu- tion of tagged Summer Flounder within the system dif- fers by year; however, little difference was observed for Summer Flounder in the region of primary detection 4 Lucy, J. A., and L. Gillingham. 2009. Virginia Game Fish Tagging Program annual report 2008. VIMS Marine Re- source Report No. 2009-4. Virginia Sea Grant Publication No. VSG-09-03, 149 p. [Available from http://web.vims.edu/ library/GreyLit/VIMS/mrr09-04.pdf.] over the 2-year study in the Mullica River-Great Bay estuary (Sackett et ah, 2008). Although adult Summer Flounder occupy a vari- ety of habitats in estuaries, sandier substrates enable these flatfish to bury themselves easily (Bigelow and Schroeder, 1953; Dahlberg, 1972; Orth and Heck, 1980; Roundtree and Able, 1992a). These substrates are of- ten found in areas with high-velocity currents, such as those currents in channels near an inlet. Fishes and crustaceans compose a large portion of the adult Sum- mer Flounder diet (Latour et ah, 2008; Buchheister and Latour, 2011), and higher current velocities most likely deliver more potential prey into an area per unit of time. Summer Flounder have been observed in deep- er areas (-8.5 m) of other MAB estuaries, presumably because of stable environmental conditions (Smith and Daiber, 1977; Sackett et ah, 2008). For future acoustic telemetry studies in the Wachapreague lagoon system and in other estuaries, the effects of release location and year on tagged fish should be considered in order to make inferences about Summer Flounder distribu- tion and habitat preferences within estuaries. The coexistence of behavioral types has been noted in other species and postulated to result in approxi- mately equal fitness among individuals (Bolnick et ah, 2003; Kobler et ah, 2009). Summer flounder appear to fit this pattern. In our study, the majority of Summer Flounder resided primarily in the upper channels, al- though a small group of fish (12 individuals) did use the lower channels. The use of the lower channels in- creased as the study period progressed, and these fish Capossela et al : Migratory and within-estuary behaviors of adult Paralichthys dentatus of the southern mid-Atlantic Bight 199 had a later mean dispersal date than the fish that did not use the lower channels. Divergent patterns of be- havior have been observed in other acoustic telemetry studies on Summer Flounder. In the Mullica River- Great Bay estuary, up to 80% of fish remained in the lower bay near the inlet where they were tagged (Sack- ett et al. 2008), but several fish did move into the river system. At an artificial reef in the Chesapeake Bay, larger Summer Flounder were more likely to stay in close proximity to the reef structure than were smaller fish (Henderson, 2012). The behavior of Summer Flounder in estuaries has been described as sedentary with only minor activity before fall emigration (Desfosse, 1995; Sackett et al., 2008; Henderson, 2012). This description characterized Summer Flounder in the Wachapreague lagoon sys- tem, where fish rarely exhibited large-scale movements (100s of meters) between receivers in the upper chan- nels. However, passive telemetry cannot capture small- scale movements adequately, and fish may have been active within smaller areas (<100s of meters). Active tracking of Summer Flounder in the Mullica River- Great Bay estuary revealed that fish were in motion within small areas (0.18 km2) for most of the time that they were observed and that small-scale movements in deeper waters (-8.5 m) were not related to tidal cur- rents or temperature (Sackett et al., 2008). Small-scale activity was attributed to feeding, competition, or ter- ritorial behaviors (Sackett et al., 2008). We did not ob- serve significant effects of temperature or tidal stage on large-scale (100s of meters) fish activity in the up- per channels during the residency period. Fish in these regions may have an ample supply of prey delivered by the currents and, therefore, may not need to make large-scale movements or use energetically beneficial tidal conditions (e.g., Wirjoatmodjo and Pitcher, 1984; Szedlmayer and Able, 1993; Miller, 2010). During the residency period, fish activity in the up- per channels of the Wachapreague lagoon system was significantly greater at night than during the day. Laboratory-based observations revealed that Sum- mer Flounder are more active during