Bull Mar Sci. 90(2):000–000. 2014 http://dx.doi.org/10.5343/bms.2012.1092 1Bulletin of Marine Science © 2014 Rosenstiel School of Marine & Atmospheric Science of the University of Miami Survival, growth, and recruitment of octocoral species (Coelenterata: Octocorallia) in Coiba National Park, Pacific Panama Catalina G Gomez 1 * Hector M Guzman 1 Andrew Gonzalez 2 Odalisca Breedy 1, 3 ABSTRACT.—Octocorals (order Alcyonacea) from the tropical eastern Pacific have been largely ignored in coral reef studies, with the exception of recent taxonomic reviews. This study is the first to examine the population dynamics of 10 shallow water species in six genera (Leptogorgia, Pacifigorgia, Muricea, Psammogorgia, Heterogorgia, and Carijoa) found in rocky coral communities in Coiba National Park, Pacific Panama. For a 17-mo period, we monitored, every 4 mo, 1445 colonies of 15 species in fixed plots at 20 m depth in four coral communities. Size distribution, survivorship, and recruitment rates were calculated. Growth rate was calculated for Leptogorgia alba Duchassaing and Michelotti, 1864, Pacifigorgia Irene Bayer, 1951, Psammogorgia arbuscula Verrill, 1868, and Muricea austera Lamouroux, 1821. Average octocoral density was 38.7 (SD 27.55) colonies m−2 (n = 1394) with a range of 1–103 colonies m−2 and 1–11 species within a study plot. An overall population decline of 25.2% was observed in 1 yr. Leptogorgia alba was the most common species; it was abundant at all sites and exhibited characteristics of an r-selected species. In contrast, M. austera showed traits of a K-selected species, with low mortality and recruitment rates. Studied species were grouped into two distinct clusters based on their distribution, average density, mortality, and recruitment rates. Five species were grouped with L. alba and six species were grouped with M. austera. Octocorals (order Alcyonacea) are sessile colonial marine invertebrates found in many environments, ranging from cold deep oceans to shallow tropical seas (Bayer 1981, Alderslade 1984). The life histories of octocorals vary greatly due to species- specific symbionts (Mosher and Watling 2009), different growth forms (Breedy 2009), depth ranges, and reproductive strategies (e.g., Lasker 1990, Jordán-Dahlgren 2002), and the presence or absence of zooxanthellae (Van Open et al. 2005). Shallow water octocorals have lower dispersal capacities than scleractinian corals (Concepcion et al. 2010) and can occur in patchy distributions (e.g., Guzman et al. 2004, Linares et al. 2007). 1 Smithsonian Tropical Research Institute, Apartado 0843-03092, Balboa, Ancon, Panama. Email: . 2 Department of Biology, McGill University, 1205 Docteur Penfield, Montréal, Québec, Canada, H3A 1B1. Email: , Phone: (1) 514-398-6444. 3 Centro de Investigación en Ciencias del Mar y Limnología, Centro de Investigación en Estructuras Microscópicas, Universidad de Costa Rica, 11501- 2060 San José, Costa Rica. Email: , Phone: (506) 2511-4468. * Corresponding author email: , Phone: (507) 212-8736. Date Submitted: 29 November, 2012. Date Accepted: 8 July, 2013. Available Online: coral reef paper FastTrack➲ publication Bulletin of Marine Science. Vol 90, No 2. 20142 For centuries, humans have exploited gorgonian octocoral species for ornamental purposes (Tsounis et al. 2010). Recently, more species, including those in the eastern Pacific, have been recognized as a source of novel natural products and active com- pounds that can be used in medicine (e.g., Blunt et al. 2004, Gutierrez et al. 2005, 2006) by those seeking to exploit the coral’s chemical defenses against predators (Epifanio et al. 2000). In addition, human activities have damaged soft coral com- munities directly, through trawling, and indirectly through sedimentation and other forms of habitat degradation (Shester and Ayers 2005). The biology and ecology of octocoral species have been well studied in the Indo- Pacific, Caribbean, Mediterranean, off Hawaii, and in the deep ocean. These studies examined symbiotic relationships with zooxanthellae algae (e.g., Lewis and Coffroth 2004, Koike et al. 2004, Van Oppen et al. 2005) and brittle stars (Mosher and Watling 2009); growth (e.g., Mistri and Ceccherelli 1992, 1994, Castanaro and Lasker 2003, Lasker et al. 2003, Cadena and Sanchez 2010, Munari et al. 2013); sexual reproduc- tion (e.g., Kahng et al. 2008, Linares et al. 2008b, Hellström et al. 2010, Kahng et al. 2011); asexual reproduction (Lasker 1990); recruitment (Jamison and Lasker 2008); predation (Lasker and Coffroth 1988); feeding (Lasker et al. 1983); life history pat- terns (Linares et al. 2007, Linares et al. 2008a); and bleaching (Prada et al. 2010). Grigg (1972, 1974, 1975, 1977) studied life history trends in Muricea species in the eastern Pacific; however, knowledge about the species in the tropical eastern Pacific (TEP) is restricted mainly to taxonomy, biogeography, and phylogeny (Breedy and Guzman 2002, 2007, 2011, Vargas et al. 2008, 2010, Guzman and Breedy 2011). Octocoral species present in the TEP are found in high-energy environments such as seamounts and rocky walls exposed to strong currents, waves, and swells (Breedy and Guzman 2002). These conditions may explain the lack of studies in these areas. Unlike octocorals in other locations, none of the 27 species studied by Van Oppen et al. (2005) in the TEP are associated with zooxanthellae, meaning that they are obli- gate heterotrophs and obtain nutrients via suspension feeding. The TEP ranges from the Sea of Cortez to northern Peru (Robertson and Cramer 2009) and includes different octocoral communities that encompass 11 genera in four families with high levels of diversity, abundance, and endemism (Bayer 1953, Guzman et al. 2008). The Gulf of Chiriquí in Panama is a biodiversity hot spot in the TEP, with 52 octocoral species in seven genera (Guzman and Breedy 2008a). Coiba National Park (CNP), which is located along the Gulf of Chiriquí, is the larg- est marine protected area on the Pacific coast of Panama; it is part of the Marine Conservation Corridor of the TEP that includes Cocos Island (Costa Rica), Malpelo and Gorgona (Colombia), and Galápagos Islands (Ecuador) (see Fig. 1). The most common octocoral genera in CNP are Pacifigorgia, Leptogorgia, Muricea, Psammogorgia, Eugorgia, Heterogorgia, and Carijoa. The park contains 35 reported species, many of which are endemic to Coiba Island and the Gulf of Chiriquí (Guzman et al. 2004, Breedy and Guzman 2011). These species are distributed in highly diverse patches, along a depth gradient, and on occasions are found in monospecies patches. At CNP, octocorals share the substratum with encrusting coralline algae, scleractin- ian corals, algal turf, sponges, and tunicates (CGG, HMG, OB pers obs). Of the 35 species present at CNP only one, Carijoa riisei (Duchassaing and Michelotti, 1860), has been studied. Until 2010, C. riisei, which inhabits reefs in the Indo-Pacific, and off Hawaii (Kahng and Grigg 2005, Kahng et al. 2008), Indonesia, Gomez et al.: Octocoral survival, growth, and recruitment in Pacific Panama 3 and Vietnam (Calcinai et al. 2004), was recognized as an alien invasive species; how- ever, Concepcion et al. (2010) confirmed that this species is native to the Pacific. It is a fast-growing azooxanthellate octocoral that prefers shaded places (Kahng and Grigg 2005) and exhibits density-dependent sexual reproduction (Kahng et al. 2008). Carijoa riisei is considered a fouling species (Bayer 1961) that often overgrows other sessile organisms (Grigg 2003). The present study describes the octocoral community composition in four highly- diverse areas and the population dynamics of 10 common species in six genera in permanent monitoring plots located in CNP. It quantifies species mortality, recruit- ment, and growth rates in order to classify species into an r to K selection gradient. The present study is the first contribution on the biology of octocoral species and the ecology of octocoral communities in the tropical eastern Pacific. Figure 1. Study sites Jicarita, Catedrales, Roca Hacha, and Frijoles, in Coiba National Park, located in Panama and part of the Marine Conservation Corridor of the tropical eastern Pacific, which also includes Cocos (Costa Rica), Gorgona, Malpelo (Colombia) and Galápagos Islands (Ecuador). Bulletin of Marine Science. Vol 90, No 2. 20144 Methods Study Site CNP is the largest marine protected area in Pacific Panama. It encompasses an archipelago with nine main islands and about 30 islets with an area of 270,125 ha, of which 80.2% are marine. The Park, including its Special Zone of Marine Protection, was declared a UNESCO World Heritage Site in 2005 because of its high levels of endemism and key ecological interactions, both in terrestrial and marine communi- ties (Guzman et al. 2004, Maté et al. 2009). A management plan designed to protect its highly diverse marine and terrestrial ecosystems took effect in 2009. This plan established a non-take zone of 1.61 km around all islands and rocks and regulated artisanal fishery and tourism activities inside the park (Maté et al. 2009). Marine communities at the CNP are considered special because they are located in the Gulf of Chiriquí, which is unaffected by upwelling and trade north winds (D’Croz and O’Dea 2007), but experiences the influence of El Niño–Southern Oscillation (ENSO) every 2–7 yrs (Enfield 2001). Study sites were chosen based on previous descriptions of the abundance and rich- ness of the octocoral communities in the protected area (Guzman et al. 2004) and the tropical eastern Pacific (Guzman and Breedy 2008a), with one site on the leeward side and three on the seaward sides of the archipelago. These sites in CNP have ba- saltic rocky substrata inhabited by coral communities; they are exposed to different levels of wave action, currents, and swell. Site descriptions follow. Frijoles (7°38´59.6˝N, 81°43´09.4˝W).—This small islet is located approximately 2.5 km offshore of the northern part of Coiba Island, about 17 km from the mainland, on the leeward side of the archipelago, protected from strong currents and swell. It is surrounded by shallow water (20 m depth) and located approximatley 30 km away from the 400 m drop-off. The islet forms a vertical basaltic rocky wall from the sur- face down to 20 m depth. Encrusting coralline algae, small scleractinian colonies, macroalgae, and sponges cover the substrate. This site is located in a medium-high diversity area for corals and octocorals (Guzman et al. 2004). However, the octocoral community on the protected southeastern side of this islet is not diverse, and the most common species is Leptogorgia alba (Duchassaing and Michelotti, 1864). The coral predator Acanthaster planci (Linnaeus, 1758) is frequently seen in the area. Roca Hacha (7°25´ 55.0˝N, 81°51´ 29.0˝W).