Author: Kyle VanderLugt
Institution: University of Hawaii
Date: April 2005
Abstract
Copepod nauplii are marine zooplankton invertebrates that have been shown to be advantageous as a food source for larval finfish ornamentals when the larvae first begin to feed. Despite this potential, copepod nauplii have up not been reared in sufficient quantities to sustain the large-scale feeding of ornamental finfish in captivity until this point. The aim of this scientific investigation was to develop a protocol for the effective cultivation and collection of eggs from the calanoid copepod, Bestiolina similis, and to compare the rates of mortality and egg production in response to additions of various food sources and concentrations to optimize animal longevity and reproductive performance.
Bestiolina similis were fed three phytoplankton strains: Chaetoceros neogracille, Isochrysis galbana Tahitian strain, and Rhodomonas sp. at various concentrations in order to compare the effect of food addition on survival rates. It was found that the highest survival rates were achieved by supplying the copepods with Rhodomonas sp. at a concentration of 5.0 104 cells ml-1. B. similis were maintained in 2L cultures using the optimal concentrations for each of the three phytoplankton species as determined from the survival experiment. Copepods were isolated from each culture every 12 hours, and egg production was monitored for periods of 12 hours. In this experiment, copepods fed with Isochrysis galbana had the highest egg productions. The highest average egg production rate was 23 eggs female-1 day-1 with the maximum measured at 30 eggs female-1 day-1.
Introduction
To support the growing interest in the aquarium trade and to limit the fishing of open ocean coral reefs, scientific researchers have begun to focus on captive cultivation of finfish ornamentals at high density. Despite a report in 1998 by the Food and Agriculture Organization (FAO) estimating the total aquaculture production from finfish to be over 20 million metric tones, progress on high-density captive production of ornamental finfish has been thwarted by a lack of natural feeds, namely copepods (Lee, 2003). Copepods are a marine invertebrate, and their nauplii have been shown to be an important food source for finfish larvae in their natural coral reef habitats. Due to an insufficient quantity of copepods, researchers have tried, with little success, other less suitable foods such as barnacle nauplii, mysids, oyster and mussel larvae, ciliates, and nematodes (Ogle et al., 2002). Because researchers have been unable to match the high production rates in large-scale rotifer cultures while using copepods, the potential advantage of using copepods for large-scale food source has not yet been realized (Støttrup et al., 1986; Schipp et al., 1999).
Copepod nauplii have been shown to be a highly advantageous food source for larval fish due to a need to culture copepods and harvest their nauplii (McMichel and Peten, 1989; Støttrup and Norsker, 1997; Payne et al., 2001). When larval ornamentals first begin to feed, the larvae are small and consumption of a suitable food is critical for healthy development. Copepod nauplii are small enough for first feeding, and they offer a higher nutritional content to ornamental finfish larvae (Ostrowski and Laidley, 2001). They have a preponderance of phospholipids rather than triacylglycerols, as well as ratios of fatty acids that more closely approximate the natural diet of marine finfish larvae (McKinnon et al., 2003; Watanabe et al., 1983; Kraul et al. 2002). The unique swimming movement of calanoid copepods, described as a cycloid path (Kittredge et al., 1974), further makes nauplii an attractive food source, as they induce a positive feeding response in finfish (Kuhlmann et al., 1983; Doi et al., 1997).
Recognition of the potential advantage in the mass rearing of copepods has led researchers to investigate the conditions required to improve large scale culturing systems. Experiments are being conducted to determine the biological parameters that maximize survival rates and reproductive success (Payne and Rippingale, 2001; Schipp et al., 1999; Ogle et al., 2002). One culturing method developed by Payne and Rippingale featured a five hundred liter culture with a semi-continuous recirculating system (Payne et al., 2001). With this system, they were able to standardize nauplius production, collecting approximately 5105 nauplii culture-1 vessel-1 day-1 from the calanoid copepod Gladioferens imparipes (Payne and Rippingale, 2001). However, each copepod species requires different culturing conditions, making it necessary to first establish the optimal biological parameters before designing an automated culturing system for another species.
