Volume 19 March 2007 ISSN; 1528-0470 GULF AND CARIBBEAN RESEARCH Gulf and Caribbean Research Volume 19 Issue 1 January 2007 Habitat Use by Juvenile Gag^ Mycteroperca microlepis (Pisces: Serranidae)^ in Subtropical Charlotte Harbor, Florida (USA) J. Patrick Casey Charlotte Harbor Field Laboratory Gregg R. Poulakis Charlotte Harbor Field Laboratory Philip W Stevens Charlotte Harbor Field Laboratory DOI: 10.18785/gcr.l901.01 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr & Part of the Marine Biology Commons Recommended Citation Casey,}., G. R. Poulakis and P. W. Stevens. 2007. Habitat Use by Juvenile Gag, Mycteroperca microlepis (Pisces: Serranidae), in Subtropical Charlotte Harbor, Florida (USA). Gulf and Caribbean Research 19 (l)= L9. Retrieved from http:// aquila.usm.edu/ gcr/voll9/ issl/ 1 This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized editor of The Aquila Digital Community. For more information, please contact Joshua.Cromwell^usm.edu. HABITAT USE BY JUVENILE GAG, MYCTEROPERCA MICROLEPIS (PISCES: SERRANIDAE), IN SUBTROPICAL CHARLOTTE HARBOR, FLORIDA (USA) J* Patrick Casey^^ Gregg M. Pci iilald^ , and Philip W* Stgvees Florida Fish md Wildlife Conservation Commission, Fish and Wldiife Research Institute, Charlotte Harbor Field Labomtory, 1481 Market Circle, Unit I, Fort Charlotte, Florida 33953- 3815 USA ^Correspondmg author Phone (941)255-7403, Fax (941)255-7400, E-mail patrictcasey@ MyFWC,cam ABSTRACT Estuaries piay a key role in the juvemle of gag (Mycteropema micmiepisf use of estuarme habitats by juvenile gag hgiK been examined in temperate estuaries, which are at the noithem limits of the range of this species, but the importance of subtropical estuaries during the ^ly life histoiy of this species has not l^en studied extensively. Gag weie collected in subtropical Charlotte Harbor, Honda, during routme monthly sampling &om January 1996 to D^iember 2CXI2. Juvenile gag were collected using a 2L3"m seine, a 183-m haul seine, and a 183-m pame seine. A total of 738 individuals mnging from 30 to 489 mm standard length (SL) were collected in 4,480 samples. Most gag (96%) were probably young-of-the-year (< 288 mm SL). The majority of juveniles were collected in polyhaline Gasparilla and Pine Island sounds from April to December, with a few lar^r individuais captured year-round. The observe period of gag settlement was similar to that reported in other subtropical and tempemte estuaries, but gag in Charlotte Harbor remained in the estuary longer and egressed at a larger sire than did gag in other ^tnaries. Relative abundance of juvenile gag within Charlotte Harbor was greatest on shallow s^grass shoals but was also high along fringing mangrave shorelines, which is a habitat not previously re^rrted for gag. Inteobuchon Estuaries play a key role in the juvenile stage of gag (Myctewpema mictvlepis) (Keener et si. 1988). The species ranges from New York to Brazil, including the entire Golf of Mexico (COM), and juveniles have been reported to occur in temperate and subtropical estuar- ies from Virgmia to the northeastern GOM (Hoese et al. 1961, Hood and Schlieder 1992, Koenig and Coleman 1998). This economically important serranid spawns in large aggregations, such as those found at tradidonal West Florida Shelf sites in. the GOM, primarily during February and March (Hood and Schlieder 1992, Collins et al. 1998). The planktonic l^ae move into estuaries and settle out at about 15 mm standard length (SL) in the first avaii- able habitat, such as polyhaline seagrass beds and oyster shell habitats near inlets and mouths of tidal creeks (Ross and Moser 1995, MuUaney and Gale 1996). As juvenile gag grow rapidly during their estuarine residence, they may also use manmade habitats like seawalls and jetties (Hastings 1979, Bullock and Smith 1991). Latitudinal differences in climate appear to affect the duration of estuarine residence and size attained by juvemle gag before they disperse to non-estuarine habi- tats (Ross and Moser 1995). Juveniles ere usually found in North Carolina estuaries from April to September and in estuaries along the northeastern GOM from April to October (Ross and Moser 1995, Koenig and Coleman 1998), The fi.rst cold front of fall is thought to trigger th.eir egress to deeper ocean water (e.g,, Ross and Moser 1995). In temperate estuaries such as Bogue Sound, North Carolina, maximum reported size does not exceed 200 mm SL (Ross and Moser 1995), but in subtropical estuaries such as Ikmpa Bay, Florida, gag can reach 360 mm SL (Hood and Schlieder 1992). Habitat use by juvenile gag has been examined in temperate estuaries, which are at the northern timits of the range of this species; however, the importance of subtropi- cal estuaries during the early life history of this species has not been studied extensively. Despite the effects of increas- ing urbanization and the resultant demands for freshwater resonrees, Charlotte Harbor, a large subtropical estuary in southwestern Florida, supports many suitable habitate for juvenile gag (e.g,, seagrass beds, oyster shell habitats). Although juveniles have been collected from seagrass beds within the estuary (Wang and Raney 1971, Hanson et al. 2CM)4, Fitzhugh et al. 2005), questions regarding aspects of habitat use, especially use of tropieal climate habitats like mangroves, have not been examined. Thus, Ihe objec- tive of tiiis study to use an estuary-wide dataset from a long-term fisheries-independent monitoring progrmn in Charlotte Harbor to examine distribution, seasonality, habitat use, and relative abundance of juvenile gag in a subtropical estuary. Matehmls and Methods S tudy loeatiinii Charlotte Harbor, located on the southwestern coast of Florida, is one of the largest estuaries in Florida and Casey et al. is separated from the GOM hy a series of barrier islands. IWo large inlets, Boca Grande Pass and San Carlos Pass, and foiir smaller inlets allow tidal water exchange. The modal depth of the estuary is 3^ m, and the deepest point is 15.5 m in Boca Grande Pass (Huang 1966). The climate is subtropical, with infrequent freezes and distinct wet and dry seasons. Mean annual rainfail is 127 cm, 60% of which fails between June and September (Taylor 1974), whereas mean water temperature is 25 ranging from 12“C to 36“ C, and mean salinity is 29 psu, ranging from 5 psu to 40 psu (PoulaMs et al. 2003, present study). Charlotte Harbor supports a variety of habitats that are used by at least 255 species of fish (PoulaJds et al. 2004). The 2 predominant habitats for fishes are seagrass flats and fringing mangroves. Red mangrove {Ekizophom fmmgle% white mangrove {Luguncularia nwemosd)^ and black mangrove {Mdcennia germinam) are the 3 species found in Charlotte Harbor, but red mangroves dominate along the shoreline (143 km^; Poulakis et al. 2003). Turtle grass ijhaiassia testudinmn), shoal grass {Ealodule wrightii), juid manatee grass {Syringodium fiUforme) the most common seagrass species in the estuary (262 km^; Sargent et al. 1995). Other habitats foimd in Charlotte Harbor include oyster bars, sandy shoals (areas away from shore that are <0.5 m deep at mean low tide), seawalls, and bridge pilings. Sample collectieii Fish abundance and habitat data collected through- out Charlotte Harbor by the Florida Fish and Wildlife Conservation Commission (FWC), Fish and Wildlife Research Institute's Fisheries-Indepeiideiit Monitoring program from January 1996 to December 2002 were ana- lyzed for this study Monthly stratified-raudom sampling was conducted during the day by using 3 different seines. Between 17 and 32 samples were completed each month for each gear, with effort distribufrd equally throughout the study area. The estuary was divided into 1x1 nautical- mile cartographic grids (1 nm^), and grids with appropri- ate water depths for each seine 1.5 m for 21.3 m seine, < 2.5 m for 183-m haul seine, ^ 3.3 m for 183-m purse seine) were selected as the samplmg universe. Using a 10 X 10 cell overlay, each cartographic grid was subdivided into 100 microgrids (0.1 x 0.1 nm), which represented the potential sample sites that were randomly selected without replacement each month. Samples were stratified by habitat type depending on gear. The ll.S-m center-bag seine (21.3m x l.B m, 3.2-mm stretch mesh) was pulled along shorelines and cm shoals away from shore (PoulaJds et al. 2003). Samples collected with the 183~m center-bag haul seine (183 m x 3 m, 37.5- mm stretch mesh) were stratified based on the presence or absence of overhanging shoreline vegetation (e.g., fring- ing mangroves). This seine was deployed by boat, in. a rectangular shape (40 m x 103 m), along shorelines and on offshore flats inside the estuary and retrieved by hand (Kupschus and Tremain 2001). The ISS-m terminal-bag purse seine (183 m x 5.2 m, SO-mm stretch mesh) was set at least 40 m from the shoreline and was retrieved with the aid of a motorized hydraulic system (Wessel and Winner 2003). All fishes were identified to the lowest possible taxon and enumerated. Up to 40 fishes were measured to the nearest millimeter SL, and aU juvemle gag were released alive in the field. The bottom type, seagrass spe- cies, shoreline vegetation species, and coverage (%) of each sample were qualitatively measured. Saliniiy (psu), dissolved oxygen (mgl"^), and water temperature (°C) were r^orded with a hand-held data sonde. A vertical profile of these parameters was taken at the surface (0.2 m below surface) and at each whole meter increment nntil reachi.ug the bottom (0,2 m above bottom). Data analyiis The locations of captured juvenile gag were plotted by gear type to examine distribution Ihroughout Ihe estuary. Three size-classes were plotted separately (< 100 mm, 101 < 250 mm, and >251 mm) to explore possible differences in ontogenetic distribution within the estuary. To examine seasonality, length-fi*equency date were divided into 10- mm size-classes and pooled by month and year for each gear to examine month of settlement, growth and relative abundance during estuarine residency, month of egress from estiiaiy, and gear selectivity. The gear type and areas within Charlotte Harbor where gag were most abundant were analyzed further by analysis of covariance (ANCOVA), which was performed by using a general, linear modeling (GLM) approach, to investigate the influence of water temperature, salin- ity, water depth, overall seagrass coverage, bottom type, shoreline type, year, month, and geographic location on the gear-specific relative abraidance of gag (PROC GLM; SAS Institute 1988). The relative abundance of gag was ln(x -hi) transformed before analysis. Water temperature, saiinity, water depth, and overall seagr^s percent coverage were the covariates (continu- ous variables) and were hi(x + 1) transformed to stabilize the variance in the data before analysis. The value of each abiotic covariate used in the model was the mean of all readings taken at each sample location. We tested for parallelism in the model by plotting each covariate against gag relative abundance to ensure similar slopes. 2 Casey et al. Bottom type sand^ mod, oyster)^ shofcHoe vs. shoal, year, month, and geographic location were the cIms variables (categorical variables) in the model Samples were excluded from ail habitat analyses when the overall seagrass coverage could not be estimated due to poor water clarity or water depth (3.5% of all samples). The geographic location variable included data west of the dot- ted line in Figure 1 and was defined as areas north of Boca Grande Pass (Gasparilla Sound) and south of Boca Grande Pass (Pine Island Sound). For the shoreline vs. shoal class variable, “shoreline” was defined as the habitat at the land-water interface (i.e., mangroves, beach, seawall), and “shoal” was defined as areas that were ^ 0.5 m deep at mean low tide and were at least 5 m from the shoreline (i.e., oyster bar, sand bar). We constructed a full model that included all class variables and covariates and then simplified it using a step- wise elimmation procedure. The variables with the highest P values were removed from the model one at a time until all remaioing variables were significant at a -0,01, The sigoificaiice level of 0.01 was used to mlnimixe the pos- sibility of Type I error. We report only significant class variables and covariates in the results. Tukey’s Studenlimi Range (HSD) test was used post-hoc to determine where differences occmxed in each significant variable (Zar 1999). Results; A total of 738 juvenile gag ranging from 30 to 489 mm SL were collected in 4,480 samples (Ihble 1). Most gag (96%) were probably young-of-the-year (< 288 mm SL), based on data presented in Hood and Schlieder (1992). Most juveniles were collected in the 183-m haul seine (u = 615) and ranged from 88 to 440 mm SL (mean^ 183 mm). The purse seine collected individuals {n = 83) ranging from 50 to 489 m m SL (mean ~ 204 mm), and the 2L3-m seine collected juveniles {n = 40) ranging from 30 to 204 mm SL (mean ~ 138 mm). Although samples were taken throughout the estuary (note locations of zero catches in Figure 1), gag were col- lected principally (95%) in polyhaline Gasparilla and Pine Island sounds (Table 1, Figure 1). Distribution plots of different size classes of gag showed that their distribution did not change as they grew, regardless of size or month, so aU. gag are included in Figure 1 . The salmity (mean ± s^) where gag were captured was relatively consistent (31.0 ± 0.4 psu, range = 13-40 psu). Gag were coEected mainly from April to December, but some individuals were captured in all months (Figure 2). Juvenile gag rangmg from 30 to 88 mm SL were cap- tured in April, and May. The cohorts grew and accumulated in numbers during June and My, with the highest number of gag collected in September. Although most individuals were captured from May to December in the haul seine, increasing numbers of individuals were collected in the purse seine in October and November as overall numbers declined in tiie estuary. Some of the previous years' cohorts (> 288 mm SL) were coEected year-round (Figure 2). The mean water temperature (±j^) where gag were captured was 27.7 ± 0.3“C (range = 14,5™33.5“C). Because 78% of aE gag were captured in ttie haul seine from May to December iu GaspariEa and Pine Island sounds (Table 1), we used only these data in the general Eneai model to examine specific habitat use in areas and times when gag were most abundant- The variables that significantly affected gag abundance were geographic location (Pine Island Sound vs. Gasparilla Sound), year, month, water depth, shoreline vs. shoal, and overall sea- grass percent cover (ANCOVA; r^ = 0.25; Thble 2). In Gasparilla and Kne Island sounds, the annual, relative abundance of juvenile gag in 2002 was at least 2.7 gag per haul greater than in other years (Thkey's Studenlized Range test; F<0.05) (Figure 3). Gag abundances in GaspariEa and Pine Island sounds were significantly lower in May and December than in June through November (Tukey's Studentized Range test; F<0.05). Gag were significantly more abundant in GaspariEa Sound than they were in Pine Island Sound (Tukey's Studentized Range test; P < 0.05). Mean water temperature and salinity varied Mttie during May to December in both Gasparilla and Pine Island sounds and did not contribute to the model. In GaspariEa and Pine Island sounds, juvenile gag were coEected priscipaEy in habitats that contained S: 50% overall seagrass coverage (Figure 4). Relative abundance on shoals was 2.9 fish per haul greater than near mangrove and beach shorelines (Tnfcey’s Studentized Range test; P < 0.05). Only 8 of the shoal samples were on oyster bars (19 gag collected); the other 30 samples were on shoals that had > 50% overall seagrass coverage (133 gag col- lected). The majority of sample sites {n ~ 280) were along mangrove shorelines that had ^ 50% overaE seagrass cov- erage, and that is where most of the JuvenEe gag {n = 226) were coEected. The cateh-per-unit-effort increased from 0.3 to 1.4 gag per haul when the bag depth was greater flian one meter. Discussion Juvemie gag are lypkaEy concentrated in poIyhaEne areas close to passes, and these areas apparently repre- sent the first suitable environments that prcsettlement 3 Casey et al. Figure 1. Distribution and relative abundance (abundance index = number of fish haul'^) of juvenile gag in Charlotte Harbor, Florida. Samples (n = 4,480) were collected throughout the estuary, but most (95%) individuals were captured in Gasparilla and Pine Island sounds (areas west of dotted line). 4 Casey et al. TABLE 1 Summary of samples collected from 1996 to 2002 in Charlotte Harbor, Florida. The 2 sample regions are separated by the dotted line in Figure 1. (f) = number of samples that captured gag. Gear Eastern Charlotte Harbor Gasparilla & Pine Island sonnds Total Total samples No. of gag (f) Total samples No. of gag (f) Total samples No. of gag (f) 21.3-m seine 1,192 7(4) 872 33 (23) 2,064 40 (27) 1 83-m haul seine 712 33 (8) 644 582 (121) 1,356 615(129) 1 83-m purse seine 530 1(1) 530 82 (34) 1,060 83 (35) Total 2,434 41 (13) 2,046 697 (178) 4,480 738 (191) gag encounter when they move into estuaries throughout their range (Keener et al. 1988, Ross and Moser 1995). Gag spawn principally during February and March in the GOM (Hood and Schlieder 1992, Collins et al. 1998). Larvae remain in the plankton for about 40 d, and juveniles typically settle in temperate estuarine habitats in April and May (Keener et al. 1988, Ross and Moser 1995, Collins et al. 1998, Fitzhugh et al. 2005). Our data indicate that juvenile gag also moved into subtropical Charlotte Harbor and settled during April and May. Due to the shape and hydrologic regime of Charlotte Harbor (rivers located far from passes and expansive poly- haline sounds), juvenile gag of various size classes were concentrated in high-salinity areas near Gasparilla and Boca Grande passes but also inhabited shallow areas in Gasparilla and Pine Island sounds several kilometers away from the GOM. In contrast, previous research determined that juvenile gag in temperate estuaries were concen- trated in tidal creeks and seagrass beds near inlets (Ross and Moser 1995, Mullaney and Gale 1996, Koenig and Coleman 1998, Heinisch and Fable 1999). Exclusive use of polyhaline areas in estuaries by different size-classes has typically been attributed to low mobility during estua- rine residency (Koenig and Coleman 1998, Heinisch and Fable 1999). However, it is unclear whether distribution is dependent solely upon settlement patterns or if survival decreases in lower salinities. Within the polyhaline areas of estuaries, gag have been collected near seagrass beds, oyster-shell habitats, mangroves, seawalls, and jetties (Hastings 1979, Bullock and Smith 1991, Mullaney and Gale 1996, Koenig and Coleman 1998, present study). In Charlotte Harbor, juve- niles were most abundant where water depths were > 1 m on seagrass-covered shoals and along mangrove-lined shorelines — the dominant habitats in Gasparilla and Pine Island sounds. In estuaries where seagrasses are absent, juveniles typically have been collected from oyster-shell habitats in high-salinity tidal creeks (Mullaney and Gale 1996). Therefore, high-salinity habitats that provide struc- ture appear to be preferred by juvenile gag during their estuarine residency throughout their range. One habitat that has received little attention, but provides considerable structure and large areas of suit- able habitat for juvenile gag, is fringing mangroves. A consistent number (low s- for shoreline, see Figure 4) of juvenile gag were collected along fringing mangroves in this study. The dominant species of fringing mangrove in Charlotte Harbor is the red mangrove, which provides structure for fish assemblages in the form of prop roots and overhanging branches that extend into the water away from the shoreline (Thayer et al. 1987, Ley et al. 1999, Poulakis et al. 2003). Because of the large area (ca. 4,120m2) and multiple habitats (e.g., seagrass beds, fringing mangroves) encompassed by the haul seine, the exact habitat where juvenile gag resided could not be determined using this gear. However, a hook-and-line study that targeted com- mon snook (Centropomus undecimalis) captured juvenile gag as bycatch (D.A. Blewett, unpublished data. Fish and Wildlife Research Institute, Charlotte Harbor Field Laboratory). During hook-and-line sampling, juvenile gag were extracted from among red mangrove prop roots, providing evidence that they use fringing red mangrove habitats. Studies conducted in temperate estuaries have indicat- ed that the passage of cold fronts in September and October trigger the egress of juvenile gag from estuarine waters to open-ocean waters (Ross and Moser 1995, Koenig and Coleman 1998). Our data showed that juveniles began to decline in abundance throughout Charlotte Harbor from October to December. Before dispersing offshore, juvenile gag appeared to first move to deeper open waters within the estuary during October and November, as indicated by the increased number of juvenile gag captured in the purse seine, which samples deeper habitats away from shore. Although Charlotte Harbor becomes affected by cold fronts in September and October, the effects of those 5 Casey et al. 30 - 25 - 20 - 15 - 10 - 5- JANUARY- MARCH n = 4 0 30 -\ APRIL 20 - 15 - 10 - 5- 0 ■ n HuFin . Ui (0 o E 3 MAY ■ n = 35 - ..HmI I In nrm.n.MFi 0 30 n JUNE 25 - a7 = 60 20 - 15 - 10 - 5- 0 30 25 ■ n . nMn . M . n . 30 n 25 20 - 15 - 10 - 5 0 30 25 20 15 10 5 AUGUST n = 119 ni L nTnn . nMMn . SEPTEMBER n = 195 0 30 - 25 - 20 - 15 - 10 - 5 - mi* n . OCTOBER n = 150 0 30 25 20 15 10 - 5 - 0 Ml ffl-, NOVEMBER n = 7^ rm. J 11(1, ■ ■■■ n . n . Standard length (mm) Figure 2. Monthly size distribution of juvenile gag in Charlotte Harbor, Florida (1996-2002). Black = 21.3-m seine, White 183-m haul seine. Hatched = 183-m purse seine. Casey et al. TABLE 2 The significant (a <0.01) variables identified by the general linear model analysis as contribnting to the abnndance of jnvenile gag captnred in the 183-m hanl seine in Gasparilla and Pine Island sonnds from May to December. Source df Sum of squares F P Model 17 56.82 8.10 <0.0001 0.25 Month 7 10.67 3.70 0.0007 Year 6 17.51 7.07 <0.0001 Geographic location 1 3.91 9.48 0.0022 Shoreline vs. shoal 1 3.79 9.18 0.0026 Water depth 1 12.62 30.59 <0.0001 Seagrass percent cover 1 8.71 21.12 <0.0001 Error 407 167.90 Corrected total 424 224.72 cold fronts are milder than at higher latitudes. Therefore, because juvenile gag remain in subtropical estuaries like Charlotte Harbor longer than they do in temperate estuar- ies, they may attain larger sizes before egressing. Movement toward the ocean is enhanced by cold fronts, but these fronts are probably not the only cue used by gag. Previous studies have suggested that a few individuals may begin egressing to open ocean waters before water temperatures are lowered by cold fronts. For example, in Bogue Sound, North Carolina, and St. Andrews Bay, Florida, juveniles were observed along jet- ties in inlets several weeks before the first cold front (Ross and Moser 1995, Heinisch and Fable 1999). Similarly, near Charlotte Harbor, juvenile gag were observed near rock outcroppings on the GOM side of Boca Grande Pass during August in waters less than 5 m deep (J.P. Casey, pers. obs.). It is unknown whether these individuals settled out in these habitats or moved there after first settling in the estuary. Year Figure 3. Annual mean relative abundance (abundance index = number of fish haul'^) of juvenile gag captured in the 183- m haul seine from May to December in Gasparilla and Pine Island sounds (mean ±S-). Although most juveniles egress to the GOM during their first winter, some individuals remain in estuaries for a second year or possibly return to their respective estuar- ies after moving into the GOM (Heinisch and Fable 1999, present study). Heinisch and Fable (1999) hypothesized that some fish remained in temperate St. Andrews Bay, Florida, during winter because of the great depth (19.8 m) in the inlet, but in Charlotte Harbor young gag have been collected during winter in relatively shallow water (< 3.3 m). Heinisch and Fable (1999) also suggested that larger juveniles nugrate offshore and then return to their respec- tive estuaries. Future studies would be necessary to under- stand the extent to which this may occur. One method that could be used to determine the movement between estuaries and offshore habitats is chemical markers within otoliths (Hanson et al. 2004). Seagrass percent cover Figure 4. Mean relative abundance (abundance index = num- ber of fisb baul'^) of juvenile gag captured with tbe 183-m baul seine from May to December in Gasparilla and Pine Island sounds by habitat (mean ±s-). Shore habitats included mangrove and beach shorelines, and shoal habitats included sandbars and oyster bars. Seagrass percent cover included up to three of the seagrass species found in Charlotte Harbor. The number in parentheses indicates the number of samples taken in each habitat. 