the day (Olla et al., 1972), but such studies considered activity on a much smaller scale (e.g., in a seawater tank that was 10.6x4.5x3.0 m). Similar large-scale (200-400 m) activ- ity of Summer Flounder in the Chesapeake Bay also was greatest at night and influenced by lunar phase (Henderson, 2012). Although benthic foragers (such as Summer Flounder) are generally more light sensitive than are other estuarine pelagic piscivores (Horodysky et al., 2010), the foraging ability of visual predators is most likely limited at night. Therefore, night-time movements may be associated with behaviors other than prey localization and feeding. Conclusions One of the benefits of acoustic telemetry is the ability to identify variation in behavior within a population that renders a species differentially vulnerable to es- tuarine conditions, predation, and harvesting. Differ- ences and similarities in behavior patterns observed for a species by multiple researchers can be used to identify factors that influence such patterns. Our study confirms that, although the life history and migration dynamics of Summer Flounder are well described, in- dividual fish are not uniform in their use of estuaries during summer residencies throughout the MAB. Residence times vary by estuary, indicating that lo- cal conditions are important to population success. Fish size may also effect how long Summer Flounder remain in an estuarine system. As was found in a northern MAB estuary, most tagged Summer Flounder in the Wachapreague lagoon system were sedentary over 100s of meters and remained in deeper (>3 m) waters near the inlet until they undertook the spawning migration (although a small number of individuals did make use of other regions). Further research is needed to consider the effects of release location and year on distribution of tagged Summer Flounder. Studies that combine acoustic moni- toring with the distribution and availability of preda- tors and prey may help explain observed distributions. Establishment of a network of strategic acoustic moni- toring stations within multiple MAB estuaries and along the continental shelf would enable monitoring of fish in these habitats and could help clarify the fate of early or temporary emigrants (Grothues et al., 2005; Able and Grothues, 2007). A better understanding of Summer Flounder habitat preferences and behaviors in estuaries along their range of distribution is essential for protecting areas that promote year-class strength and spawning success. Acknowledgments We thank the following individuals for their assistance with this study: R Bushnell, D. Gauthier, M. Hender- son, J. Smith, and L. Smith. We acknowledge M. Luck- enbach, S. Fate, R. Bonniwell, and the support staff of the Virginia Institute of Marine Science Eastern Shore Laboratory. We also thank T. Targett for his comments on earlier drafts of this manuscript and D. Fox for sharing detections from Delaware Bay. Funding for this project was provided by the Oceanside Conserva- tion Co., Inc., Student Research Grant, and the Eastern Shore Graduate Research Grant. This is contribution 3251 from the Virginia Institute of Marine Science, College of William & Mary. 200 Fishery Bulletin 111(2) Literature cited Able, K. W., R, E. Matheson, W. W. Morse, and G. Shepherd. 1990. Patterns of summer flounder Paralichthys denta- tus early life history in the Mid-Atlantic Bight and New Jersey estuaries. Fish. Bull. 88:1-12. Able, K. W., and S. C. Kaiser. 1994. Synthesis of summer flounder habitat parameters. NOAA Coastal Ocean Program Decision Analysis Series No.l, 68 p. NOAA Coastal Ocean Office, Silver Spring, MD. Able, K. W., and T. M. Grothues. 2007. An approach to understanding habitat dynamics of flatfishes: advantages of biotelemetry. J. Sea Res. 58:1-7. Agresti, A. 2007. An introduction to categorical data analysis, 2nd ed., 372 p. John Wiley & Sons, Inc., Hoboken, NJ. Beck, M. W., K. L. Heck, K. W. Able, D. L. Childers, D. B. Egg- leston, B. M. Gillanders, B. Halpern, C. G. Hays, K. Hoshino, T. J. Minello, R. J. Orth, P. F. Sheridan, and M. P. Weinstein. 2001. The identification, conservation, and management of estuarine and marine nurseries for fish and inverte- brates. BioScience 51:633-641. Bennetts, R. E., J. D. Nichols, J. Lebreton, R. Pradel, J. E. Hines, and W. M. Kitchens. 