—This basaltic rocky outcrop is located approximatley 0.7 km offshore from the western seaward side of Coiba Island and is exposed to a strong swell. It is located in an area where natural mudslides are common, which create occasional sedimentation events. It is surrounded by shallow water (20 m depth) and located approximately 5 km away from the 400 m deep drop- off. This area was described as having medium-high diversity of corals and octo- corals (Guzman et al. 2004). Roca Hacha forms a vertical rocky wall with a decrease in slope at 20 m depth. A highly diverse octocoral community covers the substrate (Guzman and Breedy 2008b) as well as encrusting coralline algae, macroalgae, algal turf, sponges, tunicates, and small scleractinian corals. Catedrales (7°13´33.7˝N, 81°49 4´5.4˝W).—This underwater basaltic rocky outcrop, which forms two peaks 10 m below the sea surface, is located in the southwest region of CNP, approximately 2.6 km off of Jicarita Island. It is surrounded by shallow water Gomez et al.: Octocoral survival, growth, and recruitment in Pacific Panama 5 (20 m depth) and located approximately 3 km away from the 400 m deep drop-off. It is located in one of the four highly diverse areas of corals and octocorals within CNP (Guzman et al. 2004). Catedrales is exposed to strong currents from different directions and has a rich octocoral community. Encrusting coralline algae, sponges, tunicates, and small scleractinian corals also inhabit the substrate. Jicarita (7°12´12.5˝N, 81°48´02.3˝W).—This island is located in the southernmost region of CNP. It has a 30 m high cliff that continues vertically down into the water and decreases in slope at 20 m depth; it is 3 km away from the 400 m deep drop-off. It is exposed to surf and a strong swell, which brings sediment up from the bottom during the rising tide. A very diverse octocoral community, as well as encrusting cor- alline algae, algal turf, sponges, and small scleractinian corals cover Jicarita’s basaltic rocky substrate. This site has the highest coral diversity in Pacific Panama (Guzman and Breedy 2008b), and like Catedrales, it is located in one of the four most highly diverse areas of corals and octocorals within CNP (Guzman et al. 2004). Population Dynamics and Colony Monitoring Nine 1-m2 fixed plots were haphazardly established at a depth of 20 m at each site described above during the first week of June 2009. Fixed plots were marked within the octocoral community by installing 2 × 2 cm stainless steel square bars in the substrate with underwater cement; these markers were arranged parallel to shore and placed 5 m apart. An aluminum 1-m2 (0.84 × 1.20 m) quadrat was designed to fit on the fixed stainless steel bars. The quadrat was divided into eight 0.42 × 0.30-m sections summing to a square meter. Each section was photographed using a digital high-resolution Nikon D-80 camera with a wide-angle lens inside an Ikelite underwa- ter case and two external digital strobe flashes. The camera was attached to a stain- less steel tripod to maintain a fixed distance (0.80 m) from the quadrat. To monitor changes in water temperature, a logger (HOBO Pro v.2) was attached to the first fixed bar on each site and programmed to record the water temperature every 30 min. Loggers were attached to the bars with plastic cable ties and replaced every monitor- ing period. Data were downloaded using manufacturer software HOBOware® Pro Onset Computers and analyzed using SigmaPlot v.11 software. Each octocoral colony within the fixed plots was identified to the species level and assigned a unique ID number. The colonies were then manually marked in the pictures using Nikon NX2 and Corel PHOTO-PAINT X3 software. Carijoa riisei colonies were identified as individual colonies when they were spatially separated from one another. Unlike the other studied species, C. riisei is stolonal and colonies are difficult to distinguish due to their vegetative growth. All colonies within the fixed plots were monitored about every 4 mo for a 17-mo period from June 2009 to November 2010, for a total of five monitoring periods: (t1) 20 June, 2009, (t2) 20 October, 2009, (t3) 28 March, 2010, (t4) 21 July, 2010, (t5) 30 November, 2010. Species abundance, local diversity, recruitment rates, colony survivorship, damage, and mor- tality for each monitoring period were quantified. Colonies present at the beginning of the study (t1) were classified into species-spe- cific relative size classes from 1 to 4, with 1 representing recruits, and 4 representing the largest colony in the study plots. Colony size in each size class is detailed in (Table 1). Data obtained were analyzed to measure differences in community composition among sites, species-specific survivorship, and recruitment rates. Bulletin of Marine Science. Vol 90, No 2. 20146 Colony Growth Colony growth was studied for four species encompassing four genera within the study sites: L. alba at Frijoles, Pacifigorgia irene (Bayer, 1951) at Jicarita, and Muricea austera (Verrill, 1868), and Psammogorgia arbuscula (Verrill, 1866) at Roca Hacha. Pacifigorgia rubicunda (Breedy and Guzman, 2003) colonies were initially studied at Catedrales; however, it was not possible to identify the studied colonies in the field, and no growth analysis was performed for this site or this species. Ten colo- nies of each species were tagged within the monitoring plots and repeatedly pho- tographed in June and October of 2009, and March, July, and November of 2010. Digital photographs were taken using a 1 × 1-cm reference grid board. Colonies were later measured to the closest millimeter with the aid of ImageJ 64 software (Lasker et al. 2003). Due to the variability of growth forms (e.g., seafan, branching), two kinds of measurements were taken: for the branched species (L. alba and M. austera), four measurements were taken: (1) colony width was measured between the farthest two points perpendicular to an imaginary y-axis intercepting the holdfast; (2) colony height was measured from the holdfast to the farthest point on the y-axis; (3) num- ber of branch tips were counted, which included small (≥0.5 cm) growing tips; and (4) branch growth involved measuring 10 individual branches at the beginning and end of the study period in five of the studied colonies. For the fan-like species (P. irene and P. arbuscula), the fan surface-area was measured as well as the maximum width and length. Data Analyses Community Composition and Species Densities.—Rank abundance distribution plots (RADs; MacArthur 1957) were used to describe octocoral communities with- in each site. RADs are common community composition descriptors that provide a graphic means for comparing the proportion of rare species among communi- ties (McGill et al. 2007). The relative abundance of each species was calculated and plotted on a logarithmic scale against the species rank in abundance, from most to least common (from 1 to n). Overall species densities were calculated by dividing the number of colonies present on each plot by the total number of study plots. This calculation included the plots on sites where the focal species was not found. Survival Analysis.—The difference in survivorship between species and within species among sites for all size classes (1–4) was analyzed using survival curves. Table 1. Colony size at each size class per species. Species Measurement Size class (cm) 1 2 3 4 Leptogorgia alba height <2 2–5 5–10 10–15 Pacifigorgia rubicunda width <2 2–7 7–14 14–21 Pacifigorgia irene width <5 5–15 15–25 25–35 Carrijoa riisei area (cm3) <2 2–50 50–100 100–150 Heterogorgia verrucosa width <2 2–4 4–8 8–12 Leptogorgia cofrini width <1 1–3 3–6 6–9 Pacifigorgia cairnsi width <2 2–5 5–10 10–15 Muricea austera width <5 5–15 15–25 25–35 Psammogorgia arbuscula width <5 5–15 15–25 25–35 Gomez et al.: Octocoral survival, growth, and recruitment in Pacific Panama 7 Kaplan-Meier log-rank survival analysis for multiple groups was used to examine for a significant difference among survival curves using the “s” statistic (Kaplan and Meier 1958). Analyses were performed using SigmaPlot 11.2 Software (Systat Software 2009). The non-parametric log-rank test uses chi-square statistics to test for differences between survival curves. It assumes equal accuracy among all data at a given time. For the purposes of the analysis, colonies that survived for the entire study period (17 mo) were marked as “censored.” Colonies that died were marked as “failure” and the survival time was the number of months that the colony remained alive or present in the study plots. For example, a colony that was recorded at t1 and t2, but not at t3, t4, and t5 had a survival time of 4 mo. Mortality rates between moni- toring periods were calculated by dividing the number of dead colonies at t n + 1 by the total number of colonies at t n . Yearly mortality indices were calculated by dividing the total number of dead colonies by the total number of monitored colonies on each study plot over the course of a year. A significant difference in survivorship among species, sites, and across colony sizes was calculated with repeated measurements analysis of variance. Mortality peaks among sampling periods and the distribution of size classes within species were identified with a chi-square analysis (χ2). When necessary, data were transformed to achieve normality (log 10 x for density and √x + 1 for mortality and recruitment data). Recruitment Rate.—The number of new colonies present in the study plots dur- ing each monitoring time was quantified. Analysis of variance (ANOVA) on ranks was performed to test for significant differences among monitoring sites. Friedman repeated measures of variance on ranks (Friedman 1937) were performed to test for significant differences among monitoring periods; the statistic for this test is repre- sented as “q.” An overall yearly recruitment rate per m2 was also calculated as well as significant differences between sites among species (ANOVA). This recruitment rate was calculated by dividing the total number of new colonies by the initial number of colonies on each study plot over a 12-mo period. Analyses were performed using SigmaPlot 11.2 Software. Colony Growth.