Invest in arms: behavioural and energetic implications of multiple autotomy in starfish (Asterias rubens)

The autotomy of body parts as a means of escaping predation or renewing damaged tissues has evolved in a number of animal groups. Starfishes are unique in that they can autotomise >75% of their body mass and continue to survive. Presumably, multiple autotomy of tissue has energetic costs in terms of potential fitness and may affect the allocation of energy reserves accordingly. We investigated arm autotomy, predatory capabilities and subsequent regeneration in common starfish, Asterias rubens, that were induced to lose one, two or three arms. Initially, both regeneration of autotomised arms and the rate of growth of intact arms was slowest in animals that had lost the most arms (i.e. three arms missing vs two arms missing vs one arm missing). However, 8 months later, the growth of intact arms since the start of the experiment was not significantly different between groups of starfish that had autotomised different numbers of arms. However, the average dry weight per regnerating arm was significantly higher in starfish that had autotomised the most arms. Arm loss decreased the ability of starfish to open mussels and those that had autotomised two arms were significantly less likely to feed successfully on a mussel in a 24-h period than intact starfish. Our data suggest that proportionally more energy is allocated to arm regeneration in starfish that have suffered multiple arm loss and this may compensate a potential decrease in fitness that results from decreased feeding capability.     

Size-related aspects of arm damage, tissue mechanics, and autotomy in the starfish Asterias rubens

 Arm damage is a widely reported but superficially investigated aspect of the biology of the starfish Asterias rubens L. In the present study, the incidence of arm damage was surveyed in populations of A.  rubens at two sites in the Firth of Clyde, Scotland, and three sites in Gullmarsfjorden, Sweden. The mean (±SD) incidence across all sites of individuals with basal arm damage (resulting from detachment at the basal autotomy plane) was 19.69 ± 8.86%, the incidence of those with distal arm damage (resulting from amputation at more distal levels) was 7.74 ± 10.01%. The mean incidence of arms with basal damage was 5.28 ± 4.12%, of those with distal damage 1.83 ± 2.45%. There was a significant negative correlation between size and the incidence of basal damage at all but one site, but no significant correlation between size and distal damage at any site. Mechanical tests on specimens of the aboral body wall from the basal region of the arm (which included the autotomy plane) and from a more distal region revealed that with increasing body size there was a significant increase in yield stress, ultimate stress and Young's modulus (stiffness) but no significant change in yield strain and ultimate strain. There was no significant difference between the relationships for basal and distal specimens. It is hypothesised that in larger individuals increased mechanical toughness replaces autotomy as an effective antipredator strategy. Using two methods to induce autotomy, a significant positive correlation between size and the delay between the onset of stimulation and arm detachment was found; this may represent a size-related decline in the efficiency of the autotomy mechanism through the relaxation of selection pressure. Since size is an unreliable indicator of age in A.  rubens, the trends identified herein can be interpreted only tentatively as age-associated phenomena. Received: 4 September 1999 / Accepted: 17 February 2000     

Distribution pattern, reproductive biology, and fishery of the wedge sole Dicologlossa cuneata in the Gulf of Cadiz, south-west Spain

The wedge sole Dicologlossa cuneata (Moreau, 1881) is a coastal species that lives along the inner shelf waters of the Gulf of Cadiz down to a depth of 115 m. Its fishery, the size composition of the exploited population, and its reproduction have been studied on the basis of data acquired from commercial statistics from 1984 to 1994 and from sampling carried out during 1993 and 1994. The relative abundance of the species and its geographic and bathymetric distribution have been examined by analysis of four trawl surveys carried out from 1992 to 1994. D. cuneata is captured mainly by the trawl fleet, and to a lesser extent by gillnet gears. Wedge sole fishery has a marked seasonality. Size-frequency distributions from commercial samples and from bottom-trawl surveys are very similar, ranging from 12 to 26 cm total body length (TL). D. cuneata is a “repeat-spawner”. Peak spawning is from January to May. Females attain sexual maturity at 18 cm TL, while sexually mature males have been observed at 15 cm TL. The flexibility of the spawning strategy of D. cuneata throughout its distribution range is discussed. The scarcity of small specimens in the monthly samples and sequential spawning throughout the year prevented the estimation of growth parameters by means of length-frequency analysis. D. cuneata is present in greatest abundances in the sandy–muddy bottoms of the Guadalquivir River mouth at 15 to 30 m depth. Its optimum habitat coincides with the principal commercial fishing grounds. The possibility of reproductive vertical migration is discussed, but more precise data are needed before a viable hypothesis can be made. Received: 1 July 1997 / Accepted: 27 November 1997     