2001. Methods for estimating dispersal probabilities and related parameters using marked animals. In Disper- sal (J. Clobert, J. D. Nichols, E. Danchin, and A. Dhondt, eds.), p. 2-17. Oxford Univ. Press, Oxford. Bigelow H. B., and W. C. Schroeder. 1953. Fishes of the Gulf of Maine. Fish. Bull. 53:1-577. Bolnick, D. I., R. Svanback, J. A. Fordyce, L. H. Yang, J. M. Davis, C. D. Husley, and M. L. Forister. 2003. The ecology of individuals: incidence and implica- tions of individual specialization. Am. Nat. 161:1-28. Buchheister, A., and R. J. Latour. 2011. Trophic ecology of summer flounder in lower Ches- apeake Bay inferred from stomach content and stable isotope analyses. Trans. Am. Fish. Soc. 140:1240-1254. Capossela, K. M., R. W. Brill, M. C. Fabrizio, and P. G. Bushnell. 2012. Metabolic and cardiorespiratory responses of sum- mer flounder Paralichthys dentatus to hypoxia at two temperatures. J. Fish Biol. 81:1043-1058. Castorlin, M. E., and W. W. Reynolds. 1982. Thermoregulatory behavior and diel activity of yearling winter flounder, Pseudopleuronectes america- nus (Walbaum). Environ. Biol. Fishes 7:177-180. Conrath, C. L. 2005. Nursery delineation, movement patterns, and mi- gration of the sandbar shark, Carcharhinus plumbeus, in the eastern shore of Virginia coastal bays and la- goons. Ph.D. diss., 148 p. College of William & Mary, Gloucester Point, VA. Dahlberg, R. M. 1972. An ecological study of Georgia coastal fish- es. Fish. Bull. 70:323-353. DeCelles, G. R., and S. X. Cadrin. 2010. Movement patterns of winter flounder ( Pseudo- pleuronectes americanus) in the southern Gulf of Maine: observations with the use of passive acoustic telemetry. Fish. Bull. 108:408-419. Desfosse, J. C. 1995. Movements and ecology of summer flounder, Para- lichthys dentatus , tagged in the southern mid-Atlantic Bight. Ph.D. diss., 187 p. College of William & Mary, Gloucester Point, VA. Fabrizio, M. C., and J. P. Pessutti. 2007. Long-term effects and recovery from surgical im- plantation of dummy transmitters in two marine fishes. J. Exp. Mar. Biol. Ecol. 351:243-254. Fleiss, J. L. 1981. Statistical methods for rates and proportions, 2nd ed., 321 p. John Wiley & Sons, Inc., New York. Gibson, R. N. 1994. Impact of habitat quality and quantity on the recruitment of juvenile flatfishes. Neth. J. Sea Res. 32:191-206. Grothues, T. M., K. W. Able, J. McDonnell, and M. M. Sisak. 2005. An estuarine observatory for real-time telemetry of migrant macrofauna: design, performance and con- straints. Limnol. Oceanogr. Methods 3:275-289. Henderson, M. J. 2012. Movements, growth, and mortality of Chesapeake Bay summer flounder based on multiple tagging tech- nologies. Ph.D. diss., 262 p. College of William & Mary, Gloucester Point, VA. Heupel, M. R., J. M. Semmens, and A. J. Hobday. 2006. Automated acoustic tracking of aquatic animals: scales, design and deployment of listening station ar- rays. Mar. Freshw. Res. 57:1-13. Horodysky, A. Z., R. W. Brill, E. J. Warrant, J. A. Musick, and R. J. Latour. 2010. Comparative visual function in four piscivo- rous fishes inhabiting Chesapeake Bay. J. Exp. Biol. 213:1751-1761. Knights, B.C., and B.A. Lasse. 1996. Effects of implanted transmitters on adult blue- gills at two temperatures. Trans. Am. Fish. Soc. 125:440-449. Kobler, A., T. Klefoth, T. Mehner, and R. Arlinghaus. 2009. Coexistence of behavioral types in an aquatic top predator: a response to resource limitation? Oecologia 161:837-847. Kraus, R. T., and J. A. Musick. 2001. A brief interpretation of summer flounder, Para- lichthys dentatus , movements and stock structure with new tagging data on juveniles. Mar. Fish. Rev. 63(3):l-6. Latour, R J., J. Gartland, C. F. Bonzek, and R. A. Johnson. 2008. The trophic dynamics of summer flounder ( Para- lichthys dentatus ) in Chesapeake Bay. Fish. Bull. 106:47-57. Littell, R. C., G. Milliken, W. W. Stroup, R. Wolfinger, and O. Schabenberger. 2006. SAS for mixed models, 2nd ed., 814 p. SAS Insti- tute, Inc., Cary, NC. Malloy, K. D., and T. E. Targett. 1991. Feeding, growth and survival of juvenile summer flounder Paralichthys dentatus : experimental analysis of the effects of temperature and salinity. Mar. Ecol. Prog. Ser. 72:213-223. 