—The difference in size (height, width, area) between monitoring periods and the total difference from the beginning to the end of the study peri- od were calculated for each monitored colony (n = 10 per species). When a colony died before the study was over, size differences were calculated only if the colony remained alive for more than two monitoring periods (≥8 mo). The difference in size was divided by the number of months over which the colony was studied, and the average monthly growth was calculated for each species. To calculate branch length growth, 10 individual branches from a single colony in five M. austera and three L. alba individuals were measured to the closest millimeter at the beginning and end of the study, and the differences were divided by the total study period (17 mo). Life History Strategies.—To identify groups of species with similar life history pat- terns we performed a hierarchical cluster analysis using the function pvclust in the R package “pvclust” (Suzuki and Shimodaira 2011) performed in R Software 2.13 (R Development Core Team 2012). The analysis was executed using Euclidean distance with the “Ward” hierarchical method. This multivarate analysis calculates P values via multiscale bootstrap resampling that indicates how strong each cluster is sup- ported by the data. The analysis compared studied species using four biological vari- ables: distribution, density, and mortality and recruitment rates. Bulletin of Marine Science. Vol 90, No 2. 20148 Results Site Comparisons Mean monthly water temperatures ranged from 25.3 to 28.9 °C. Monthly standard deviations were relatively high (up to 1.65 °C) from February to May 2010. According to the National Oceanic and Atmospheric Administration’s (2011) sea surface height records, the study period overlapped with an ENSO warming event from June 2009 to April 2010 and a La Niña–Southern Oscillation (LNSO) cooling event from July 2010 until the end of the study (Fig. 2). The average monthly water temperature did not differ significantly among sites (ANOVA: F = 1.62, P = 0.19). Temperature for monitoring period 4 (July–November 2010), however, which overlapped with the LNSO event, differed significantly from the rest of the monitoring periods after al- lowing for the differences in temperature between sites (two-way ANOVA: F = 40.78, P < 0.001). The interaction between sites and monitoring periods was not significant (F = 0.12, P = 0.99; Fig. 2). The highest recorded temperatures were 30–30.8 °C dur- ing March and April 2010 at all four sites. The coldest temperatures were 20–20.6 °C at Catedrales and Jicarita in November 2010 and at Roca Hacha in March 2009. The lowest temperature, at Frijoles, was 21.7 °C in October 2010. In total, 1445 colonies of 15 species were monitored. There was an overall average density of 38.7 (SD 27.5) colonies m−2 (n = 1394) with a range of 1–103 colonies m−2 and 1–11 species within a study plot. An overall annual population decline of 25.2% was observed. There was a significant difference in colony density between study sites (ANOVA: P < 0.001). This difference was significant between all sites except Jicarita and Catedrales (Holm-Sidak: P > 0.1). Roca Hacha contained 14 species, with a mean density of 77.6 (SD 19) colonies m−2 (n = 678) and a 1-yr population decline of 17%. Catedrales had 12 species, a mean density of 44.3 (SD 18) colonies m−2 (n = 350), and a 1-yr population decline of 14%. Jicarita contained 12 species, with a mean Figure 2. Monthly seawater temperature profiles for study sites in Coiba National Park, Panama. Average temperature (straight lines) and maximum and minimum values (dotted lines), with El Niño (ENSO) and La Niña (LNSO) events overlapping the study period. Gomez et al.: Octocoral survival, growth, and recruitment in Pacific Panama 9 density of 30.3 (SD 14) colonies m−2 (n = 268) and a 1-yr population decline of 24%, and Frijoles had five species, a mean density of 11.2 (SD 5) colonies m−2 (n = 98), and a 1-yr population decline of 30%. There was higher species dominance at Frijoles, where L. alba was the dominant species, and there was a high occurrence of rare species in Catedrales, Jicarita, and Roca Hacha (Fig. 3). A site comparison analysis that included all species present at each site revealed a significant and strong positive relationship between density and yearly mean recruit- ment (Pearson’s correlation: n = 4, r2 = 0.99, P < 0.01). At the species level, a positive relationship was found between species density and recruitment (Pearson’s correla- tion: n = 13, r2 = 0.70, P < 0.01), but not a significant relationship between density and mortality. A positive relationship was also found between species recruitment and mortality rates (Pearson’s correlation: n = 13, r2 = 0.56, P < 0.05) Survival curves The global Kaplan-Meier log rank survival analysis, which included all species at all sites, resulted in survival curves that differed significantly among colony size groups (s = 144.55, P < 0.001). Larger sized colonies had higher survivorship than smaller colonies (Fig. 4A). Similarly, survival curves differed significantly among sites (s = 17.23, P < 0.001) due to low survivorship at Frijoles (Fig. 4B). A significant difference in survivorship curves among species also was detected (s = 225.32, P < 0.001). L. alba and C. riisei had the lowest survivorship and M. austera and P. rubicunda had the highest (Fig. 4C, Table 2). Overall colony mortality did not differ among moni- toring periods after allowing for the differences in mortality among sites (F = 0.43, P = 0.72). Recruitment An overall comparison, in which all species were pooled, showed that recruit- ment peaks differed among sites. Frijoles and Jicarita had a peak during the first Figure 3. Rank abundance distribution (RAD) for each study site in Coiba National Park, Panama during t1 (June 20, 2009). Bulletin of Marine Science. Vol 90, No 2. 201410 monitoring period (June–October 2009), with a total of 77 and 81 recruits, respec- tively. Catedrales exhibited a peak during t2 (October 2009–March 2010) with a total of 106 recruits, and Roca Hacha showed a peak during t5 (July 2010–November 2010) with a total of 184 recruits (Fig. 5). Yearly recruitment rates for L. alba, H. verrucosa, and P. rubicunda were calculated using information from all monitored plots (36 m2 plots). Study plots from Frijoles (9 m2) were not used to calculate recruitment rates for the rest of the species, which were absent or very rare at this site. Leptogorgia alba recruits were the most abundant (3.55 colonies m−2), followed by P. rubicunda and P. irene (1.07 and 1.05 colonies m−2, respectively). Recruits of M. austera, Pacifigorgia stenobrochis (Valenciennes, 1846) (an azooxathellate species), and Leptogorgia taboguilla (Hickson, 1928) were not observed during the study pe- riod (Table 2). Figure 4. Global survivorship plots for octocorals in Coiba National Park, Panama. (A) Among size classes, with all species and sites together, larger colonies (4) had an overall higher survi- vorship than smaller colonies (1). (B) Among sites, with all species and size classes together, colonies at Frijoles had an overall lower survivorship. (C) Among species, wide survivorship variation among species. Gomez et al.: Octocoral survival, growth, and recruitment in Pacific Panama 11 Species-Specific Analyses Leptogorgia alba colonies were found at all of the study sites, and a total of 607 colonies were monitored. The species’ overall mean density was 11.8 (SD 9.0) colonies m−2 (n = 426) and ranged between 0 to 40 colonies m−2. It had an average mortality rate of 0.4 (SD 0.1 and an annual recruitment rate of 0.3 (SD 0.3 (Table 2). Mortality rates were not significantly different among monitoring periods. However, recruit- ment was not constant over the study period; there was recruitment peak during t3 and t4 (χ2 = 13.5, P < 0.01). Size classes were not randomly distributed (χ2 = 33.8, P < 0.001); sizes 2 and 3 were the most common at every site (Table 3). Survival curves were significantly different for all size classes (s = 44.98, P < 0.001); larger colonies had higher survivorship than smaller colonies, and the order of survivorship by size class was 4 > 3 > 2 > 1. In terms of growth measurements, only 3 of the 10 colo- nies that were initially measured and marked for monitoring survived to the end of the study period. The average monthly growth of these three colonies was 0.34 (SD 0.14) cm in height and 0.37 (SD 0.32) cm in width, with a monthly increase of 1 (SD 1.2) for branch tips and an increase in branch length of 0.19 (SD 0.02) cm. An in- crease in branch length was followed by an increase in branch tips (at t4 and t5). The maximum-recorded monthly growth was an increase in branch length of 0.41 cm (between t3 and t4) and an increase of 3.2 new branches between t4 and t5 (Table 4). Leptogorgia alba was a dominant species at Frijoles and very common at the other three sites, with a mean density of 12.6 (SD 6.4) colonies m−2 (n = 113) at Catedrales; 8.4 (SD 3.5) colonies m−2 (n = 76) at Frijoles; 5.3 (SD 3.8) colonies m−2 (n = 48) at Jicarita; and 21 (SD 11.4) colonies m−2 (n = 189) at Roca Hacha (Table 2). When ana- lyzing the population at each site, a significant difference in survivorship among size classes was detected only at Roca Hacha, where all size classes differed (s = 26.29, P < 0.001), and at Frijoles between size class 1 and 4 (p = 0.040, s = 8.31). Overall Figure 5. Recruitment rates for each study site in Coiba National Park, Panama. Peaks occurred during different monitoring periods. Bulletin of Marine Science. Vol 90, No 2. 