Discard analysis and damage to non-target species in the "rapido" trawl fishery

This paper describes the catch composition in the rapido trawl fishery and the direct effects on non-target species. All data were collected on commercial fishing vessels so as to reflect commercial rapido-trawling practice. The effects on non-target species were measured using two different damage scales (three- and seven-level scales) depending on the morphology of the taxa. Damage assessment was performed taking into account the whole fishing process by collecting individuals that passed through the cod-end, individuals that were retained in the cod-end and dropped onto the deck and individuals that were collected at the end of the sorting operation just before their return to the sea. Due to differences in the habitat and spatial distribution of target species, discard/commercial ratio was very different among the three different target species fisheries: 1:6 in the queen scallop (Aequipecten opercularis) fishery, 2:1 in the flatfish (Solea spp., Platichthys flesus, Psetta maximus and Scophthalmus rhombus) fishery and 9:1 in the scallop (Pecten jacobaeus) fishery. Damage sustained by non-target species was species-specific and related to the morphology of different organisms. The sorting operation produced similar levels of injury to those of the gear itself: all discarded animals showed higher levels of damage after the sorting than before. Damage to animals that had passed through the cod-end followed the same pattern, and these data could give an estimate of the "unobserved mortality". Our observations indicated a higher impact on non-target species caused by the queen scallop fishery than that caused by the flatfish fishery. This could be due to the total amount of hard-shelled species (in the queen scallop fishery, A. opercularis accounted for 87% of the total catch biomass) in any given haul, since shells macerated the catch during towing. Discarded animals from the queen scallop fishery showed higher levels of damage than those collected in the flatfish fishery. The rapido trawl fishery seemed to exert a strong selective pressure on the macrobenthic community, being able to modify the epibenthic fauna structure which, in heavily exploited fishing grounds, was dominated by bivalves, gastropods, crabs, starfish and brittlestars.     

Effect of arm regeneration on energy storage and gonad production in Ophiocoma echinata (Echinodermata: Ophiuroidea)

Two laboratory experiments were conducted on Ophiocoma echinata (Lamarck, 1816) collected from Missouri Key, Florida, USA, to ascertain the effects of arm regeneration on energy storage and gonad production. In each experiment individuals were divided into non-regenerating (all arms intact) and regenerating (three arms autotomized) groups and maintained together in a re-circulating seawater system for 2 mo. In Experiment 1, individuals were fed once a week, and in Experiment 2 once every 2 wk. The amount of food energy required for maintenance was estimated to be 0.172 kJ d⁻¹. The lower feeding level (Experiment 2) was below maintenance ration and both groups lost material relative to an initial group. At the higher feeding level (Experiment 1), regeneration of three arms resulted in less storage of organic material, mainly lipid in the stomach, and less gonad production. Thus, regeneration may reduce energy reserves and reproductive output. Received: 23 April 1998 / Accepted: 11 January 1999     

Incidental capture,direct mortality and delayed mortality of sea turtles in Australia's Northern Prawn Fishery