1994. The use of RNA:DNA ratios to predict growth limitation of juvenile summer flounder (Paralichthys dentatus) from Delaware and North Carolina estuaries. Mar. Biol. 118:367-375. Miller, M. H. 2010. Movement of summer flounder (Paralichthys den- Capossela et al.: Migratory and within-estuary behaviors of adult Parahchthys dentatus of the southern mid-Atlantic Bight 201 tatus) in relation to hypoxia in an estuarine tributary. Master’s thesis, 103 p. Univ. Delaware, Lewes, DE. Morse, W. W. 1981. Reproduction of the summer flounder, Paralichthys dentatus (L. ). J. Fish Biol. 19:189-203. Murdy, E. O., R. S. Birdsong, and J. A. Musick. 1997. Fishes of the Chesapeake Bay, 324 p. Smithson- ian Inst. Press, Washington, D.C. Necaise, A. M. D., S. W. Ross, and J. M. Miller. 2005. Estuarine habitat evaluation measured by growth of juvenile summer flounder Paralichthys dentatus in a North Carolina estuary. Mar. Ecol. Prog. Ser. 285:157-168. Norcross, B. L., and D. M. Wyanski. 1994. Interannual variation in the recruitment pattern and abundance of age-0 summer flounder, Paralichthys dentatus , in Virginia estuaries. Fish. Bull. 92:591-598. Olla, B. L., C. E. Samet, and A. L. Studholme. 1972. Activity and feeding behavior of the summer floun- der ( Paralichthys dentatus) under controlled laboratory conditions. Fish. Bull. 70:1127-1136. Orth, R. J., and K. L. Heck Jr. 1980. Structural components of eelgrass ( Zostera ma- rina) meadows in the lower Chesapeake Bay — fishes. Estuaries 3:278-288. Pollock, K. H., S. R. Winterstein, C. M. Bunck, and P. D. Curtis. 1989. Survival analysis in telemetry studies: the stag- gered entry design. J. Wild 1. Manage. 53:7-15. Poole, J. C. 1962. The fluke population of Great South Bay and their significance to the sport fishery. N.Y. Fish Game J. 9:93-117. Ray, G. C. 2005. Connectivities of estuarine fishes to the coastal realm. Estuar. Coast. Shelf Sci. 64:18-32. Richards, C. E., and M. Castagna. 1970. Marine fishes of Virginia’s Eastern Shore (in- let and marsh, seaside waters). Chesapeake Sci. 11:235-248. Rogers, K. B., and G. C. White. 2007. Analysis of movement and habitat use from telem- etry data. In Analysis and interpretation of freshwa- ter fisheries data (M. L. Brown, and C. S. Guy, eds.), p. 625-676. Am. Fish. Soc., Bethesda, MD. Roundtree, R. A., and K. W. Able. 1992a. Fauna of polyhaline marsh creeks in southern New Jersey: composition, abundance and biomass. Es- tuaries 15:171-186. 1992b. Foraging habits, growth, and temporal patterns of salt-marsh creek habitat use by young-of-the-year summer flounder in New Jersey. Trans. Am. Fish. Soc. 121:765-776. Ryan, S. E., L. S. Porth, and C. A. Troendle. 2007. Defining phases of bedload transport using piecewise re- gression. Earth Surf. Process. Landforms 27: 971-990. Sackett, D. K., K. W. Able, and T. M. Grothues. 2007. Dynamics of summer flounder Paralichthys denta- tus, seasonal migrations based on ultrasonic telemetry. Estuar. Coast. Shelf Sci. 74:119-130. 2008. Habitat dynamics of summer flounder Paralich- thys dentatus within a shallow USA estuary based on multiple approaches using acoustic telemetry. Mar. Ecol. Prog. Ser. 364:199-212. Schwartz, F. J. 1964. Fishes of Isle of Wight and Assawoman Bays near Ocean City, Maryland. Chesapeake Sci. 5:172-193. Smith, W. G. 1973. The distribution of summer flounder, Paralichthys dentatus, eggs and larvae on the continental shelf be- tween Cape Cod and Cape Lookout, 1965-66. Fish. Bull. 71:527-548. Smith, R. W., and F. C. Daiber. 1977. Biology of summer flounder, Paralichthys dentatus, in Delaware Bay. Fish. Bull. 75:823-830. Stierhoff, K. L., T. E. Targett, and K. Miller. 2006. Ecophysiological responses of juvenile summer and winter flounder to hypoxia: experimental and mod- eling analyses of effects on estuarine nursery quality. Mar. Ecol. Prog. Ser. 325:255-266. Stierhoff, K. L., T. E. Targett, and J. H. Power. 2009. Hypoxia-induced growth limitation of juvenile fishes in an estuarine nursery: assessment of small- scale temporal dynamics using RNA:DNA. Can. J. Fish. Aquat. Sci. 66:1033-1047. Szedlmayer, S. T., and K. W. Able. 1993. Ultrasonic telemetry of age-0 summer flounder, Paralichthys dentatus, movements in a southern New Jersey estuary. Copeia 3:728-736. Tyler, R. M. 2004. Distribution and avoidance patterns of juvenile summer flounder ( Paralichthys dentatus) and weakfish ( Cynoscion regalis) in relation to hypoxia: field studies in a temperate coastal lagoon tributary and laboratory choice-trial experiments. Ph.D. diss., 183 p. Univ. Delaware, Newark, DE. Wiijoatmodjo, S., and T. J. Pitcher. 