201412 Table 2. Mean and standard deviation (m2 yr−1) of density, mortality, and recruitment for species found at study sites in Coiba National Park, Panama. A diamond indicates a significant difference between sites, an asterisk a significant difference between monitoring periods, and a triangle is a significant difference in survivorship between size classes, “n1” is the number of colonies at monitoring period 1, and “n2” the number of colonies during all monitored period. Not seen during the study period = n/s, not applicable = na. Species/site n1/n2 Density (n1) SD Range Mortality (n2) SD Recruitment (n1) SD Carrijoa riisei Overall 117/144 3.3 7.5 0–41 0.5 0.3 0.20 0.50 Catedrales 25/35 2.8 4.3 0–12 0.6 0.1 0.30 0.90 Frijoles 0/0 0.0 na na na na na na Jicarita 16/18 1.8 2.6 0–20 0.9 0.4 0.10 0.20 Roca Hacha 76/90 8.4 13.3 0–41 0.5* 0.2 0.06 0.10 Heterogorgia verrucosa▲ Overall 93/111 2.6♦ 2.9 0–9 0.2 0.2 0.10 0.20 Catedrales 26/36 2.9 2.5 0–6 0.1 0.9 0.90 0.20 Frijoles 11/12 1.2 2.9 0–9 0.4 0.4 0.00 0.00 Jicarita 8/9 0.9 1.5 0–4 0.04 0.1 0.03 0.10 Roca Hacha 48/54 5.3 2.3 1–9 0.2* 0.2 0.07 0.10 Leptogorgia alba▲ Overall 426/607 11.8♦ 9.0 0–40 0.4 0.1 0.30 0.30 Catedrales 113/150 12.6 6.4 3–22 0.4 0.1 0.30 0.20 Frijoles 76/107 8.4 3.5 5–15 0.5* 0.3 0.40 0.30 Jicarita 48/68 5.3 3.8 0–12 0.5 0.4 0.20 0.20 Roca Hacha 189/282 21.0 11.4 5–40 0.6 0.3 0.40* 0.30 Leptogorgia cofrini▲ Overall 82/92 2.3 3.7 0–16 0.4 0.2 0.10 0.20 Catedrales 24/27 2.7 2.5 0–7 0.5 0.2 0.10 0.10 Frijoles 0/0 0.0 na na na na na na Jicarita 39/44 4.3 6.4 0–16 0.2* 0.3 0.02 0.10 Roca Hacha▲ 19/26 2.1 1.8 0–5 0.3 0.2 0.20 0.30 Leptogorgia rigida Overall 25/26 0.7 2.3 0–11 0.3 0.3 0.01 0.04 Catedrales 0/0 0.0 na na na na na na Frijoles 1/1 0.1 0.3 0–1 n/s na n/s na Jicarita 24/25 2.7 4.2 0–11 0.4 0.1 0.01 0.04 Roca Hacha 0/0 0.0 na na na na na na Leptogorgia pumilla Overall 9/11 0.3 0.6 0–2 0.2 0.3 0.10 0.20 Catedrales 1/1 0.1 0.3 0–1 n/s na n/s na Frijoles 2/1 0.0 na na 0.4 0.5 0.20 0.30 Jicarita 7/5 0.8 0.8 0–2 0.1 0.3 n/s na RH 1/3 0.1 0.3 0–1 n/s na 0.10 0.20 Leptogorgia taboguilla Overall 5/6 0.1 0.5 0–3 na na na na Catedrales 4/5 0.4 0.5 0–3 na na na na Frijoles 0/0 0.0 na na na na na na Jicarita 0/0 0.0 na na na na na na Roca Hacha 1/1 0.1 0.3 0–1 n/s na n/s na Gomez et al.: Octocoral survival, growth, and recruitment in Pacific Panama 13 Table 2. Continued. Species/site n1/n2 Density (n1) SD Range Mortality (n2) SD Recruitment (n1) SD Muricea austera Overall 22/22 0.6♦ 1.3 0–6 0.20 0.30 ns na Catedrales 1/1 0.1 na 0–1 n/s na n/s na Frijoles 0/0 0.0 na na na na na na Jicarita 0/0 0.0 na na na na na na Roca Hacha 21/21 2.3 1.7 1–6 0.20 0.30 n/s n/s Pacifigorgia cairnsi▲ Overall 43/50 1.2 1.4 0–6 0.20 0.20 0.20 0.40 Catedrales 14/17 1.6 1.5 0–4 0.40 0.30 0.20 0.30 Frijoles 0/0 0.0 na na na na na na Jicarita 13/15 1.4 1.0 0–3 0.20 0.30 0.20 0.50 Roca Hacha 19/18 1.8 1.7 0–6 0.20 0.30 0.20 0.30 Pacifigorgia eximia Overall 6/8 0.2 0.4 0–1 0.10 0.30 0.10 0.30 Catedrales 2/1 0.2 0.4 0–1 0.40 0.50 ns na Frijoles 0/0 0.0 na na na na na na Jicarita 0/0 0.0 na na na na na na Roca Hacha 4/6 0.4 0.5 0–1 n/s na 0.20 0.50 Pacifigorgia firma Overall 7/10 0.2 0.4 0–1 0.10 0.30 0.10 0.10 Catedrales 1/1 0.1 0.3 0–1 n/s na n/s na Frijoles 0/0 0.0 na na na na na na Jicarita 2/4 0.2 0.4 0–1 0.10 0.10 0.20 0.10 Roca Hacha 4/5 0.4 0.5 0–1 0.20 0.40 0.10 0.20 Pacifigorgia irene Overall 193/232 5.4♦ 5.8 0–24 0.10 0.10 0.20 0.20 Catedrales 54/70 6.0 2.8 2–11 0.20 0.20 0.20 0.20 Frijoles 0/0 0.0 na na na na na na Jicarita 23/30 2.6 2.3 0–7 0.04 0.10 0.20 0.30 Roca Hacha 116/132 12.9 5.2 6–24 0.20 0.09 0.10 0.10 Pacifigorgia rubicunda▲ Overall 291/332 8.1♦ 7.0 0–26 0.20♦ 0.20 0.10 0.20 Catedrales 64/83 7.1 2.4 3–11 0.20 0.20 0.10* 0.20 Frijoles 7/9 0.8 0.8 0–2 0.30 0.40 0.20 0.30 Jicarita 78/84 8.7 5.7 2–20 0.05 0.06 0.04 0.06 Roca Hacha 142/156 15.8 6.7 7–26 0.20* 0.05 0.07 0.05 Psammogorgia arbuscula▲ Overall 25/32 0.7 1.5 0–7 0.20 0.30 0.10 0.30 Catedrales 5/6 0.6 0.8 0–2 n/s na 0.10 0.20 Frijoles 1/1 0.1 0.3 0–1 n/s na n/s na Jicarita 2/3 0.2 0.4 0–1 0.50 0.40 0.10 0.20 Roca Hacha 17/22 1.9 2.5 0–7 0.20 0.20 0.20 0.50 Pacifigorgia stenobrochis Overall 2/2 0.1 0.3 0–1 n/s na n/s na Catedrales 0/0 0.0 na na na na na na Frijoles 0/0 0.0 na na na na na na Jicarita 1/1 0.1 0.3 0–1 n/s na n/s na Roca Hacha 1/1 0.1 0.3 0–1 n/s na n/s na Bulletin of Marine Science. Vol 90, No 2. 201414 Table 3. Size class distributions of octocoral species at study sites in Coiba National Park, Panama. Relative class sizes: 1 represents small recruits, 2 small colonies, 3 medium size colonies, and 4 large colonies. Asterisks indicate species for which the survivorship curves significantly differed between class sizes. Not applicable = na. Size class (%) Species/site n 1 2 3 4 Carrijoa riisei Overall 115 39.1 23.5 25.2 12.2 Catedrales 25 52.0 12.0 28.0 8.0 Frijoles 0 na na na na Jicarita 16 25.0 56.3 18.8 0.0 Roca Hacha 74 37.8 20.3 25.7 16.2 Heterogorgia verrucosa * Overall 91 27.5 36.3 25.3 11.0 Catedrales 26 30.8 15.4 30.8 23.1 Frijoles 11 54.5 27.3 18.2 0.0 Jicarita 7 0.0 57.1 28.6 14.3 Roca Hacha 47 23.4 46.8 23.4 6.4 Leptogorgia alba * Overall 406 19.2 37.9 34.5 8.4 Catedrales 104 11.5 33.7 34.6 20.2 Frijoles 72 25.0 29.2 38.9 6.9 Jicarita 46 26.1 39.1 32.6 2.2 Roca Hacha 184 19.6 43.5 33.2 3.8 Leptogorgia cofrini * Overall 78 9.0 38.5 46.1 6.4 Catedrales 22 4.5 27.3 59.1 9.1 Frijoles 0 na na na na Jicarita 38 13.2 44.7 34.2 7.9 Roca Hacha 18 5.6 38.9 55.6 0.0 Muricea austera Overall 20 0.0 15.0 30.0 55.0 Catedrales 0 na na na na Frijoles 0 na na na na Jicarita 0 na na na na Roca Hacha 20 0.0 15.0 30.0 55.0 Pacifigorgia cairnsi * Overall 38 13.2 26.3 39.5 21.1 Catedrales 11 0.0 36.4 36.4 27.3 Frijoles 0 na na na na Jicarita 11 0.0 18.2 45.5 36.4 Roca Hacha 15 33.3 26.7 33.3 6.7 Pacifigorgia irene Overall 178 8.4 24.7 36.5 30.3 Catedrales 51 11.8 19.6 27.5 41.2 Frijoles 0 na na na na Jicarita 23 13.0 8.7 30.4 47.8 Roca Hacha 104 5.8 30.8 42.3 21.2 Gomez et al.: Octocoral survival, growth, and recruitment in Pacific Panama 15 survivorship did not differ among sites. Recruitment was only significantly higher at Roca Hacha during t3 (q = 4.39, P = 0.001) and t5 (q = 3.92, P = 0.021). Pacifigorgia rubicunda was common at all sites except Frijoles. A total of 332 colo- nies were monitored. Overall mean density was 8.1 (SD 7.0) colonies m−2 (n = 291), with a range of 0 to 26 colonies within a square meter, a mean annual mortality rate of 0.2 (SD 0.2) and an annual recruitment rate of 0.1 (SD 0.2) (Table 2). There was a significantly higher mortality during t5 (χ2 = 39.37, P < 0.001). However, there was not a significant difference in recruitment rates between sampling periods. Colony size was distributed unevenly with 53.6% of the colonies in size class 3 (χ2 = 145.8, P < 0.01). There was a significant difference in survivorship between size classes (P < 0.01). Size 1 had lower survivorship than sizes 2, 3, and 4. Pacifigorgia rubicunda colony density was significantly different among sites (F = 6.88, P < 0.001): there was an average of 15.8 (SD 6.7) colonies m−2 (n = 142) at Roca Hacha; 8.7 (SD 5.7) colonies m−2 (n = 78) at Jicarita; 7.1 (SD 2.4) colonies m−2 (n = 64) at Catedrales; and 0.8 (SD 0.8) colonies m−2 (n = 7) at Frijoles (Table 2). Mortality rates were significantly different among sites (P < 0.01) and ranged from 0.05 (SD 0.06) in Jicarita to 0.3 (SD 0.4) in Frijoles. There was not a significant difference in recruit- ment rate among sites for this species. Size class 3 was more frequent at Catedrales, Jicarita, and Roca Hacha, and size class 2 was more frequent at Frijoles (Table 3). The survivorship curve was higher at Jicarita (s = 13.01, P < 0.001), and there was an overall (all sites together) difference in survivorship among size classes (s = 18.12, P < 0.001), with size 1 having a lower survivorship than sizes 3 and 4 and size 2 hav- ing a lower survivorship than size 4. A difference in survivorship among monitoring periods occurred only at Roca Hacha, where it was higher at t2 than t5 (q = 4.77, P = 0.001). Recruitment differed among monitoring periods only at Catedrales, where it was higher at t4 than t5 (q = 4.75, P = 0.01). Recruitment did not differ among sites (Table 2). Pacifigorgia irene colonies were frequently found in monospecific patches, in which most of the colonies shared the same orientation. A total of 233 colonies were moni- tored, with a mean density of 5.4 (SD 5.8) colonies m−2 (n = 193), ranging from 0 to 24 colonies m−2. This species had an annual mortality rate of 0.1 (SD 0.1) and an annual recruitment rate of 0.2 (SD 0.2) (Table 2). Size class was not randomly distributed (χ2 Table 3. Continued. Size class (%) Species/site n 1 2 3 4 Pacifigorgia rubicunda * Overall 276 5.1 26.4 53.6 14.9 Catedrales 57 1.8 19.3 54.4 24.6 Frijoles 7 28.6 71.4 0.0 0.0 Jicarita 75 2.7 20.0 58.7 18.7 Roca Hacha 137 6.6 30.7 53.3 9.5 Psammogorgia arbuscula * Overall 25 8.0 24.0 28.0 40.0 Catedrales 5 0.0 20.0 80.0 0.0 Frijoles 0 na na na na Jicarita 0 na na na na Roca Hacha 17 5.9 23.5 11.8 58.8 Bulletin of Marine Science. Vol 90, No 2. 201416 = 31.03, P = 0.01); only 0.8% of the population was in size class 1 (Table 3). Survival curves did not differ among size classes or monitoring periods. Recruitment was not significantly different between monitoring periods. Colonies used for growth measurements were in size classes 3 and 4. There was an average monthly increase in colony area of 0.84 (SD 5.64) cm2 (n = 10). Area measurements provided a bet- ter growth estimator than height and width because of colony breakage and uneven growth (Table 4). Colonies had frequent fan breakages, but these did not cause colony death. Density differed significantly among sites (F = 18.62, P < 0.001), with a mean density of 12.9 (SD 5.2) colonies m−2 (n = 116) at Roca Hacha; 6 (SD 2.8) colonies m−2 (n = 54) at Catedrales; and 2.6 (SD 2.3) colonies m−2 (n = 23) at Jicarita. There were no P. irene colonies in the study plots at Frijoles at the beginning of the study, but there was one recruit during t3. Size class 4 was more common at Jicarita and Catedrales, and size class 3 was more common at Roca Hacha (Table 3). There was no significant difference in survival curves or recruitment rates among sites. Carijoa riisei colonies exhibited rapid growth, during which small recruits merged into larger adjacent colonies, making it difficult to distinguish individual colonies. A total of 144 colonies was monitored with an average density of 3.3 (SD 7.5) colonies m−2 (n = 117), ranging from 0 to 41 colonies m−2. Colonies had a mortality rate of 0.5 (SD 0.3) and a recruitment rate of 0.2 (SD 0.5) (Table 2). Size classes were not evenly distributed (χ2 = 16.6, P < 001,): almost 40% of the monitored colonies were in size class 1 (Table 3). Survivorship did not differ significantly among size classes, even when the colonies were grouped in two size categories. Mortality and recruitment rates did not differ between monitoring periods. Densities of C. riisei colonies varied between study sites: there was a mean density of 8.4 (SD 13.3) colonies m−2 (n = 76) at Roca Hacha; 2.8 (SD 4.3) colonies m−2 (n = 25) at Catedrales; and 1.8 (SD 2.6) colonies m−2 (n = 16) at Jicarita (Table 2). Survivorship curves differed among sites; colonies at Roca Hacha had higher survivorship than colonies at Catedrales and Frijoles (s = 31.69, P = 0.001). Survivorship differed among monitoring periods only at Roca Hacha, with higher values during t2. There was no significant difference in recruitment rates among sites. Heterogorgia verrucosa (Verril, 1868): One hundred eleven colonies were moni- tored, with a mean density of 2.6 (SD 2.9) colonies m−2 (n = 93) which ranged from 0 to 9 colonies per m2. This species had an overall mortality rate of 0.2 (SD 0.2) and a recruitment rate of 0.1 (SD 0.2) (Table 2). Size classes were not randomly distributed (χ2 = 11.98, P < 0.01), only 11% of the colonies were in the larger size class (Table 3). Survivorship was different among size classes (s = 15.82, P = 0.001); size class 1 had lower survivorship than size classes 3 or 4. Neither recruitment