7 Casey et al. In. conclnsioti, the gag is m ecoiiomi.cal1.y important reef species that is dependent on estuarine habitats during its early “life stages (Keener et al., 1988), Juvenile gag are distributed in the high-salinity areas of estuaries, and the period of settlement is s imil ar in temperate and subtropi- cal areas. However, gag remained in subtropical Charlotte Harbor longer and egressed at a larger size than in estuar- ies at higher latitudes. Habitat use by juvenile gag within the high-salinity areas of Charlotte Harbor was greatest on shallow seagrass shoals, but red mangrove-lined shorelines represent a suitable habitat not previously reported for this species. Interannual variability in gag abundance was evident in Cliariotte Harbor, with 2002 having a stronger year-class than the other years of this study. Variability in yoimg-of-the-year abundances may be attributed to flnctu- adons in factors such as fecundity of the offshore popula- tion, larval mortality, larval transport to the estuary due to winds and associated ciurenLs, and survival rates within the estuary (Keener et al 1988, Epifanio and Garvine 2001, Papemo 2002, Fitdiugh et al 2(X)5), This study describes specific locations, habitat types, and interammal, patterns of abundance within the Charlotte Harbor estuaiy that can be used to gauge future changes that may result from natu- ral (e.g., hurricanes) or anthropogenic alterations to water quality and habitat. Acknowledgments We thank D. Blewett for assistance with SAS pro- gramming and helpriil commeiits on earlier drafts of the manuscript, J. Buhrman for help in obtaining literature, A. Chonco for assistance with Figure 1, M. Greenwood for his statistical advice, and the Charlotte Harbor Field Laboratory staff for sampling assistance. We also thank A. Acosta, L. Bullock, S. Carlton, L. Hallock, J. Jackson, J. Leiby, A. Willis, and 3 anonymous reviewers for pro- viding useful comments on the manuscript. This work was snppoited by funding from Florida saltwater fi.shmg license sales and the Department of loteriar, US Fish and Wildlife Service, Federal Aid for Sportfish Restoration Project Number F-43 to the Honda Fish and Wildlife Conservation Commission. Literatuee citeo Bullock, L,H:, and G.B, Smith. 1991. Seabasses (Pisces: Senaaidae) Memoirs of the Houi^lass Cruises, Vol 8, part 2. Florida Marine Research Snstlttite, Ifepartment of Natural Resoiirces, St Petembarg, FL, USA, 243 p. Coffins, L.A,, A.G, Johnson, C.C. loeaig, and M,S, Baker Jr, 1998, RepFToductive patterns, sex ratio, md fecuudily io gag, Myciempercu micmlepis (Serranidae), a protogyuous grou- per from the northeastern Gulf of Fishery Bulletin 96:415-427. Epif^o, C.E, R.W. Garvins. 2001. Larval transport on the Atlantic contisenM shelf of North America: A review. Estuarine, Coastal and Shelf Science 52:51-77. Fitzhugh, G.R., C.C. F.C. Coleman, C.B. Grimes, and W.S. Stuiges in. 2005. Spatial and tempos p^tems in fertilization and settiement of young gag (Myctewpert’a mkwlepis) along the West Florida Shelf. Bulletin of Marine Sdence 77:377-396. Hanson, RL, C.C. Koenig, and V.S, Zdanowicz, 2004. Elen^ntal composition of otoliths used to trace estuarine habitats of juvenile gag Myctemperca micwiepis along the west of FlorMa, Mmne Ecology Rogress Smes 267:253-265. Hastings, R.W. 1979. The origin ^d seasonality of the tish fauna on a new jetty in the northeastern Gulf of Mexico. Bulletin of the Flmida State Museum, Biological Science 24:1-122. Heinisch, B.V. and W.A. Fable Jr. 1999. Movement of gag, Myctempema micmlepis, in and from SL Andrew Bay, Florida. BuHetin of Marine Science 64:501-508. Hoese, H.D., C.E. Richards, and M. C^tagna. 1961, Ap^arance of the gag, Myctemperca micwiepis, in coastal waters in ■Virginia, Chesape^ Science 2:104-105, Hood. P.B. and R.A. Schlieder, 1992. Age, growth, and rqsroduc- tion of gag, Mycteropema micmlepis (Pisces: Serranidae), in die eastern Gulf of Mexico, Bulletin of Marine Science 51:337-352. Huang, T.C. 1966. A Sedimentologk study of Charlotte Harbor, southwestern Florida. M.S. thesis, Flcmda State Univemty, Tallah^s^, FL, USA, 97 p. Keener, R, G.D. Johnson, B.W, Steoder, E.B, Brothem, and H,R, Beatty. 1988. Ingress of postlarval gag, Mycterperca micro- iepis, (Pisces: Serranidae) through a South Carolina barrier island inlet. Bulletin of Marine Science 42:376-396. Krienig, C.C. EC. Coleman. 1998. Absolute abundance and survival of juvenile gags in sea grass beds of the northeast- ern Golf of Mexico. Transactions of the American Fisheries Society 127:44-55. Kupschus, S.R, and D.M. Tiemain. 2(X)1, Associations between fish assemblages and environmental factors in nearshore habitats of a subtropical estuary. Journal of Fish Biology 58:1383-1403, Ley, J,A., C.C. Melvor, C.L, Monta^, 1999, Fishes in mangrove prop-root habitats of northeastern Florida Bay: Distinct assemblages across an estuarine gradient, Estuarme, Coastal and Shelf Science 48:701-723. Mullaney, M,D, Jr L,D, Gale, 1996. Ecomorphological relationships in ontogeny: Anatomy and diet of gag, Myctemperca micmlepis (Pisces: Serranidae). Copeia 1996:167-180, FapemOj R. 2002. Age-0 spot (Leiostomm xanthurm) from two estuaries along central Florida’s east (x>ast: Comparisons of the timiitg of i^mitment, seasonal changes in abnn- dancx, and rates of ^wth and mortality. Florida Scientist 65:85-99. Foulakis, G.R., D.A. Blewett, and M,E. Mitchell. 2003. The elfecte of season and prosiniity to fringing mangroves on seagiass-associated fish commimities in Charlotte Harbor, Florida. Gulf of Mexico Science 21 :1?1-I84. Foulakis G.R., R.E. Matheson Jr., M.E. Mitehell, D.A. Blewett, and C.F. Idelberger. 2004. Fishes of the Charlotte Harbor estuarine systein, Florida. Gulf of Mexico Scienix 22:117- 150. Ross, S.W. and M.L. Moser. 1995. Life Mstoy of juvende gag, Myctsmperca micmiepis, in North Camlina estuaries. Bulletin of Marine Science 56:222-237. Sargent, F.J., T.J. Leary, D.W. Crewz, and C.R. Kruer. 1995. Scarring of Florida’s seagrass; Assessment and manage- ment options. FMRI Technical RqmrtTR-l. Florida Marine Research Institute, St Petersburg, FL, USA, 37 p. SAS Institute. 1988. Froc GLM procedures SAS/Stat user’s guide, VoL 2, release 6.03 ed. SAS iMtitute, Inc., Cary, NC, USA, p. 893-993. Tkylor, JX. 1974. The Charlotte Harbor estuarine system. Florida Scientist 37:205-216. Thayer, G.W., D.C. Colby, and W.F. Hettler. 1987. Utilization of the red mangrove prop root habitat by fishes in sooth Florida. Marine Ecology Fhogress Series 35:25-38. Wang, i.C.S. and E.C. Raney. 1971. Distribution and fluc- tuations in the fish :^una of the Charlotte Harbor estuary, Florida. Charlotte Harbor Estuarine Stedies 3. Mote Marine Laboratory, Sarasota, FL, USA, 56 p. W^sel, M.R. and B.L. Winner. 21K)3. Using a modified purse seine to collect and monitor estuarine fishes. Gulf and Caribbean Research 15:61-71 . Zai, J.H. 1999. Biostatistical. Analysis. 4th ed. Ptentiix-F^l, Englewood Cliffs, Ni, USA, 663 p. 9 Gulf and Caribbean Research Volume 19 Issue 1 January 2007 Breeding Season and Molt Cycle of the Fiddler Crab Uca rapax (Brachyura^ Ocypodidae) in a Subtropical Estuary Brazil^ South America Daniela da Silva CastigHoni Instituto de Biociencias, Brazil Maria Lucia Negreiros-Fransozo Instituto de Biociencias, Brazil Rosana Carina Flores Cardoso Instituto de Biociencias, Brazil DOI: 10.18785/gcr.l90L02 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr & Part of the Marine Biology Commons Recommended Citation da Silva Castiglioni, D., M. Negreiros-Fransozo and R. C. Flores Cardoso. 2007. Breeding Season and Molt Cycle of the Fiddler Crab Uca rapax (Brachyura; Ocypodidae) in a Subtropical Estuary Brazil; South America. Gulf and Caribbean Research 19 (l): 1 1-20. Retrieved from http://aquila.usm.edu/gcr/voll9/issl/2 This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized editor of The Aquila Digital Community. For more information, please contact Joshua.Cromwell^usm.edu. BREEDING SEASON AND MOLT CYCLE OF THE FIDDLER CRAB UCA MAPAX (BRACHYUKA, OCYPODIDAE) IN A SUBTROPICAL ESTUARY, BRAZIL, SOUTH AMERICA Daolela da Sil¥a Caullpom^ Btoia Luda and Eosaiia Carina Fleres Cardeso (MLNF; RCFC) NEBECC (Gmiq) of SUidies on Crustacean Biology, Ecology and Cultu?^), Depariamento de Zoologia, Imtituto de Biociencias, Caixa Postal 510, UNESF, 1 861 8-000, Botucatu, Sdo Riulo, Bmsil; (DSC) Graduation Course in 'Oology, lnstituU> de Biocieiwim, UNESP, Botucatu, Sdo Paulo, Bmsil "^Corresponding author, E-nmil mlnf@ihhimesp.br ABSTRACT TMs is the fet of the brmiiiig season and molt cycle of Uca rapax cm the Brasiliaii coast during a cme-year period (April 2(M)l“Mai€li ^)02). At 2 sites, the Ihamambaca and Ubatamirim mangrove 2 collators captured crabs once a month at low tide Tot 15 miantes by the catch-per-imit-effort pmc^ixre, digging mto the sediment and removing the cmbs. The gonad-development stages of cmbs of both seKes were determined, by direct observation* and the molt stege was estimated from the hardness of the tegument Crabs with develof^ gonads were found mostly in warmer seasons, with ovlg^ons females occurring nMinly in summer and autumn. The repmductlve cycle is seasonal. Recently nmltoi individixals were collects in relatively higher numbem for juveniles than for adult crabs. Introduction The reproductive strategy of brachynraii crabs is extremely diversified, ultimately shaped to maximize egg production and offspring survivorship, thus increas- ing the chances for preservatioii of the species (HartnoU and Gould 1988). The reproductive period for many brachyurans can be estimated by observing gonad matu- ration at the macro- or micfoscopic levels and from the frequency of ovigerous females in the population (e.g., Wolfrath 1993, Emmerson 1994, Mouton and Felder 1995, Costa and Negrefros-Fransozo 2003, Colpo and Negreiros-Fransozo 2003, Lilulo 2004, 2005a, 2005b). The reproductive period results from a complex interaetion between internal and external factors, leading to intra- and inter-spedfic variation in the duration of the reproductive season (Sastry 19B3). Many ocypodid crabs have seasonal reproduction, as observed in Uca lactea (de Haan, 1835) studied by Yamaguchi (1971), in Uca pugilator (Bose, 1802) and Uca pugnax (Smith, 1870) studied by Christy (1982), and Uca tfmyeri Rathbun, 1900 studied by Salmon (1987). The limited reproductive season in semi-terrestrial crabs might be related to changes in the temperature and pho- toperiod and the availability of food resources (Plllay and Ono 1978). The availability of food for body maintjenance, somatic growth and reproduction of the adult crabs, and for growth and survival of the larval and/or juvenile stages is assumed to be the most important factor (Sastry 1983) in synchronization and coordmation of reprodnetive activity in a given habitat. Mangrove ecosystems are highly productive environ- ments, hospitable for feeding, growth, and reproduction of many species of crabs, shrimps, fishes, and other animals (Schaeffer-Novelli 1995). In these ecosystems, fallen man- grove leaves provide most of the organic matter deposited in the sediments. Deposit-feeding ocyptMlid embs of Ihe genus Uca feed on the organic matter, including the endo- fauna that are sorted out from the substrate. Their actual food supply depends on the ecosystem productivity, micro- bial activity, substrate texture, and tidal action (Mural et al. 1982, Twilley et al. 1995, Monia et al. 1998). There are about 100 species of fiddler crabs worldwide, most of them included in 2 distinct morphological categories with different morphology, zoogeography, and behavior: the broad-front species and the narrow-front species (Crane 1975, Christy and Salmon 1984, Rosenberg 2001). There are about 10 species of fiddler crabs typically found in Brazilian mangroves: Uca burgerd Holthuis, 1967; Uca cwmlanta Crane, 1943; Uca leptodactyla Rathbim, 1898; Uca mordax (Smith, 1870); Uca rapax (Smith, 1870); Uca umgmsyensis Nobili, 1901; Uca victoriana (von Hagen, 1987); Uca vocator (Herbst, 1804); Uca thayeri Rathbun, 1900; and Uca maracoam (Latreille, 1802-1803) (Melo 1996). A peculiar morphological feature separating these species into 2 groups is the size of the carapace front. Uca hurgersi, U. cumulmta, U, leptodactyla, U. mordax, U, rapm., U, urugmyensis, U. vlctorima, U. vocator and U thayeri have a broad front, while U. maracoam has a narrow front. Among the species studied here, U, rapax is one of the most abundant species of the genus Uca in the Brazilian 11 Castiglioni et al* mangroves. It burrows into mud or muddy sand and feeds on organic matter in the sediment of mangroves along the northern coast of the State of Sao Paulo, Brazil, According to Melo (1996), U. rapax is distributed throughout Florida, the Gulf of Mexico, the Antilles, Venezuela, and Brazil (from the states of Paid to Santa Catarina). This study investigates the breeding season and molt cycle of U. rapax from the Itamambuca and Ubatumirim mangrove forests near Ubatuba on the northern coast of Sao Paulo, Brazil, The studies are based on the frequency of gonad and molt stages and the ratio of ovigerous females. Although located close to each other, the study sites have distinct landscapes and hydrological features that determine the vegetation community. The vegetation in the Itamambuca mangrove (23°24'4"S, 45”00'7"W) consists of only Laguncularia racemosa (Linnaeus). In the Ubatumirim mangrove (23°20'17.8"S. 44‘‘53'22"W) the vegetation is mostly L racemosa with some Avicennia shaueriana Stapf, and Leech (Negreiros-Fransozo, pers. comm.). The Itamambuca mangrove is a highly productive ecosystem with hydrology and sediment characteristics that retain minerals and a rich environment suitable for development of fiddler crab populations (Colpo 2001). As described by Castiglioni and Negreiros-Fransozo (2004), these 2 mangroves have distinctly different sediment organic matter content and texture as well as river and burrow s alini ties. Material and Methods (iabs were collected monthly by 2 people from April 2001 through March 2002 in both mangroves, using the procedure of catch per unit effort (cpue). Over 15 min time periods during low tides (spring tide), crabs were removed from their burrows by digging to the end of each burrrow with diving knives. Additional collections were made in August through December 2002, using the same procedure attempting to locate ovigerous females. Crabs were counted, sexed, and measured (carapace width (CW) to the nearest 0.01 mm), and females were checked for eggs. Ovigerous females were preserved indi- vidually in 70% e than ol. We determined the relative frequency of ovigerous females over the course of the year. The stages of embry- onic development were classified as initial, intermedi- ate, or final, according to the relative proportion of yolk content and the appearance of eyes and appendage buds in the embryo (see Costa and Negreiros-Fransozo 1998 for details). A multinomial proportions analysis (Curi and Moraes 1981) with a 5% significance level was used to examine reproductive seasonality. From this analysis we considered autumn as April, May, and June; winter as July, August, and September; spring as October, November, and December; and summer as January, February, and March. The size of ovigerous females was compared between the populations by Student’s t test (a= 0.05; Zar 1996). The gonad development stages were analyzed in each sex. The carapace in the dorsal region was removed, and the shape, size, and color of the gonads were observed under a stereomicroscope. The female gonads were classified in 6 developmental stages: spent = SP; advanced = AD; developed = DE; developing = DI; rudimentary = RU; and immature = IM. Five stages were used for males: spent = SP; developed = DE; developing = DI; rudimentary = RU and immature = IM. This procedure was modified from Haefher (1976), Abelld (1988), Choy (1988) and Costa and Negreiros-Fransozo (1998). Comparisons of the gonad proportion between seasons in each sex were performed using a multinomial proportions analysis (Curi and Moraes 1981) to determine the reproductive period. Fiddler crabs were arranged in 2 groups: juvenile or immature crabs (specimens with immature or rudimentary gonads) and adult or mature crabs (specimens with develop- ing, developed, advanced, or spent gonads). Comparisons of the immature and mature crabs between seasons in each sex were performed using a multinomial proportions anal- ysis (Curi and Moraes 1981). The reproductive period was determined using data for the frequency of mature males and females over the year (Costa and Negreiros-Fransozo 1998, Mantelatto and Fransozo 1999). The air and burrow temperatures were measured monthly, with 3 lephcates at each site, and were compared by ANOVA among seasons and sites (a = 0.05; Zar 1996). The degree of association among crabs with developed gonads and environmental factors (air and burrow temper- atures) was assessed using Pearson’s correlation (a = 0.05; Zar 1996). The molt stages were described based on Warner (1977) and Abello (1988) as follows: A) post-recent molt = carapace very flexible and without calcification; B) post- advanced molt = onset of calcification; brittle carapace but more resistant and with a consistency similar to cardboard; C) intermolt = carapace fiiUy calcified, with a leathery con- sistency; D) pie-molt = a new exoskeleton present inside the old one and the molt lines emerging in the pterigosto- mial region; and E) molt = exact moment of the change or exit of the animal from the old exoskeleton. Following this scheme, the molt stages were grouped in 2 phases: molt activity (A, B, and D stages) and inter- molt (C stage). The proportions of the 2 stages were com- pared between seasons using the multinomial proportions analysis (Curi and Moraes 1981). We analyzed the molt 12 Castiglioni et al. feequency in each sex by size class (CW) for both man- groves. Results During the study period, a total of 1,294 fiddler crabs were collected at Itamambuca: 667 males and 627 females. Eight ovigeroas females were collected in the autumn which corresponds to 2.22% of all adult females. In the Ubatiimirim mangrove, during the summer, a total of 2,107 specimens were collected: 1,117 males and 990 females with 20 ovigerous females (3.03%). Most of the oviger- ous females collected from both mangroves bore eggs in the final embryonic developmental stage. The additional coEection taken in No%^ember 2002 comprised 27 oviger- ous females from Itamambuca aurl 67 from Ubatumirim. Most of the ovigerous crabs eggs were in the final devel- opmental stage. When collected, the crabs had emerged from burrows and were moving iteely on, the substrata. Ovigerous femal.es ranged from. 14,2 to 24.2 mm CW at Itamambuca and 10.2 to 21.3 mm CW at Ubatomiiim. The mean size of ovigerous females from Itamambuca (19,5 ± 3.3 mm; mean ± s) was larger than ovigerous females from Ubatumirim (16,2 ±3.2 mm; mean ± s) (Student t = test; F < 0.05). Males with developed gonads were found in all seasons, but significantly more males in this condition were found in the summer and autumn at Itamambuca (P < 0,05) (Ihble 1). Males with developing gonads were found in all seasons, but significantly more were found in the spring and summer at Ubatumirim (ANOVA; P < 0.05) (Table 2). Females with developmg and developed gonads were found in most of the samples, except during the winter at Ubatumirim. Females with advanced gonads occurred only during the summer at both sites (P < 0.05). In both populations there were many females with spent gonads in the autumn and winter (ANOVA; P < 0,05), unlike in the other seasons (Tables I and 2), In the Itamambuca mangrove, mature males were found in higher frequencies during autumn, spring, and smnmer, whereas mature females increased only duimg summer (Table 3) (ANOVA; P < 0.05). At Ubatumirim, mature males were frequent in all seasons of the year, and mature females were frequent during spring and summer (Table 4) (ANOVA; P < 0.05). The mean air temperature was similar between man- grove sites and did not differ markedly through seasons (ANOVA; P < 0.05). The lower temperatures were regis- tered duriiig autunin (ANOVA; F < 0.05) (Ikble 5). The mean temperature inside the burrows of U. rapax did not TABLE 1 Frequency (%) of the gonad stages (SF = spent; AD = advanced; DE = developed; DI = developing; RU = nsdUmentary; IM = Imniatere) for male and female Uca mpax dimng the seasons (Au -Aatam; W = Winter; Spr = Spring; Sn = SninmeF) In the Itamambuca mangrove. Seasons Stagei An W Spr Sn Males SP 17.0 a 18.0 a 15.0 a 15.0 a AB A BC BC DE 15.0 a 5.0 be 0.5 c 12.0 ab C C C BC DI 34.0 a 35.0 a 35.0 a 37.0 a A A B A RU 3.0 c 15.0 b 27.0 ab 30.0 a C BC AB AB M 29.0 a 27.0 a 22.0 a 5.0 b AB AB AB C Females SP 55.0 a 41.0 ab 15.0 c 35.0 b A A B A AD 0.6 b 0,0 b 0,0 b ll.Oa D BC C BC DE 2.0 b 3,0 c 7,0 a 9.0 a CD BC B BC DI 8.0 ab 3,0 b 15.0 a 23.0 a C BC B AB RU 2.0 c 6.0 be 29.0 ab 15.0 a CD C A BC IM 31.0 a 46.0 a 32.0 a 7.0 b B A A C Note: lower case letters correspond to the comparisons in each gonad stage among the seasons; capital letters correspond to the comparisons ^ong the gonad stages in each season. Values with at least one letter in common did not differ statistically (OL^OffS). differ between sites in a same season but reached maii- mum values during the summer at both sites (ANOVA; P < 0.05) (Table 5). The relative frequency of mature males and females tended to increase with increasing air temperature (Itamambuca: - 0.71 in males; ” 0,53 iu females; Ubatumkim: - 0.9S in males; = 0.66 in females; P < 0.05) (Figure 1), Throughout the year, significantly more (ANOVA; F < 0.05) specimens were in the intermoit than molt stages for 13 Castiglioni et al. TABLE 2 Frequency (%) ©f the gonad stages (SF = spent; AD = advanced; DE = developed; DI = developing; EU = mdlmeiitaiy; IM = innnatnre) for male and female Deo mpm diiriiig die seasons (An -Antom; W = Winter; Spr = Spring; Su = Summer) in the Itamainliiiea mangrove. Seasons Stages An W Spr Sii Males SP 37.0 ab 43.0 a 28.0 be 24,0 c A A A AB DE 3.0 a 0.7 a 2.0 a 12,0 a B C C BC DI 140 b 20.0 b 33.0 a 33,0 a C B A A RU 12.0 b 17.0 ab 23.0 a 21.0 ab C B AB ABC IM 33.0 a 20.0 b 13.0 b 10.0 b A B B C Females SP ILO ab 20.0 a S.Ob S.Ob B C B B AD 0.0 a 0.0 a 0.0 a 6.0 b C B C B DE 3.0 ab 0.0 b 8.0 a iS.Oa C B B BC DI 6.0 b 0.4 c 28.0 a 19.0 a BC B A C RU 43.0 ab 41.0 a 31.0 b 8.0 c A A A A M 36.0 ab 39.0 a 24.0 b 11.0 c A A A BC Note; lower case letters correspond to the comparisom in each gonad stage ^ong the i^a^ns; coital letters correspond to the comparisom among the gonad stages in e^h season, Valnes widi at least one letter in common did not differ statistically (a ^0.05), both populations (Figures 2 and 3). There were no signifi- cant differences between sexes in the molt activi.ty* except for males in Itamambuca (Figure 2). Smaller crabs of both sexes molted more often (Figure 4), Figme 5 shows the frequency of molt activity of U, rspnx in relation to gonadal development stages for each sex. The recently molted stage A (see Materials and Methods) was observed in male crabs that had immature gonads and females with immature and mdimentaiy gonads in the Itamambnca mangrove popnlatioji. At Ubatumirim, TABLE 3 Frequency (%) of iiiiiiiatiire and nsature Uca rapnx toy seaison iM the Itamamtoeca mangrove. Seasons Mates Female Inmiatiine Matare Mature Autumn 37.7 ab 623 a 38.8 b 61.2 b Winter 50.6 a 49.4 b 61.4 a 38.6 c Spring 37.0 ab 63.0 a 46.2 b 53.8 b Summer 35.6 b 54.4 a 21.9 c 78J. a Note: lower case letters cormspond to the comparison between each demographic catego^ for each sampling season. Yalnes with at least one letter in common did not differ statistically (a = 0,05). TABLE 4 Frequency (%) of iimnatnre and matiire Uea rsipisx toy sei^on in the Itemaiiibiicii mm^mve. Males Females Seasons Autumn liiiiinatiiie 45.0a Mature 55.0b Immatiire 79.3a Matuine 20.7b Winter 36.6ab 63.4ab ?9.9a 20.1b Spring 36.8ab 63.2ab 55.5b 44.5a Summer 31.2b 68.8a 59.5b 40.5a Note: lower case letters correspond to the comparison between each demographic categoiy in each sampling season. Values with at least one letter in common did not differ statistically (a=0.05). this molt stage was observed in males with mdimentaiy and spent gonads, but it was not observed in any females. Male crabs in the post-advanced molt stage (B) have been found in all gonad stages from both mangroves. However^ this molt stage was not observed in female crabs with advanced gonads from Itamambuca, or in females with advanced and developed gonads from. Ubatumirim. Discussion In the mam systematic sampling period, a small number of ovigerous females were found in Itamambuca and Ubatumirim mangroves. However, additional ovig- erous females were collected in the additional samples (August-December 2002), and most of them had emerged from and were moving around the burrows and carrying eggs in the late embiyfordc stage. According to Saknon (1987), females of broad-fronted fiddler crabs such as U. rapax can incubate their eggs imdergroimd to protect them frx>m extreme enviromnentai conditions. This provides 14 Castiglioni et al. TABLE 5 Comparison of mean (± 5 ) temperature in ”C between year seasons in the mangroves, s = standard deviation. Seasons Sites Itamambuca Ubatumirim Air temperature Burrows temperature Air temperature Burrow temperature Autunm 24.6 ± 5.36 b 25.00 ±4.52 be 25.00 ± 1.10 b 24.70 ± 2.15 b Winter 30.10 ±6.02 a 22.80 ± 0.79 c 28.20 ± 2.79 ab 26.80 ± 2.53 b Spring 31.10 ±5.76 a 27.50 ±0.91 be 27.80 ± 2.73 ab 25.10 ± 3.03 b Sununer 32.90 ±1.78 a 31.60 ± 2.50 ab 32.40 ±2.91 a 31.60 ±4.45 a Note: Values with at least one letter in conunon within a column did not differ statistically (ANOVA; a = 0.05). a uniform environment, thus promoting synchrony in embryonic development and larval hatching. Christy and Salmon (1984), Murai et al. (1987), and Henmi (2003) also observed such behavior in Uca pugilator (Bose, 1802), Uca lactea (DeHaan, 1875), and Uca perplexa (H. Milne Edwards, 1837), respectively. The ovigerous females had large broods, remained in their burrows during the entire incubation period, and did not feed during this phase. We infer that the individuals of U. rapax in the mangroves studied may have been searching for a hatching area. To maximize the probability of larval survival, where their eggs are ripe for hatching, many mangrove crabs travel the long distance to the water’s edge during the night (Gifford 1962). Only ovigerous females of U. rapax with eggs in the final embryonic development stage were found during this study, probably because these females are more active or exit their burrows to liberate the larvae dur- ing spring tide periods. Christy (1978) suggested that the synchronization of reproduction with tidal cycles in Uca species could be an adaptation to increase the probability that the planktonic larvae are carried back to the adults’ environment by tidal currents. In species of Uca with large broods, the females pro- duce more eggs in a single spawn, but they cannot carry eggs continuously. This is probably because of the vulner- ability of the large egg masses to stress and desiccation. Females of these species do not feed enough during the incubation period to develop new oocytes internally. On the other hand, species with small broods produce fewer eggs each spawn but can develop broods continuously; their egg mass is protected, and the females can feed during incubation to develop a new brood (Henmi 1989, (/) .a 5 o » o c c> 3 C P Season Males I I Females Figure 1. Frequency of mature fiddler crabs (males and females) and air temperature ("C; mean ± 5 ^) in both mangroves seasonally. = standard error. 