The species composition, catch and mortality rates of sea turtles captured incidentally by the tiger prawn fishery on Australia's northern coast in 1989 and 1990 were estimated by monitoring the fishery's catch. In 1990, the delayed rate of mortality from damage was estimated and the size composition was measured. Five species of turtles were captured: the flatback (Natator depressa, 59% of the total), loggerhead (Caretta caretta, 10%), olive ridley (Lepidochelys olivacea, 12%), green turtle (Chelonia mydas, 8%) and hawksbill (Eretmochelys imbricata, 5%). The turtle catches varied with water depth: the highest catch rates (0.068±0.006 turtles per trawl) were from trawls in water between 20 and 30 m deep, relatively few turtles (10%) were captured in water deeper than 40 m (25% of trawls). Catch rates varied with time of year: the highest catch rates were 0.098 (±0.013) turtles per trawl in winter. There was no significant difference in the overall catch rate (²= 0.047; p=0.8111; df=1) but a significant difference in mortality rate (²= 3.99; p<0.05; df=1) between the two years. The incidence of capture in the commercial fishery was 0.051 (±0.003) turtles per trawl towed for about 180 min, with 0.007 (±0.001) turtles per trawl drowning in the nets. There were no significant differences in the catch and mortality rates between the two years for any of the turtle species except the loggerhead, which had a significantly (² = 11.029; p=0.0013; df=1) lower catch rate in 1990 (0.002±0.001 turtles per trawl) than in 1989 (0.008±0.002 turtles per trawl), and a significantly higher mortality in 1990 (33%) than in 1989 (19%). Catch rates and mortality varied between the species: the flatback had the highest catch rate (0.030±0.002 turtles per trawl) but the lowest mortality (10.9%); the loggerhead had a catch rate of 0.005±0.001 turtles per trawl, and high mortality (21.9%); the olive ridley had a catch rate of 0.006±0.001 turtles per trawl and a low mortality (12.5%); the green turtle's catch rate was 0.004±0.001 per trawl and mortality 12.0%; the hawksbill had the lowest catch rate (0.002±0.001 turtles per trawl) but highest mortality (26.4%). Based on the fishing effort (27 049 d for 1989 and 25 746 d for 1990), we estimate that 5 503 (±424) turtles were caught and returned to the sea in 1989 and 5 238 (±404) in 1990, of which 567±140 drowned in 1989 and 943±187 in 1990. In 1990, an estimated 25% of all captured turtles suffered some non-lethal damage; an estimated 21% of turltes were captured comatose and 4% were injured. We conclude that, considering other threats, trawl-induced drowning is not the major impact on turtle populations in northern Australia, but that measures to reduce drowning and delayed mortality would be desirable.     

Validation of daily increment deposition in otoliths. Age and growth determination of Aphia minuta (Pisces: Gobiidae) from the northwest Mediterranean

The transparent goby Aphia minuta (Risso, 1810) is one of the main target species of the small-scale fishery off the Island of Majorca. Otolith microstructure and length-frequency analysis were used to study the age and growth of this species during the 1982/1983 and 1992/1993 fishing seasons. Daily periodicity of increment formation was determined by experiments with marked otoliths in individuals maintained in captivity. The length range of the catches during the 11 yr period was between 12 and 49 mm, with a main distribution (89%) between 24 and 40 mm. Otolith age-readings indicate that the population exploited in the commercial fishery consists of seven age-groups (2 to 8 mo old), with a very high proportion of individuals (95%) between 3 and 6 mo old. Population growth-curves revealed no differences between males and females. The growth parameters for the whole population are: asymptotic length, L _∞ = 53.69 mm; growth coefficient, K = 2.23 yr⁻¹; theoretical age at length zero, t ₀ = −0.005 yr. Those individuals of A. minuta caught in Majorca during the winter period reached a maximum age of 7 or 8 mo. Received: 30 December 1996 / Accepted: 16 April 1997     

Tiger shark (Galeocerdo cuvier) abundance and growth in a subtropical embayment: evidence from 7 years of standardized fishing effort

The tiger shark (Galeocerdo cuvier Peron and Lesueur 1822) is a widely distributed predator with a broad diet and the potential to affect marine community structure, yet information on local patterns of abundance for this species is lacking. Tiger shark catch data were gathered over 7 years of tag and release research fishing (1991–2000, 2002–2004) in Shark Bay, Western Australia (25°45′S, 113°44′E). Sharks were caught using drumlines deployed in six permanent zones (~3 km² in area). Fishing effort was standardized across days and months, and catch rates on hooks were expressed as the number of sharks caught h⁻¹. A total of 449 individual tiger sharks was captured; 29 were recaptured. Tiger shark catch rate showed seasonal periodicity, being higher during the warm season (Sep–May) than during the cold season (Jun–Aug), and was marked by inter-annual variability. The most striking feature of the catch data was a consistent pattern of slow, continuous variation within each year from a peak during the height of the warm season (February) to a trough in the cold season (July). Annual growth rates of recaptured individuals were generally consistent with estimates from other regions, but exceeded those for populations elsewhere for sharks >275 cm fork length (FL), perhaps because mature sharks in the study area rely heavily on large prey. The data suggest that (1) the threat of predation faced by animals consumed by tiger sharks fluctuates dramatically within and between years, and (2) efforts to monitor large shark abundance should be extensive enough to detect inter-annual variation and sufficiently intensive to account for intra-annual trends.     