nor mortality rates differed significantly among sampling periods. Heterogorgia verrucosa colony density was significantly different between sites (F = 9.75, P < 0.001); there was an average of 5.3 (SD 2.3) colonies m−2 (n = 48) at Roca Hacha; 2.9 (SD 2.5) colonies m−2 (n = 26) at Catedrales; 1.2 (SD 2.9) colonies m−2 (n = 11) at Frijoles; and 0.9 (SD 1.5) colonies m−2 (n = 8) at Jicarita (Table 2). Size classes 1 and 3 were equally abundant at Catedrales, whereas size 1 was most abun- dant at Frijoles and size 2 was most abundant at Jicarita and Roca Hacha (Table 3). Survivorship curves did not differ significantly among sites. A size class difference in survivorship was not found when analyzing sites separately. Survivorship differed between monitoring periods only at Roca Hacha, where it was higher at t2 than t4 (q = 3.87, P < 0.05). There was not a significant difference in recruitment among sites. Gomez et al.: Octocoral survival, growth, and recruitment in Pacific Panama 17 Leptogorgia cofrini (Breedy and Guzman, 2005): Ninety-two L. cofrini colonies were monitored. The species occurred at a mean density of 2.3 (SD 3.7) colonies m−2 (n = 82) with a range of 0 to 16 colonies m−2. Its mean mortality rate was 0.2 (SD 0.1) and its mean recruitment rate was 0.2 (SD 0.2) (Table 2). Morality and recruitment rates did not differ among monitoring periods. Size classes were not randomly dis- tributed (χ2 = 38.41, P < 0.01): most of the colonies (84.7%) were in size classes 2 and 3 (Table 3). There was a significant difference in survivorship between size classes (s = 8.95, P = 0.03); size class 4 had higher survivorship than size class 2. Leptogorgia cofrini density was not significantly different among sites: it had an av- erage density of 4.3 (SD 6.4) colonies m−2 (n = 39) at Jicarita; 2.7 (SD 2.5) colonies m−2 (n = 24) at Catedrales; 2.1 (SD 1.8) colonies m−2 (n = 19) at Roca Hacha; and was not present in the monitored plots at Frijoles (Table 2). Survivorship curves did not differ among sites, and when analyzed separately, a difference in survivorship between size classes was only observed at Roca Hacha (s = 7.99, P = 0.02). Survivorship curves dif- fered among monitoring periods only at Jicarita (q = 3.83, P = 0.02), where they were higher at t2 than t5. Recruitment did not differ significantly among sites. Pacifigorgia cairnsi (Breedy and Guzman, 2003): Fifty colonies were monitored, with an average density of 1.2 (SD 1.4) colonies m−2 (n=43), an average mortality rate of 0.2 (SD 0.2), and a recruitment rate of 0.2 (SD 0.4) (Table 2). There was not a sig- nificant difference in the distribution of size classes. However, survivorship curves differed among size classes (s = 9.16, P = 0.03); size class 4 had higher survivorship than size class 2. There was not a significant difference between mortality and re- cruitment rates among sampling periods. Pacifigorgia cairnsi densities did not differ between sites: the mean density at Jicarita was 1.4 (SD 1) colonies m−2 (n = 13); 1.6 (SD 1.5) colonies m−2 (n = 14) at Catedrales; and 1.8 (SD 1.7) colonies m−2 (n = 19) at Roca Hacha. There were no P. cairnsi colonies at study plots in Frijoles (Table 2). The difference in survivorship between size classes was not observed when the sites were analyzed separately. There was no difference in recruitment or survivorship among sites. Leptogorgia rigida (Verril, 1864) was only found at Jicarita, with the exception of one small colony at Frijoles. Twenty-six colonies were monitored, with a mean abun- dance of 2.7 (SD 4.2) colonies m−2 (n = 24) at Jicarita (Table 2). The estimated mortal- ity rate for this specie is 0.3 (SD 0.3). There was only one recruit during t3 at Jicarita, which was used to estimate a recruitment rate of 0.01(SD 0.04) (Table 2). Twenty-two colonies of Muricea austera were monitored with a mean density of 0.6 (SD 1.3) colonies m−2 (n = 22) with a range of 0 to 6 colonies m−2. It had a mortality rate of 0.2 (SD 0.3), which did not differ between monitoring periods or size classes. No recruits were seen over the duration of the study (Table 2). Size class was not distributed randomly among colonies (χ2 = 13.2, P < 0.01), with 55% of the colonies in size class 4 (Table 3). M. austera colonies had slow growth with a monthly net increment of 0.03 (SD 0.1) cm in height, 0.04 (SD 0.2) cm in width, and an increase in the number of branch tips of 0.33 (SD 0.4) cm. The net branch length growth was negative due to bites from possible predators in individual branches [–0.01 (SD 0.06) cm]. The maximum-recorded monthly growth was an increase of 0.95 cm in height, 1.55 cm in width, 3 new branches, and an increase in branch length of 0.14 cm (Table 4). This species was only common at Roca Hacha, were it had a mean abundance of 2.3 (SD 1.7) colonies m−2 (n = 21) (Table 2). This species was rare at Catedrales, with a mean density of 0.1 (n = 1), and was absent in the study plots at Frijoles and Jicarita. Bulletin of Marine Science. Vol 90, No 2. 201418 Psammogorgia arbuscula had a mean density of 0.7 (SD 1.2) (n = 25). A total of 32 colonies were monitored, presenting a mortality rate of 0.2 (SD 0.3) and a re- cruitment rate of 0.1 (SD 0.3) (Table 2). These rates were not significantly different between sampling periods. Colonies were randomly distributed in size classes. A significant difference among size class survivorship curves was detected (s = 18.77, P < 0.001,), with size class 1 having lower survivorship than size classes 3 and 4 (P = 0.01 and P = 0.01, respectively) and size class 2 having a lower value than size class 4 (P = 0.01) (Table 3). Measured colonies had a net monthly growth of 0.08 (SD 0.17) cm in height and 0.08 (SD 0.32) cm in width. The maximum growth was an increase of 0.84 cm in height and 0.93 cm in width. Nine of the ten monitored colonies remained alive at the end of the study (Table 4). Psammogorgia arbuscula was common at Roca Hacha and very rare at Jicarita, Frijoles, and Catedrales, with a mean abundance of 1.9 (SD 2.5) colonies m−2 at Roca Hacha (n = 17); 0.6 (SD 0.8) colonies m−2 at Catedrales (n = 5); 0.1 (SD 0.3) colo- nies m−2 at Frijoles (n = 1); and 0.2 (SD 0.4) colonies m−2 at Jicarita (n = 2) (Table 2). Recruits (n = 5) were only observed at Roca Hacha during t4 and t5 (Table 2). Life History Strategies Based on four biological variables (species distribution, average density, recruit- ment, and mortality rates), the cluster analysis divided species in two significantly distinct groups (P < 0.05) (Fig. 6). Group 1 clustered six species: C. riisei, H. ver- rucosa, L. alba, L. cofrini, P. irene and P. rubicunda. The average density of these species was 5.6 colonies m−2 (SD 3.7), average mortality rate of 0.3 (SD 0.2), average recruitment of 0.2 (SD 0.1) and an average distribution among study sites of 3.7 (SD 0.5). Group 2 clustered seven species: L. rigida, Leptogorgia pumilla (Verrill, 1868), M. austera, P. cairnsi, Pacifigorgia eximia (Verrill, 1868), Pacifigorgia firma (Breedy and Guzman, 2003), and P. arbuscula. These species had an average density of 0.6 colonies m−2 (SD 0.4), average mortality of 0.2 (SD 0.1), average recruitment of 0.1 (SD 0.1), and an average distribution among study sites of 2.7 (SD 0.1). Table 4. Monthly mean growth and maximum growth recorded for four species at Coiba National Park. Colony superficial area was only measured for fan-like species. Height, width, and branch length values given in centimeters, area given in centimeters squared. Value with asterisk is n = 5. Species n Height (SD) Width (SD) Area (SD) Number of branches (SD) Branch length (SD) Average monthly net growth Leptogorgia alba 3 0.34 (0.14) 0.37 (0.32) – 1.00 (1.02) 0.19 (0.02) Pacifigorgia irene 10 −0.05 (0.18) −0.01 (0.21) 0.84 (5.64) – – Psammogorgia arbuscula 9 0.08 (0.17) 0.08 (0.32) – – – Muricea austera 10 0.03 (0.12) 0.04 (0.17) – 0.33 (0.42) −0.01 (0.06)* Maximum recorded monthly growth Leptogorgia alba 0.86 0.79 – 3.50 0.41 Pacifigorgia irene 1.12 1.17 20.6 – – Psammogorgia arbuscula 0.84 0.93 – – – Muricea austera 0.95 1.55 – 3.2 0.14 Gomez et al.: Octocoral survival, growth, and recruitment in Pacific Panama 19 Discussion Inter-Regional Differences The octocoral densities (up to 106 colonies m−2 with a mean of 38.7 colonies m−2) found at CNP were higher than those reported in the Caribbean: 25.1 colonies m−2 in south Florida (Goldberg 1973); 17.6 colonies m−2 in Carre Bow Cay, Belize (Lasker and Coffroth 1983); 9.98 colonies m−2 in Providencia Island, Colombia (Sanchez et al. 1998); 3.6–5.9 colonies m−2 in the Florida Keys (Chiappone and Suvillan 1994); and 62.3 colonies m−2 in southwest Puerto Rico (Yoshioka 1996). Octocorals were cer- tainly dominant species in these communities, especially at Roca Hacha, Jicarita, and Catedrales. These findings support statements of Bayer (1953) in the tropical eastern Pacific, who described these species as the most characteristic components of rocky communities. At CNP, octocorals shared the rocky substrata with sponges, tuni- cates, and encrusting coralline algae; however, the corals were the only organisms creating complex three-dimensional structures, hosting a variety of invertebrates, and serving as aggregation areas for fish. Octocoral species richness at CNP was found to be high compared to other islands along the Marine Conservation Corridor of the tropical eastern Pacific. There are a total of 34 reported species in CNP (Guzman et al. 2004); 15 of those species were present in the study plots, compared to 12 reported in Cocos (Breedy and Cortes Figure 6. Hierarchical cluster analysis of similarities between species explained by four variables (species distribution, average density, and mortality and recruitment rates) Group 1 clusters r- selected species and Group 2 clusters K-selected species. Bulletin of Marine Science. Vol 90, No 2. 