15 Castiglioni et al. A 100- 80- ^ 60- I 20- Z 0 - o Seasons EZIlntermolt Molt Figure 2. Uca rapax in Itamambuca mangrove: Plot of mean seasonal frequency of molt activity and intermolt for males (%) (A), females (B). All seasonal comparisons between molt and intermolt did not differ significantly (P > 0.05). A 100-1 80- 60- Autumn Winter Spring Summer Season □□intermolt H Molt Figure 3. Uca rapax in Ubatumirim mangrove: Plot of mean seasonal frequency of molt activity and intermolt for males (%) (A), females (B). All seasonal comparisons between molt and intermolt did not differ significantly {P > 0.05). Henmi and Kaneto 1989), Because the egg mass of U. rapax does not remain completely covered by the abdo- men, and also because no females were caught with eggs in initial and intermediate embryonic development, we suppose that the ovigerous females remain in their bur- rows during the incubation period. The ovigerous females collected had spent and empty gonads. These females probably do not bear a new brood inunediately after larval hatching as in other Uca species as they probably do not feed during the incubation period and have no resources to produce new egg masses. Fiddler crabs are typically adapted to live in hot climates. In the tropics, they are active year-round, and reproductively active crabs are found during all months, since environmental conditions are permanently favorable for feeding, gonad development, and larval release (Sastry 1983, Thurman 1985). In the subtropics, reproduction in some species is limited more by the dry season than by temperature. For a few species, usually those living in tem- perate zones, reproduction is controlled by temperature, by their distributional limits, or in some cases by intertidal zonation, e.g., the ocypodid crab Macrophthalmus gran- didieri A. Milne Edwards, 1867 studied by Emmerson (1994). Reproduction is restricted to the warmer months (summer in the south hemisphere) in fiddler crabs, whereas during the colder months (winter in the south henusphere) they hibernate in their burrows (Crane 1975). Some inves- tigators (Yamaguchi 1971, Christy 1982, Salmon 1987, Rodriguez et al. 1997) have observed the occurrence of reproduction during warmer months in Uca species. In both mangrove populations that we studied, the period of high reproductive activity in U. rapax occurred in the summer, but this species could also be found reproduc- ing during other months, except in winter. Associations between temperature and reproduction may be related to better conditions for larval development, in terms of the availability of food or more favorable conditions for lar- val growth. However, in the case of tropical species with year-round procreation, reproduction may be associated with other factors, not only with temperature (Santos and Negreiros-Fransozo 1999, Costa and Negreiros-Fransozo 2003). Factors such as day length, food availability, rain- fall and photoperiod have been indicated as other major modulators of reproduction in brachyurans (Conde and Diaz 1989, Zimmerman and Felder 1991, Flores and Negreiros-Fransozo 1998, Leme and Negreiros-Fransozo 1998, Negreiros-Fransozo et al. 2002, Cobo and Fransozo 2003, Litulo 2004). The U. rapax populations of the Itamambuca and Ubatumirim mangroves had high proportions of crabs with developed and advanced gonads during the warmer seasons, which may be related to adequate conditions for 16 Castiglioni et al. Itamambuca Ubatumirim J Size class (mm) Male I I Female Figure 4. Frequency of molt activity (%) for male and female Uca rapax by size class (carapace width) in both mangroves. in (0 o (D 3 W E >* o c (D 3 D- 0) 40 30 20 10 40-1 30- 20 - Itamambuca Ubatumirim 10 0 Dl Gonad stage Male I I Female Figure 5. Frequency of molt activity (%) in relation to gonad stages for male and female Uca rapax in both mangroves. IM immature, RU = rudimentary, DI = developing, DE = developed, and SP = spent. 17 Castiglioni et al. the deyelopment and survival of the brood at these times. Haley (1970), studying the ocypodid Ocypode quadmta (Fabricius 1787) in Texas, observed that females with mature ovaries occimred in Mgh proportioiis from April to August (spring-suiomer), suggesting that this period is one of intense reproductive activity. This may be related to day length, which stimulates ovarian maturatioii in immature females. Negreiros-Fransoso et al. (2002) investigated the biology of O. quadrata in a sandy beach at Ubatuba, Brazil and observed high reproductive intensity from October to May (spring, summer, and autumn). The higher abundance of females with fully developed gonads in this period was positively correlated with abiotic factors such as air tem- perature, water surface temperature, and precipitation. The high intensity of molt activity in the first size classes can be explained by the direction of energy into growth until the crabs aitain sexual maturity. After this phase, the frequency of the crabs in molt activity decreases, because energy resources become divided between molt- ing and reproduction (Hartooll 1988), After the pubertal molt we observed many crabs in the process of molting, which mitigates against the hypothesis that a tenomal molt occurs in 17. rqpax soon after sexual maturity. The low inddeiice (under 30%) of sexually mature crabs in molt activity is usual for semi-terrestrial crabs. Moreover, fiddler crabs molt underground (Hyatt and Salmon 1978, Christy and Salmon 1984, Salmon 1987, Atkinson and Ikylor 1988, Koga et al. 2000). The antagonism between the reproductive process and growth is well known io crustaceans. The competition for energy resources required by one process or another leads to wide diversity in patterns of growth and reproduction (Hartnoll 1985, Ldpez-Greco and Rodriguez 1999). These patterns allow each spedes to maximize its reproduc- tive potential within the Hmits of its genotypic plasticity (Hartnoll 1985). It can be assumed that the pattern of t/. rapwc results from the interaction between growth and reproduction. However, several aspects concerning the manner in which these antagonistic processes Interact still requires investigation. In aU crustaceans, reproduction is dynamically related to the physical and chemical conditions of the organisms and to enviromneiital conditions, food availabiiity, and the presence of competitors or predators. The relative impor- tance of the proximity of the factors that control reproduc- tive activity can vary for different species m the same envi- ronment or in habitats with different characteristics (Sastiy 1983). Differences like sediment organic matter content, river and burrow water salimly, and granular composition of the substrate (see Castiglioni and Negrekos-Fransozo 2005) appear to act directly or indirectly on aspects of the populations. Rqjroduction is especially affected leading to variations in the process in different populations. This is the first account of the breeding season and molt cycle of Uca rapax in Brazil. Further studies on fecundity, fertility, larval migrations, reproductive behav- ior, and feeding wiU add to our comprehension of the reproductive strategies of this fiddler crab. Acknowledgements We are grateful to the Fundagao de Amparo a Pesquisa do Estado de Sao Paulo-FAPESP, for a feUowship to DSC (#01/01810-9) and financial supportfor fieldwork to MLNF (98/03134-6). The authors would like to acknowledge C. Tudge for his valuable comments on the manuscript. We also thank the NEBECC staff for tiieir help during field and laboratory activities. All sampling in this study was conducted in compliance with current applicable state and federal laws. Literatume cited Abelld, F, 1988, Reproductive biology of Mmmpipus tuber- culatum (Roux, 1830) in the uorihwestera Medtteiranean, Ophelia 30:47-53. AtkiasoB, R J,A.. and A.C. Ihylor 1988, Physiological, ecology of burrowing decapods. Symposium of tee Zoological Soci.ety of London 59:201-226, Castiglioni, D.S. aodM,L,NegmiTOS"Frafisozo. 2004. Comparative analysis of relative growth of Uca mpwc (Smith) (Crustacea, Qcypodidae) from two mangroves i.o Sot Paulo, Brazil Revlsta Brasileira de Zoologia 21:137-144. Castiglioni, D.S. and M.L. Negreiros-Fransozo. 2CM}S, Comparative population biology of Uca mpax (Smith, 1870) (Bracbyura, Oeypodidae) from, Itaxnambuca and ITbatomirim mangroves in Ubatuba littoral, Brazil. Journal of Natural History 39:1627-1640, Choy, S.C. 1988. Reproductive biology of IJacamnm puber and .L hohatus (Decapoda, Bfacb 5 Fura, Fortunidae) from the Gower Peninsula, South Wales. Marisie Biology 9:227- 241, Christy, J.H. 1978. Adaptive si.gmficOTce of reproductive cycles in tee ti.ddler crab Uca pugUatoK A hypoteesis. Science 199:453^55, Christy, J.H, 1982. Adaptive significance of semi lunar cycles of larval, release in fiddler crabs (Genus Uca): Test of an hypothesis, Bi.ological Bulletin 163:251.-^3, Christy, J.H. and M. Salmon. 1984. Ecology and evolution of mating systems of fiddler crabs (genus Uca). Biological Review 59:483-509, Cobo, V.J, and A. Fransozo. 2003. Extemat tectofs deterra.i.ning breeding season in tee red mangrove crab Goniopsis cru~ enfata (Latreille) (Crustacea, Brachyura, Grapsidae) on tee Sao Paulo State norteorn coast, Brazil.. .Revistm Brasileim de Zoologia 20:213-217. 18 CASTfGLTOm ET AL. Colpo, K.D. 2001. Biologia populacional comparativa de llca VQcator (Herbst, 1804) (Brachyiira, Ocypodidae) em tres localidades do litoral noite paolista M.S. Thesis. Universidade Estadiial Faulista-Batucatii, Sao Paulo, SP, BR, 104 p. Colpo, K.D. and M.L. Negisiros-Fransozo. 2003. Reproducdve outpat of Uca vocator (Herbst, 1804) (Braebyura, Ocypodidae) from three subtropical, mangroves in Brazil. Cmstaceana 76:l“l 1 . Conde, J.E. and H. Diaz. 1989. The mangrove tee crab Aratus pismii in a tropical estuarine coastal region. Estuarine, Coastel and Shelf Science 28:639-^50. Costa, T.M, and M.L. Negreiros-Fransozo. 1998. The repiodnc- tive cycle of Callimctes danae Smith, 1B69 (D^a|K)da, Fortimidae) in the Ubaliiba region, Brazil. Cmstaceana 71:615-627. Costa, T.M. and M.L. Negreiros-Fransozo. 2(K)3. Fopnlation Mol- o^ of Vca th^eri Rathbim, 1^0 (Brachynra, Ocyp3didae) in a subtropical South American masgrove area: Results from transit and catch-per-unlt-effort techniques. Cmstaceana 75:1201-1218. Cmne, J. 1975. Fiddler crabs of die world. OsyiKididae: Genus Uca. Princeton University Press, Princeton, NJ, USA, 736 p. Cuii, P.R. and R.V. Moraes. 1981. Associagao, horaogeneidade e contrast ente proporgoes em tabelas contendo distri- bulges mnltinomiais. Ciencia e Cultara 33:712-722. Emmerson, W.D. 1994. Seasonal breeding cycles and ses: ratios of eight species of crabs from Mgazana, a mangrove estuary in Transkei, sonthem Africa. Journal of Cmstarssan Biology 14:568-578. Flores, A.A.V. and M.L. Negieiros-Fransozo. 1998. Extenml fac- tors determining seasonal breeding in a subtmpica! popula- tion of the shore crab ^^M:hygmpms tmnsversus (Gibbes, 1850) (Brachynra: Grapsidae), Invertebrate Reproduction and Development 34:149-155. Gifford, C.,A. 1962. Some observations on the general biology of the land crab CardLmma guanhumi (Lateille), in South Florida. Biological Bulletin 123:207-223. Haefeer Jr, ,F.A. 1976. Distribution, reproduction and moulting of the rock crab, Cancer irmmtm Say, 1917, in the Mid- Atlantic Bight Journal of Natural History 10:377-397. Haley, S.R. 1970. Reproductive cycling in the ghost crab, Ocypade qimdrata (Fabr.) (Brachyura, Ocypodidae). Cmstaceana 23:1-1 1 . Haitnob, R.G. 1985. Growth, sexual maturity and raproductive ontpnt in: A.M. Wenner, ed. Factom in Adult Growth. A.A. Balfcema, Rotterdam, The Netherlands, 362 p. HartnoJl, R.G. 1988. Evolution, systematic and geographi- cal distribution. In: W.W. Butggren and B.R. McMahon, eds. Biology of Land Crabs. Cambridge University Press, Cambridge, UK, 479 p. Haitnoll, R.G. and P. Gould. 1988. Brachyuran life history strate- gies and the optimization of egg production. Symposimn of the Zoological ScK:iety of London 59:1-9. Henmi, Y. 1989. Life-history patterns In two forms of Macrophthalmus japonicus (Cnistacea: Brachyura). Marine Biology 101:53-60. Henmi, Y. 2(K)3. Trade-off between bnxid size and bnxid interval and the evolution of underground incubation in three tiddler crabs (Uca perplem, U vocans and U. dussumien). Journal of Cmstacean Biology 23:46-54. Henmi, Y. and M. Kaireto. 1989. E^roductive ecology of three ocypodid crabs 1. The Influence of privity differences on repmductive traits. Ecological Research 4:17-29. Hyatt, G.W. and M. Salmon. 1978. Combat In the fiddler crabs Uca pugilator and U. pugnax: A quantitative analysis. Behavior 65 : 1 82-21 1 . Koga, T, M. Mural, S. Goshima, and S. Poovachiranon. 2CW0. Undergronnd matiiig in the fiddler crab Uca tetragon: The association between female life history traite and mate mat- ing tactics. Journal of Experimental Marine Biology and Ecology 248:35-52. Leme, M.H.A. andM.L. Ne^iios-Fransozo. 1998. Reproductive patterns of Aratus pisonii (Decapoda: Grapsidae) ibom an estuarine area of Sao Paulo northern coast, Brazil. Revista de Biologia Tropical 46:673-678. Litolo, C. 2(X)4. Reproductive aspects of a tropical population of the tiddler crab Uca annulipes (H. Milne Edwards, 1837) (Brachyura: Ocypsdidae) at Costa do Sol Mangrove, Maputo Bay, southern Mozambique. Hydrobiologia 525:167-173. Litnio, C. ^DSa. Population stmcture and rq^roductive biology of the tiddler crab Uca imersa (Hoffman, 1874) (Brachyura: Ocypodidae). Acta Oecxsiogica 27:135-141. Litulo, C. 2005b. PcqmJation biology of the fiddler crab Uca annuUpes (Brachyura: OcypcMiidae) In a tropical East Affican mangrove (Mozambique). Estuarine, Coastal and Shelf Science 62:283-290. Lopez-Gmco, L.S. and E.M. Rodriguez. 1999. Annual reprixiuc- tion and growth of adulte crabs Chmmagnathm gmnu- iata (Crustacea, Brachyura, Grapsidae). Cahiers de Biologic Marine 40:155-164. Mantelatto, F.L,M. and A. Frausozo. 1999. Reproductive biol- ogy and molting cycle of the crab Calimectes orna- tus (Decapoda, Fortunidae) from the Ubatuba region, Sao Paulo, Brazil. Cmstaceana 72:63-76. Melo, G.A.S. 1996. Manual de Identificagao dos B^hyura (Carangoejos e Siris) do Litoral Brasileiro. Ed. Plelade, Sao Paulo, SF, Brazil, 603 p. Mouia, D.E., C.C. Lampareili, RO. Rodrignes, and R.C. \5ncent. 1^8. Decomposigao de folhas em mangoezais na regiao de Bertioga, Sao Paulo, Brasil. Proceedings of the IV SImposio de Ecossistemas Brasileiros 1:130-148. Mouton Jr, E.C. and D.L. Felder. 1995. Reproduction of the fid- dler crabs Uca iongisigtmiis and Uca spinicarpa in a Gulf of Mexico salt marsh. Estuaries 1 8:469-481. Murai, M., S. Goshima, and Y. Henmi. 1987. Analysis of the mat- ing system of the fiddler crab, Uca lactea. Animal Behavior 35:1334-1342. Mural, M., S. Goshiim, and Y Nalrasooe. 1982. Some behavioral characteristics related to tisod supply and soil texture of burrowing habitats observed on Uca vocans and U. iactea perpiexa. Marine Biology 66:191-197. Negreiros-Frasso^, M.L., A. Fransozo, and G. Bertinl. 2(K)2. Reproductive cycle and recruitment period of Ocypade quadrata (D^apoda, Ocypodidae) at a sandy beach in southeastern Brazil. Journal of Cmstacean Biology 22:157- 161. 19 CASTfGLTOm ET AL. Fiiky, K.K. and Y. Ono. 197S. ITie bimiing cycles of two se- cies of grapsld crabs (Cmstacea: Decapoda) from the North coast of Kyushu, Japan. Marine Biology 45:273-248. RtKirigues, A., P. Drake, and A.M. Arias. 1997. Reproductive l^riods and larval abundance patterns of the crabs Fanopeus afriamus and Uca tangeri in a shallow inlet (SW Sp^). Marine Ecology Process Series 149; 133-142. Rosenberg, G.M.S. 2fK}l. The systematics and taxonomy of iddler crabs: A phytogeny of the genus Uca, Jonma! of Cmsl^:ean Biology 21:839-869. Salmon, M. 1987. On the reproductive behavior of the fiddler crab Uca ihayeri, with cxsmparisons to U, pugiiator and U. vocans'. Evidence for behavioral convergence. Joumal of Crasl^sean Biology 7:25-44. Santos, S. and M.L. Negieiros-Fransozo. 1999. Reproductive cycle of the swimming emb PorPmus spinimemm l^treille (Crustacea, Decapoda, Bracbynra) from the Ubatuba, Sao Paulo, Bi^il. Revista Bmsileira de Zoologia 16:1183- 1193. Sastry, A.N'. 1983. Ecological aspects of reproduction. In: F.J. Vemberg and W.B. Vemberg, cds. The Biology of Crasl^sea: Environmental Adaptetions, Vol. 8, Academic Press, New York, NY, USA, p. 179-269. Schaeffer-Novelli, Y. 1995. Manguezal, Ecosistema entre a Iferra e o Mar. Caribbean Ecolo^cai Reseach, Sao ^nlo, SP, BR, 64 p, Thiirman 0, C.L. 1985. Reproductive biology and population structure of the fiddle crab Uca mbcylindrica (Stimpson). Biological Bulletin 169:215-229. Twiliey, R.R., S.C. Snedaker, A. Yanez-Arandbia, and E. Medina. 1995. Mangrove systems. In: V.H. Heywood, ed. Global Biodiversity Assessment, Biodiversity and Ecosystem Function: Ecosystem Analyses. Cambridge University Press, Cambridge, UK, p. 387-393. Warner, G.F. 1977. The Biology of Crabs. Elek Science, 202 p. Wolfrath, B. 1993. Observations on the bebaviour of the European fiiddier crab Uca tangeri. Marine Jkx)logy Process Series 100:111-118. Yamaguchl, T. 1971. Conrtehip behavior of a fiddler crab, Uca lactea, Kumamoto ioumal. of Science Biology 10:13-37. Zar, J.H. 1996. Biostalistical Analysis. Prentice-Hall, Up^r Saddle River, NJ, USA, 663 p. Zimmerman, T.L. and D.L- Felder. 1991. Reprodoefive ecol- ogy of an inteitidal brachyuian crab Sesarma sp. (nr. reticMiatumX from die Gulf of Mexico. Biological Bulletin 181:387_401. 20 Gulf and Caribbean Research Volume 19 Issue 1 January 2007 A Comparison of Fish Assemblages Among Five Habitat Types Within a Caribbean Lagoonal System Ivan Mateo USVJ Division of Fish and Wildlife William J. Tobias USVI Division of Fish and Wildlife DOI; 10.18785/gcr.l901.03 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr & Part of the Marine Biology Commons Recommended Citation MateO; 1. and W. J. Tobias. 2007. A Comparison of Fish Assemblages Among Five Habitat Types Within a Caribbean Lagoonal System. Gulf and Caribbean Research 19 (l): 21-31. Retrieved from http:/ / aquila.usm.edu/ gcr/voll9/ issl/3 This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized editor of The Aquila Digital Community. For more information, please contact Joshua.Cromwell^usm.edu. A COMPARISON OF FISH ASSEMBLAGES AMONG FIVE HABITAT TYPES WITHIN A CARIBBEAN LAGOONAL SYSTEM Ivan Mateo* and 'VMIliam J. IbMas US¥I Division of Fish ami WilMifs, Rmnbow Fhmi 45 Marshili, Frsdericksted, St Cmix USVI 00840 ^Current Address. Department qf Fisheries, Animal and Veterinary Science, University of Rhode Island, Kingston, M 02881 USA. Phone (401) 398-1742, Fax (401) 398-1742, E-rmii imat€o32m fmtmmLcom. ABSTRACT Fish assemblage associate with patch nibble, seagmss, algal plain, and sandy habitats ty|^s were studied at St Croix’s Sontheasterii barrier reef lagoon using underwater visual census techniques. Higher species richness and fish density were observed over patch i^fs/rubble habitat followed by seapass, algal plain, and onvegelated sandy habitat types, Thalmsonui hifasclatum, Eaerntdon flmoUmatum, and Acemthurm chirur- gm were the most coimnon fishes in highly structured habitat types (patch reef, rubble). HalichDems hmttatm, Sparisoma radirms, newly settled gmnts (i.e., Haenmkm spp.), and juveniles of Ocyums chrysurm were mainly asscKjiared with vegetated habitat types (seagrass, algal beds), while Xyrichtys martinicensis and Coryphopterm ghmcofraenum were common over unvegeMted sandy habitat types. Cluster analysis among baekreef lagoon habitat types based on the entire fish density data showed distinct associations of fish assemblages by habitat type, regard- less of season. Fish assemblages in the more structured habitat types were similar to each other but different from unstructured vegetated, and unvegetat^ habitat types. These results suggest that differences in fish sp^es richness and density in the backreef lagoon are related to habitat type. The ^ologlcal impoitooce and need for protoJtion of backr^f lagoon habitat types are discussed in relation to their potential role as mnaeries for many fish species. IhrmooucTiON Nearshore ecosystems such as seagrass meadows, marshes, and mangrove lagoons supply many vital eco- logical functions in coastal waters, including shoreline protection, and nutrient cycling (Ogden and Gladfelter 19B3, Parrish 1989). Most notably, these ecosystems pro- vide food and reflige that supports a great abundance and diversity of fishes as well as shrimp, oysters, crabs, and other invertebrates (Ogden and Ziematin 1977, Shnlman 1984, Shuhnan 1985, Parrish 1989). In the Caribbean, it has generally been accepted that mangroves and sea- grass meadows form important nurseries for juveniles of several reef fish species (Ogden and Gkdfelter 1983, Parrish 1989, Nagelkerken et el. 2000), and juvenile coral reef fishes have been fiequently observed in mangroves and seagrass meadows in the Caribbean (e.g., Baelde 1990, Sedberry and Carter 1993, Appeldoom et al. 1997, Lifideman et al. 1998, Nagelkerken et al 2000), The adults of these species have been observed on reef environments or in offshore waters, suggesting the migration of juvenile from the mangroves mid seagrass beds to the reef or deeper waters at a certain age (Ogden and Ehrlich 1977, Weinstein and Heck 1979, Rooker and Dennis 1991, Appeldoom et al. 1997, Lindeman et al. 1998, Nagelkerken et al. 2000). Although numerous studies have been done on mangrove, seagrass, and coral reef systems, only recently researehers have investigated the connectivity among these coasM ecosystems (Sedberry and Carter 1993, Nagelkerken et id. 2000, Adams and Ebersole 2002). Comparisons of nursery value among nearshore habitat types have usually focused on a single habitat (i.e., mangrove or seagrass) (Robblee and Zieman 1984, Baelde 1990, Rooker and Dennis 1991) even though individual species may use many different habitats. Furthermore, embayments and lagoons often not only contain mangroves and seagrass meadows, but a vari- ety of other shaUow-water habitats like algal plains, areas with bare sediment, sand-rubble zones, or patch reefs. Seagrass meadows and mangroves may be less important as nurseries in regions where animals use alternative habi- tats successftilly. Few studies have quantified the propor- tions of reef fishes that pass through these nursery habitats, and information concerning other habitats that can be used as alternative nurseries are lacking. Thus, linkages of fishes between these backreef lagoon habiMs remain largely unknown (Ogden and Gladfelter 1983, Birkeland 19B5, Pmish 1989). Therefore, this study addresses the foUowmg questions: 1) Does species composition and abundance differ among backreef lagoon habitat types? 2) Which habitat types are used as nurseries by selected fish species? 3) Which backreef lagoon habitat types are uti- lized more by a fish species when multiple nursefry habitat t 3 ^es are present? 4) Do fish species show an ontogenetic shift from nursery habitat types to other backreef lagoon habitat types? 5) Do closely related fish spedes show simi- lar seasonal patterns in habitat use? Mateo and Tobias Figure 1. Location of the southeastern hackreef lagoon of St. Croix, USVI. The 3 embayments studied, Great Pond Bay (GPB), Robin Bay (RB) and Ihmer Hole Bay (THB) are shown. Materials and Methods The nearshore nursery habitat types in 3 protected hackreef lagoon embayments on St. Croix’s southeast coast (Turner Hole Bay, Robin Bay and Great Pond Bay) (Figure 1) were sampled monthly from July 2000 to July 2001. For each bay, a 20 m x 20 m grid pattern was laid over a nautical chart. Grid intersecting points were labeled with consecutive numbers and were the basis for selecting transect starting points for each embayment. The number of starting points surveyed (10 in each of the 3 embay- ments) was based on a preliminary fish census (Rogers et al. 1994), and each starting point was selected randomly each month. At each of the 10 starting points, a single 50 m transect fine (marked at 1 cm intervals) was laid out on a compass bearing randomly selected for each transect. At each starting point, one weighted end of the 50 m transect line tape was dropped and was laid by a diver in the direc- tion of the compass bearing. On each transect, 100 m^ were surveyed visually for fish, with 2 parallel 1 m x 50 m belt transects surveyed by 2 divers swinuning on opposite sides of the transect line. At each transect site, a fish census and a benthic sur- vey was done. Each diver recorded fish species and size class of individuals for each species and size class. Fish size classes were characterized as < 5 cm, 5-10 cm, and > 10 cm in total length (TL). For most species, juveniles < 5 cm were recorded as recruits. For smaller species. such as wrasses, grunts, and damselfishes, juveniles < 3 cm were considered as recruits. Larger individuals in size classes 5-10 cm and >10 cm were considered as sub-adults. To minimize the potential bias of counting the same individual twice along the belt transect, divers conferred with each other using hand signals to make sure fish were counted only once (Eberhardt 1978), and divers were trained to maintain constant swimming speed along the transect, and not to count fish that entered the census area after the visual census had started (Samoylis and Carlos 2000). For the benthic survey, each diver recorded the domi- nant habitat type at the beginning and end of changes in habitat type (to the nearest cm) under the transect line. Five benthic habitat types were identified: Patch reef: isolated, high calcareous structure (not part of the contiguous reef) with a vertical profile that often, but not always, contained live coral cover. Rubble: low-relief calcareous structure composed primarily of dead/dying coral fragments that were not attached to the habitat type. Sand: areas of open sand with little (< 10% cover) or no plants or coralline material, mostly unvegetated. Algal plain: sandy bottom dominated by (>60% cover) Dictyota spp., Halimeda spp., Penicillus spp., Acanthophora spp., and/or Udotea spp., which may have include sparse stands of Syringodium fiUforme and Thalassia testudinum. 