Species composition, biomass and vertical distribution of micronekton over the mid-slope region off southern Tasmania, Australia

Seasonal sampling was carried out based on day/night, vertically stratified tows (100 or 125 m strata) in the upper 900 m of the water column over the mid-slope commercial fishing grounds south of Tasmania. A large midwater trawl (105 m² mouth area) was used with an opening/closing cod-end. Subtropical convergence and subtropical species dominated the fauna, but many less abundant, more widely-distributed species were also present. Fishes, which contributed 89% of micronekton biomass and 135 of 178 species, were dominated by the Myctophidae (48% biomass and 48 species). Twenty micronekton species made up 80% of the total biomass. Overall, the micronekton fish biomass in this region was 2.2 g m⁻² wet weight. A pronounced day/night shift in the distribution of biomass was attributable to diel migratory species. During the day, <0.2% of the total micronekton biomass was found in 0 to 300 m; most biomass was below 400 m, with peaks at 400 to 525 m and 775 to 900 m. At night, 53% of the biomass was found in 0 to 300 m, with progressively less in each deeper stratum. The vertical ranges of individual species typically exceeded 400 to 500 m during the day and night and were non-coincident, although nyctoepipelagic migrators were concentrated in the surface 200 m at night. Distinct epipelagic, lower and upper mesopelagic assemblages were identified, and patterns of epipelagic migration, limited migration and non-migration were categorised for species from each of the lower and upper mesopelagic assemblages. The vertical distribution of these assemblages was coincident with the primary water masses: subantarctic mode water (∼250 to 600 m) and antarctic intermediate water (below ∼700 m). The flux of migrating micronekton, estimated at 0.94 to 3.36 g C m⁻² yr⁻¹ to the lower mesopelagic and 1.14 to 4.06 g C m⁻² yr⁻¹ to the upper mesopelagic, appeared to be considerably outweighed by the consumption needs of aggregated mid-slope benthopelagic predators. We suggest that advection of mesopelagic prey in antarctic intermediate water may sustain aggregated populations of orange roughy (Hoplostethus atlanticus) and other predators on the micronekton in mid-slope depths at this site. Received: 2 April 1997 / Accepted: 21 August 1997     

Predation on Euphausia pacifica by demersal fishes: predation impact and influence of physical variability

The importance of Euphausia pacifica Hansen to the demersal fish community off Sendai Bay, northern Japan, was studied based on a total of 256 trawl samples collected during May, October and November, from 1989 to 1992. The samples were classified into 16 assemblages according to year, month and depth. The importance of E. pacifica to the total diet of each of the assemblages was measured using an index considering both fish diet and species composition. A total of 24 fish species out of 87 were found to ingest E. pacifica. Of these, Gadus macrocephaus and Theragra chalcogramma had the highest predation impact due to their dominance in the fish assemblages and the high proportion of E. pacifica in their diets. The contribution of E. pacifica to the total diet of the fish assemblage was highest in the shallow (≤300 m) regions during May, accounting for an average of 38.5% for the 4 years. However, during May 1990, when the warm Kuroshio Extension prevailed, the contribution was higher in the deep (>300 m; 22.6%) region, reflecting active and/or passive movements of E. pacifica. The average predation impact was maximal in the May/shallow assemblage (mean ± SE; 4.6 ± 1.4 kg wet wt km² d⁻¹) and was minimal in the November/shallow assemblage (0.4 ± 0.3). Annual estimates of consumption by demersal fish fishes ranged from 43 to 128 metric tons, representing 15 to 64% of the annual commercial catch of E. pacifica by local fisheries. It is therefore suggested that the fisheries on E. pacifica have the potential to considerably impact the demersal fish assemblages. Received: 3 December 1997 / Accepted: 10 June 1998     

Alarm signal response in the brittle star Amphiura filiformis

 Flat fishes, such as the dab Limanda limanda, commonly prey on arms of Amphiura filiformis. We demonstrate in flume experiments that A. filiformis showed a more or less simultaneous response to an up-stream predation by dab on conspecifics: the arms, which are stretched up into the water column when filter feeding, responded by bending down to the sediment surface, and some arms submerged, at least partly, into the sediment. We interpret this as an alarm response. A similar, but significantly weaker response in A. filiformis was also recorded when dab ate Amphiura chiajei. Homogenates of the two Amphiura species and arm pinching – to simulate partial predation – elicited a similar response in A. filiformis, the arms bent to the sediment surface. In contrast, however, the behavioural response of the arms to the homogenates was to protrude further out of the sediment and sweep the surface vigorously. We interpret this as a feeding behaviour. Response to arm pinching was less clear. The ecological consequences and origin of alarm signal response in A. filiformis are discussed. Received: 8 February 1999 / Accepted: 16 July 1999     