201420 2008), 10 in Colombia’s Pacific (Prahl et al. 1986), and 7 in Galápagos (Hickson 1928, Williams and Breedy 2004, Breedy and Guzman 2007, Breedy et al. 2009). This level of species richness was similar to that reported in Bocas del Toro (29 species) in Caribbean Panama (Guzman and Guevara 1999) and Costa Rica, with 26 species found in the Caribbean Sea and 30 in the Pacific Ocean (Breedy 2009). It was also similar to the levels found in Japan, where there were 27 species above 40 meters (Matsumoto et al. 2007). Octocoral species richness at CNP was low, however, com- pared to the the Indo-Pacific region, which reports 90 genera in 23 families in shallow waters (Fabricius and Alderslade 2001); the Mediterranean region, with 43 reported species (Koukouras et al. 2001); the Jaragua National Park, Dominican Republic, with 47 species in 15 genera (Weil 2006); and Hong Kong, which has 42 species in 23 gen- era (Fabricius and McCorry 2006). Octocoral colonies were distributed with patches of very high densities (106 colo- nies m−2) and low densities (5 colonies m−2). This patchy distribution is reflected in the high standard deviation in colony density in the overall analysis (average density 38.7 ± 27.5 colonies m−2) and for each site and species separately (see Table 2). There are two potential reasons for this type of uneven distribution: asexual reproduction due to branch breakage, which is thought to be common in octocorals (e.g., Lasker 1990), and a possible reproductive strategy in which the new recruit settles near the parental colony. Mortality rates ranged between 1% and 6% yr−1, which is similar to the rates found in protected areas in the Mediterranean with low (2.7%) and high (7.4%) diving ac- tivities (Coma et al. 2004), and relatively low compared to the 8% mortality rate re- ported for the Caribbean (e.g. Yoshioka and Yoshioka 1991, Yoshioka 1994). Smaller colonies (sizes 1 and 2) had higher mortality rates than larger colonies; the relation- ship between size and survivorship has been reported in a variety of environments, including the genus Muricea in the Gulf of Mexico (Grigg 1977), and shallow water Pseudopterogorgia spp. from Puerto Rico, where larger colonies had 96% survivorship compared to 62% in smaller colonies (Yoshioka 1994). Mortality in larger colonies was mainly due to detachment, which was seen in the temperate octocoral Paramuricea clavata in areas frequented by tourist divers (Coma et al. 2004) and Caribbean oc- tocorals in shallow and exposed areas (Wahle 1985, Yoshioka and Yoshioka 1991). Mortality in smaller colonies and among recruits was mainly due to overgrowth by macroalgae, sponges, or other octocorals, especially C. riisei. Detachment and abra- sion, rather than predation, were the main causes of mortality for two Muricea spe- cies in Baja California (Grigg 1977). Grigg (1977) explained that high mortality rates in M. californica were due to bioerosion by bivalves and other invertebrates, which weakened the basal attachment of older colonies. The distribution of octocorals is limited by the availability of substrata (Grigg 1977, Preston and Preston 1975, Birkeland 1974, Opresko 1974) and functional larvae (Jordan-Dahlgren 2002). Free rocky space available in our sites suggests that the dis- tribution in the study site is controlled by recruitment of new larvae. At the species level recruitment rates were lower than mortality rates, reflected in the decline in natural population observed during the study (25.2%). This de- cline was relatively low, however, compared to the 1999 mass mortality event in the Mediterranean Sea, which caused a decrease of >50% of the Corallium rubrum popu- lation (Cerrano et al. 2000) and 48% of the Paramuricea clavata population (Linares et al. 2005). This mass mortality event was probably related to a positive anomaly in Gomez et al.: Octocoral survival, growth, and recruitment in Pacific Panama 21 sea surface temperatures of 4 °C followed by extensive colony attacks by opportunis- tic pathogens like protozoans and fungi (Cerrano et al. 2000), and subsequent colony death (Linares et al. 2005). Conversely, although there were ENSO and LNSO events, no abrupt changes in temperature were recorded during the present study and the presence of opportunistic pathogens was not observed. Net growth measurements reported in the present study include negative growth caused by predation or colony breakage. During the monitoring periods we observed sea turtles and Scaridae and Acanthurus fish biting at colonies on several occasions. We attempted to study the effect of predation on mortality and growth, but unfortu- nately, the exclusion cages were lost twice from the study site, and it was not possible to collect this type of data. Due to negative growth and further tissue regeneration, measurements of the fan area were a better measure of growth than fan height and width alone, especially for P. irene. Another barrier to studying growth was the high colony mortality rate in L. alba colonies, with only three out of ten monitored colo- nies surviving to the end of the study. Growth rates (measured by colony height) in species in Puerto Rico averaged about 2.0 cm yr−1 (0.8 to 4.5 cm yr−1) (Yoshioka and Yoshioka 1991), which is relatively low compared to our calculations for L. alba (4.0 cm yr−1) and high compared to our calculations for P. arbuscula (0.1 cm yr−1) and M. austera (0.3 cm yr−1). The Puerto Rican study only measured colony height, how- ever, and in an effort to avoid negative growth, did not include broken or predated colonies. Grigg (1974) found that colony growth decreased consistently with colony height, but this relationship was not significant in Puerto Rico’s octocorals (Yoshioka and Yoshioka 1991). In the 5-yr study in Puerto Rico, there was high intraspecific variability, which supported the idea that variability was not just an “artifact” of short-term observations (Yoshioka and Yoshioka 1991). Inter-Site Differences Differences among the study sites in species richness, abundance, mortality, and recruitment could be the result of a combination of factors. The low diversity and abundance of octocorals in Frijoles, relative to the other three study places, could be explained by its geographic position. Frijoles is located on the leeward side of the archipelago and closer to the mainland (about 17 km) than the other three sites. These sites are located on the seaward side of the archipelago, farther away from the mainland (>50 km) with no land protection, leaving them exposed to a variety of cur- rents, breaking waves, and surges (see Fig. 1). Frijoles is surrounded by shallow water (20 m deep); it is approximately 30 km away from deep water (>400 m), which is not the case for the other three sites: Roca Hacha is approximately 5 km, and Jicarita and Caterdales are only approximately 3 km from the >400 m drop-off (see Fig. 1). The Frijoles location is closer to the mainland, protected from strong currents, and surrounded by shallow water, which could be why this site had the warmest mean monthly temperature profile. Although its temperature was not significantly different from the other sites, its higher temperature may indicate that this site is less influenced by the cold deep water near Coiba. Fabricius and De’ath (2008) reported higher octocoral richness in locations with high water column productivity, greater depth, and more water flow, which may also be the case in Roca Hacha, Jicarita, and Catedrales, but not in Frijoles. This idea would support Jordan-Dahlgren’s (2002) pro- posal that octocoral distribution is controlled by the availability of functional larvae, Bulletin of Marine Science. Vol 90, No 2. 201422 which could be reduced at Frijoles if the larvae come from cold and productive deep water or off shore currents bringing larvae from more populated communities. In addition, the invasive coral predator, A. planci, was seen at Frijoles during every monitoring period but never seen at the other three locations. Acanthaster planci is known to feed on soft corals and gorgonians when there is low hard coral cover (Moran 1990), which is the case in the studied rocky coral communities. Predation by this sea star could decrease the possibility of octocoral recruits surviving to adult- hood. Interestingly, this coral predator seems to affect species other than L. alba, which was a dominant species at Frijoles, and recruitment and mortality rates of this species did not differ among sites. Inter-Species Differences Life history traits varied widely among studied species; however, the cluster analy- sis suggested two main life history strategies among studied species, one group of species resembling K selection and another group resembling r selection. The inter- action between species density, distribution among study sites, and recruitment and mortality rates in species found in group no. 1 (Fig. 6) resembled r-selected species. These species were commonly found in most of the sites where highly dense patches were common (with up to 41 colonies m−2). These species also had higher recruitment and mortality rates than species in group 2. Two of these species were widely distributed and had the highest densities and growth rates (L. alba and P. irene) of the studied species. Out of these species, L. alba, which is widely distributed at CNP, and present in Cocos Island, Costa Rica´ s mainland shores (Breedy and Cortés 2008), El Salvador (Bielschowsky 1929), Colombia (Prahl et al 1986), Galápagos (Breedy and Guzman 2007), and continen- tal Ecuador (Bielschowsky 1929), had the most active life history as shown through high recruitment and mortality rates. Pacifigorgia irene and P. rubicunda had rela- tively high recruitment values but lower mortality indices relative to L. alba. These species are widely distributed at CNP and also reported in Costa Rica (Breedy and Guzman 2003). Although C. riisei was not as common as these other three species, it showed high distribution and high mortality and recruitment values. Species in this group also exhibited a patchy distribution pattern, with some plots being very densely populated (24–41 colonies m−2) and others having only one or two colonies, as reflected in the large density ranges and SD values in Table 2. Additionally, most P. irene colonies in a patch shared the same fan orientation, which has been described for Muricea californica (Aurivillius, 1931) and Muricea fruticosa (Verril, 1869) as a response to water flow direction (Grigg 1972). Group 2 resembles K-selected species (Fig. 6). These species had a lower density, very low or no recruitment during the study period, low mortality and where absent in some of the study sites. Additionally growth studies for M. austera and P. arbus- cula showed low growth rates compared to species in Group 1. These species where rare compared to species in Group 1 and where not found in highly dense patches, with a maximum of 11 colonies m−2 (L. rigida). In general, the genus Pacifigorgia had a greater survivorship than the genus Leptogorgia, at least among the studied species in the studied sites (average mor- tality rates of 0.17 (SD 0.11) for five Pacifigorgia species and 0.29 (SD 0.11) for four Leptogorgia species). These genera are among the most diverse and abundant of shal- low water octocoral fauna in the eastern Pacific (Breedy and Guzman 2002, Breedy Gomez et al.: Octocoral survival, growth, and recruitment in Pacific Panama 23 and Guzman 2007). These differences could be due to colony morphology. In gen- eral, Pacifigorgia