1984. Flounders follow the tides to feed: evidence from ultrasonic tracking in an estuary. Estuar. Coast. Shelf Sci. 19:231-241. Wyanski, D.M. 1990. Patterns of habitat utilization in 0-age summer flounder. Master’s thesis, 55 p. College of William & Mary, Gloucester Point, VA. 202 Fishery Bulletin 111(2) Fishery Bulletin Guidelines for authors Manuscript preparation Contributions published in Fishery Bulletin describe original research in marine fishery science, fishery en- gineering and economics, as well as the areas of ma- rine environmental and ecological sciences (including modeling). Preference will be given to manuscripts that examine processes and underlying patterns. Descriptive reports, surveys, and observational papers may occa- sionally be published but should appeal to an audience outside the locale in which the study was conducted. Although all contributions are subject to peer review, responsibility for the contents of papers rests upon the authors and not on the editor or publisher. Submission of an article implies that the article is original and is not being considered for publication elsewhere. Articles may range from relatively short contributions (10-15 typed, double-spaced pages [tables and figures not in- cluded!) to extensive contributions (20-30 typed pages). Manuscripts must be written in English; authors whose native language is not English are strongly advised to have their manuscripts checked by English-speaking colleagues before submission. Title page should include authors’ full names and mailing addresses and the senior author’s telephone, fax number, and e-mail address. Abstract should be limited to 250 words (one-half typed page), state the main scope of the research, and emphasize the authors conclusions and relevant findings. Do not review the methods of the study or list the contents of the paper. Because abstracts are circulated by abstracting agen- cies, it is important that they represent the research clearly and concisely. General text must be typed in 12-point Times New Roman font throughout. A brief introduction should convey the broad significance of the paper; the remain- der of the paper should be divided into the following sections: Materials and methods, Results, Discus- sion, Conclusions, and Acknowledgments. Headings within each section must be short, reflect a logical se- quence, and follow the rules of subdivision (i.e., there can be no subdivision without at least two subhead- ings). The entire text should be intelligible to interdisci- plinary readers; therefore, all acronyms, abbreviations, and technical terms should be written out in full the first time they are mentioned. For general style, follow the U.S. Government Print- ing Office Style Manual (2008) [available at http://www. gpoaccess.gov/stylemanual/index.htmlI and Scientific Style and Format: the CSE Manual for Authors , Edi- tors, and. Publishers (2006, 7th ed.) published by the Council of Science Editors. For scientific nomenclature, use the current edition of the American Fisheries So- ciety’s Common and Scientific Names of Fishes from the United States, Canada, and Mexico and its compan- ion volumes (Decapod Crustaceans, Mollusks, Cnidaria and Ctenophora, and World Fishes Important to North Americans). For species not found in the above men- tioned AFS publications and for more recent changes in nomenclature, use the Integrated Taxonomic Informa- tion System (ITIS) (available at http://itis.gov/), or, sec- ondarily, the California Academy of Sciences Catalog of Fishes (available at http://researcharchive.calacademy. org/research/ichthyology/catalog/fishcatmain.asp) for species names not included in ITIS. Citations must be given of taxonomic references used for the identification of specimens. For example, “Fishes were identified by using Collette and Klein-MacPhee (2002); sponges were identified by using Stone et al. (2011).” Dates should be written as follows: 11 November 2000. Measurements should be expressed in metric units, e.g., 58 metric tons (t); if other units of measure- ment are used, please make this fact explicit to the reader. Use numerals, not words, to express whole and decimal numbers in the general text, tables, and fig- ure captions (except at the beginning of a sentence). For example: We considered 3 hypotheses. We collected 7 samples in this