22 Mateo and Tobias Habitat types Figure 2. Plot of the total percent of habitat cover found on southeastern barrier reef lagoons in St Croix when all bays were pooled. Number of samples per habitat N = 24. Total area surveyed = 70,000 m^ Seagrass: monospecific or nearly monospecific stands of, with varying densities of S. filiforme. Percent cover of each habitat type was estimated from linear coverage along belt transects. The proportional com- position of each habitat type covered in each belt transect was estimated by measuring the length of line overlying each habitat type and dividing it by the total length of the transect. Prior to conducting data analyses, fish density estimates from both divers were checked for independence with a Pearson product-moment correlation coefficient. This was used to test for independence between the paired diver observations (Zar 1984). If uncorrelated, the paired tran- sects could be used as potentially independent samples. We considered r < 0.50 to indicate independence. Correlation between paired divers was low (r = 0.43, P = 0.343, n = 330), and we interpreted the data generated from the 2 divers as separate and independent census data sets. Data were standardized by month by pooling belt tran- sects of all 3 embayments by habitat type. This allowed for equal sample size {N = 24) for the one year study. Monthly data were further pooled into winter (December, January, February), spring (March, April, May), summer (June, July, August), and fall (September, October, November) seasons. The seasonal grouping was based on previous temperature measurements done by the USVI Division Fish and Wildlife. Normality of the number of fish per transect, number of species per transect, size class distribution, and fish density of the most abundant species were verified with the Kolmogorov-Lilliefors Normality test (Zar 1984). Since the estimates failed the normality test even after logjQ (jc - 1 - 1 ) or square root transformation, non-paramet- ric statistics were used to analyze the data. Overall fish density, species richness, density of the most abundant species recorded on transects within habitat type, density of economically important species among habitat types and size classes were examined with two-way ANOVA on ranks (Sokal and Rohlf 1981) unless specified. If the overall F- value was significant, Tukey’s pair-wise multiple comparison procedure was used to separate mean values. Finally, fish density delineated by habitat type, season and bay were analyzed using the Bray-Curtis similarity mea- sure (Krebs 1999) and clustered using the average linkage method with the PRIMER software package (Plymouth Marine Laboratories, UK). Results Species composition among habitat types Seventy-one fish species were recorded within the St. Croix southeastern backreef lagoonal system. The estimat- ed percent cover of backreef lagoon habitat types pooled by bay was dominated by seagrass and sand, whereas patch-reef and rubble covered the least area of the bottom (Figure 2). Species richness was highest on patch reef (54), followed by rubble (41) and seagrass beds (39) habitats (Table 1), and lowest on algal beds (28) and sandy habitats (10). Thirty-five species occurred over both patch reef and rubble habitats, while the overlap in species between seagrass and the other habitat types was low (Table 1). The most abundant taxa per habitat type were; Patch reef: Thalassoma bifasciatum, Haemulon fla- volineatum, Halichoeres bivittatus, Acanthurus chirur- gus, and newly settled grunts {Haemulon spp.), which all together made up 56% of the total number of fishes recorded from the patch reef habitats. Rubble: newly settled grunts {Haemulon spp.), H bivittatus, A. chirurgus, S. leucostictus and T. bifasciatum, which together made up 76% of the total number of indi- viduals at this habitat type. Seagrass: newly settled grunts {Haemulon spp.), Sparisoma radians, H. bivittatus, H. flavolineatum, and A. chirurgus. Those species comprised 90.0% of the total fishes recorded in the seagrass beds. Algal plain: newly settled grunts {Haemulon spp.), H bivittatus, H. flavolineatum, S. radians, and A. chirurgus, which made up 96% of the total number of individuals. Sand: newly settled grunts {Haemulon spp.), Xyrichtys martinicensis, Caranx ruber, H. bivittatus, and Coryphopterus glaucofraenum. Those species comprised 98.0% of the fishes recorded in sandy habitats. Variation in species richness Species richness differed significantly among habitat types (ANOVA, P < 0.001, df = 4, 20) but not among sea- 23 Mateo and Tobias TABLE 1 :Fisli spedes atandaiiee m iieti-stioire liabttet types In a tropical lagoon in soatlieastem coast of St, Crotej US¥I, Spedes Patch Meef EnfeWe Seagrass Algal Plain Sand Total Haemuhn spp. 337 646 2599 3824 256 7662 Halichoeres bivittatus 406 31S 615 241 27 1607 Spansoma radians 39 71 758 53 0 921 Haemulon JfavoUneatum 446 59 248 156 0 909 Acanthurus chimrgi4s 383 134 139 6 1 663 Umlassorm bifasciaium 464 117 21 0 0 602 Siegastes ieucostictus 349 144 31 5 0 529 Acanthums bahianus 210 144 84 36 0 474 Scarus iseri 224 78 21 10 0 333 Siegastes partitas 164 44 3 0 0 211 Xyrichtys martinicemis 11 0 29 25 134 199 Ocyurus chrysurus 31 5 76 28 0 140 Latjanus mahogonl 25 32 44 17 0 118 Haemulon plumierii 78 2 20 16 0 116 Pseudupeneus maculatus 12 7 71 0 0 90 Cryptatomus mseus 0 0 55 22 4 80 Sparisoma viride 54 22 0 0 0 76 Holocentms adscemionis 62 8 0 0 0 70 Siegastes adustus 51 8 0 0 0 59 Sphoemides spengteri 7 4 17 28 0 56 Halichaeres poeyi 4 1 45 5 0 55 Cmthigaster mstrata 26 4 8 7 0 45 Acanthurus coemleus 24 3 14 0 0 41 Camwc ruber 4 0 j 1 28 40 Epinephelus guuatus 32 7 0 0 0 39 Sparisoma chrysopterum 25 9 0 0 0 34 Sparisoma auropenatum 32 4 0 0 0 36 Chmtodon stnatus 4 11 2 10 0 27 CephaiophoUs julva 17 10 0 0 0 27 Myripnsds Jacobm 24 0 2 0 0 26 Coryphopterus glaucopaenum 0 17 0 0 9 26 Monacmihus ciUaius 0 1 16 2 0 19 Abudefduf samtiUs 14 1 0 0 0 15 Chmmis multHineam 13 2 0 0 0 15 Holomnthus ciliaris 12 0 1 1 0 14 Micfuspathodon chrysurus 8 4 0 0 0 12 Halichaeres radians 10 2 0 0 0 12 Caranx crysos 0 0 7 4 0 11 Chaetodon capistmtus 7 1 2 0 0 10 Pareques acuminaius 10 0 0 0 0 10 Stegmtes pianijmns 6 4 0 0 0 10 Stegastes variabilis 3 6 0 0 0 9 Xyrichtys splendens 0 0 3 3 1 7 Apogon spp. 2 2 0 2 0 6 Senmius tigrinus 5 1 0 0 0 6 Faraiichthys tropicus 0 0 3 1 1 5 24 Mateo and Tobias TABLE 1 (csut) Fisli species abEodaiice within ii.eaishore habitat types in a tropical lagoon in southeastern coast of St. Croix, USVT, Species Patch Reef Rubble Seagrass Algal Plain Sand Total Batistes vetula 3 2 0 0 0 5 Sphymem harracisda 0 0 4 0 0 4 Lutjanus symgris 1 0 2 1 0 4 Lactaphrys triqueter 0 0 1 1 2 4 Chromis cymea 3 0 0 0 0 3 Auiostomus maculatus 0 0 2 0 0 2 Calamus bajonado 0 0 2 0 0 2 Lutjanus analis 0 0 2 0 0 2 Bothus lumiius 0 0 1 1 0 2 Cosmocampus elucens 0 0 1 1 0 2 Bodiamus mfm 1 1 0 0 0 2 Mulloidichthys martmiciM 2 0 0 0 0 2 Pomaemthus paru 2 0 0 0 0 2 Scambemmorus regalis 0 0 1 0 0 1 Sphymena picudiila 0 0 1 0 0 1 Diodon hystrix 1 0 0 0 0 1 Domtonotus megalepsis 1 0 0 0 0 1 Cephaiophoii^ cruentata 1 0 0 0 0 1 Germs cinereus 1 0 0 0 0 1 Gobiosoma spp. 1 0 0 0 0 1 . Hypoplectrm gummigutta 1 0 0 0 0 1 Hypoplectrus nigricans 1 0 0 0 0 1 Lactaphrys spp. 1 0 0 0 0 1 Scaq^aena spp. 0 1 0 0 0 1 Synodus foetens 0 1 0 0 0 1 sons (ANOVA, 0,113, df == 3, 20; Figure 3). Post hoc tests showed that species richness in patch reefs and rubble was significaoriy higher than seagrass, algal beds, and sandy bottom types (Tukey’s test, P < 0,001). There was no sig- nificant interaction between habitat type and season in spe- cies richness (ANOVA, P = 0.082, df = 12, 20; Figure 3). Varistlon in fish densi^ Fish density was significantly different among habi- tat types (ANOVA, P< 0.001 df = 4, 20) and seasons (ANOVA, ,F< 0.002, df= 3, 20), but there was no inter- action between the main effects (ANOVA, P<0360, df = 12, 20; Figure 4). Fish density was significantly high- est in pateh reefs and mbble relative to seagrass, algal beds and sandy habitat types across all seasons (Ibkey's test, F < 0.001). Comparisons of fish density pcx^led by habitat type indicated that fish density observed in summer was statistically higher than in the spring and winter season (Tukey^s test, F < 0.05). Similarity of fish faeims immiig hshllnl; type end sea- son Fish assemblages recorded frem each habitat type tended to cluster together (Figure 5) showing a compara- tively high degree of similarity; season appeared to have no effect. Fish assemblages observed over patch reefs and mbble were more similar to each other than to those observed in vegetated and unvegetated habitat types. Unvegetated habitat types clustered together separately from all other habitat types. Yaristloii m desisily of econoimcally importaiit spedes coiiiErioii species by habitat type Acmthurus chimrgus reemits (< 5 cm) and sub-adults /m^ (5-10 cm, > 10 cm) were higher in patch reefs and mbble (Figure 6a) than any other habitat types (ANOVA, P < 0.001; df = 4, 8; Ihkey’s test, P < 0.001). Density of newly settled grunts/m^ {Haemulon spp., < 3 cm) were higher on algal plains compared with other habitat types (ANOVA, F< 0.001; df:::4, 8; Tukey^s test, P< 0.009; 25 Mateo and Tobias □ Patch reef □ Rubble 0Seagrass Figure 3. Plot of the mean (s) species richness by habitat type Figure 4. Plot of the mean (s) fish density by habitat type observed during different seasons, s = s tandar d deviation. observed during different seasons, s = standard deviation. Figure 6b). Scams iseri recruits/m^ (< 5 cm) were more dominant in patch reefs (Figure 6c), than juveniles (5-10 cm) and sub-adults (> 10 cm)/m^ (ANOVA, P<0.01; df = 4, 8; Tukey’s test, P < 0.001). Ocyurus chrysurus recruits (<5 cm)/m^ were highest on algal plains and seagrass while larger individuals (> 10 cm) were most abundant on patch reefs (ANOVA, F<0.01; df = 4, 8; Tukey’s test, P < 0.02; Figure 6d). Density (ind/m^) of size class > 5 cm was almost exclusively found in patch reefs for Haemulon plumierii (ANOVA, P< 0.001, df = 4, 8; Tukey’s test, P < 0.003; Figure 6e) and H. flavolineatum (ANOVA, P< 0.001, df = 4, 8; Tukey’s test, P< 0.002; Figure 6f) . There were no significant interaction terms among habitat type and size class for fish density for all species studied (ANOVA, P < 0.05, df = 8, 20). Fall Rubble Summer Rubble Spring Rubble Winter Rubble Fall Coral Summer Coral Spring Coral Winter Coral Fall Algae Summer Algae Spring Algae Winter Algae Fall Seagrass Summer Seagras! Spring Seagrass Winter Seagrass Fall Sand Summer Sand Spring Sand Winter Sand Dissimilarity Figure 5. Cluster analysis of fish density by habitat type and season based on visual census using the Bray-Curtis similar- ity metric and average linkage clustering techniques. Occurrence and density of recruits and non-recruits of common fish species Recruit density of H. bivittatus (ind/m^) was highest in patch reef habitat in summer relative to other habitat types (ANOVA, df = 4, 20, P< 0.001; Tukey’s test, P< 0.001; Figure 7a). Haemulon bivittatus sub-adults (5-10 cm, >10 cm) showed significant seasonal differences in den- sity (ind/m^) (ANOVA, P < 0.004, df = 3, 20; Figure 7b), with the highest mean density being recorded in winter over patch reefs for both size classes (Tukey’s test, df = 4, P < 0.001; Figure 7c). There were no significant interac- tion terms between habitat type and season for H. bivit- tatus recruits (ANOVA, P = < 0.686, df = 12, 20) and H. bivittatus sub-adults (ANOVA, P < 0.902, df = 12, 20). A higher density (ind/m^) of newly settled grunts {Haemulon spp.) was observed over algal beds during summer and fall seasons (ANOVA, P < 0.01; df = 4, 20; Tukey’s test, P < 0.009; Figure 7d). No significant interaction terms between habitat type and season were found for newly settled grunts Haemulon spp. density (ANOVA, P < 0.578, df = 12, 20). Non-recruit grunt {Haemulon spp) density (ind/m^) (5-10 cm) was significantly highest in spring (ANOVA, P< 0.004, df = 4, 20; Tukey’s test, P<0.01; Figure 7e) over patch reef habitats, whereas the density (ind/m^) of grunt >10 cm in patch reefs were highest during sununer (Figure 7f). There were no significant interaction terms between habitat type and season in non-recruit Haemulon spp. density (ANOVA, P < 0.563, df = 12, 20). Discussion Differences in species richness and fish density The nearshore environment within St. Croix’s south- east bank barrier reef lagoon exhibited distinct patterns in the distribution of fish assemblages among seagrass, algal plains, patch reefs, rubble, and sandy habitat types. The 26 Mateo and Tobias Size class (cm) Size class (cm) Patch Reef □ Rubble jSeagrass jAlgal Plain I Sand Figure 6. Plot of density of economically important species by size class and habitat types: 6a) A. chirurgus, 6b) O. chrysurus, 6c) Haemulon spp., 6d) S. iseri, 6e) H. flavolineatum, 6f) H. plumierii. seagrass and algal plain were dominated by small resident species, such as Halichoeres spp. and S. radians, and by juveniles of non-resident species like Haemulon spp. and O. chrysurus. Rubble and patch reefs harbored higher spe- cies richness and were mostly dominated by small juvenile damselfishes, parrotfishes, grunts, and doctorfishes. The highest species richness occurred over patch reefs and rubble, than over vegetated habitat type types (seagrass, algal plains) and unvegetated sandy habitat types during all 4 seasons. Fish densities were also higher in patch reefs and rubble than over seagrass, algal plains, and sandy bot- toms. Fish assemblages in physically ‘structured environ- ments’ (patch reef and rubble) tend to be more similar to each other than to those in vegetated (seagrass/algal beds) and unvegetated habitat types (sand) (Nagelkerken et al. 2000, Adams and Ebersole 2002). The cluster analysis based on our data confirms this pattern, illustrating that fish assemblages from patch reef, and rubble habitat types were more similar to each other, but different from assem- blages associated with seagrass, algal plain, and sandy habitat types, regardless of season. Nagelkerken et al. (2000) and Adams and Ebersole (2002) similarly observed a hierarchy in fish abundance within highly structured hab- itat types (patch reefs, rubble, cobbler) possessing higher density followed by seagrass and then algal beds, and over unvegetated sand bottom. Seasonal Distribution Seasonal changes in species composition and density of fish populations were major characteristics in our near- shore lagoonal habitat types. Peak density of total fishes occurred during summer, with a secondary peak in fall and the lowest density of total fishes in winter. Although 27 Mateo and Tobias H. bivittatus (< 5 cm) H. bivittatus (5 -10 cm) Haemulon spp. (5-10 cm) B Patch reef □ Rubble Seagrass Algal plain Sand Figure 7. Plot of seasonal occurrence of common species by habitat type: 7a-c) H. bivittatus; 7d-f) Haemulon spp.; < 5 cm, 5-10 cm, > 10 cm, respectively. recruits were mainly abundant in the latter half of the summer and in the fall, the pattern of habitat type use by settlers and juveniles differed by species and season. For example, 50% of all recruits occurred in late sum- mer whereas 42% occurred in Fall and 8% in the rest of the year. Tropical fish assemblages have shown seasonal fluctuation characterized both by a higher species richness and by a higher abundance of fishes during the summer and fall, which is influenced by recruitment of juveniles, the increase of food availability, and spawning patterns (Ogden and Gladfelter 1983, Baelde 1990). Considering that much of the research of settling and juvenile fishes in the Caribbean has taken place during the summer, future research should focus on the relative mechanism of trans- port processes throughout the year. Habitat type distribution Distinct patterns of species among habitat types were observed at St. Croix’s southeastern barrier reef lagoon with some species being found exclusively or predomi- nantly in one of the 5 habitat types. The scarid, C. roseus, and the labrid, H. poeii, for example, were more associated with seagrass, whereas other labrids such as T. bifascia- tum and H. radians, the scarids S. aurofrenatum, S. iseri, and S. viride, and the squirrelfishes H. adcensionis and M. jacobus were found predominantly over patch reefs. Fish species found predominantly over bare sand were X. martinicensis and C. glaucofraenum. In contrast, H. bivittatus, S radians, newly settled grunts Haemulon spp., H. flavolineatum, H. plumierii, the lutjanids O. chrysurus and L. mahogoni, and the acanthurids A. chirurgus and A. bahianus were commonly associated with more than one habitat type. The differences in fish size distributions among habitat types suggested different ontogenic distri- bution patterns by species. For example, many recruits of economically important species (e.g., S. iseri, A. chirurgus) were on patch reefs and rubble whereas recruits of newly settled Haemulon spp. and O. chrysurus were mostly in 28 Mateo and Tobias seagrass beds and algal plains. Density of larger (> 10 cm) fish (e.g,, K piumieni, S. iseri, H, JkvoUmatum, aod O, chrysurus) weie low iti patch reefs and seagrass beds, how- ever. These results reflect the temporary and successive use of seagrass and patch reef habitats by juveniles of various species that move elsewhere as they grow, as it is the case for O. chrysurm, and S. iseri (Qifton 1991, Tolimieri 199S, Nagelkerken et al. 2000, Cocheret de la Morini^e et al. 2002). Changes in habitat type use might also be expected to coincide with size-related changes in individ- ual fitness or physiological requirements (Appeldoom et al. 1997, Lindeman et al. 1998), or ontogenetic changes in diet, mortality, and/or competitive mteractioos (Shulman and Ogden 1987, Nagelkerken et al. 2000, Cocheret de la Mormifere et al. 2002). For example, some snappers, grunts, and parrotfishes progressively change habitat type with size (Appeldoom et al. 1997, Lindeman et al. 1998, Adams and Ebersole 2CK)2). Ocyums chrysurus^ H. fla- voUmatum, and K plimikm settle onto seagrass beds and algal plains and migrate to nearby reefs at larger sizes (Shulman and Ogden 1987, Appeldoom et al.. 1997, Lindeman et al. 1998), Distiibutioii of S. iseri appears to be determined by habitat-based distribution of primary food sources (CHfton 1991, Tolimieri 1998). As their for- aging eSicieiicy and home range increases ontogenetically, these nursery areas no longer provide adequate shelter and food sources; thus, the species migrate to deeper habitat like fore-reef and mid-shelf reef to meet their ecological requirements (Clifton 1991, Tolimieri 1998). Our study suggests that nurseiy habitats are not limi ted to seagrass and mangrove systems, but include other nearshore habitats like patch reefs, rubble areas, and algal plains. Patch reefs and rubble habitats had the highest density of fish recruits; however, total counts of fish recruits were higher on seagrass beds and algal plains because of the areal coverage of these latter habitat types in the bays surveyed (Figure 2). In the Caribbean, recent studies of nearshore fish assemblages suggest that patch reefs and rubble areas appear to be important shelter sites for juveuile fishes in mangrove and seagrass dominated lagoons (Risk 1997, Nagelkerken et al. 2CKX1, Adams and Ebersole 2002). For example in Curasao, shallow- water coral reefs were utilized more by K chrysargymum, L. muhogoni^ A. bahianus, and A. samtilis ^ nurse^ are^ than seagrass and mangrove lagoons (Nagelkerken et al. 2000). In southwestern Puerto Rico, H. flavolmeatum, O, chrysurus, and H, plumierU showed greater preference for shallow coral reefs as nursery areas than mangrove and seagrass (Murphy 2001). Finally, similar patterns of habitat use were noted for K fiavolimatum, A. bahimm\ and A. chirurgus as they were not strictly dependant on mangroves and seagrass as nurseries but used alternative nursery areas like shallow coral reefs and rubble areas in St Croix (Risk 1997, Adams and Ebersole 2002), However, to evaluate the role of nearshore habitat types as fish nursery, more studies should be done focus- ing in understanding how these habitat types may provide a nursery function role to reef fishes. During a study of seagrass and mangrove fishes in Belize, CMttaro et al. (2005) found that based on density, assemblage composi- tion, and relative rates of predation, not all mangrove and seagr^s beds appeared to offer nursery fiinction. Their study highlighted the need to avoid generalizations about mangroves and seagrass having nursery related functions, if estimates of density are the only method to confhm nurs- ery potential. Additionally, in the Indo-Pacific, the nursery value of mangroves and seagrass have been questioned as juvenile fishes did not show evidence of using mangroves as shelter (Thollot, 1992, Huxham et al. 2004). Therefore, many factors like density, survival, growth, and movemem among habitat types have to be examined simultaneously in order to support a particular habitat type as being a nurs- ery (Beck et al, 2001), Impllcutioiis for numageuieet Knowledge of habitat type use patterns by different fish life stages along a cross-shelf gradient is needed to unckmt^d tiie importance of nearshore habitats as nurs- ery areas and presumable ontogenetic shifts in habitat type requirements. Based on these data, it would then be possible to infer connectivity of reef fish migrating among habitat types from inshore to offshore duiiag post-settle- ment ontogeny (Appeldoom et ai. 1997, lindeman et al. 1998, Nagelkerken et al. 2000). Determinatton of nursery value of nearshore habitats and ontogenetic shifts in fish habitat type use (Appeldoom et al. 1997, Lindeman et al. 1998, Cocheret de la Morin&re et al. 2002) would facilitate fisheries conservation and coastal zone management plans. For example, designing marine protected areas (MPAs) and improving the efficacy of the proposed MPAs in areas adjacent to the St Croix East End Marine Park, As our study suggests, there is increasing evidence that many reef fish are dependent on nearshore systems that comprise a mosaic of habitat types including not only coral habitat structure but also a mixture of seagrass, algal plains, and rubble. Each of these habitat types contains unique biotic communities that vary differently depend- ing on the scale at which individual or community level processes are observed. Strong linkages exist between fish and habitat and successful implementation of marine reserves requires knowledge of location, distribution, and extent of habitat ty^s necessary for successful recruit- 29 Mateo and Tobias ment* gmwthj feedings and reprodnctiois. To mcasuto the efficacy of a marine reserve to enhance fish abundance, it is critical to develop a baseline against which fntore estimates can be compared. The lesnits found in this study provide a foundation upon which an ecosystem approach could be developed. As resource managers address the placement and design of MPA"s, information on habitat type use and life-history characteristics of coral reef fishes are required and will become a vital part of the decision- making process. The use of lagoon habitat types as nurser- ies by reef-associated fishes must be incorporated into any management plan. Acknowledgments We thank W. Ventura, H. Rivera, and K. Barnes for their logistical support during the field survey and A. Adams, A. Sabat, R. Appeldoom, J. Holmquist, and D. Durant for their valuable comments on earlier drafts of the manuscripts. This study was funded by US Fish and Wildlife Service Federal Aid Grant, F-7 imder the Dingell" Johnsoffl Sport Fish Restoration Act, Litematuse Cited Adams, A, and J, Ebersole, 2002, Use of lagoon habitats and back reef by coral reef fishes. Marine Ecology Progress Series 228:213-226. Appeldoom, C.W, Recksieck, R,L, Hill, RE, Pagan, and G.D. Dennis. 1997, Marine protected areas and reef fish movements: The role of habitat in contiollmg ontogenetic movements. Proceedings 8th Internationa Coral Reef Symposium 2:1917-1922. Baelde, P. 1990. Differences in the structures of fish assem- blages in. Thalmsia iestudimm beds in Guadeloupe, French West Indies, and their ecological significance. Marine Biology 105:163-173. Beck, M, W., K.L, Heck, Jr, K.W, Able, D.L, Childers, D.B. Eggleston, B.M. GiUanders, B. Halpem, C.G. Hays, K. Hoshlno, TJ. Minello, RJ. Orth, P,E Sheridan, and M,E Wemste.m. 2001, The identification, conservation and management of estuarine and marine nurseries for fish and invertebrates. BioScience 51:633-641. Birkeland, C. 1985, Ecological interactions between tropi- cal coastal ecosystems. UNEP Regional Seas Reports and Studies 73:1-26. Chittaro, P,M., E Usseglio, and P,F, Sale, 2005, Variation in fish density, assemblage composition and relative rates of predation among mangroves, seagrass ^d coral reef habitats. Euyiromnental Biology of Fishes 72:175-187 Clifton, K,E, 1991. Subordinate group members act as food finders within striped panotfish territories. Journal of Experimental Marine Biology and Ecology 145:141™148, Cocheret de la Moriniere E., I Nagelkerken, B.J. A, PoUux, M.A Hemminga, M.H.L. Huiskes, and G. van der Velde, 2002, Ontogenetic dietary changes of coral reef fishes in the mangrove-seagrass-reef continuum; Stable isotopes and gut-content analysis. Marine Ecology Progress Series 245:279"2S9. Eberhardt, L,L, 1978, Transect methods for popniaticm studies. Journal of Wildlife Management 42:1-31. Huxham, M., E. Kimani, and J, Augley. 