Sustainable fishing gear: the case of modified circle hooks in a Costa Rican longline fishery

Our research aims to identify longline fishing gear modifications that can improve fishing selectivity and reduce incidental capture of non-target species. Catch rates and anatomical hook locations (AHL) were compared when using a 14/0 standard ??control?? circle hook with a 0° offset and an experimental ??appendage?? hook in a Costa Rican longline fishery. With the appendage, the maximum dimension of the appendage hook was increased by 10% and the minimum dimension of the hook by 19%. A total of 1,811 marine animals were captured during five fishing trips. By taxonomic groups, sea turtles represented the largest total catch (27%), followed by sharks (26%), rays (25%), mahimahi (Coryphaena hippurus) (12%), and tunas and billfish (10%). Non-target and discard species, such as rays and sea turtles, accounted for over half of the total catch. Catch per unit effort (CPUE; number of individuals per 1,000 hooks) was higher with control hooks compared to appendage hooks for all species?? categories except rays; appendage hooks caught 52% fewer sea turtles and 23% fewer tunas and billfish than standard hooks, which represents a significant reduction in bycatch of endangered and other species. No differences were found in the AHL for sea turtles, suggesting use of the appendage may not incur additional advantages regarding turtles?? post-release survivorship. Despite lower catch rates for marketable species, such as sharks and mahimahi, use of the appendage resulted in dramatic reductions in catch rates of sea turtles. The results suggest that large scale adoption of hooks with a significantly wider hook dimension could be an effective conservation measure to maintain marine biodiversity while allowing for continued fishing.     

Abundance and arm loss of the infaunal brittlestar Ophiophragmus filograneus (Echinodermata: Ophiuroidea), with an experimental determination of regeneration rates in natural and planted seagrass beds

The regeneration rate of arms of the infaunal brittlestar Ophiophragmus filograneus was determined in natural and planted beds of the seagrass (Halodule wrighttii) in Tampa Bay, Florida, USA in 1990. Annually, brittlestar densities did not differ on average between the two beds, but they did show marked differences on individual dates. Rates of arm loss could not be determined for the two sites separately, but were estimated for the entire area. Between 52 and 94% of the brittlestars collected showed evidence of arm regeneration. Approximately one-third of those collected were regenerating one arm, with smaller proportions regenerating two or more arms. Regeneration rate was greater overall for individuals that had lost more tissue; however, the rate of soft-tissue regeneration per arm regenerated was significantly greater for those that had lost one arm compared to those that had lost three arms. The rate of soft-tissue regeneration did not differe between individuals from the two seagrass beds, and regeneration rate was lower than reported for other infaunal brittlestars. Skeletal regeneration per arm was similar in both arm-removal treatments. This result supports the hypothesis that echinoderm skeleton is energetically inexpensive and may help to lengthen arms during early regeneration. The rate of skeletal production was three-fold greater in individuals from the planted seagrass than in those from the natural. bed. As increased skeletal regeneration requires some additional energy, this indicates a greater energy expenditure and differential allocation by brittlestars regenerating in the planted seagrass areas.     

Biomasses,catch rates and abundances of demersal fishes,particularly predators of prawns,in a tropical bay in the Gulf of Carpentaria,Australia