location. Use American spelling. Re- frain from using the shorthand slash (/), an ambiguous symbol, in the general text. Word usage and grammar that may be useful are the following: Aging For our journal the word aging is used to mean both age determination and the aging process (se- nescence). The author should make clear which meaning is intended where ambiguity may arise. Fish and fishes For papers on taxonomy and biodiver- sity, the plural of fish is fishes, by convention. In all other instances, the plural is fish. Examples: The fishes of Puget Sound [biodiversity is indicated]; The number of fish caught that season [no emphasis on biodiversity]; The fish were caught in trawl nets [no emphasis on biodiversity]. The same logic applies to the use of the words crab and crabs, squid and squids, etc. Sex For the meaning of male and female, use the word sex, not gender. Participles As adjectives, participles must modify a specific noun or pronoun and make sense with that noun or pronoun. Incorrect: Using the recruitment model, estimates of age-1 recruitment were determined. [Estimates did not use the recruitment model.] Correct: Using the recruitment model, we deter- mined age-1 estimates of recruitment. [The participle now modifies the word we, those who were using the model.] Guidelines for authors 203 Incorrect: Based on the collected data, we concluded that the mortality rate for these fish had increased. [We were not based on the col- lected data.] Correct: We concluded on the basis of the collected data that the mortality rate for these fish had increased. [Eliminate the participle and replace it with an adverbial phrased Equations and mathematical symbols should be set from a standard mathematical program (MathType) or tool (Equation Editor in MS Word). LaTex is accept- able for more advanced computations. For mathemati- cal symbols in the general text (a, %2, n, ±, etc.), use the symbols provided by the MS Word program and itali- cize all variables. Do not use photo mode when creating these symbols in the general text. Literature cited section comprises published works and those accepted for publication in peer-reviewed journals (in press). Follow the name and year system for citation format in the “Literature cited” section (that is to say, citations should be listed alphabetically by the authors’ last names, and then by year if there is more than one citation with the same authorship. Abbrevia- tions of serials should conform to abbreviations given in Cambridge Scientific Abstracts (http://www.csa.com/ ids70/serials_source_list.php?db=aquclust-set-c). Authors are responsible for the accuracy and com- pleteness of all citations. Literature citation format: Author (last name, followed by first-name initials). Year. Title of article. Abbreviated title of the journal in which it was published. Always include number of pages. If there is a sequence of citations in the text, list chrono- logically: (Smith, 1932: Green. 1947; Smith and Jones, 1985). Digital object identifier (doi) code ensures that a publication has a permanent location online. Doi code should be included at the end of citations of published literature. Do not punctuate the code with a final sen- tence period, e.g., doi.lO.xxxxx.xx.x. Cite all software and special equipment or chemical solutions used in the study within parentheses in the text (e.g., SAS, vers. 6.03, SAS Inst., Inc., Cary, NC). Footnotes are used for all documents that have not been formally peer reviewed and for observations and communications. These types of references should he cited sparingly in manuscripts submitted to the journal. All reference documents, administrative reports, inter- nal reports, progress reports, project reports, contract reports, personal observations, personal communica- tions, unpublished data, manuscripts in review, and council meeting notes are footnoted in 9 pt font and placed at the bottom of the page on which they are first cited. Footnote format is the same as that for formal literature citations. A link to the online source (e.g., [http://www/ , accessed July 2007.]), or the mail- ing address of the agency or department holding the document, should be provided so that readers may ob- tain a copy of the document. Tables are often overused in scientific papers; it is seldom necessary or even desirable to present all the data associated