2004. Mangrove fish: A comparison of commimity structure between forested and cleared habitats. Estuarine, Coastal Shelf Science ^:637“647, Krebs, C.J. 1999. Ecological Methodology. 2nd edition, Addison Wesley, Longman Inc., Menlo Park, IL, USA, 620 p. Lindeman, K.C., G. Diaz, J.E. Serafy, and J.S. Ault. 1998. A spatial framework for assessing cross-shelf habitat use among newly settled grunts and snappers. Proceedings of the Gulf Caribbean Fisheries Institute 50:385-416. Murphy, B,R. 200L Comparison offish communities with- in mangrove, seagrass, and shallow coral reef habitats In southwestern Puerto Rico using a cross-shelf habitat classification system, MS Thesis, University of Rhode Island, Kingston, RI, USA, 151 p. Nagelkerken, L, M. Doienbosch, W.C.E.P. Verbek, E. Cocheret de la Moriniere, and G. van der Velde, 2000, Importance of shallow-water biotopes of a Caribbean bay for juvenile coral mef fishes: Patterns in biotoj^ association, commum.ty structure, and spatial distribu- tion. Marine Ecology Progress Series 202:175-192. Ogden, J.C, and P.R. Ehrlich. 1977. The behaviour of heterotypic resting schools of juvenile grunts (Pomadasyidae). Marine Biology 42:273-280. Ogden, J.C. and J.C, Zieman. 1977. Ecological aspects of coral reef, seagrass beds in the Caribbean, Proceedings 3rd Intemational Coral Reef Symposium 1:377-389. Ogden, J.C, and E.H. Gladfelter. 1983. Coral reefs, sea- grass beds, and mangroves: Their interactions in the coastal zones of the Caribbean, UNESCO Reports in Marine Science 23. 133 p. Parrish, J,D, 1989, Fish communities of interacting shal- low-water habitats in tropical oceanic regions. Marine Ecology Progress Series 58:143-60. Risk, A, 1997, Effects of habitat on the settlement and post-settlement success of the ocean surgeonfish Aamthurus bahianus. Marine Ecology ftogress Series 61:51™59. Robblee, M,B. and J.C. Zieman. 1984. Diel variation in the fish fauna of a tropical seagrass feeding ground. Bulletin Marine Sciences 34:335-"345, 30 Mateo and Tobias Rooker, J.R. and Dennis* 1.991* Diel lunar and sea- sonal changes in a mangrove fish assemblages off southwestern Puerto Rico. BuJletm Marine Scieiice 49:6B4»69B* Samoilys, M. and G. Carlos. 2000. Deteonimng methods of underwater visual census for estimating the abun- dance of coral reef fishes. Environmental Biology of Fishes 57:287-309. Sedberry, G.R. and J. Carter. 1993. The fish commu- nity of a shallow tropical in Belize, Central America. Estuaries 16:198-215. Shulman, M.J. 1984. Resource limitation and recruitment patterns in a coral reef fi.sh assemblages. Journal of Experimentai Marine Biology and Ecology 74:85- 109. Shulman, M.J. 1985. Recruitment of coral reef fish- es: Effects of distribution of predators and shelter. Ecology 66:1056-1066. Shulman, MJ. and J.C. Ogden. 1987. What controls tropical reef fish populations: Recruitoieiit or benthic mortality? An example in the Caribbean reef fish Hmmulon flavolineatmt. Marine Ecology Progress Series 39:223-242. Sokal, R.R. and FJ. Rohlfi 1981. Biometry. W.H. Freeman and Company, New York, NY, USA, 959 p. Tbollot, R 1992. Importance of mangroves for Pacific reef fish species, myth or reality? Proceedings 6th International Coral Reef Symposium 2:934-941. Tolimieri, N. 1998. Contrasting effects of roicrohabitat use on large scale adult abundance in two families of Caribbean reef fishes. Marine Ecology Progress Series 111:1-6. Weinstein, M.P. and K.L. Heck, Jr. 1979. Ichthyofauna of seagrass meadows along the Caribbean coast of Panama and In the Gulf of Mexico: Composition, structure and community ecology. Marine Biology 50:97-107. Zar, J.H. 1984. Biostatistical Analysis. 2nd edition. Prentice Hall, NJ, USA, 718 p. 31 Gulf and Caribbean Research Volume 19 Issue 1 January 2007 Abundance and Ecological Distribution of the "Sete-Barbas" Shrimp Xiphopenaeus kroyeri (Heller^ 1862) (Decapoda: Penaeoidea) in Three Bays of the Ubatuba Region^ Southeastern Brazil Rogerio C. Costa Universidade Estadual Paulista, Brazil Adilson Fransozo Universidade Estadual Paulista, Brazil Fulvio A.M. Freire Universidade Estadual Paulista, Brazil Antonio L. Castilho Universidade Estadual Paulista, Brazil DOI: 10.18785/gcr.l901.04 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr & Part of the Marine Biology Commons Recommended Citation Costa, R. C., A. Fransozo, F. A. Freire and A. L. Castilho. 2007. Abundance and Ecological Distribution of the "Sete-Barbas" Shrimp Xiphopenaeus kroyeri (Heller, 1862) (Decapoda: Penaeoidea) in Three Bays of the Ubatuba Region, Southeastern Brazil. Gulf and Caribbean Research 19 (l): 33-41. Retrieved from http : // aquila.usm.edu/ gcr/vol 1 9/ iss 1 /4 This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized editor of The Aquila Digital Community. For more information, please contactJoshua.Cromwell@usm.edu. ABUNDANCE AND ECOLOGICAL DISTRIBUTION OF THE “SETE-BAR- BAS” SHRIMP XIPHOPENAEUS KROYERI (HELLER, 1862) (DECAPODAj PENAEOIDEA) IN THREE BAYS OF THE UBATUBA REGION, SOUTH- EASTERN BRAZIL Moglii© C, Adilsoii FaiYi© A,M, Frelre^s Aatoaio L. Castillio^ ^^^NEBECC (Gmup of Studies on Crustacean Bioiogy, Ecology and Culture) ^Departamento de Ciincias Bioldgice^, Facuidade de Ciincim, Universidade Estadual Faulistaf UNESP, 17033-360, Bauru, Sdo F^ulo, Brazil E-mail rccosta@fc,imesp.br ^Departamento de Zoologia, Instituto de Biocieiwias, JJniversidade Estadual Faulista, 18.618.000— Botucatu, Sdo Paido, Brazil Email fransozo^ ibb.unesp.br ABSr^CTThe influence of environimEtal factors on the abundance and spatiai~tem|sorai distribution of die shrimp Xiphopenaeus kmyeri was investigated in southeastern Brazil over 2 years. Monthly colections were conducted in Mar Vlrado, Ubatoba and Ubatumitim Bays using a commercial shrimp fishing boat equipped with 2 “double-rig” nets. Each bay was divided into 6 sampling stations, all of which were less than 25 m deep. The spatial distribution of X hYzyeri differed among Bays. Highest abundance values were recorded in areas where silt and clay comprised mote than 70% of the bottom sediment Abundance of X kroyeri followed a seasonal trend, being higher during fall and winter, when intrusions of tropical waters are frequent, causing an increase in salinity (>35%<}) and temperature (> 21® C). A clear decease in shrimp abundance followed a decrease in bottom temperattim (< 20® C) during spring and s ummer due to the influence of cold water currents, particularly the South Atlantic Central Water. These result suggest that sediment type, salinity, and tempemture are among the most important variables affecting the spatial and seasonal distribution of this species. iNTEODUCnON The seabob shrimp Xiphopenmas kmyeri (Heller, 1862), coimnonly known in Brazil as “camarao sete- barbas,” is widely distributed in ttie Western Adentic from Cape Hatteras (North Carolina, USA) through the CairibbeaB region to southern Brazil (State of Rio Grande do Sui) (Dlucao 1995, Costa et aL 2000, Castro et aL 2005). TMs species is the second most important fishery resource iu southeastern Brazil and is the most heavily exploited benthic shrimp species on the coast of the slate of Sao Paulo (DTucao et aL 2002, Castm et al 2005). In addition, X, kroyeri plays m important ecological role in maietainiTig the stabili.ty of trophic relationships in ben- iMc co mm unities (Piles 1992, Nakag^ and Negreiros- Fransozo 1998). Xiphopenaeus kroyeri has been heavily exploited over the past few decades, at times accounting for 90% of aU penaeoid shrimps caught in shallow waters down to 20 m (Costa 2002, Fransozo et al. 2002). During the 1980s and early 1990s, their combined biomass aver- aged over 10, OCX) t/yr but declined to less than 5, OCX) t/yr in the late 1990s (D'lncao et al. 2002). Studies on X kmyeri to date have focused on aspects of its geographical and bathymetric distributions (Williams 1984, DTncao 1995, Boschi 2000), faunal surveys along die Sao Paolo coast (NakagaM et al. 1995, Costa et aL 2000, 2003), or abundance and diversity patterns withm the benthic community (Fires 1992). However, virtually nothing is known about the ecological distribution of X kroyeri along the Brazilian coast or elsewhere in the Western Atiantic. An important aspect of the area investigated is its hydrographic stmctnre (Fires 1992). According to Castro- Filho et al. (1987), 3 water masses are present on the marine continental shelf, with different distribution pat- terns in summer and winter. Coastal Water (CW) has high temperature and low salinity (T > 20® C, S < 36 psn). Tropical Water (TW) has both high temperature and salin- ity (T > 20® C, S > 36 psn), and Sonth Atlantic Central Water (SACW) has both low temperature and salinity (T < 18® C, S < 36 psn). These water masses interact to modify the temperature, salinity, and food availability dur- ing the course of the year. We analyzed the spatial and temporal distribution patterns of X. kroyeri in 3 Bays in the Ubatoba region, Sao Paulo State, Brazil. Abundance patterns are related to Yariation in sahsity, temperature, depth, sediment compo- sition, and organic-matter content. Matemial and Methods Shrimp were collected monthly from January 1998 to December 1999 in Mar Virado (MV), Ubatuba (UBA), and Ubatumirim (UBM) Bays, located in the Ubatuba region, state of Sao Paolo, Each bay was classified into 6 stations, which were selected to include the full range Costa et al. of environmental conditions where X kroyeri is found. These conditions included; their position relative to the bay mouth, depth, and sediment type; the presence of a rocky wall or beach along the bay shore; the inflow of fresh water; and the proximity of offshore water, i.e., open areas with higher salinity. Four of the stations were located at mean depths of 5 (IV), 10 (III), 15 (II) and 20 m (I), and 2 were established adjacent to rocky shores (an exposed and a sheltered shore, stations V (9 m) and VI (6.5 m), respec- tively) (Figure 1). A shrimp-fishing boat equipped with 2 double-rig nets (mesh size 20 mm and 15 nun in the cod end) was used for trawling. At each station we trawled over a 30-min period, covering 18,000 m^. At each station, salinity and temperature (bottom and surface water), depth, organic matter content (%), and grain size of sediments were measured. The sampling methods and the analysis of environmental factors during the same period have been described by Bertini et al. 2001. Bottom water was sampled using a Nansen bottle. Salinity (psu) was measured with a optic refractometer Atago S/1000 and temperature (° C) using a thermometer attached to the bottle. An ecobathymeter coupled with a GPS was used to record depth (m) at sampling stations. Sediment samples were collected at each station with a Van Veen grab (0.06 m^). In the laboratory, the sediment was oven-dried at 70° C for 72 h. For the analysis of grain size composition, two 50-g sub-samples were treated with 250 ml of a 0.2 N NaOH solution, stirred for 5 min to separate the silt and clay particles, and then rinsed on a 0.063-mm sieve. Sediments were sieved through 2 mm (gravel); 2.0- 1.01 mm (very coarse sand); 1.0-0.51 mm (coarse sand); 0.50-0.26 mm (medium sand); 0.25-0.126 mm (fine sand); and 0.125-0.063 mm (very fine sand); smaller particles were classified as silt-clay. Grain size catego- ries followed the American standard (Wentworth 1922), and fi'actions were expressed on the phi (0) scale, thus accounting for the central tendency of sediment samples, e.g., -1 = phi < 0 (gravel); 0 = phi < 1 (coarse sand); 1 = phi < 2 (intermediate sand); 2 = phi < 3 (fine sand); 3 = phi < 4 (very fine sand) and phi > 4 (silt + clay) (Tucker 1988). Cumulative particle-size curves were plotted on a computer using the phi scale, with values corresponding to 16th, 50th, and 84th percentiles being used to determine the mean diameter of the sediment using the formula Md = (^i 6 ^^ 50 ^^ 84 V^- Filially? phi was calculated using the formula phi = -log 2 d, where d = grain diameter (mm) (Tucker 1988). The organic matter content (%) was obtained by ash- weighing: 3 aliquots of 10 g each per station were placed in porcelain crucibles and burned for 3 h at 500° C, and the samples were then weighed again (see Mantelatto and Fransozo 1999). The abundances of shrimps were compared among years, bays, stations, and seasons (summer (January- March), autumn (April-June), winter (July-September), and spring (October-December)) of the year using an anal- ysis of variance factorial model (ANOVA, P < 0.05). The homoscedasticity (Levene test) and normality assumptions Figure 1. Map of the Ubatuba region with the indication of stations in each bay. 34 Costa et al. Months Figure 2. Boxplot showing mean (±s), maximum and minimum salinity values for each month during 1998 and 1999. 5= stan- dard deviation. were examined and the data were log^Q-transformed prior to the analysis (Zar 1999). The influence of environmental factors on the species abundance was evaluated by multiple linear regression and also compared through analysis of variance (ANOVA, P<0.05) (Zar 1999). Results Monthly and among-station variations in mean bottom salinities are shown in Figures 2 and 3. In general, highest salinity values (>35 psu) were found during the autumn (May and June in 1998 and April-June in 1999), whereas the lowest values occurred during early spring (September and October) in the first year and during winter and spring (except November) during the second year. Temperature within each season was significantly lower in the second year. Clear temperature differences were found among stations during spring and summer, with stations I-III being cooler than stations IV-VI (Figure 4). Mean temperature values were homogeneous among Stations Figure 3. Boxplot showing mean (s), maximum and minimum salinity values for each station in the hays during 1998 and 1999. MV, Mar Virado; UBA, Ubatuba; and UBM, Ubatumirim. «= standard deviation. 35 Costa et al. 32 30 28 26 24 22 O 20 ^ 18j B 16] (0 o 14J & Summer i T T 1 T T ^ tUU ^ Min-Max IZZI Mean ±SD ■=■ Mean Stations Figure 4. Boxplot showing mean (±s), maximum and minimum temperature values for each station and season during 1998 and 1999. s = standard deviation. stations in other seasons. Organic matter contents at each station of each bay are shown in Table 1. Differences in mean organic-matter content levels were found among embayments, with deeper stations located near the bay mouth (I and II) showing the lowest levels. The amount of mud in the sediments decreased north- ward, i.e., from Mar Virado Bay to Ubatumirim Bay (Table 1, Figure 5). In Mar Virado Bay, the silt + clay fraction (phi > 4) was the most prevalent at the majority of stations, with values above 70% at stations II through V. A predomi- nance of fine and very fine sand, associated with silt and clay, was observed in Ubatuba, particularly in Ubatumirim Bay (Table 1), except for stations VI in Ubatuba Bay and station I in Ubatumirim Bay. A total of 563,636 individuals were collected during the present study; 324,861 during the first and 238,775 during the second year. In both years, the abundance of X kroyeri was higher in Mar Virado Bay (248,792), compared to Ubatuba Bay (206,284) and Ubatumirim Bay (108,560). The differences among bay and year were statistically siginificant {P < 0.05, Table 2). The highest shrimp abundance occurred during fall and winter in 1998 and during fall in 1999 (Figure 6), periods when shrimp abundance was significantly higher than in other seasons (P < 0.01). Conversely, lowest abun- dance occurred during sununer and spring, particularly in 1999. The interaction between year and season was also significant (Table 2). About 82% of all shrimps were caught in shallow areas, i.e., depths < 15 m (Figures 7 and 8, Table 1), except at station VI in Ubatuba Bay. Substantial differences in abundance were found among stations (P < 0.001) and between its interactions with bay (P < 0.001) and sea- son (P < 0.05; see Table 2); no other interactions were observed (P > 0.05). The correlation (r = 0.49) between abiotic factors and variation in shrimp abundance indicated that more individuals were collected in conditions of medium bot- tom temperature (22-24® C) and high salinity (36-38 psu). With respect to the substrate, shrimp abundance increased in areas with high organic matter content and high percent- age of silt and clay (high phi values). Also, in spring and 36 Costa et al. TABLE 1 Mean values of sediment parameters (diameter = phi; mud content = % silt + clay; organic matter content = o.m.), and number of individuals (N) for each station in each sampled bay from 1998 to 1999. Mar Virado Bay Ubatuba Bay Ubatumirim Bay phi mud o.m. phi mud o.m. phi mud o.m. STA (^) (%) (%) N (%) (%) N (%) (%) N I 4.3 46.8 3.0 17,613 3.2 16 3.6 5,657 1.5 2.6 2.1 3,011 II 5.7 75.3 4.6 47,632 4.0 21.2 4.2 19,240 3.8 23.9 3.4 10,162 III 6.2 88.3 5.4 42,946 5.3 61.9 8.0 52,096 4.4 35.7 5.2 43,357 IV 5.9 81.2 5.6 52,944 5.7 76.3 5.7 49,074 4.9 49.6 4.2 15,607 V 5.8 79.7 4.2 50,753 4.8 47.3 7.5 79,481 4.0 22.2 2.4 16,788 VI 5.4 64.4 4.4 36,904 3.6 36.8 6.1 736 4.4 33.4 4.2 19,635 Total 248,792 206,284 108,560 summer the number of individuals decreased at stations with depths over 20 m (Figure 8). The same periods and stations had low temperatures (Figure 4). In other seasons, particularly autumn, the spatial distribution of X kroyeri was more homogeneous. There was a good fit of the multiple regression analy- sis using significant {P < 0.05) environmental variables and the abundance of X. kroyeri (r = 0.49, P < 0.001, F = 69.95, N = 432), which can be expressed as: A = -176.447 + 18.432s + 42.347 phi where: A = abundance; s = bottom salinity (partial cor- relation = 0.11, P < 0.05); phi = phi (partial correlation = 0.49, P < 0.05). The abiotic factors such as sediment (phi) and salinity were positively correlated with the number of collected Stations Figure 5. Mean diameter of sediment grains (phi) at each sampled station in the bays studied. MV, Mar Virado; UBA, Ubatuba; UBM, Ubatumirim. individuals. No significant relationship was observed between bottom temperature or water depth and abundance (P > 0.05). Discussion The most important variables affecting the spatial and seasonal distribution of X. kroyeri in this study were sedi- ment type, salinity, and temperature. This was exemplified by the high abundance of the species in areas characterized by muddy substrates, and high salinity and temperature. The northern coast of Sao Paulo state is strongly influenced by 2 water masses: CW and TW. The effects of these water masses are felt most during autunrn and winter, when temperature and salinity levels increase to over 21° C TABLE 2 Results of the analysis of variance (factorial ANOVA) of the number of individuals collected (log^Q-trans- formed) of Xiphopenaeus kroyeri by year, bay, or season and station. Source df MS F P Year 1 2.48 4.66 0.03 Bay 2 16.36 30.73 0.00 Station 5 31.15 58.51 0.00 Bay X Station 10 5.71 10.73 0.00 Season 3 16.31 30.63 0.00 Season x Year 3 1.47 2.76 0.04 Season x Station 15 0.93 1.75 0.04 37 Costa et al. Season Figure 6. Number of individuals of X kroyeri obtained by season in the sampled Bays during the 2-yr period. and 35 psu, respectively. Another water mass, the SACW, intrudes throughout late spring and summer causing decreases in temperature (< 20® C) and bottom salinity (< 5 psu). The incursion of the TW into the uppermost water layers and the dislocation of the CW toward the ocean during the fall and winter causes vertical mixing and thus eliminates the existing seasonal thermocline causing the SACW to recede towards the offshore region (Castro-Filho et al. 1987, Castilho et al. in press). In addition to corroborating the scenario described above, our results indicate that fluctuations in the seasonal and bathymetric distribution of X kroyeri were influenced by variation in temperature (summer) and salinity levels caused by these water masses. Therefore, the influence of the SACW in the spring and summer most likely led to a decrease in the number of collected individuals. The retreat of this water mass and the incursion of TW during autumn and winter considerably increased the abundance of X kroyeri. Similar results were found by Fransozo et al, (2002) for X kroyeri in Fortaleza Bay, by Costa and Fransozo (2004) for Rimapenaeus constrictus (Stimpson, 1874), and by Costa et al. (2005a) for Sicyonia dorsalis Kingsley, 1878, all in the Ubatuba region. In addition, the number of captured individuals in these studies was smaller in these periods when compared with autunm and winter. Several authors (Rodrigues et al. 1993, Nakagaki and Negreiros-Fransozo 1998, Castro et al. 2005) have hypothesized that individuals of X kroyeri migrate to deeper regions to spawn, given that their main reproductive period occurs during spring and summer. However, even after 3 yr of sampling stations up to 40 m deep in Ubatuba Bay, Costa (2002) and Pinheiro (2004) did not And a single individual of X kroyeri deeper than 25 m. Therefore, one can infer that these shrimp migrate to the northernmost region of southeastern Brazil upon the arrival of the cold water currents. According to Castro-Filho et al. (1987), the SACW reaches its northern limit off the state of Rio de Janeiro, north of which temperatures are markedly higher than off southern Brazil. There was a marked increase in abundance during winter, even though bottom temperature dropped consid- erably. This inverse oscillation in abundance caused by variation in temperature during summer and winter may have masked the association between temperature and shrimp abundance, minimizing its impact on the analysis. Nevertheless, our results suggest that temperatures below 21® C may be limiting for this species. High abundance of X kroyeri were almost invariably associated with high salinity. This finding is similar to that of Castro et al. (2005), who investigated the popula- tion structure of this species in Ubatuba Bay, concluding (MV □ UBA □ UBM III IV V VI I 98 Stations V( Figure 7. Number of individuals of X kroyeri obtained by station in the sampled bays during the 2-yr period. MV, Mar Virado; UBA, Ubatuba; UBM, Ubatumirim. 38 Costa et al. Silt+Clay (%) 2500 n 2000 1500 1000 500 0 16[-19 □ S DA m\N HSp 19[-22 22[-25 25[-28 28[-31 Depth (m) Environmental factors Temperature (-C) Figure 8. Distribution of the mean number of individuals of X. kroyeri in relation to the environmental factors (botton salin- ity, silt + clay, organic matter, granulometric classes, depth and bottom temperature). S, summer; A, autumn; W, winter; Sp, spring. that juveniles are not dependent on estuarine regions and complete their life-cycles in salinities above 30 psu. In contrast, our results contradict other studies that suggest that X kroyeri is euryhaline but only tolerates salinities between 21.2 psu and 36 psu in many parts of its geo- graphical range, such as the coast of Texas, USA (Gunter et al,1964), the Laguna de Terminos, Mexico (Signoret 1974), and the Caribbean coast of Colombia (Cortes and Newmark 1992). These contradictory observations may result from the presence of only small estuaries in the Ubatuba region (Costa and Fransozo 1999). The abundance of X. kroyeri was strongly associated with the mud content of the substrate in each bay, which increased southward from Ubatumirim Bay to Mar Virado Bay. Therefore, the high abundance of X. kroyeri in Mar Virado Bay is probably a result of the high silt and clay content. The increasingly mixed sediments in other embay- ments, such as at station VI in Ubatuba Bay and station I in Ubatumirim Bay, seem to be avoided by this species. Given that penaeoid shrimps usually prefer substrates with higher mud and silt content, probably to facilitate their burrowing behavior, this characteristic may affect their distribution (Dali et al. 1990). However, a prefer- ence for a given kind of sediment seems to be species specific. In another study at the same site, Costa et al. (2004) found the same spatial distribution in the shrimp Pleoticus muelleri (Bate, 1888), diTid Artemesia longinaris Bate, 1888 was found at sites with higher percentages of fine and very fine sands (Fransozo et al. 2004, Costa et al. 2005b). On the other hand, the shrimp R. constrictus 39 Costa et al. showed a stronger preference for much coarser substrates (Costa and Franso^o 20D4). According to Penn (1984), preference for a given substrate in the case of penaeoids is associated with their capacity to perform gas exchange when burrowed. Several authors have suggested that the distribution of penaeoid shrimps is strongly modulated by the texture and organic content of the substrate (Rulifson 1981, Somers 1987, Stoner 1988, Ddl et al 1990, Sanchez 1997). In the present study, the distinctive characteristics of each bay determined to a large extent the differences in the abun- dance of Z kmyerl However, no correlation was detected between shrinip abundance and the organic content of the substrate. Even though water depth was not identified as a sig- nificant predictor of the abundance of Z. kmyeri in mul- tiple regression analyses in our study, this result should be interpreted with caution, given that this abiotic factor is usually co-lineat wi.th bottom temperature and the type of sediment. Pi.res (1992) studied the decapod community in the Caraguatatuba region, the southemmost part of the north shore of the state of Sao Paulo and encountered specimens of Z. kmyeri in depths between 50 and 60 m. High mud contents were found at these depths compared to other sites farther north. When viewed in the light of our study, these results suggest diat the sediment with higher mud content allowed an exp^sion of their bathymetiic distribution and thus might represent an essential factor for the establishment of this species. Acknowledgements The authors are grateful to FAPESP for providing financial support (Grant Nos. 94/4878-8, 97/12108-6, 97/12106-3, 9712107-0 and 98/3134-6). We are thank- ful to J. Reid (Virginia Museum of Natural History), M. Pie (Boston University), and M.L. Negreiros-Fransozo (UNESP-Botucatu, Sao Paulo) for constructive comments on early drafts of the manuscript and great help with the English language. We are also thankful to the NEBECC co-workers for their help during fieldwork. All sampling in tMs study has been conducted in compliance with current applicable state and federal laws. Literatuee citeo Beitini, G., A, Fransozo, and R.C. Costa. 