The demersal fish fauna of Albatross Bay, in the eastern Gulf of Carpentaria, northern Australia, was sampled on seven cruises from August 1986 to November 1988, using a random stratified trawl survey. Four depth zones between 7 and 45 m were sampled during both day and night. The mean biomass of fish from all seven cruises was 297 kg ha^–1 for days trawls and 128 kg ha^–1 for night trawls. The overall mean catch rates were 922 kg h^–1 for day trawls and 412 kg h^–1 for night trawls. There were marked differences between cruises in both the biomass and catch rate. Approx 890 000 fish of 237 species were collected. Of these, 25 species comprised 82% of the total biomass and 74% of the overall catch rate. The dominant families were Leiognathidae, Haemulidae and Clupeidae, with Sciaenidae and Dasyatidae important at night.Leiognathus bindus was the most abundant species. Twenty-five species occurred in more than 50% of trawls, withCaranx bucculentus the most frequently caught (96% of all trawls). Thirty four species were predators on prawns; their absolute mean biomass was 50 kg ha^–1 during the day and 39 kg ha^–1 at night. The corresponding catch rates were 171 and 125 kg h^–1. Multiple-regression analyses were used to discriminate the effects of diel, seasonal, depth and cruise patterns. Of the 31 most abundant species, 15 showed diel patterns of abundance; 11 species showed seasonal patterns of abundance; 23 species had differential depth distribution; and 13 species showed significant cruise-to-cruise variation in abundance. Cruise variations in abundance were tested against salinity, temperature, tidal exchange, plankton biomass and prawn abundances as well as periods (and lags) of total rainfall prior to sampling. Only total rainfall showed any significant correlation. Total rainfall over a period of 6 wk immediately prior to sampling showed significant positive correlations with the abundances of five species, with overall daytime catch rates, and with the suite of 34 prawn predators. Rainfall and river runoff into Albatross Bay were significantly correlated. In Albatross Bay, the complex of factors affecting fish abundances and the magnitude of between-cruise differences indicate that such tropical communities may be unpredictable and are not seasonally constant. The high catch rates in Albatross Bay relative to similar tropical areas elsewhere are discussed and attributed to the light exploitation of the Albatross Bay stocks. Other than a prawn fishery, there is no commercial trawling in Albatross Bay. Hence, the only fishing mortality is a result of by-catch from prawn trawling. The annual total of such fish by-catch is probably less than 10% of the estimated standing stock of 93 000 tonnes.     

Possible biases in experiments evaluating the consumption of discards by seabirds in the North Sea

Discard utilization by seabirds in the North Sea has been quantified by so-called discard experiments on board fishery research vessels. In this paper we evaluate possible biases in this method, although discard practices vary considerably between commercial fishing fleets, seasons and areas. Whereas the total consumption of discards remained relatively unaffected by the different phases of the fish catch procedure, the length of fish discarded, the different methods of compiling the data and the quantities of fish discarded at once all significantly influence the results. The differences were particularly apparent between single- and multi-item trials in rapidly sinking fish. Correction factors for single-item experiments are suggested. Even more obvious were differences between the methods when focusing on the proportions obtained by the different bird species. Correcting the results of discard experiments on research vessels as presented in this paper appears necessary when quantifying the consumption of discards by seabirds in the North Sea. Received: 10 September 1997 / Accepted: 9 March 1998     

Born to lose. I. Measures of tissue loss and regeneration by the brittlestar Microphiopholis gracillima (Echinodermata: Ophiuroidea)

To measure amounts of tissue lost in natural populations of the burrowing amphiurid ophiuroid Microphiopholis gracillima (Stimpson), individuals were collected from subtidal mud flats in North Inlet, South Carolina, USA, at monthly intervals between February 1985–February 1987 and December 1989–November 1990. Between 20 and 70% of all individuals were regenerating the disc, and 85% of the 2045 arms examined had regeneration scars; >50% had one scar and some arms had up to 4 scars. Fewer individuals were regenerating discs in warmer months, but there was no seasonality in arm-loss frequency. To quantify rates of arm regeneration in the field, individuals which had 1, 2, or 3 arms removed were placed in mud-filled cores in the field in late July and November 1988 and in March and May 1989, and recovered after periods of about one month. Another set of cores was held in a running seawater laboratory during the May 1989 experiment. No regeneration occurred during the cooler times of year (November and March), and rates of regeneration were slower in May (total: 0.13 mg/d; tissue: 0.03 mg/d) than July (total: 0.17 mg/d; tissue: 0.05 mg/d). These rates indicate complete replacement of lost tissue in 100 to 120 d during the growth season. Within experiments, per arm regeneration rates were similar regardless of the number of arms removed. This finding is complicated by small sample size, high variability and low statistical power, but in general individuals which lost 2 or 3 arms regenerated proportionally more tissue than individuals which lost 1 arm. Individuals held in the laboratory regenerated the same amount of tissue but 30% less skeleton than individuals in the field. Sublethal tissue loss is common in this population, and M. gracillima is capable of regenerating at least 50% (each arm=17% of total body weight x 3) of its standing crop in a single growing season. Burrowing brittlestars probably constitute a significant renewable energy source for higher trophic levels in areas where they occur in dense populations.     