with a study. Tables should not be ex- cessive in size and must be cited in numerical order in the text. Headings should be short but ample enough to allow the table to be intelligible on its own. All un- usual symbols must be explained in the table legend. Other incidental comments may be footnoted with italic numeral footnote markers. Use asterisks only to indi- cate significance in statistical data. Do not type table legends on a separate page; place them above the table data. Do not submit tables in photo mode. Figures must be cited in numerical order in the text. Graphics should aid in the comprehension of the text, but they should be limited to presenting patterns rather than raw data. Figures should not exceed one figure for every four pages of text. Figures must be la- beled with the number of the figure. Place labels A, B, C, etc. within the upper left area of graphs and photos. Avoid placing labels vertically (except for the y axis). Figure legends should explain all symbols and abbre- viations seen in the figure and should be double-spaced on a separate page at the end of the manuscript. Color is allowed in figures to show morphological differences among species (for species identification), to show stain reactions, and to show gradations in temperature con- tours within maps. Color is discouraged in graphs, and for the few instances where color may be allowed, the use of color will be determined by the Managing Editor. • Notate probability with a capital, italic P. • Provide a zero before ail decimal points for values less than one (e.g., 0.07). • Capitalize the first letter of the first word in all la- bels within figures. • Do not use overly large font sizes in maps and for units of measurements along axes in figures. • Do not use bold fonts or bold lines in figures. • Do not place outline rules around graphs. • Use a comma in numbers of five digits or more (e.g., 13,000 but 3000). • Place a North arrow and label degrees latitude and longitude (e.g., 170°E) in maps. • Use symbols, shadings, or patterns (not clip art) in maps and graphs. Failure to follow these guidelines and failure to correspond with editors in a timely manner will delay publication of a manuscript. Copyright law does not apply to Fishery Bulletin, which falls within the public domain. However, if an author reproduces any part of an article from Fishery Bulletin in his or her work, reference to source is con- sidered correct form (e.g.. Source: Fish. Bull 97:105). 204 Fishery Bulletin 111(2) Submission Submit manuscript online at http://mc.manuscriptcentral. com/fisherybulletin. Commerce Department authors should submit papers under a completed NOAA Form 25-700. For further details on electronic submission, please contact the Associate Editor, Kathryn Dennis, at kathryn.dennis@noaa.gov When requested, the text and tables should be submit- ted in Word format. Figures should be sent as PDF files (preferred), Windows metafiles, TIFF files, or EPS files. Send a copy of figures in the original software if con- version to any of these formats yields a degraded ver- sion of the figure Questions? If you have questions regarding these guidelines, please contact the Managing Editor, Sharyn Matriotti, at sharyn.matriotti@noaa.gov Questions regarding manuscripts under review should be addressed to Kathryn Dennis, Associate Editor. Fishery Bulletin Subscription form Superintendent of Documents Publications Order Form *5178 1 ! YES, please send me the following publications: Subscriptions to Fishery Bulletin for $32.00 per year ($44.80 foreign) The total cost of my order is $ . Prices include regular domestic postage and handling and are subject to change. (Company or Personal Name) (Please type or print) (Additional address/attention line) (Street address) (City, State, ZIP Code) (Daytime phone including area code) (Purchase Order No.) Charge your order. IT’S EASY! Please Choose Method of Payment: ] Check Payable to the Superintendent of Documents Q GPO Deposit Account -□ ] VISA or MasterCard Account To fax your orders (202) 512-2104 (Credit card expiration date) (Authorizing Signature) Mail To: U.S. Government Printing Office P.O. Box 979050, St. Louis, MO 63197-9000 Thank you for your order! Also available online at http://bookstore.gpo.gov/actions/GetPublication.do7stocknumber =703-023-00000-2 SMITHSONIAN INSTITUTION LIBRARIES