species have thicker branches, which create a relatively strong com- pound or single network that is securely attached to the substratum by a holdfast of different dimensions and forms, and in some cases the colonies are strengthened by midribs (Breedy and Guzman 2002). Conversely, Leptogorgia species have slen- der branches and are attached to the substrate by a single mother branch (Breedy and Guzman 2007), possibly making them more vulnerable to colony damage and detachment. Branch thickness has been described as an adaptation to strong water movement on at least three occasions (Grigg 1972, Velimirov 1976, Kin et al. 2004). Although these studies have compared intra-species adaptations, we hypothesize this is also the case when comparing different species. As reported for four species in the tropical eastern Pacific (Patton 1972, Cantera et al. 1987, Neira et al. 1992, Ramos 1995), octocorals at CNP were highly associ- ated with symbiotic invertebrates. These kinds of associations have been described as obligated to the point that the two individuals are “partners for life” (Mosher and Watling 2009). Therefore, the range of different life histories among octocoral spe- cies found in this study could also be associated with the life history of associated taxa; as Grigg (1975) stated, the stability of these foundation species reflects the suit- ability and stability of their associated taxa. The present study is the first contribution on the biology of these species, which are the main components of rocky coral communities at CNP and in similar environ- ments in the tropical eastern Pacific. The information presented here serves as base- line knowledge in the event of future exploitation of any of these species, which are known to be a source of active compounds (Maia et al. 2000, Gutierrez et al. 2005, Gutierrez et al. 2006, Reimão et al. 2008). This baseline knowledge will also inform management of octocoral communities inside and outside marine protected areas. Acknowledgments The authors thank C Guevara for laborious assistance in the field; the Government of Panama (ANAM), which provided permits to work in Coiba National Park; and the reviewers for their comments. This research was partially sponsored by Secretaría Nacional de Ciencia y Tecnología de Panama (SENACYT- Grant No, PNCOIBA-08-024), the Smithsonian Tropical Research Institute, and the McGill-STRI NEO Program. Literature Cited Alderslade P. 1984. Subclass Alcyonaria. In: Mather P, Benett I, editors. A coral reef handbook. The Australian Coral Reef Society. p. 45–48. Bayer FM. 1953. Zoogeography and evolution in the octocotallian family Gorgoniidae. Bull Mar Sci Gulf Caribb. 3:100–119. Bayer FM. 1961. The shallow water Octocorallia of the West Indian Region. Martinus Nijhoff. The Hague. http://dx.doi.org/10.1039/b821918b Bayer FM. 1981. Status of knowledge of octocorals of world seas. Academia Brasileira de Ciências, Rio de Janeiro. Birkeland C. 1974. The effect of wave action on the population dynamics of Gorgonia ventalina Linnaeus. Stud Trop Oceanogr. 12:115–126. Blunt JW, Copp BR, Munro MHG, Northcote PT, Prinsep MR. 2004. Marine natural products. Nat Prod Rep. 21:1–49. http://dx.doi.org/10.1039/b006897g Bulletin of Marine Science. Vol 90, No 2. 201424 Breedy O. 2009. Octocorals. In: Wehrtmann IS, Cortés J, editors. Marine biodiver- sity of Costa Rica, Central America. Springer Netherlands. p. 161–167. http://dx.doi. org/10.1007/978-1-4020-8278-8_13 Breedy O, Cortes J. 2008. Octocorals (Coelenterata: Anthozoa: Octocorallia) of Isla del Coco, Costa Rica. Rev Biol Trop. 56(Suppl 2):71–77. Breedy O, Guzman HM. 2002. A revision of the genus Pacifigorgia (Coelenterata: Octocorallia: Gorgoniidae). P Biol Soc Wash. 115:782–839. Breedy O, Guzman HM. 2003. Octocorals from Costa Rica: The genus Pacifigorgia (Colenterara: Octocorallia: Gorgoniidae). Zootaxa. 281:1. Breedy O, Guzman HM. 2007. A revision of the genus Leptogorgia Milne Edwards and Haime, 1857 (Coelenterata: Octocoralia: Gorgonidae) in the eastern Pacific. Zootaxa. 1419:1–90. Breedy O, Guzman HM. 2011. A revision of the genus Heterogorgia Verrill, 1868 (Anthozoa: Octocorallia: Plexauridae). Zootaxa. 2995:27–44. Breedy O, Guzman HM, Vargas S. 2009. A revision of the genus Eugorgia Verrill, 1868 (Colenterata: Octocorallia: Gorgoniidae). Zootaxa. 2151:1–46. Cadena NJ, Sánchez JA. 2010. Colony growth in the harvested octocoral Pseudopterogorgia acerosa in a Caribbean coral reef. Mar Ecol. 31:566–573. http://dx.doi. org/10.1111/j.1439-0485.2010.00397.x Calcinai B, Bavestrello G, Cerrano C. 2004. Dispersal and association of two alien species in the Indonesian coral reefs: the octocoral Carijoa riisei and the demosponge Desmapsamma anchorata. J Mar Biol Assoc UK. 84:937. http://dx.doi.org/10.1017/S0025315404010227h Cantera JR, Prahl VH, Neira OR. 1987. Moluscos, crustaceos y equinodermos asociados a la gorgonia Lephogorgia alba Duchassaing y Micheloti 1864, en la isla de Gorgona, Colombia. Bol Ecotropica. 17:3–23. Castanaro J, Lasker HR. 2003. Colony growth responses of the Caribbean octocoral, Pseudopterogorgia elisabethae, to harvesting. Invertebr Biol. 122:299–307. http://dx.doi. org/10.1111/j.1744-7410.2003.tb00094.x Cerrano C, Bavestrello G, Bianchi CN, Cattaneo-vietti R, Bava S, Morganti C, Morri C, Picco P, Sara G, Schiaparelli S, et al. 2000. A catastrophic mass-mortality episode of gorgonians and other organisms in the Ligurian Sea (northwestern Mediterranean), summer 1999. Ecol Lett. 3:284–293. http://dx.doi.org/10.1046/j.1461-0248.2000.00152.x Chiappone M, Sullivan KM. 1994. Ecological structure and dynamics of nearshore hardbottom communities in the Florida Keys. Bull Mar Sci. 54:747–756. Coma R, Pola E, Ribes M, Zabala M. 2004. Long-term assessment of temperate octocoral mor- tality patterns, protected vs unprotected areas. Ecol Appl. 14:1466–1478. http://dx.doi. org/10.1890/03-5176 Concepcion GT, Kahng SE, Crepeau MW, Franklin EC, Coles SL, Toonen RJ. 2010. Resolving natural ranges and marine invasions in a globally distributed octocoral (genus Carrijoa). Mar Ecol Prog Ser. 401:113–127. http://dx.doi.org/10.3354/meps08364 D’Croz L, O’Dea A. 2007. Variability in upwelling along the Pacific shelf of Panama and impli- cations for the distribution of nutrients and chlorophyll. Est Coast Shelf Sci. 73:325–340. http://dx.doi.org/10.1016/j.ecss.2007.01.013 Enfield DB. 2001. Evolution and historical perspective of the 1997-1998 El Nino-Southern Oscillation event. Bull Mar Sci. 69:7–25. Epifanio RDA, Maia LF, Fenical W. 2000. Chemical Defenses of the Endemic Brazilian Gorgonian Lophogorgia violacea Pallas (Octocorallia, Gorgonacea). J Brazil Chem Soc. 11:584–591. http://dx.doi.org/10.1016/S1872-2032(08)60048-7 Fabricius K, Alderslade P. 2001. Soft corals and sea fans: a comprehensive guide to the tropi- cal shallow water genera of the central-west Pacific, the Indian Ocean and the Red Sea. Australian Institute of Marine Science, Townsville, Australia. p. 264. Fabricius K, De’Ath G. 2008. Photosynthetic symbionts and energy supply determine octocoral biodiversity in coral reefs. Ecology. 89:3163–3173. http://dx.doi.org/10.1890/08-0005.1 Gomez et al.: Octocoral survival, growth, and recruitment in Pacific Panama 25 Fabricius KE, McCorry D. 2006. Changes in octocoral communities and benthic cover along a water quality gradient in the reefs of Hong Kong. Mar Pollut Bull. 52:22–33. http://dx.doi. org/10.1016/j.marpolbul.2005.08.004 Friedman M. 1937. The use of ranks to avoid the assumption of normality implicit in the analy- sis of variance. J Am Stat Assoc. 32:675–701. http://dx.doi.org/10.2307/2279169 Goldberg WM. 1973. The ecology of the coral-octocoral communities off the southeast Florida coast: geomorphology, species composition, and zonation. Bull Mar Sci. 23:465–488. Grigg RW. 1972. Orientation and growth from of sea fans. Limnol Oceanogr. 17:185–192. http://dx.doi.org/10.4319/lo.1972.17.2.0185 Grigg RW. 1974. Growth rings: annual periodicity in two gorgonian corals. Ecology. 55:876– 881. http://dx.doi.org/10.2307/1934424 Grigg RW. 1975. Age structure of a longevous coral: a relative index of habitat suitability and stability. Am Nat.109:647–657. http://dx.doi.org/10.1086/283035 Grigg RW. 1977. Population dynamics of two gorgonian corals. Ecology. 58:278–290. http:// dx.doi.org/10.2307/1935603 Grigg RW. 2003. Invasion of a deep black coral bed by an alien species, Carijoa riisei, off Maui, Hawaii. Coral Reefs. 22:121–122. http://dx.doi.org/10.1007/s00338-003-0306-5 Gutierrez M, Capson TL, Guzman HM, González J, Ortega-Barria E, Quiñoa E, Riguera R. 2005. Leptolide, a new furanocembranolide diterpene from Leptogorgia alba. J Nat Prod. 68:614–616. http://dx.doi.org/10.1021/np049745z Gutierrez M, Capson TL, Guzman HM, Gonzalez J, Ortega-Barria E, Quiñoa E, Riguera R. 2006. Antiplasmodial metabolites isolated from the marine octocoral Muricea austera. J Nat Prod. 69:1379–1383. http://dx.doi.org/10.1021/np060007f Guzman HM, Benfield S, Breedy O, Mair JM. 2008. Broadening reef protection across the ma- rine conservation corridor of the eastern tropical Pacific: distribution and diversity of reefs in Las Perlas Archipelago, Panama. Environ Conserv. 35:46–54. http://dx.doi.org/10.1017/ S0376892908004542 Guzman HM, Breedy O. 2008a. Distribución de la Diversidad y Estado de Conservación de los Arrecifes Coralinos y Comunidades Coralinas del Pacífico Occidental de Panamá (Punta Mala-Punta Burica). The Nature Conservancy. Guzman HM, Breedy O. 2008b. Leptogorgia christiae (Octocorallia : Gorgoniidae) a new shal- low water gorgonian from Pacific Panama. J Mar Biol Assoc UK. 88:719–722. http://dx.doi. org/10.1017/S0025315408001240 Guzman HM, Breedy O. 2011. Pacifigorgia marviva (Anthozoa: Octocorallia) a new species from Coiba National Park, Pacific Panama. J Mar Biol Assoc. 92:693–698. http://dx.doi. org/10.1017/S0025315411000373 Guzman HM, Guevara CA. 1999. Coral reefs of Bocas del Toro, Panama: III Distribution, structure, diversity and conservation status of reefs in Pastores, Cristobal, Popa and Cayo Agua islands. Rev Biol Trop. 47:659–676. Guzman HM, Guevara CA, Breedy O. 2004. Distribution, diversity, and conservation of coral reefs and coral communities in the largest marine protected area of the Pacific Panama (Coiba Island). Environ Conserv. 31:111–121. http://dx.doi.org/10.1017/S0376892904001250 Hellström M, Kavanagh K, Benzie J. 2010. Multiple spawning events and sexual reproduction in the octocoral Sarcophyton elegans (Cnidaria: Alcyonacea) on Lizard Island, Great Barrier Reef. Mar Biol. 157:383–392. http://dx.doi.org/10.1007/s00227-009-1325-8 Hickson SJ. 1928. The Gorgonacea of Panama Bay together with a description of one species from the Galápagos Islands and one one of Trinidad. Videnskavelige Meddelelser fra den naturhistoriske Forening i Kovenhavn for Aarene. 