Redistribution of benefits but not detection in a fisheries bycatch‐reduction management initiative

Reducing the capture of small fish, discarded fish, and bycatch is a primary concern of fisheries managers who propose to maintain high yields, species diversity, and ecosystem functions. Modified fishing gear is one of the primary ways to reduce by‐catch and capture of small fish. The outcomes of gear modification may depend on c ompetition among fishers using other similar resources and other gears in the same fishing grounds and the subsequent adoption or abandonment of modified gears by fishers. We evaluated adoption of modified gear, catch size, catch per unit effort (CPUE), yield, and fisher incomes in a coral reef fishery in which a 3‐cm escape gap was introduced into traditional traps. There were 26.1 (SD 4.9) fishers who used the experimental landing sites and 228(SD 15.7) fishers who used the control landing sites annually over 7 years. The size of fish increased by 10.6% in the modified traps, but the catch of smaller fish increased by 11.2% among the other gears. There was no change in the overall CPUE, yields, or per area incomes; rather, yield benefits were redistributed in favor of the unmodified gears. For example, estimated incomes of fishers who adopted the modified traps remained unchanged but increased for net and spear fishers. Fishers using escape‐gap traps had a high proportion of income from larger fish, which may have led to a perception of benefits, high status, and no abandonment of the modified traps. The commensal rather than competitive outcome may explain the continued use of escape‐gap traps 3 years after their introduction. Trap fishers showed an interest in negotiating other management improvements, such as increased mesh sizes for nets, which could ultimately catalyze community‐level decisions and restrictions that could increase their profits.     

Assessment of the state and management of the cephalopod trawl fishery resources in Greek waters

The cephalopod trawl fishery in Greek waters is reviewed in relation to fishing effort for 1964–1981. The mean annual (1964–81) catch of cephalopods amounted to 1348 tons. Total landings of cephalopods are very well described by the exponential⁹ and linear²⁹ surplus yield models, according to which optimum fishing effort, maximum sustained yield and optimum catch per unit of fishing effort range between 120000–150000HP, 1586–1630 tons and 10–13kgr/HP respectively. In addition, the analysis revealed that cephalopod resources were slightly overfished (according to the linear model) during 1975–1981. Hence, authorities must take immediate measures for the protection and rational management of cephalopod resources in Greek waters. Possible alternative measures are discussed in the text.     

Trace Metal Studies on the Starfish Asterias rubens L. From the Western Baltic Sea

In 1984, 115 samples of the starfish Asterias rubens L. collected in the south-eastern part of Cadet Trench (Mecklenburg Bay/Western Baltic Sea) were analyzed for their contents of a few major (calcium, magnesium) and trace elements (cadmium, copper, iron, mercury, manganese, nickel, lead, selenium and zinc). Distinct differences were found between starfish from different stations, and these are attributed to the composition of the sediments acting as a substrate for their prey (mussels, snails). Except for cadmium, the concentrations of the elements studied all correlated negatively with the diameter and weight of the starfish.

Parallel analyses of starfish arms and the central discs of the same animals showed that iron, zinc, copper and mercury levels were 16 to 30% higher, and selenium, manganese, magnesium, calcium and lead were 4 to 9% higher in the arms. Cadmium concentrations were 20% higher in the central discs than in the arms.

Stock estimations (about 52,000 tons fresh weight) show that starfish play a significant role in the benthic ecosystem of the western Baltic Sea. They can consume up to 200,000 tons of mussels and clams (Mytilus edulis, Macoma baltica) a year and may therefore represent a notable factor in the trace metal balance of the region.     

Life history and fishery of Lepidopus caudatus (Pisces: Trichiuridae) in the Catalan Sea (Northwestern Mediterranean)

The frostfish Lepidopus caudatus (Euphrasen, 1788) is a mesopelagic species; it occurs along the shelf and slope down to 450 m in the Catalan Sea. Its' fishery, population structure, growth, diet and reproduction was studied on the basis of commercial statistics from 1984 to 1990 and on samplings made in 1988 and 1989. L. caudatus is fished by two types of commercial method, trawl and long-line; catches by the latter method display a marked seasonality. The minimum size of fish caught is 26 cm total length by trawling and 71 cm by long-line. Maximum sizes are similar for both gears: 196 and 188 cm, respectively. L. caudatus exhibits fast growth and attains an age of 8 yr. It feeds on small mesopelagic prey. It is a partial spawner reproductive activity occurring from April to November. Males attain sexual maturity at 97 cm and females at 111 cm total length.