200L Ecological distributioo of three species of PerBephom (Brachyura: Ixucosiidae) m the- Ufeatuba region, Sao Paulo, Brazil. NaupMus 9:31-42, Boschi, E.E, 2000. Species of decapod crustaceans and their dis- tribution in the marine zoog^gmphic proviiices. Revista de Investigacidn y DesanoUo Pesquero 13:7-136. Castilho, A.L., R.C. Costa, A. Fraasozo, and E.E. Boschi, in press. Reproductive pattern of the South American endemic shrimp Artemesia tongimris (Decapoda, Penaeidae), off the coast of Sao Paulo state, Brazil, Revisla de Biologfa Tropical, Castro, R.H., R.C, Costa, A, Fransozo, and RL.M, Mantelatto, 2005. Population structure ofllie seabob shnmpXiphopemeus krayeri (Heller, 1862) (Crustacea, Penaeoidea) in the littoral of Sao Paulo, Brazil. Sdentia Marina 69:105-112. Castro-Filho, B.M., L.B. Miranda, and S.Y, Myao, 1987, Cxsndlg^s MdrogrMc^ na plaMorma continental ao largo de Ubatuba: Varia^^s sazonais e em mddla escala. Bolethn do iastituto Oceanogrifico 35:13-51, Cort6s, M.L, and B. Newmark, 1992. Distribuddn y abun- dascia dbl canwdn titi Xiph&pemmus kmyeri en Q>sta Verde (Cienaga) Caribe colomblano, Boletin feotropica 25:15-27. Costa, R.C. 2002, Biologia e Distribuiglo Ecoldgica das Espdcles de Camaroes Deadrobranchiata (Crustacea: Decapoda) na Regiao de Ubamba (SP), Ph.D. Thesis, Universldade fetadual FauHsta, Botucatu, Sio ^ulo, BR, 1B6 p. Costa, R.C. and A, Fransozo, 1999. A nursery ground for two tropical pink-shrimp Pemem species: ITbatiiba Bay, north- ern coast of Sao Paulo, Brazil. NaupHus 7:73-81, Costa, R.C, and A. Franso^, 2(XH. Abundance and ecologic dis- tribution of the shrimp Rimapenaem comtrictm (Crustacea: Penaeidae) in the norihem coa.st of Sao Paulo State, Brazil. Journal of Nature! History 38:901-912, Costa, R,C., A. Fmnsoio, RL.M. Mantelatto, and R.H, Castro, 2000. Occurrence of shrimp species (Natantia: Penaeidea: C.mdea) in Ubatuba Bay, Ubatuba, SP, Brazil, Proceedings of the Biological Society of Washington 113:776-781. Costa, R.C,, A. Fransozo, G.A.S. Melo, and F,A.M, Freiie, 2(XI3, An illustrated key for Dendrobranchiata shrimps from the northern coast of Sao Paulo state, Brazil. Biota Neotrdpica 3:1-12. Costa, R.C., A, Fransozo, and A.P, Pinheiro. 2(X)4, Ecological distribution of the shrimp Pie&ticus mueUeri (Bate, 1888) (Decapoda; Penaeoidea) of Southeastern Brezil, Hydrobiologia 529:195-203, Costa, R.C., A. Fransozo, and M.L. Negreiros-Fr^sozo, 2(K)Sa, Ecology of rock shrimp Sicyenia dorsalis Kingsley, 1878 (Crustacea: Sicyoniidae) in a Subtropical region of Brazil. Gulf and Caribbean Research 17:49-56. Costa, R.C., A. Fransozo, A.L. Castilho, and RA.M. Freire, 2005b. Annual, seasonal and spatial variation of abundance of Artemesia hngimris (Decapoda, Penaeoidea) in a south- eastern region of Brazil, Journal of the Marine Biological Association of the United Kingdom 85:107-112, 40 Costa et al. Dali, W., BJ. Hill, EC. RotMisbes^, and D.I. Sba^les. 19W. The biology of tbs Penaeidae. In: J.H.S. Blaxtsr and A.J. Southward, eds. Advances in Marine Biology. San Diego: Academic Press, San Diego, CA, USA, 27:1-489. D^Incao, F, 1995. Ikxonomia, padioes dktribudonais e ecologi- cos dos DendrobrancMata (Crustacea: D^apoda) do litoral brasileiio. PhD Thesis, Universidade Federal do Parana, Curitiba, Parana, BR, 365 p. D^Incao, F., H. Valentini, and L.F. Rodrigues. 2^2. Avalij^ao da pesca de camaroes nss legioes Sudeste e Sul do Brasil. Atlantica 24:103-116. Fransozo, A., R.C. Costa, A.L. Castilho, and F.L.M. Manteialto. 2004. Ecological distribution of the shrimp "%arba-mga” Artemesia hngimins (Dfecapc^: Pea^idas) in Foitaleza Bay, Ubahiba, Brazil. Revista de Investigacidn y Desarrollo Pesquero 16:45-53. Fransozo, A,, R.C. Costa, F.L.M- Mantelatto, M.A.A. Flnheiro, and S. Santos. 2002. Composidon and abundance of shrimp species (Penaeidea and Caridea) in Fortaleza Bay, Ubatuba, SSo Paulo, Br^l. In: E.E. Briones and F. Alvarez, eds. Modem Approaches to the Study of Cmstacea, New York, NY, USA, p. 117-125. Gimter, G., i.Y, Christmas, and R, Killebrew. 1964. Some mla- tlons of salinity to population distributions of motile estua- rine organism, with special refemnce to penaeid shrimp. Ecology 45:181-185. Mantektto, F.L.M. and A. Fransozo. 1999. Characterization of the physical and chemical parameters of Ubatuba Bay, northern coast of Sao Paulo state, Brazil. Revista Brasileira de Blologia 59:23-31. Nakagaki, J.M. and M-L. Negreiros-Fransozo. 1998. Population biology of Xiphopenaeus kmyeri (Heller, 1862) (I^apoda: Penaeidae) fiom Ubatuba Bay, Sao Paulo, Brazil, ioumal of Shellfish Research 17:93-935. Nakagaki, J.M., M .L. Negreiros-Fransozo, and A. Fmnsozo. 1995. Composigao e aboodlncia de canmrdes niarinhos (Ciustocea: Decj^oda: Penaeidae) na Ens^a de Obatiiba, Ubatuba, Brazil. Aiquivos de Blologia e Tecnologia 38:583- 591. Penn, J.W. 1984. The behavior and caSchabilily of some com- mercially exploited peoaeids and their relationship to stock and recmltment. In: J.A. Golland and BJ. Rothschild, eds. Penaeid Shrimps — Their Biology and Management. Fishing News Books, Faroham, UK, p. 173-186. Pinheim, A.P. 2004. Biodivemidade de Fenaeoidea (Cmstacea: D^apc^) do Litoral Norte do Estado de Sao Paulo. Master’s Thesis, Universidade Estadual Paniista, Botacalu, SIo Paulo, BR, 89 p. Fires, A.M.S, 1992. Stmctuie and dynamics of benthic megafauna on the continental shelf ofehoie of Ubatuba, southeastern, Brazil. Marine Ecology Progress Series 86:63-76. RcKlrigues, E.S., J.B. Pita, R. Graga-Lopes, J.A. Coelho, and A. Puzzi. 1993. Aspectos biologieos e pesqueiros do camarao sete-barbas (Xiphopenaem kr&yeri) capturados pela i^sca artesanal no litoral do estado ck Sao Paulo. Boletim do Im^titutoi de Pesca 19:67-81. Rulifson, R.A. 1981, Substrate preference of juvenile penaeid shrimp in estuarine habitats. Contributions of Marine Science 24:33-52. Sanchez, A.J. 1997. Habitat preference of Penmm dMomrum Burkenmad (Crustacea: Deeapoda) in a tropical coastal lagoon, southwest Gulf of Mexico. Journal of Experimental Marine Biolo^ and Ecolo^ 217:107-1 17. Signoret, M. 1974. Abuadancia, tamano y distribucioo de canm- rones (Crustacea, Penaeidae) de la laguna de Terminos Campeche y su relacion con aigunos fostores hidioMgicos. Anales del Instituto de Blologia, Universidad Nacional Autdnoma de Mexico. Serie 2kiol6glca 45:119-140. Somers, IF. 1987. Sediment type as a factor in the distribution of corameicial prawn s|^les in the Western Gulf, Australia. Australian Journal of Marine and Freshwater Research 38:133-149- Stoner, A.W, 1988. A nursery ground for Ibui tropical Fenaeus species: Laguna loyuda, Puerto Rico. Marine Ecology fejgress Series 42: 133-141. Thcker, M. 1988. T^hnlqoes in Sedimentology. Blackwell Scientitic :Publieations, Boston, MA, USA, 394 p. Wentworth, C.K. 1922. A scale of gracfe and terns for cladistic sedimenbi. Journal of Geology 30:377-392. Williams, A.B. 1984. Shrimps, lobsters and crabs of the Atlantic coast of the eastern Unitol States, Maine to Florida. Smithsonian Instltetion Press, Washington, DC, USA, 550 p. Zai, J.H. 1999. Biostatistical Analysis. Prentice-Hall, NJ, USA, 663 p. 41 Gulf and Caribbean Research Volume 19 Issue 1 January 2007 Growth Phases and Differential Growth Between Sexes ofUca maracoani Latreille^ 1802-1803 (Grustacea^ Brachyura^ Ocypodidae) Gustavo L. Hirose Group of Studies on Crustacean Biology^ Ecology^ and Culture Maria L. Negreiros-Fransozo Instituto de Biociencias, Brazil DOI: 10.18785/gcr.l901.05 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr & Part of the Marine Biology Commons Recommended Citation HirosC; G. L. and M. L. Negreiros-Fransozo. 2007. Growth Phases and Differential Growth Between Sexes of Uca maracoani Latreilk; 1802-1803 (Crustacea; Brachyura; Ocypodidae). Gulf and Caribbean Research 19 (l); 43-50. Retrieved from http://aquila.usm.edu/gcr/voll9/issl/5 This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized editor of The Aquila Digital Community. For more information, please contact Joshua.Cromwell^usm.edu. GROWTH PHASES AND DIFFERENTIAL GROWTH BETWEEN SEXES OF UCA MARACOAM LATREILLE, 1802-1803 (CRUSTACEA, BRACHYURA, OCYPODIDAE) GiistsYO L. EBrose^ and Marta L. Negreiros-Frai[iEQi0^»^* ^NEBECC (Gmup of Studies on Crustacean Bioiogyt Ecology and Culture). ^Departamento de Zoologm, Imduao de Biocimdas, Caijxi E}stal 510, UNESP, 18618-WO Botucatu, SF, Brazil ^Correspondmg author E-mail mlnf@ibhunesp.br ABSIKACT Among crustacean decapocb, fiddler crabs of the genus JJca are the most characteristic of the intertidal zones of tropical and subtropical estuaries. The pmsent study detemdaed the growth phases and the beginning of differential growth between the sejces, based on analyses of mlative growth of Uca maracomiL Collections were made in the Jabaquam mangrove* located in Paraty* Rio de Janeiro* Brazil. Specimens were collected manually during low tide periods. In the laboratory, crabs were seaed and measured. We measumd carapace width (CW), carapace lengdi (CL)* propodus lengib (PL; the right one for females and the major one for males), propodus height (PH), abdomen width (AW), and gooopod length (GL; for males). The begmnmg of differential growth between the sexes could be identified by the allometric technique. Males and females showed 3 distinct growth phases. Diffemntiai growth betw^n sexes began when males reached 7 mm and females 9.2 mm CW. The growth pattern among dtff^nt phases and the beginning of differential growth seemed to maintam a strict relationship with the ontogenetic changes* mainly those related to behavioral and reprsniiictive as|^ts. This information is important for ^nerai biological knowl^ge of this species* mainly concerning aspects of its growth. IpfrMOBUCTiON Fiddler crabs of the genus Uca Leach* 1814 show a distmct sexual dimorphism. The chelipeds of females are symmetrical and small; both are adapted for catching and parsing substrate particles to the mouth appendages. In males, one cheliped is more developed than the other, and is used for interaction with other maies and defense of territory; only the smaller cheliped is ntilimi to scoop a small amoimt of material Jiom the substrate and transport it to the mouth for feeding (Yamaguchi 1977, Christy and Salmon 1984, Rosenberg 2002), like those of the female. The study of growth in crabs is facili.tated by the hard tegument and the periodic change in the exoskeleton at molting. This makes possible exact measurements and observations of ontogenetic subdivisions in the body size at the beginmng of morphological sexual maturity (Huxley and Teissier 1936). Several aspects of the minor and major chelipeds of fiddler crabs have been studied. These studies have treated the morphological stmctnies (Crane 1975, Yamaguchi 2001), variations in shape (Rosenberg 1997, 2002), sexual differences (Yamaguchi, 2005), handedness (Jones et al. 1982, Williams et al. 1981, Yamaguchi et al. 2001), and growth (Hartnoll 1974* 1978, 1982). The differential growth of the large chelipeds of males, as well as other body parts such as male gouopods and the abdomen of females (Hartnoll 1974, 1978), reveal the transition from juvenile to adult phases (Negreiros-Fransozo et aJ..* 2003). Allometric growth has been studied in several species of ocypodid crabs, including Uca tangeri (see Von Hagen 1987, Colpo et al. 2003), Uca thayerl (see Negreiros- Fransozo et al. 2003), Uca hurgersi (see Benetti and Negreiios-Fransozo 2003), Uca mpax (see Castiglioni end Negreiros-Fransozo 2004), Uca mor^ (see Masiinari et al. 2005), and Uca mamcoani (see Masumri et al. 2005). We investigated a population of U. muracoani Latreille, 1802-1803 from the mud beach of the Jabaquara mangrove at Faraly, Rio de Janeiro, Brazil focusing on the detemfination of growth phases and the beginnings of differential growth between the sexes, based on relative growth. Material and Methods Uca maracoani occurs along coastlines in the Western Atlantic, including the Antilles, Venezuela, and the Guyanas. It is found along nearly the entire coast of Brazil, from the state of Maranhio south to Parand (Melo 1996). Sampling was carried out monthly on the mud beach of the Jabaquara mangrove (23“12T0.0'^S, 44“43H4.r^W) from January to July 2003. The crabs were removed from their burrows manually during low tide. Each month, a capture effort of 60 min by 2 collectors was employed The crabs were packed In plastic bags and frozen for about 2 h. In order to ensure that the smallest classes were also coUecled, additional sampling was carried out hy 2 people for 20 min, utilizing a small spoon. These small crabs were isolated in small containers, labeled* ^d transported to the laboratory. Hirose and Fransozo AW males Figure 1. The body parts of Uca maracoani measured in this study. (CW = Carapace Width; CL = Carapace Length; AW = Abdomen Width; GL = Gonopod Length; PH = Cheliped Propodus Height; PL = Cheliped Propodus Length.) In the laboratory, crabs in the intermolt stage were sexed and measured with a digital caliper (0.01 mm), for the following dimensions: carapace width (CW), carapace length (CL), propodus length of the cheliped (PL; right in females and major in males), propodus height of the cheliped (PH), abdomen width (AW) and gonopod length (GL; in males). The measured dimensions are illustrated in Figure 1. Crabs with an imperfect carapace or with body parts in regeneration were not used for analysis. The smallest individuals, obtained in the additional samplings, were maintained alive in plastic containers with about 20 ml of sea water and fed daily with nauplii of Anemia sp. They were monitored daily for the presence of molts until they reached the juvenile stage when they could be identified to species. The exuvia from each of the first 3 molts only (to nunimize the effect of laboratory conditions) were sexed by counting the number of pleopods. Each exuvia was measured under a stereomicroscope for the same body dimensions described above for the adult crabs, except for gonopod length. Statistical analyses Relative growth was analyzed based on the allometric technique, for observation of changes in growth of certain body parts in relation to others (Huxley 1950). The data were plotted in dispersion graphics. Next, the equation of function type Y = aX^ was adjusted to the empirical points and linearized to the form logF = a + b logX, where X = the independent variable, using the carapace width (CW); Y = dependent variables, utilizing the other body dimen- sions; b = allometric constant that expresses the allometric coefficient of body parts in the study. Growth could thus be characterized as positive allometry, when b > I’, negative allometry, when b < 1 ; or isometric, when b = \ (Huxley 1950). The value found was tested by Student’s t test with significance level a = 0.05. To test the similarity of slopes and the intercepts of lines for each phase of devel- opment and between sexes, we used a covariance analyses (ANCOVA) (a = 0.05) (Zar, 1996). The programs Mature I and II (Somerton 1980a, b) were used to estimate the size at which males and females changed growth phase, based on the regression analysis of relative growth. To determine the size at which the differentiation between sexes began, a series of successive covariance analyses was carried out, using intervals of 0.1 mm CW. This proceeded until the point where the lines of growth diverged, or the point at which the males and females attained growth represented by distinct lines (a = 0.05). 44 Hirose and Fransozo Eesui.ts A total of 563 cmb^ was collected (253 males and 310 females). Their sizes ranged from 4,9 to 43.7 mm CW (28.5 ± 10.1 mm) for males and from 5,1 to 38.7 mm CW (24.9 ± 7.9) for females. We obtained 92 e^cuvia from the small individuals raised. The sizes of exuvia ranged from 1.2 to 5,1 mm CW (2.8 ± 0.93); of these, only 7 specimens conld he identified as males and 12 as females. The sizes of males ranged from 3.7 to 5.1 mm CW (4.4 ± 0.48) and of females from 3.5 to 5.0 mm CW (4.0 ± 0.51). The change In growth phases and the beginning of differentiation in growth between the sexes conld be expressed by PL vs. CW for males and AW vs. CW for females. The inflexion points of the growth lines related to the change of phases were estimated by means of the Mature I and II programs. Males (< 9.4; 9.4 < CW < 21.2; > 21,2 mm) end females (< 10.3; 10,3 < CW < 19.4; > 19.4 mm) showed 3 distinct growth phases. For females. Mature I indicated the CWSO (19,4 mm) eqmvalent to the interval of superposition of lilies (16,5 ^ X < 20,9 min CW) for juvenile md adult. The regression equations obtained for the growth ph^s and between sexes were submitted to ANCOVA, which verified that the pattern of growth differed (P < 0.05) between sexes and among phases. Thus, we can assume ttiat the lines for different phases (undifferentiated Juvenile, and adnit phase) and for sexes (males and females) could ^ better adjusted to iso- lated data than in a single line. Exceptions were obtained only for some phases in 'the relationships CL vs. CW and PHvsXWJIkMe 1) Males and females showed different gmwth patterns of the carapace. For the relationship CL vs. CW, males showed negative allometry for all phases, whereas the females showed negative allometry only for the juvemles {b = 0.819), passing to d!.screte positive allometri.c growth in subsequent phases. A slight diffemnce in the shape of the adult male carapace compared with females explained this. The cheliped relationships PH vs. CW and PL vs, CW showed similar growth patterns. The males, for all age groups, showed positive aliometry beginning in the nndif- ferentiated phase {b ™ 1.195; h = 1.147 for PL vs. CW and PH vs. CW, respectively), passing to a more positive slope in the later phases (h = 2.047 and b = 1329). The females showed less positive allometry compared with the males (b = 1.056 for the rclatioBship PL vs. CW), passing to isometry or negative allometry in the case of the juvenile phase and adult females in the relationship PH vs. CW (b = 0.831). For the relation AW vs, CW, sexual dimorphism in growth was very evident However, iu this case, the females showed positive allometry in aD. age group catego- ries, mcreasing the slope in the juvenile-phase females (b ~ 1.887). The males showed isometric growth (undifferenti- ated b ~ 0,958), passing to a positive allometfic (jttvenile phase b = 1.174) and later to a negative afiometric growth (adult h = 0.944). Males (smalier than 4 mm CW) were not included in the analyses of the relationship GL vs. CW. The crabs weiie distributed in 2 age group categories, with different growth patterns (b = 1.646 and b = 0.909 juvenile phase and adults, respectively). The onset of differential growth between sexes was at 7 mm for males and 9.2 mm CW for females (Figure 2D). Below these points, there was a single line of growth for both sexes (a = 0.05). The patterns of growth found for relationships per- formed with the data on sex differentiation determined by ANCOVA were the same among the ontogenetic phases, showing positive allometry for the 2 relationships analyzed in both sexes for the undifferentiated and juvenile phases. (Table 2) Discussion Morphometric data are widely utilized in papers on crustaceans for the study of relative growth (HartnoU 1974, 1982), especially for detection of changes in the level of aUometiy, which can be related to certain biologicai fea- tures of the species. Most of the studies on morphological stoictuies in brachyuran crabs have used the dhnensions of the carapace, abdomen, and cheMpeds as a reference and found distinctive changes in such structures between sexes or growth phases. M the population studied, U. mamcoani showed 3 phases of growth. Similar patterns have been reported for other species of the genus: U, tangeri studied by von Hagen (1987), and U. tkayeri studied by Negreiros- Fransozo et al. (2003). Other brachimran crabs, mainly in the superfamily Majoidea, show similar patterns, although the growth phases are known by different names (Sainte- Marie et ai. 1995, Almmo-Bmscia and Sainte-Marie 1998, Sampedro et al. 1999). The first phase of gmwth found fm U. mamcoani is represented by morphologically undifferentiated crabs, which only show visible secondary ®xual characters in the lar^st classes. They may not have initiated gonad develop- ment; this phase can be called the undifferentiated period. The first point of the transition is from undifferenti- ated to a second phase of growth that, probably, is the 45 Hirose and Fransozo TABLE 1 Regression analyses of morphometric data of Uca maracoani. Carapace width (CW) was nsed as the indepen- dent variable. CW = Carapace Width; CL = Carapace Length; AW = Abdomen Width; PL = Cheliped Propodns Length; PH = Cheliped Propodns Height; and GL = Gonopod Length. UMF = nndifferentiated males and females; UJF = nndifferentiated and jnvenile female; JAM = jnvenile and adnlt males; JAF = jnvenile and adnlt female UF = nndifferentiated females; JF = jnvenile females; AF = adnlt females; UM = nndifferentiated males; JM = jnvenile males; AM = adnlt males; + and - = allometry; 0 = isometry.) Intercep Resnlts of (log) Slope Mature I and II Relationship Sex N a b H II P Allometry F value Change of phases UMF 196 0.240 0.819 0.981 22.60 0.00 - CL vs. CW JAM 255 -0.108 0.967 0.997 29.73 0.00 - JF 56 -0.216 1.060 0.978 2.82 0.00 + AF 136 -0.198 1.043 0.992 5.66 0.00 + UM 91 -0.626 0.958 0.854 0.97 0.00 0 JM 64 -0.818 1.174 0.936 4.47 0.00 + AW vs. CW AM 189 -0.495 0.924 0.944 5.00 0.00 - UF 109 -0.680 1.105 0.937 3.74 0.00 + 16.94 10.30 mm JF 65 -1.481 1.887 0.927 13.43 0.00 + 72.94 19.30 mm AF 221 -0.808 1.377 0.968 22.35 0.00 + UM 86 -0.500 1.195 0.916 5.00 0.00 + 20.62 9.40 mm JM 61 -0.819 1.576 0.969 16.00 0.00 + 140.13 21.20 mm PL vs. CW AM 180 -1.454 2.047 0.982 50.00 0.00 + UJF 157 -0.453 1.056 0.986 6.60 0.00 + AF 134 -0.377 1.003 0.969 0.00 1.00 0 UM 75 -0.927 1.147 0.855 2.67 0.00 + JM 75 -0.851 1.329 0.967 11.38 0.00 + PH vs. CW AM 179 -1.239 1.613 0.976 32.10 0.00 + UF 99 -0.858 0.948 0.923 1.88 0.00 0 JAF 190 -0.768 0.831 0.945 12.14 0.00 - GL vs. CW JM 71 -1.177 1.646 0.872 8.53 0.00 + AM 189 -0.269 0.909 0.955 6.43 0.00 - beginning of gonad maturation (juvenile phase). In a study on U. tangeri, von Hagen (1987) considered the interval between the points of inflexion as a phase of transition or maturation, which extends to the second inflexion point that identifies the transition of individuals to the adult phase. Adult fiddler crabs can successfully copulate (func- tional maturity) only when their gonads are mature, i.e., producing gametes, and when they can display specific behaviors, such as the male “waving” display to attract females. Sexual maturation is an extended process that involves gradual ontogenetic changes, rather than a pre- cise moment, such as after the puberty molt (Luppi et al. 2004). Uca maracoani showed positive allometric growth in the change of phases for sexes and age group catego- ries. The chelipeds of males showed a gradual increase in allometry among phases, reaching the highest level in adult males. Females showed low, positive allometry in the juvenile phase, considerable positive allometry after the first inflexion point, and then a subsequent decrease after they reached morphological maturity. This growth pattern may reflect the higher energetic investment of females in reproduction during the adult phase. According to Hartnoll (2006), the major energetic needs are for ripening of gonads and formation of associated reproductive products. Slower growth can also reflect reduced energy intake because of restrictions on feeding. In females, a more general phenomenon is a restriction on feeding during incubation. A further limitation on growth in reproducing females is that they cannot molt while incubating eggs (Hartnoll 2006). The increase in allometry of chelipeds, in the case of fiddler crab males, just after they reach sexual maturity. 