85:325–422. Jamison JL, Lasker HR. 2008. Identification of octocoral recruits using microsatellite prim- ers: relationships between recruitment and adult distribution of Pseudopterogorgia spp. American Society of Limnology and Oceanography, Waco, TX, USA. http://www.jstor.org/ stable/40058312 Bulletin of Marine Science. Vol 90, No 2. 201426 Jordan-Dahlgren E. 2002. Gorgonian distribution patterns in coral reef environments of the Gulf of Mexico: evidence of sporadic ecological connectivity? Coral Reefs. 21:205–215. Kahng SE, Benayahu Y, Wagner D, Rothe N. 2008. Sexual reproduction in the invasive octo- coral Carijoa riisei in Hawaii. Bull Mar Sci. 82:1–17. Kahng SE, Benayahu Y, Lasker HR. 2011. Sexual reproduction in octocorals. Mar Ecol Prog Ser. 443:265. http://dx.doi.org/10.3354/meps09414 Kahng SE, Grigg W. 2005. Impact of an alien octocoral, Carijoa riisei, on black corals in Hawaii. Coral Reefs. 24:556–562. http://dx.doi.org/10.1007/s00338-005-0026-0 Kaplan EL, Meier P. 1958. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 53:457–481. http://dx.doi.org/10.1080/01621459.1958.10501452 Kim E, Lasker HR, Coffroth MA, Kim K. 2004. Morphological and genetic variation across reef habitats in a broadcast-spawning octocoral. Hydrobiologia. 530:423–432. http://dx.doi. org/10.1007/s10750-004-2646-8 Koike K, Jimbo M, Sakai R, Kaeriyama M, Muramoto K, Ogata T, Maruyama T, Kamiya H. 2004. Octocoral chemical signaling selects and controls dinoflagellate symbionts. Biol Bull. 207:80–86. PMid:15501849. http://dx.doi.org/10.2307/1543582 Koukouras A, Voultsiadou E, Kitsos MS, Doulgeraki S. 2001. Macrobenthic fauna diversity in the Aegean Sea, affinities with other Mediterranean regions and the Black Sea. Bios (Macedonia, Greece). 6:61–76. Lasker H. 1990. Clonal propagation and population dynamics of a gorgonian coral. Ecology. 71:1578–1589. http://dx.doi.org/10.2307/1938293 Lasker HR, Boller ML, Castanaro J, Sánchez JA. 2003. Determinate growth and modular- ity in a gorgonian octocoral. Biol Bull. 205:319–330. PMid:14672986. http://dx.doi. org/10.2307/1543295 Lasker HR, Coffroth MA. 1983. Octocoral distributions at Carrie Bow Cay, Belize. Mar Ecol Prog Ser.13:21–28. http://dx.doi.org/10.3354/meps013021 Lasker HR, Coffroth MA. 1988. Temporal and spatial variability among grazers: variability in the distribution of the gastropod Cyphoma gibbosum on octocorals. Mar Ecol Prog Ser. 43:285–295. http://dx.doi.org/10.3354/meps043285 Lasker HR, Gottfried MD, Coffroth MA. 1983. Effects of depth on the feeding capabilities of two octocorals. Mar Biol. 73:73–78. http://dx.doi.org/10.1007/BF00396287 Lewis CL, Coffroth MA. 2004. The acquisition of exogenous algal symbionts by an octocorol after bleaching. Science. 304:1490–1492. PMid:15178798. http://dx.doi.org/10.1126/ science.1097323 Linares C, Coma R, Diaz D, Zabala M, Hereu B, Dantart L. 2005. Immediate and delayed ef- fects of a mass mortality event on gorgonian population dynamics and benthic commu- nity structure in the NW Mediterranean. Mar. Ecol. Prog. Ser. 305:127–137. http://dx.doi. org/10.3354/meps305127 Linares C, Coma R, Garrabou J, Díaz D, Zabala M. 2008a. Size distribution, density and distur- bance in two Mediterranean gorgonians: Paramuricea clavata and Eunicella singularis. J Appl Ecol. 45:688–699. http://dx.doi.org/10.1111/j.1365-2664.2007.01419.x Linares C, Coma R, Zabala M. 2008b. Effects of a mass mortality event on gorgonian reproduc- tion. Coral Reefs. 27:27–34. http://dx.doi.org/10.1007/s00338-007-0285-z Linares C, Doak DF, Coma R, Diaz D, Zabala M. 2007. Life history and viability of a long- lived marine invertebrate: the octocoral Paramuricea clavata. Ecology. 88:918–928. PMid:17536708. http://dx.doi.org/10.1890/05-1931 Macarthur RH. 1957. On the relative abundance of bird species. Proc Natl Acad Sci USA. 43:293–295. http://dx.doi.org/10.1073/pnas.43.3.293 Maia LF, Epifanio RDA, Fenical W. 2000. New cytotoxic sterol glycosides from the octocoral Carijoa (Telesto) riisei. J Nat Prod. 63:1427–1430. http://dx.doi.org/10.1021/np9906323 Maté JL, Tovar D, Arcia E, Hidalgo Y, editors. 2009. Plan de Manejo del Parque Nacional Coiba, Autoridad Nacional del Ambiente. Panama. Gomez et al.: Octocoral survival, growth, and recruitment in Pacific Panama 27 Matsumoto AK, Iwase F, Imahara Y, Namikawa H. 2007. Bathymetric distribution and biodi- versity of cold-water octocorals (Coelenterata: Octocorallia) in Sagami Bay and adjacent waters of Japan. Bull Mar Sci. 81:231–251. McGill BJ, Etienne RS, Gray JS, Alonso D, Anderson MJ, Benecha HK, Dornelas M, Enquist BJ, Green JL, Fangliang H, et al. 2007. Species abundance distributions: moving beyond single prediction theories to integration within an ecological framework. Ecol Lett. 10:995–1015. http://dx.doi.org/10.1111/j.1461-0248.2007.01094.x Mistri M, Ceccherelli VU. 1993. Growth of the Mediterranean gorgonian Lophogorgia cera- tophyta (L., 1758). Mar Ecol. 14:329–340. Mistri M, Ceccherelli VU. 1994. Growth and secondary production of the Mediterranean gor- gonian Paramuricea clavata. Mar. Ecol-Prog. Ser. 103: 291-291. http://dx.doi.org/10.3354/ meps103291 Moran PJ. 1990. Acanthaster planci (L.): biographical data. Coral Reefs. 9:95–96. http://dx.doi. org/10.1007/BF00258218 Mosher C, Watling L. 2009. Partners for life: a brittle star and its octocoral host. Mar Ecol Prog Ser. 397:81–88. http://dx.doi.org/10.3354/meps08113 Munari C, Serafin G, Mistri M. 2013. Structure, growth and secondary production of two Tyrrhenian populations of the white gorgonian Eunicella singularis (Esper 1791). Estuar Coast Shelf Sci. 119:162–166. http://dx.doi.org/10.1016/j.ecss.2013.01.007 Neira OR, Barba PR, Pardo RA. 1992. Equinodermos del Parque Nacional Natural Ensenada de Utria (Pacifico Colombiano). An Inst Invest Mar Punta Betin. 21:77–83. NOAA. 2011. ENSO Cycle: Recent evolution, current status and predictions. Climate Prediction Center, NCEP. Opresko DM. 1974. Recolonization and regrowth of a population of the gorgonian Plexaura homomalla. In: Bayer FM, Weinheimer AJ, editors. Studies in Tropical Oceanography. University of Miami Press, Coral Gables, FL. p. 165. Patton WK. 1972. Studies on the animal symbionts of the gorgonian coral, Leptogorgia virgu- lata (Lamarck). Bull Mar Sci. 22:419–431. Prada C, Weil E, Yoshioka P. 2010. Octocoral bleaching during unusual thermal stress. Coral Reefs. 29:41–45. http://dx.doi.org/10.1007/s00338-009-0547-z Prahl HV, Escobar D, Molina G. 1986. Octocorales (Octocorallia: Gorgoniidae y Plexauridae) de aguas someras del Pacífico Colombiano. Rev Biol Trop. 34:13–33. Preston EM, Preston LJ. 1975. Ecological structure in a West Indian gorgonian fauna. Bull Mar Sci. 25:248–258. R Development Core Team. 2012. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0. Available from: http://www.R-project.org/ Ramos GE. 1995. Neopontonides henryvonprahli, una nueva especie de camarón pontoníno del Pacífico de Colombia (Decapoda: Palaemonidae) simbionte de las gorgonias Muricea robusta y Lophogorgia alba. Rev Biol Trop. 43:231–237. Reimão J, Migotto A, Kossuga M, Berlinckv R, Tempone A. 2008. Antiprotozoan activity of Brazilian marine cnidarian extracts and of a modified steroid from the octocoral Carijoa riisei. Parasitol Res. 103:1445–1450. http://dx.doi.org/10.1007/s00436-008-1154-6 Robertson DR, Cramer KL. 2009. Shore fishes and biogeographic subdivisions of the tropical eastern Pacific. Mar Ecol Prog Ser. 380:1–17. http://dx.doi.org/10.3354/meps07925 Sánchez JA, Zea S, Díaz JM. 1998. Patterns of octocoral and black coral distribution in the oceanic barrier reef-complex of Providencia Island, southwestern Caribbean. Caribb J Sci. 34:250–264. Shester G, Ayers J. 2005. A cost effective approach to protecting deep-sea coral and sponge ecosystems with an application to Alaska’s Aleutian Islands region. In: Freiwald A, Roberts JM, editors. Cold-water corals and ecosystems. Springer Berlin Heidelberg. p. 1151–1169. http://dx.doi.org/10.1007/3-540-27673-4_59 Bulletin of Marine Science. Vol 90, No 2. 201428 Suzuki R, Shimodaira H. 2011. pvclust: hierarchical clustering with P-values via multiscale bootstrap resampling. R package version 1.2-2. Available from: http://CRAN.Rproject.org/ package=pvclust Systat Software I. 2009. SigmaPlot® 11.2 User’s Guide. Systat Software, Inc, San Jose, CA. Tsounis G, Rossi S, Grigg RW, Santangelo G, Bramanti L, Gill JM. 2010. The exploitation and conservation of precious corals. In: Gibson RN, Atkinson RJA, Gordon JDM, editors. Oceanography and marine biology: an annual review. Taylor & Francis. 48:161–212. http:// dx.doi.org/10.1201/EBK1439821169-c3 Van Oppen MJH, Miegog JC, Sanchez CA, Fabricius KE. 2005. Diversity of algal endosymbi- onts (zooxanthellae) in octocorals: the roles of geography and host relationships. Mol Ecol. 14:2403–2417. http://dx.doi.org/10.1111/j.1365-294X.2005.02545.x Vargas S, Breedy O, Guzman HM. 2010. The phylogeny of Pacifigorgia (Coelenterata, Octocorallia, Gorgoniidae): a case study of the use of continuous characters in the system- atic of the Octocorallia. Zoosystema. 32:5–18. http://dx.doi.org/10.5252/z2010n1a1 Vargas S, Guzman HM, Breedy O. 2008. Distribution patterns of the genus Pacifigorgia (Octocorallia: Gorgoniidae): track compatibility analysis and parsimony analysis of ende- micity. J Biogeogr. 35:241–247. http://dx.doi.org/10.1111/j.1365-2699.2007.01810.x Velimirov B. 1976. Variations in growth forms of Eunicella cavolinh Koch (Octocorallia) re- lated to intensity of water movement. J Exp Mar Biol Ecol. 21:109–117. http://dx.doi. org/10.1016/0022-0981(76)90032-0 Wahle CM. 1985. Habitat-related patterns of injury and mortality among Jamaican gorgonians. Bull Mar Sci. 37:905–927. Weil E. 2006. Diversity and relative abundance of corals, octocorals and sponges at Jaragua National Park, Dominican Republic. Rev Biol Trop. 54:423–443. PMid:18494313. Williams GC, Breedy O. 2004. The Panamic gorgonian genus Pacifigorgia (Octocorallia: Gorgoniidae) in the Galápagos Archipelago, with descriptions of three new species. Proc Calif Acad Sci. 55:55–88. Yoshioka PM. 1994. Size-specific life history pattern of shallow-water gorgonian. J Exp Mar Biol Ecol. 184:111–122. http://dx.doi.org/10.1016/0022-0981(94)90169-4 Yoshioka PM. 1996. Variable recruitment and its effects on the population and community structure of shallow-water gorgonians. Bull Mar Sci. 59:433–443. Yoshioka PM, Yoshioka BB. 1991. A comparison of the survivorship and growth of shallow- water gorgonian species of Puerto Rico. Mar Ecol Prog Ser. 69:253–260. http://dx.doi. org/10.3354/meps069253 B M S