46 Hirose and Fransozo Figure 2. Uca maracoani. Dispersion points of relations between the carapace width and the dependent variables. PL = Cheliped Propodus Length; AW = Abdomen Width. A and B are dispersion points related to the changes of growth phases. C and D are dispersion points related to the onset of differential growth between the sexes. can be very important, because the behavior of cohorts in this genus is predominantly visual. An experimental study with U. tangeri by Oliveira and Custodio (1998) found that females spend more time near males with larger claws, in binary choice tests. Large chelipeds are possibly more eas- ily seen by females, increasing the chances that a male’s burrow will be visited and he will be chosen for reproduc- tion (Crane 1975, Yamaguchi 1971, Latruffe et al. 1999). In some species, U. vocans vocans studied by Salmon (1984) and U. bebei studied by Christy (1987), for instance, there is no apparent preference of females for males with larger chelipeds. In such cases, the advantage may be related to the results of fights among males (Crane 1975). Contests consist of a series of behavioral elements in which the major claw of males plays the principal role (Crane 1975, Salmon and Hyatt 1983, Pratt et al. 2003). In fiddler crabs, the contest duration is expected to reflect the endurance of the weaker of 2 contestants, typically the smaller individual (Hyatt and Salmon 1978, Jennions and Backwell 1996, Pratt et al. 2003). Fighting ability is correlated with carapace width and size of the claw (Hyatt and Salmon 1978), and represents an important feature. Although some contest elements appear dangerous, death or serious injury seldom result (Pratt et al. 2003). In females, the abdomen widens during growth. This may be related to protection of the gonopores and the mass of eggs during incubation (Hartnoll 1982). The female carapace also widens in relation to that of males, which increases the capacity of the incubatory chamber where the eggs will be carried. The sexes differed in size at maturity. The larger size of males may reflect strong competition among them, con- sidering that larger males can exert strong influence over smaller males. According to Crane (1975), U. maracoani shows a hierarchical behavior, where smaller males avoid combat with the larger, dominant males, leaving the inter- action area. In this context, it could be more advantageous for young males to initially invest in growth rather than in reproduction. Comparing the Paraty population of U. maracoani (present study) with the population from Guaratuba stud- ied by Masunari et al. (2005), it is easily realized that there exist significant population differences in relation to growth and morphological sexual maturity. (Table 3) For both sexes, the size at sexual maturity found for the U. maracoani population at Jabaquara was greater than for the population at Guaratuba studied by Masunari et al. (2005). This was also seen when the size at differentiation of chehped (males) 47 Hirose and Fransozo TABLE 2 Regression analyses of morphometric data, based on the growth differentiation between the sexes of Uca mara- coani. CW = Carapace width; AW = Abdomen Width; PL = Cheliped Propodns Length. UT = nndifferentiated males and females; JM = jnvenile males; JF = jnvenile females; + positive allometry. Relationship Sex N Intercept (log) a Slope b r2 T(b=l) P Allometrv Size at differentiation Results of ANCOVA (P value) UT 191 -0.663 1.055 0.89 2.20 0.00 + 9.2 mm 0.06 AW vs. CW JM 65 -0.950 1.281 0.90 13.56 0.00 + JF 70 -1.343 1.773 0.94 15.77 0.00 + UT 167 -0.458 1.079 0.88 2.63 0.00 + 7.0 mm 0.06 PL vs. CW JM 65 -0.737 1.509 0.97 18.17 0.00 + JF 72 -0.542 1.133 0.95 4.92 0.00 + is analyzed. Probably, the large difference between the populations may be responsible for these differences in size at sexual maturity and chehped differentiation. According to Masunari et al. (2005), the size at differentiation of the cheliped in fiddler crabs apparently is correlated with the maximum size that the species can reach. The population of U. maracoani studied by Masunari et al. (2005) represents, according to Melo (1996), the southern limit of distribution for this species on the Brazilian coast. Consequently, the environmental con- ditions such as temperature, salinity, food, and size of sediment particles, to which these crabs are exposed, are different from for the population at Jabaquara. This may reflect the differential growth, development, and size of those individuals. Size variations are common and may reflect the phenotypic plasticity of the organisms or the influence of environmental factors such as photoperiod, temperature, and food availability (Campbell and Eagles 1983). Such factors can explain the larger CW of specimens of U. maracoani from Jabaquara compared with the population at Guaratuba. For a precise detemfination of which fac- tor or factors are more important for the size difference between these populations, a further, more detailed study would be necessary. The growth patterns during the different phases, as well as the beginning of differential growth, and mor- phological sexual maturity seem to maintain a strict relationship with ontogenetic changes, mainly related to behavioral or reproductive aspects and/or environmental factors. This information is important for general biologi- cal knowledge of the species, mainly for understanding its growth processes. CW (mm) Figure 3. Uca maracoani. A - Lines representing the relationships between carapace width (CW) and abdomen width for females. The arrow shows the size at which the puberty molt occurs. B - The logistic equation indicating the size in which 50 % of females are mature. 48 Hirose and Fransozo TABLE 3 Growth features in 2 populations of Uca maracoani. Dimensions are in mm. s = standard deviation. Population (reference) Sex Maximum size Mean size ±s Growth phase Size at differentiation Sexual maturity Paraty, RJ male 43.7 28.50 ± 10.1 3 7.0 21.20 (present study) female 38.7 24.90 ± 7.9 3 9.2 19.30 Guaratuba, PR (Masunari et al. male 34.12 21.53 2 3.27 17.85 2005 ) female 29.20 19.13 2 — 11.75 Acknowledgments To FAPESP (94/4878-4; 98/3136-4) and CAPES for financial supports. To NEBECC members for their help during the fieldwork. The specimens in this study were collected according to Brazilian state and federal laws concerning sampling wild animals. References Alunno-Bmscia, M. and B. Sainte-Marie. 1998. Abdomen allom- etry, ovary development, and growth of female snow crab, Chionoecetes opilio (Brachyura, Majidae), in the northwest- ern Gulf of Saint Lawrence. Canadian Journal of Fisheries and Aquatic Sciences 55:459-477. Benetti, A.S. and M.L. Negreiros-Fransozo. 2003. Madurez sex- ual morfologica del cangrejo Uca burgersi Holthuis, 1967 (Brachyura, Ocypodidae) de una area de manglar subtropi- cal de Brasil. Universidad y Ciencia 19:9-16, Campbell, A. and M.D. Eagles. 1983. Size at maturity and fecundity of rock crabs. Cancer irroratus, from the Bay of Fundy and southwestern Nova Scotia. Fisheries Bulletin 81(2):357-361. Castiglioni,D.S. andM.L. Negreiros-Fransozo. 2004. Comparative analysis of the relative growth of Uca rapax (Smith) (Cmstacea, Ocypodidae) from two mangroves in Sao Paulo, Brazil. Revista Brasileira de Zoologia 21(1): 137-144, Christy, J.H. 1987. Competitive mating, mate choice and mat- ing associations of brachyuran crabs. Bulletin of Marine Science 41(2):177-191. Christy, J.H. and M, Salmon. 1984, Ecology and evolution of mating systems of fiddler crabs (Genus Uca). Biological Review 59:483-509, Colpo, K.D., Rodriguez, A., Lopez, J., and A. Moreno. 2003. Allometric patterns in Uca tangeri males from an exploited population at San Pedro River, Andalucia, Spain. Nauplius 11(1):57-61. Crane, J. 1975. Fiddler Crabs of the World. Ocypodidae: Genus Uca. New Jersey: Princeton University Press, MA, USA, 736 p. Hartnoll, R.G, 1974, Variation in growth pattern between some secondary sexual characters in crabs (Decapoda, Brachyura). Crustaceana 27:130-136. Hartnoll, R.G. 1978. The determination of relative growth in Crustacea, Crustaceana 34(3):281-293. Hartnoll, R.G., 1982. Growth. In: D.E. Bliss, ed. The Biology of Crustacea: Embryology, morphology and genetics. Vol. 2, Academic Press, Inc., New York, NY, USA, p. 111-185. Hartnoll, R.G. 2006. Reproductive investment in Brachyura, Hydrobiologia 557:31^0. Huxley, J.S, 1950. Relative growth and form transformation. Proceedings of the Royal Society of London 137(b):465- 469. Huxley, J.S. and G. Teissier. 1936. Terminologie et notation dans la description de la croissance relative. Comptes Rendus Hebdomadaires de la Societe de Biologic 121:934-36. Hyatt, G.W, and M. Salmon, 1978, Combat in the fiddler crabs Uca pugilator and U. pugnax: A quantitative analysis. Behaviour 65 : 1 82-2 1 1 . Jennions, M.D. and P.R.Y. Backwell. 1996. Residency and size affect fight duration and outcome in the fiddler crab Uca annulipes. Biological Journal of the Linnean Society 57:293-306. Jones, D.S. and R.W. George. 1982. Handedness in fiddler crabs as an aid in taxonomic grouping of the genus Uca (Decapoda, Ocypodidae). Crustaceana 43(1): 100-102. Latruffe, C., P.K. Mcgregor, and R.F, Oliveira. 1999. Visual signaling and sexual selection in male fiddler crabs Uca tangeri. Marine Ecology Progress Series 189:233-240, Luppi, T.A., E.D. Spivak, C.C. Bas, and K. Anger. 2004. Molt and growth of an estuarine crab, Chasmagnathus granula- tus (Brachyura: Varunidae) in Mar Chiquita coastal lagoon, Argentina. Journal of Applied Ichthyology 20:333-344, Masunari, S. and N. Dissenha. 2005. Alometria no crescimento de Uca mordax (Smith) (Crustacea, Deacapoda, Ocypodidae) na Baia de Guaratuba, Parana, Brasil. Revista Brasileira de Zoologia 22(4):984-990. Masunari, S., N. Dissenha, and R.C. Falcao. 2005. Crescimento relativo e destreza dos quehpodos de Uca maracoani (Crustacea, Deacapoda, Ocypodidae) no Baixio Mirim, Baia de Guaratuba, Parana, Brasil, Revista Brasileira de Zoologia 22(4):974-983. Melo, G.A.S., 1996. Manual de identificagao dos Brachyura (caranguejos e siris) do litoral Brasileiro. Ed. Pleiade/ FAPESP, Sao Paulo, BR, 603 p. Negreiros-Fransozo, M.L., K.D. Colpo, and T.M. Costa. 2003, Allometric growth in the fiddler crab Uca thayeri (Brachyura, Ocypodidae) from a subtropical mangrove. Journal of Crustacean Biology 23(2):273-279, 49 Hirose and Fransozo Oliveira, R,E and M.R. Custodio. 1998. Claw size, waving dis- play and female choice in the European fiddler crab, Vca tm,g€rL Ethology, Ecology and Evolution 10:241-251. Pratt, A.E., D.K. Mdaio, and G.R. Lathmp. 2003. The assess- ment game in sand fiddler crab contest for breeding bur- rows. Animal Behaviour 65:945-955. Rosenbe^, M.S. 1997. Evolution of shape differences between the major and minor chelip^ of Vca pugnax (Itecapoda: Ocypodidae). iouma]. of Cmstacean Biology 17(1):52-S9. Rosenbeig, M.S. 2002. Fiddler crab slmpe variation: A geometric morphometric analysis across the genus Vca (Cmstoxa: Brachyura: Ocypodidae). Biological Journal of the Liimeao Society 75:147-162. Sainte-Marie, B., S. Raymond, and J.C. Brethes. 1995. Growth and maturation of the benthic stages of male snow cmb Chiofmecetes opilia (Brachyura; Majidae). Canadian loomal of Fisheries and Aquatic Sciences 52:903-924. Salmon, M. 19H4. The courtship, aggression and mating system of a ‘"primitive” fiddler crab {Vca voams\ C^ypodidae). Transactions of the Zoological Society of London 37:1-50. Salmon, M. and G.W. HyatL 1983. Spatial and temporal aspects of repnxiuctioo in Nordi Carolina fiddler crabs (Ucii pugi~ loXor Bose), Journal of Experimental Marine Biology and Ecology 70:21-43. Sampedro, M., E. Gonzalez-Gurriai^, i, Fieiie, and R. Muino. 1999. Morphomeliy mid sexual nmtiirity in the spider crab Maja squimM (Decapoda: Majidae) in Galicia, Spain. Journal of Cmstacean Biology 19:578-592. Somerton, D. 19B0a. A computer technique for estimating the size of sexual nmturity in crabs. Canadian Journal of Fisheries and Aquatic Sciences 37:1488-1494. Somerton, D. 1980b. Fitting straight lines to Hiatt growth dia- grams: A re-eva!,uation. Journal du Conseil Intenational pour FExpioratlon cb la Mer 39:15-19. Von Hagen, H-0. 1987. Ailometric growth of two populations of Vca tangesi from the Guadalquivir estuary (Andalusia). Investigi^ones Fesqueras 51(sopl. 1);443-4S2. Yamaguchi, T. 1971. Courtship feeimvior of a fiddler crab, Vca iactea. Kumamoto Journal of Science, Biology 10:13-37. Yamaguchi, T. 1977. Studies on the hand^luess of the fiddler cmb Uca Iactea. Biologi,cal Bulletin 152:424-236. Yamaguchi, T. 2{M)1. Dimorphism of chellpeds in the fiddler crab, Uca arcuata. Cmstoxana, 74(9):913-923. Yamaguchi, T, Y. Heumi, andR. Ogata. 2(B5. Sexual differences of the feeding claws and mouthpaits of the fiddler crab, Uca arcuata (De Haan, lB33)(Brachyura, Ch^fpodldae). Cmstaceana 78(10):! 233-1263. Williams, M.J. and RK. Heug. 1981 Hand^luess in male Uca vocam (Linnaens, 1758) (Decapods, Cteypodidae). Cmstaceana 40(2):21 5-216. Zar, J.H. 1996. Biostatistical Analysis. 3rd Edition, ftentice-Hall, Upi^r Saddle River, NJ, USA, 915 p. 50 Gulf and Caribbean Research Volume 19 Issue 1 January 2007 First Record in Honduras of the Halfbeak Hypor/zampws roberti hildebrandij ordain and Everman 1927^ (Hemiramphidae) Collected in an Inland Reservoir Wilfredo A. Matamoros University of Southern Mississippi Julio E. Merida University of Southern Mississippi Janise Palmer Scheda Ecological Associates, Inc. DOI: 10.18785/gcr.l901.06 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr Part of the Marine Biology Commons Recommended Citation Matamoros, W. A., J. E. Merida andj. Palmer. 2007. First Record in Honduras of the UdXtbcdkHyporhampus roberti hildebrandi, Jordan and Everman 1927, (Hemiramphidae) Collected in an Inland Reservoir. Gulf and Caribbean Research 19 (l): 51-53. Retrieved from http://aquila.usm.edu/ gcr/voll9/ issl/6 This Short Communication is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized editor of The Aquila Digital Community. For more information, please contact Joshua.Cromwell(®usm.edu. SHORT COMMUNICATION FIRST RECORD IN HONDURAS OF THE HALFBEAK HYPORHAMPHUS ROBERTI HILDEBRAND!, JORDAN AND EVERMAN 1927, (HEMIRAMPHIDAE) COLLECTED IN AN INLAND RESERVOIR Wilfredo A. Matamoros^*, Julio £. Merida, and Janise Palmer^ ^Department of Biological Sciences, The University of Southern Mississippi, 118 College Dr. # 5028, Hattiesburg, Mississippi 39406-Q(X)l USA, ^Scheda Ecological Associates, lnc„ I486-E Skees Road, West Palm Beach, Florida 3341 1 USA ^Corresponding author. E-mail wilfredo.matamoros@usm.edu Introduction An inhabitant of tropical America, the Central American halfbeak {Hyporhamphus roberti hildebrandi), is one of 2 subspecies of halfbeaks of the subgenus Hyporhamphus (Collette 2003, Collette 2004) that belong to the family Hemiramphidae. This family has repre- sentatives from the Atlantic, Pacific, and Indian Oceans (Greenfield and Thomerson 1997, Berra 2001, Collette 2004), and nearly all species are marine; however, some Hemiramphidae species in the Indo-Austrafian region are restricted to freshwater (Greenfield and Thomerson 1997), The distributional range of H, r. hildebrandi extends along the Caribbean coast of Central America from Mexico to the Gulf of Uraba in Colombia (CoUette 2004, Miller et al. 2005). The sub-species is considered marine and estua- rine, commonly found in mangrove forests (Greenfield and Thomerson 1997), and coastal lagoons (Schmitter- Soto 1998, Diaz-Ruiz et al. 2003, Collette 2004). For example, they have been collected in the Laguna de Bacalar in southern Mexico (Schmitter-Soto 1998) and the Tortuguero National Park in Costa Rica (Winemiller and LesUe 1992). Previous specimens collected in Honduras have been from estuarine and marine systems or from freshwater sys- tems with a direct connection to brackish or marine water (see NeoDat, http://www.neodat.org). Here, we report the first record of H. r. hildebrandi in Honduran freshwater (see Reis et al. 2003) as well as in a landlocked freshwater body of water. Materials and Methods In January 2003, three specimens of H. r. hildebrandi were collected by personnel from the Honduras Center of Studies of Contaminants Control (CESCCO) along the shore of (he Francisco Morazdn El Caj6n (“El Cajdn”) reservoir. El Caj6n is located in northeastern Honduras (Figure 1) at 15W'N, 87‘’33'W (Central Offices), at an altitude of 285 m above sea level, between the munici- palities of Santa Cruz de Yojoa in the department of Cort6s, Victoria in (he department of Yoro, and Lagos, La Libertad, Minas de Oro y Mctobar in the department of Comayagua. The reservoir spans an area of 94 km^ with a perimeter of 469.7 km, and a maximum depth of 185 m. Specimens were collected as part of a field assess- ment following a fish-kill and were dead upon collection. Specimens were immediately fixed in 10% formalin, rinsed in tap water, and then transferred to 70% ethanol for preservation. Standard length (SL, mm) was measured using digital calipers and specimens were weighed (wet weight, g). Fish were identified based on standard char- acters (Greenfield and Thomerson 1997) and deposited in The University of Southern Mississippi Museum of Ichthyology (voucher number: USM 31216). To determine the presence of the sub-species in other Central America freshwater bodies of water, we queried NeoDat (http://www.neodat.ofg), an internet database spe- cializing in collections of neotropical fishes. Results Three specimens of H. r, hildebrandi were collected and identified based on the following combination of characteristics (Greenfield and Thomerson 1997): lower jaw much longer than rest of head but shorter than half of the standard length; scales absent or only a few present on anterior part of the dorsal and anal fins, total gUl rakers on first arch more than 38; dorsal plus anal rays usually total more than 30. Specimen identification was confirmed by B. Collette (Collette pers. comm.. National Marine Fisheries Service Systematics Laboratory, Washington, DC, USA). Standard length and wet weight of fish were 124.4, 122.5, and 116.7 mm; and 7.6, 6.8, and 5.4 g, respectively. Results from the database queries conducted showed that H. r. hildebrandi has been collected in Mexico, Belize, Guatemala, Honduras, Costa Rica, and Panama. Most of these collections came from coastal ecosystems. Unreported collections of H. r, hildebrandi from fresh- Matamoros et al. Figure 1. Map of Honduras showing the location of the hydroelectrical dam “El Cajon” and Lake Yojoa. Circled stars depict locations of previous unreported collections of H. r. hildebrandi in Honduras and Guatemala. water systems that have direct connections to brackish waters including the following localities in Guatemala: Lake Izabal (USNM-1 14261, USNM-134705), Rio Dulce (UMMZ-197188), Rio Polochic (AMNH-35043, CAS- 45429), and El Estor (USNM-1 34706). An additional site was located in the Rio Patuca (UMMZ- 199576) in Honduras and a site in Honduras located 30 km east of Trujillo (B. Collette, pers. comm., see Figure 1). Discussion In Central America, Honduras is quite possibly the country with the least studied ichthyofauna. The need for systematic studies of Honduras freshwater fishes has been clearly recognized (Carr and Giovannoli 1950, Miller 1966). Recently, Lyons (2005) who focused on a disjunct distribution of the genus Sicydium in Mexico and Central America emphasized the need for a stronger knowledge base of the ichthyofauna in Honduras. Accordingly, it would not be surprising to discover range expansions of fishes in Honduras or other regions where ichthyological research is scarce. The novelty, however, is to encounter in an inland freshwater body a species from a family of fishes that is thought to primarily exist in coastal and marine waters. Museum records demonstrate the presence of H. r. hildebrandi in coastal lagoons and marine environ- ments of Honduras, as well as freshwater in Guatemala. Specimens collected in Lake Izabal and along river sys- tems in the region, document the presence of the species in Guatemalan freshwater systems with connections to brackish water (Miller 1966). Yet, our report documents the first record of H. r hildebrandi from an inland, land- locked, freshwater system, indicating the species may be established and recruiting in freshwater. It is unknown at what point H. r hildebrandi may have been established in El Cajon, since the reservoir was constructed in 1985, 52 Matamoros et al. and no data exist in relation to the states of ichthyotaima diversity both before and after the dam was built. Lake Yojoa, another freshwater body near El Cajon, was surveyed by Maitiii (1972) and Cruz (1985), but the presence of H. k hiMehraryii was not reported in their sur- veys. Currently this s^ies is common in the littoral zone of the laise (W. Matamoros, pers. obs.), and we infer that the arrival of H. r. hildebrandi in the area happened after 19B5. However, the means of dispersion employed by the species H. r. hildebrmdi to expand its range is unknown. Acknowledgments We are grateful to C. ScMoss from GCRL for assis- tance in the bibliographic research, M. Espinal from USAID/MIRA Honduras for providing the map, C. Zelaya and H. Pordllo for editing the map, B. Collette for identify- ing the specimens and J. Schaefer for providing comments on earlier versions of the manuscript. LinnRiMATiriiE Cited Berra, T.M. 2001. Freshwater Fish Distribution. Academic Press, New York, NY, USA and London, UK, 604 p. Cmt, A.F. and L. Giovamioli, 1950. Hie fishes of the Choluteca drainage of Southern Honduras. Occasional Papers of the Museum Zoology, The University of Michigan 523:1-38. Collette, B.B. 2003. Hemirhamphidae, halfbeaks. In: K.E. C^enter, ed. The living marine resources of the western cend^ Atiantic. FAO Species Identifrcation Guide for Fishery pujposes and American Society of Ichthyology and Herpetology Special Publication S. FAO, Rome. Y. 2:I135--1144, Collette, B.B. 2(M)4. Family HemimmpMdae Gil! lBS9-half- l^aks. California Academy of Sciences. Annotated Check List of Fishes No. 22, 35 p. Cruz, G.A. 1985. Biology of the black bass {Miempterm mlmoi- des) in Yojoa Late, Hondums. Revista Latinoamerlcana de Acuicultura 23:12-25. Diaz-Ruiz, S., M.A. Ferez-Hemandez, and A. Agoirre-Ledn. 21X)3. Characterization of fish assemblages io a tropi- cal coastel lagcHsn in the northwest Gulf of Mexico. Car^^terizacidn de los coujuntos de en una laguna costem tropical del noroeste del Goifo de Mexico. Cieitcias Marinas 29:631-644. Greenfield, D.W. and J.E. Thomerson. 1997. Fishes of the continental waters of Belize. University Press of Florida, Gainesville, Florida, USA, 311 p. Lyons, J. 2(^5. Distribution of Sicydium Valenciennes 1837 (Pisces: Gobiidae) in Mexico and Centra! America, ffidrobioldgica 15:239-243. Martin, M. 1972. A biogeogmphic analysis of the freshwater fish- es of Honduras. PhD Dissertation. University of Southern California, Los Angeles, CA, USA. Miller, R.R. 1966. Geographical distribution of Central American freshwater fishes. Copek 19^(4);773-802. Miller, R.R., W.L. Mlnckley, and S.M. Norris. 2005. Freshwater Fishes of Mexico. The University of Chicago Press, Chicago, II, USA, 652 p. Reis, R.E., S.O. Kullander, and C.I. Feiraris Jr. 2003. Check list of the freshwater fishes of South and Central America. EDIFUCRS, Porto Alegre, BR, 735 p. Schmitter-Soto, J.J. 1998. Catalogo los paces continentales de Quintana Roo. El Colegio de k Frontera Snr, San Cristdbal de Las Casas, CIHS, 239 p. Winemiller, K.O. and M.A. Leslie. 1992. Fish assemblag- es across a complex, tropical freshwater/maiine ecotoue. Emimnmental Biology of Fishes 34:29-50. 53 Gulf and Caribbean Research Volume 19 Issue 1 January 2007 Use ofDiadema antillarum Spines by Juvenile Fish and Mysid Shrimp Taryn Townsend Montclair State University Paul A.X. Bologna Montclair State University DOI; 10.18785/gcr.l901.07 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr & Part of the Marine Biology Commons Recommended Citation Townsend, T. and P. A. Bologna. 2007. Use of Diadema antillarum Spines by Juvenile Fish and Mysid Shrimp. Gulf and Caribbean Research 19 (l): 55-58. Retrieved from http://aquila.usm.edu/gcr/voll9/issl/7 This Short Communication is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized editor of The Aquila Digital Community. For more information, please contact Joshua.Cromwell(®usm.edu. SHORT COMMUNICATION USE OF DIADEMA ANTILLARUM SPINES BY JUVENILE FISH AND MYSID SHRIMP Taryn Townsend^ and Paul Bologna^ ^Department of Biology and Molecular Biology, Montclair State University. Montclair, New Jersey 07043 USA ^Corresponding Author. Aquatic and Coastal Sciences Program, Department of Biology and Molecular Biology, Montclair State University, Montclair, New Jersey 07043 USA. Phone {973) 655-4112, Fax (973) 655-7047, E-mail hologmyj@mail.montclair.edu Introduction The long-spined sea urchin {Diadema antillarum Phillipi) is an important element in the structure and function of coral reef communities. Regarded as a key herbivore in reef communities, grazing by D. antillarum shifts community dominance &om macroalgal cover to live coral (Lessios 2005, Tdya et al. 2004). Diadema antillarum is primarily found in shallow coral reef and seagrass environments but can reside in a wide variety of habitats (Lessios 1998). This animal generally remains in sheltered areas during the day and moves to grazing sites during the evening. Its activities can create grazing halos around reefs (Ogden et al. 1973). Additionally, urchins represent potential biogenic structure and refiagia for fish and invertebrates. The availability of shelter influences the survivor- ship and recruitment of juvenile reef fishes (Shuhnan 1985). Structurally complex habitats allow prey to escape predation as they utilize small spaces for refuge (Caley and St. John 1996). Literature suggests that the urchin spines represent a complex three-dimensional structure in which small fish can evade predators. Consequently, the utilization of urchin spines as a structural habitat has been shown to increase survival of juvenile fish (Hartney and Gromd 2002). Studies have also noted that swarms of mysid shrimp (Mysidium sp.) associate with D. antillarum as a source of protection against fish predation (TXvinmg el al. 2000). Additionally, urchin size may affect how frequently fish use this complex biogenic habitat (Hartney and Grorud 2002). Some fish have been shown to associate with urchins which possess spines that are proportionate to fish body size (Lissner 1978), and D. antillarum have the unique ability to regulate body size in response to resource availability (Levitan 1988, 1989). Therefore, explora- tion of how urchin test size correlates with the presence of juvenile fish and invertebrates may be important for assessing the urchin-fish/invertebrate relationship. This research has the following objectives: 1) identify the size structure of D. antillarum among 3 coastal bays of St. John. United States \firgin Islands and 2) determine relationships between urchin presence and spine utilization by fish and mysid shrimp. Study Site Field studies were conducted at 3 coastal bays in St John, United States Virgin Islands including: Hurricane Hole, a fringing mangrove and seagrass community; Greater Lameshur Bay, a predominantly hard substrate- coral reef community; and Litde Lameshur Bay, a hard substrate-coral reef community interspersed with seagrass and unvegetated habitats. Methods To assess how D. antillarum size influenced fish and invertebrate presence, urchins were counted and mea- sured and associated mysids and fish were recorded in 2 surveys. First, 410 D. antillarum were counted and test diameter was measured and classified into size categories (0-30 mm, 30-60 mm, 60-70 mm, 90-20 mm, 120-150 mm, >150 mm). Data were recorded in the field, and all measurements were taken on individual days at each site to eliminate the possibility of measuring the same urchin twice. Urchin size-frequency data were analyzed using a non-parametric Kruskal Wallis Rank Analysis to determine whether urchin size differences existed among sites. When present, samples of fish were collected using a slurp gun, then enumerated and identified to species. An additional 628 urchins were surveyed in Little Lameshur Bay to determine the utilization of D. antillarum spines by mysids and fish. Fish and mysid presence was then tabu- lated to propose utilization of spines as refuge/biogenic habitat. In some cases, surveys did not identify certain fish species during sampling a priori, therefore they were not included within tabulated results. Results and Discussion The D. antillarum size-frequency distributions found in Greater Lameshur and Hurricane Hole indicate a rela- tively normal size distribution, with a modal test diameter Townsend and Bologna Urchin Test Size (mm) Figure 1. Diadema antillarum size frequency distribution. Bars represent total number of urchins within each test diameter size class (mm) at each of the 3 study sites. within the 60-90 mm size class (Figure 1), The Little Lameshur site differed, however, with a mode at 0-30 mm, which may reflect the presence of recently recruited juveniles and a different year class present in the samples. This disparity resulted in a significant difference in mean test size (H = 151, P < 0.001) between Greater Lameshur (sample mean = 85.5 mm) and Hurricane Hole (80.5 mm) compared to Little Lameshur (47.2 mm). Additionally, the largest size classes of urchins occurred only at Greater Lameshur (Figure 1). Our exploration of the Mysidium sp. and D. antilla- rum association suggests that urchin test diameter influ- ences mysid presence. Whereas swarms occurred over groups of smaller urchins, they were only observed over individual urchins in the 90-120 mm size class. Mysidium sp. swarms were found over urchins at all sites, with fre- quencies of occurrence of 3.8% at Hurricane Hole, 6.8% at Greater Lameshur, and 2.8% at Little Lameshur (Table 1). Mysidium sp. is known to occur in swarms just off the bottom of the sea floor near structurally complex, three- dimensional substrata, including D. antillarum (Hahn and Itzkowitz 1986). In a previous study on the homing behavior of M. gracile, it was found that mysids swarm at given sites during the day and disperse during the night. It was suggested that mysid shrimp use some type of homing behavior to re-coalesce into discrete schools after noctur- nal dispersal (Twining et al. 2000). The most abundant fish associated with sea urchin spines was Haemulon flavolineatum (French grunt). This species was only collected from urchins of a test size > 60-90 nun and only in Little Lameshur Bay (Table 1). This does not mean that they were not present in the other Table 1 Mysid shrimp and fish utilization of Diadema antillarum spines. (NA = not assessed during specific site survey). Sites Diadema antillarum (N) Mysidium sp. Schools Haemulon flavolineatu Schools (mean#/school) Canthigaster rostrata Pareques acuminatus Hurricane Hole 26 1 0 0 0 Greater Lameshur 117 8 0 NA 0 Little Lameshur East 628 17 NA 10 0 Little Lameshur West 259 8 9(11.2) NA 1 56 Townsend and Bologna A B Urchin Size (mm) Figure 2. Mysidium sp. utilization of Diadema antillarum spines. Bars represent percent frequency of occurrence of mysids among spines of urchins in each size class. A) Hurricane Hole. B) Little Lameshur. C) Greater Lameshur. bays, but rather that our survey did not identify utilization of D. antillarum in these bays. McFarland and Kotchian (1982) showed that H. flavolineatum commonly forms mixed schools with mysid shrimp (genus Mysidium). The schooling behavior of these different organisms into large complexes may relate to the morphological similarities that grunts have with mysid shrimp. The postulated ben- efits for the fish include protection (at smaller sizes) and use of mysids as food (at larger sizes). In addition to H. flavolineatum, we observed 2 other species that have not been previously recorded from D. antillarum spines. We observed and collected Canthigaster rostrata (sharpnose puffer) from within the spines of urchins in Little Lameshur Bay (Table 1). The puffers were observed and collected deep within the spines of urchin groups comprised of 3-6 individuals (observation Bologna and Townsend). Previous studies of C. rostrata in St. Thomas, USVI indicate that this species is signifi- cantly more abundant where predators are more abundant, in comparison to other prey species (Shulman 1985). Sharpnose puffers are toxic and may survive in predator rich areas because they are not potential prey for piscivores (Hixon and Beets 1993). Canthigaster rostrata is, how- ever, preyed upon by some bony fish, including Sphyraena barracuda (Randal 1967), which was frequently observed in Little Lameshur Bay (Bologna observation). Therefore, it is possible that juvenile sharpnose puffers utilize urchin spines as refuge from predators, enabling this species to co-exist in these reef communities. A Pareques acumina- tus (high-hat) was also observed and collected from deep within the spines of D. antillarum. This fish was very cryp- tic, and its body form and pigmentation resembled urchin spines. Although we only collected one P. acuminatus, we believe that its presence and its cryptic appearance among the urchin spines may suggest Batesian mimicry between P. acuminatus and D. antillarum. Through our research, we were able to identify 3 types of juvenile fish within urchin spines, as well as determine that urchin test size plays a role in juvenile fish/mysid shrimp utilization. Future investigations of the association between H. flavolineatum, C. rostrata, and D. antillarum are necessary to understand the relationship between these juvenile fish and urchins. Additionally, further studies with P. acuminatus need to be pursued in order to determine whether our observation was a random occurrence or if this species uses crypsis and mimicry of urchin spines as a predation refuge during early juvenile stages. 57 Townsend and Bologna Acknowledgeivients The authors wish to thank C. Dale, M. Gizas, C. Koritos, R KoeI, R. Pappagian, S. Regelz, and D. Ward for field assistance in collection and measmement of organ- isms. We would also like to thank the staff of the Vkgm Mand Environmental Resource Station for logistical sup- port during this research. Lastly, we wonld Uke to thank the National Park Service for allowing us to pursue this research in the Virgin Islands Coral Reef Monument and the Virgin Islands Natioiiai Park. LmSEATlOEE CiTEP Caley, M.J. and J. St John. 1996. Refuge Availability Stmctims Assemblages of Tropical R^f Fishes, Journal of Animal Ecology 65:414-428. Hahn, P. and M. Itricowitz, 19B6. Site preference and homing behavior In the mysid shrimp Mysidium gracile, Cnistaeeana 51:215-218. Hartney, K.B. and K.A. Gromd. 2W)2. Tf]^ effect of sea mchins as biogenic stmctuies on the local ahnisdance of a temperate reef fish. Oecologia 131:506-513. Hixon, M.A. and J.P. Beets. 1993. Pi^iatioii, prey refuges, and the stnichire of coral-reef fish assemblages. Ecological Monographs 63:77-101. Lesslos, H.A. 1998. Population dynamics of Diadema antiliamm (Echmodennata: Echinoidea) following mass mortality in Panama. Marine Biology 99:515-526. Lessios, H.A, 2(M)5. Disdema antillarmn ^spuJatiom in Panama twenty years following mass modality. ComI Reefs 24:125- 127. Levitan, D.R. 1988, Density-dependent size regulation and nega- tive giowdi in the sea urchin Diadema imtillaritm Philippi. Oecolo^a 76:627-429. Levitan, D.R. 1989. Density-dependent size regulation In Diadema antiUafwn: Effects on fecundity and survivorship. Ecology 70:1414-1424. Lissno", A. 1978. Factors affecting the distribution and abundance of the sea urchin Cemmstepham comnatm Verrill at Santo Catolina island. PhD DiSvSertotion, University of Southern Califoania, Los Angeles, CA, USA. McFarland, W.N. andN,M. Kotchian, 1982. Inter^tion between schools of fish and mysids. Behavioral Ecology and Sociobiology 11:71-76, Ogden, J.C., R.A. Brown, and N. Salemky. 1973. Grazing by the ecMnoid Diadema antillarum: Formation of halos around West-lndian patch reefs. Science 182:715-717. Randall, i.E. 1967. Food habits of reef fishes of the West indies. Studies in Tmpicai Oceanography 5:665-847. Shulman, M.J. 1985. Recruitnient of coral reef fishes: Effects of distribution of predators and shelter. Ecology 66:1055- 1066. Twining, B.S., J.J. Gilbert, and N.S. Fisher. MM. Evidence of homing behavior in the coral reef mysid Mysidium gmeile. Limnology and Ck:eanography 45:1845-1849. Ttiya, F., A. Boyra, P. Sanchez-ierez, R, Haroun, and C. Barbera. 2{XM. Can one species determine the stmctare of a rocky benthic community on a temperate itjcky reef: The case of tbe long-spined ^a nichin Diadema antilla- rum (Echmodennata: Echinoidea) in the eastern Atlantic. Hydrobiologia 5 19:21 1-214. 58 Gulf and Caribbean Research Volume 19 Issue 1 January 2007 Color Variation in the Caribbean Crab Platypodiella spectahilis (Herbst^ 1794) (Decapoda^ Brachyura^ Xanthidae) Joel W Martin Natural History Museum of Los Angeles County Todd L. Zimmerman Long Island University DOI; 10.18785/gcr.l901.08 Follow this and additional works at: http:/ / aquila.usm.edu/ gcr & Part of the Marine Biology Commons Recommended Citation Martin,}. W. and T. L. Zimmerman. 2007. Color Variation in the Caribbean Crab Platypodiella spectahilis (Herbst, 1794) (Decapoda, Brachyura, Xanthidae). Gulf and Caribbean Research 19 (l): 59-63. Retrieved from http://aquila.usm.edu/gcr/voll9/issl/8 This Short Communication is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Gulf and Caribbean Research by an authorized editor of The Aquila Digital Community. For more information, please contact Joshua.Cromwell(®usm.edu. SHORT COMMUNICATION COLOR VARIATION IN THE CARIBBEAN CRAB PLATYPODIELLA SPEC- TABILIS (HERBST, 1794) (DECAPODA, BRACHYURA, XANTHIDAE) Joel W. Martin^ and Todd L. ZUnmerman^ ^Natural History Museum of Los Angeles County, 900 Exposition Boulevard, Los Angeles, California 90007 USA. E-mail jmartin@nhm,org ^Long Island University, C. W. Post Campus, 720 Nortlwm Blvd, Brookville, New York 11548-1300 USA Introduction Platypodiella spectabilis (Herbst, 1794) is a relatively small crab (about 10 min carapace width) found in or near coral reefs and rocky shorelines throughout most of the Caribbean and tropical western Atlantic. Records are known from as far north as Bermuda (Chace et al. 1986, in Sterrer 1986, as Platypodia spectabilis) and as far south as Rio de Janeiro, Brazil, including the Fernando de Noronha Archipelago and Trinidade Island (Melo 1998: 490) and the state of Sao Paulo (Coelho and Ramos 1972, Fransozo et al. 2001). Thus, the range of the species is some 7,000 km from northern to southern extent. Distributional records within this range, and extending westward into the Gulf of Mexico, can be found in Rathbun (1930), Felder (1973), Powers (1977), and Abele and Kim (1986). Because of its spectacular coloration and color pattern, P. spectabilis is often depicted in faunistic guide books (e.g., Humaim 1992, Humann and DeLoach 2002), and the spe- cies is sometimes referred to as the calico crab (e.g., Chace et al. 1986) or gaudy clown crab (Williams et al. 1989, McLaughlin et al. 2005). The use of color in brachyuran crab systematic s, and in particular the use of subtle color differences to suggest or differentiate cryptic or morphologically similar species, is now well documented (e.g., see Campbell and Mahon 1974 for species of Leptograpsus, WiUiams and Felder 1989 for species of Menippe, Zimme rman and Felder 1991 for species of Sesarma). Less clearly understood is why color patterns and intensities can sometimes vary appre- ciably within a species, even within narrowly restricted geographic regions. An appreciation of color patterns is critical to correctly identifying species for conservation and resource management purposes, yet often color pat- terns and ranges are unreported, causing confusion and sometimes misidentifications. Here we document a wide range of color patterns in a small Caribbean xanihid crab based on specimens collected in essentially the same habi- tat at the same time of year. Materials and Methods As part of a biodiversity survey of Caribbean crypto- faunal invertebrates, we sampled several habitats from shal- low waters off Guana Island, British Virgin Islands, during the summers (June-August) of 1 999-2(K)2. Specimens of P. spectabilis collected during that survey came almost exclusively from an area of a few square meters at our North Bay collecting site, where they were found in inter- stices of dead coral (mostly clumps of dead Porites) in shal- low water (^ 1 m) (Station 65 of the Ziinmerman/Martin survey of Guana Island in 2000 and 2001). Collections were made by crushing clumps of dead coral and removing by hand the invertebrates they contained. In this manner, a large number of specimens of P. spectabilis were col- lected (including all of the photographed specimens except Figure 2d), especially in the year 2(X)1 . Similarly-sized crabs wem photographed while alive or just after immersion in ice water, a technique that rapidly kiUs tropical crabs while preserving their color fer up to 12 hr. Photography was done by T.L. Zim m erman (Figures la-f; Figure 2a-c) and Leslie Harris (Figure 2d). Associated field voucher numbers are listed in the figure captions. After being photographed, crabs were preserved in either 70% or 95% ethanol for eventual transfer to and storage in the Crustacea collections at the Natural History Museum of Los Angeles County. One ovigerous female from the North Bay site (photographic voucher number VcO-796, Figure lb), with a carapace width 10.5 mm and carapace length 7.1 mm^ was the parental female that formed the basis of the first description of larvae in this genus (Fransozo et al. 2001). Results As is evident from the accompanying figures (Figures 1 and 2), specimens of Platypodiella spectabilis collected in the British Virgin Islands (and presumably elsewhere) exhibit a wide range of color patterns. The background color varies from a light cream or beige (Figure la-d) to an intense yellow (Figures le, f; 2a-c). Regularly-spaced and nearly circular black spots ringed with white may be present on the carapace (Figure Ic, e), or the carapace can Martin AND Zimmerman 1. Color mlation In the Muthld jrah gpectMtig (Qertist, a: muted badEEraund yellow, no spots, no brown patches (female^ VcO-799); b: muted bacfcEnxuid yefflow with many mated lai^e brawn patches (female, Vc0-79ti); c: mated bfHhEraund ycOow with spots, no brown potdies (male, VctMtOO); d: nmted background yellow with intense lanse brawn patches;, no spots (niale> VcO-792); e: intense backgraimd ydhrw with few small brown patches with spote (mal^ Vc0^79fl); f: intense backgroond with few small brawn patches;, no spots (mal^ Vc0-801>. AH photogrophs by T. Zimnunnan* lack any semblance cf similar-sized spots (Figures la; 2b). Inegulaily-sliaped brown patches sometmies occur cm the carapace, and are often found on the cardiac and gastric regions (Figures Id; 2a-c); in some specimens (e.g., Figure 2c) the brown patches are coalesced to fonn the majority of the color on the dorsal carapace, with only limited areas displaying white or yeUow. Some individuals eadiibitcd both irregular brown patches and more regularly- spaced and evenly-sized spots (Figure le). Fbrdopcxls typ- ically bear yellow, while and brown patches so as to ^ipear banded. Usually the cnder of color on a given perraopodal article is, feom the proKimal to distal end, yellow followed by brown followed by wtule, such that the distal end C3f the article is white or light beige (e.g.. Figures la, b, d; 2b). However, in some specimens the segments of the legs are almost cximpletely brown (Figure 2c), and in others the yellow area is bordered on both sides (rather than only cm the distal side) by brown (Figure Id). Sex of the crab does ncit appear to have any notice- able effect on the cmlor pattern; figures include both males (Hguies Ic-f ; 2a-b) and females (all others except Hgure 2d, a juvenile). The juvenile we collected was filter 60 MARim AND Zimmerman Fienie 2. Color wiatkn In the xanlliid crab Platypo^/^ aptcUibQia (Herbst, 1794), contbuwiL a: conlescfiU solid brown paAwdies with some mnted Easons (and fHMsfbfy spots) orcrdifbue intense jellciwbBGk£rod(n]ale^VcO-797); b: distinct brown patdies with distinct polclieB of intense ydlowback^ound odor Onale^VcO-794); c: cookscent solid brawn pokfaes over dilEbae intense yellow bncfcgroiuid (female, Vo0^793); d: colonitioa pattem in a JuTenfle (H-2037) bdbie canpace regions aro dearly demarcateiL Fluitographs a^-c hy Tl Zimmerman^ d by Ledie Harris. Dvera]!, with a backgraund can^aoe caloratian that was off-white and wilfa a more orange (less brown) color pat- tern dorsally (Figure 2d). Dkcussion Color plays an in^ortant role in spedes identifica- tion and presumably in spedes-spedes Eecognition among decapods. Additionally, knowledge of coloration and color variability can assist in llie recpgnitiDn of species by conservationists and resource managers without extensive taxonomic training. Coral-associated crabs are among the most colorful of all decapods, with trapeziids and tetial- liids well known in this icgaid; coral-associated carpiHids and eriphiids arc also often v^y ooiorfijL However, few crab species exhibit coloration as striking as in spectu- bilis, and there are few reports of colors or color patterns that vary to the extent shown here within a known species and within a very small geographic range (in our case, within meters). A similar situation exists for a tropical her- mit crab, Calcvttts gaimardi, in the western Pticific with at least 2 distiiict color morphs that occur sympatricaUy (C. Djdge, pens, comm., Morgan 1991, Ihdge 1995), Although it was stated earlier that the wide range of color patterns in this spedes mi^ indicate a fbrmmiy unrecognized species complex (Frasozo et bL 2001), there is virtually no doubt that all specimens illustrated here belong to the same species. Even at the larval level, the widespread P ^ectabilis does not appear to vary much across its rather large range. Fransozo et al. (2001) docmnenled sHght difieroices in the morphology of reared larvae of this spedes from the Caribbean and feom Brazil, but overall, despite the geographic distance separalmg the parental fdnales, larvae from the 2 areas were found to be quite similar. The genus PfoOV’odkifa was erected by Guinot (Guinot 1967: 562) to accommodate 4 spedes fonnerly treated as Plat^wdia. Two of the spedes, P, gemmata (Rathbun, 1902) and P rvtmdata (Stimpson, 1860) are known from the eastern Pticific. Garth (1991: 131), in discussing the Galapagos crab fauna, pointed out that the 2 Ptidfrc spe^ cies are possibly the adult (described as P. rotundata) and the juvenile (P. gemmata) of the same spedes, with the name P, rvtundata having precedence. The other 2 spe^ cies are Atlantic, with P picta (A. Milne Edwards, 1869) restricted to the eastern Atlantic and P, spectabUis (Herbst, 61 Martin and Zimmerman 1794) known from the western Atlantic. Gninot (1.967) remarked on the similarities between species of thi s gemis and those of Phtymtaea and, to a lesser extent, to species of Atergatis and Atergatopsis, as well as to members of Platypodia, The latter 3 genera are sometimes considered members of the xanthid subfamily Zosindnae (e.g., see Serene 1984, Clark and Ng 1998). It would be interesting to examine the range of color patterns exhibited by species in these supposedly related genera to see if color variabil- ity has a phylogenetic component. Acknowledgements This work was funded in part by the Biotic Surveys and Inventories program of the U.S. National Science Foundation (NSF) via grant DEB 9972100 to XL. Zimmerman and J.W. Martin, by a FEET grant (DEB 9978193) from NSF to J.W. Martin and D. Jacobs, and by the Decapod Cmstacea '‘Assembling the Tree of Life*’ grant to J.W. Martin (DEB 0531616, one of 4 awards for the decapod AToL project). We thank NSF and especially D, Causey for support and encouragement during our time on Guana Maud. We also thank other members of the Guana Island field team: L. Harris, D. Cadien, R. Ware, T Haney, K. Fitzhugh, and G. Hendler. Finally, we thank the Falconwood Corporation and especially L. Jarecki for allowing us to conduct this work on Guana Island. The manuscript was improved substantially by the constructive comments of C. Tudge and an anonymous reviewer, LlTEEATTiEE CiTED Abele, L.G, and W. Kou. 1986. An illastraled guide to the maime decapod cmst^ans of Florida. Sfete of Florida Department of Enviroumental Regulation Technical Series, Voi. S, no. 1 (in 2 parts). 760 p. Campbell, N.A. and RJ. Mahon. 1974. A multivariate study of variation in two species of rock crab of the genus Leptograpsus. Australian Journal of Zoology 22:417“425, Chace, RA., Jr., JJ. McDermolt, P.A. McLaughlin, and R.B. Manning, 1986. D^apoda. In: W. Steirer, ed. Marine Fauna and Ficna of Bermuda. TOey-Intmcience, John Wiley and Sons, New York, NY, USA, p. 312-358. Clark, P.R and P.K.L. Ng. 1998. The larval development of the poisonous mosaic cmb, lMpkozo:^ymous pictor (Fabricius, 1798) (Crustacea, Decapoda, Brachyura, Xanthidae, Zosiminae), with ooxmnents on familial charsu:tie^ for first stage zoeas. Zoosystema 20:201-220. Coelho, P.A. and M. de Araiijo Ramos. 1972. A coostitin^ao e a distiibuigio da fauna de decdpodos do litoral leste da AmMca do Sul entre as latitudes ife S'^N e 39®S. Trabalhos do lostituto Oceanogrdfico da Univiversidade Federal Pernambuco, Recife 13:133-236. Felder, D.L. 1973. An annotated key to crabs and lobstem (Decapoda, Reptantia) from coastal waters of the north- western Gulf of Mexico. Center for Wetland Resources, Louisiana State University, Baton Rouge, LA, USA, 103 p. Fransozo, A., M.L. Negreiros-Fransozo, J.W. Martin, and S.E. Trantwein. 2(K)1. Morphology of the fimt zoea! stage of Fiatypodieiia spect£^ilis (Herbst, 1794) (Decapoda, Brachyura, Xanthidae) ob^ned in the laboratory. Gulf and Caribbean Research 1.3:79-85. Garth. J.S. 1991. Tbxonomy, distribution, and ecology of Galapagos Brachyura. In: M.J. James, ed. Topics in Geobiology, Mslume 8: Galapagos Marine Invertebrate. Plenum Press, New York, N Y, USA, p. 123-145. Guinot, D. 1967. Recherehes preiiminaiies .sur les groupements naturels chez les Cmstacfe Decapodes Br^hyonres BL A propos des affinitos des genres Daimides Stebbing et Daira de Bulletin do Museum National. d’Histoire Naturelle, 2e Serie, 39:540-563. Hiiinann,P. 1 992. Reef Creature Identification: Florida, Caribtean, Bahamas. New World Publications, Inc., Vaughan Press, Orlando, FL, USA. Humann, P. and N. EteLoach. 20CS. Reef Creatore Identification: Florida, Caribbean, Bahamas. New World Publications Inc., Jacteonvilie, FL, USA. McLaughlin, F.A., D.K. Camp, L.G. Eldredge, D.L. Felder, J.W. Goy, H.H. Hobbs, III, B. Keusley, R. Lemaitre, and J.W. Martin. 2005. Order Da;:apoda. In: D. Ttirgeon, ed. Common and Scientific Names of Aquatic Invertebrates of the United States and Canada, Names of Crustaceans Special Ihiblications. VoL 31. American Fisheries Society Special Publication, Bethesda, MD, USA, p. 209-326. Melo, G.A.S. 1998. Malacostraca — ^Eucarlda. Brachyura. Oxyrhyncha and Brachyrhyncha. In: F. Young, ed. Catalogue of Crustacea of Brazil. Museu Nacional (Serie Livros no. 6), Rio de Janeiro, BR, p. 455-515. Morgan, G.J. 1991. A review of the hermit crab genus Calcmus Dana (Cmstacea: Dec^^oda; Diogeuidae) from Australia, with descriptions of two new spates. Inveitebrate Thxonomy 5:869-913. Powers, L. 1977. A catalogue and bibliography to the crabs (Brachyura) of the Gulf of Mexico. Contributions in Marine Science 20 (supplement): 1-190. Rathbun, M.J. 1930. The cancroid crabs of America of the famili^ Euiyalidae, Portunidae, Atelecydidae, Cancridae, and Xanthidae. Bulletin of the US National Museum 152:1-609. Serene, R. 1984. Cmstaces Decapodes Brachyoures de FCteean Indien Occidental et de la Mer Rouge, Xanthoidea: Xanthidae et Trapeziidae, Avec un addendum par Crosnier, A.: Caipiliidae et Meni^^idae. Fauoe Tropicale 24:1-400. Sterrer, W. (editor). 1986. Marine Fauna and Flora of Bermuda. John Wiley and Sons, New York, NY, USA. Ttidge, C.C. 1995. Hermit crabs of the Great Barrier Reef and coastal Queensland. Backhuys Publishers, l^iden. The Netherlands, 40 p. A.B. and D.L. Felder. 1989. Analysis of stone crabs: Menippe mewemrla (Say), restricted, and a previously unrecognized species describe (Decapoda: Xanthidae). lYoceedings of the Biological Society of Washington 99:517-543. 62 MAmN AND Zimmerman WiMiams, A.B., L.G. Abde, D.L. Felder, H.H. Hobbs, Jr., R.B. Maiming, RA. McLau|blin, and L Perez Farfante. 1989. Common and sdentific names of aquatic nwertebrates ftom the Uoiled States and Canada: Decapod Crastaceans. American Fisheries Scjciety, Special Publication 17:1-77, Zimmemiaii, XL, and D,L, Felder, 1991, Reproductive ecol- ogy of an intertidal brachyuraa crab, Sesarma sp. (nr. iieticuhtMm% from tl^ Golf of Mexico. Biological Biilleliii 181:387-401, 63