The aim of this scientific investigation was to develop a protocol for the effective cultivation and collection of eggs from the calanoid copepod Bestiolina similis. The primary objective was to compare the rates of mortality and egg production in response to additions of various food sources, food concentration, and water quality to determine the biological parameters that optimize animal longevity and reproductive performance.
B. similis has demonstrated high potential as a food source because of their small size. Calanoids that are less then 1 mm frequently produce nauplii that are less then 100 μm in the first and second stages of development. The initial stages of nauplius development are believed to be critical because ornamentals prefer a food size of 50 to 100 μm (Ostrowski and Laidley, 2001). B. similis has an excellent nutritional profile, and it is considered a good candidate for culturing because of the low cannibalistic predation of the adult on nauplii (McKinnon et al., 2003). Bestiolina similis isolated from coastal waters off Queensland, Australia can produce up to 48 eggs female-1day-1 in the presence of large algal cells of Heterocapsa niei Although it has been shown that Bestiolina is the best candidate for larval fish feed, additional studies on optimizing culturing conditions are needed (McKinnon et al., 2003). Furthermore, the Hawaiian population is genetically isolated from the Australian one, and differences in life history can be expected. Thus variances in survivorship and egg production are expected to occur in response to feeding strategies between the Hawaiian and Australian strains. Results from this project on an indigenous species will contribute to the development of live feeds, which will ultimately permit large-scale cultivation of finfish in captivity. Successful rearing of ornamental finfish will benefit the growing aquarium industry and reduce or even eliminate the large-scale harvesting of finfish from coral reefs.
Materials & Methods
Maintenance of Phytoplankton Cultures
Three phytoplankton species were used as live food for copepods; Chaetoceros neogracille, a diatom (apical axis 10 - 15 µm); Rhodomonas sp., a planktonic marine flagellate (diameter 5 - 10 µm) and Isochrysis galbana Tahitian strain, a dinoflagellate (diameter 5 - 6 µm).
The seawater in which the phytoplankton was maintained was collected from Kaneohe Bay in 20 L Nalgene containers each day of sampling copepods. It was slot glass fibre size C (GF/C) (diameter 42.5 mm) and sterilized by microwaving for 8 min (1.5 L) at 1100 Watts. To promote algae growth, the nutrients (Micro Algae Grow®, Florida Aqua Farms Inc.) were added at a concentration of 0.5 mL/L (F/2 media). Because Chaetoceros neogracille is a diatom and thus requires sodium silicate for shell growth, 200 µL/L sodium-silicate was added in addition to Algae Grow®. Chaetoceros neogracille and Isochrysis galbana were on incandescent 12:12 light cycles and aerated with minimal bubble agitation. Rhodomonas sp. was maintained in ambient conditions with a 15:9 light-dark cycle without aeration.
Phytoplankton cell densities were determined periodically by counting cells with a hemacytometer under an inverted compound microscope (25x) (Olympus, Japan). To maintain cultures in active growth phase, 90% of the Rhodomonas and Chaetoceros and 65% of the Isochrysis culturing volume were discarded every 8 to 10 days. The remaining volume was used to inoculate new cultures using the methods previously described. The average phytoplankton densities in the stock cultures were Chaetoceros neogracille (2.6 ± 0.3) 106 cells ml-1, Isochrysis galbana Tahitian strain (2.9 ± 0.50) 106 cells ml-1, and Rhodomonas sp. (.54 ± .20) 106 cells ml-1.
Copepod Collection
The calanoid copepod Bestiolina similis was collected from Kaneohe Bay, Oahu. Fully mature adults grow up to approximately 600-700 μm (total length); thus for their capture, a planktonic net with a mesh size of 150 μm was used for trawls. Since Bestiolina is positively phototactic, we obtained high catches with subsurface tows (50 to 100 cm depth) of 1 to 2 min.
Copepods were concentrated in a 1L cod end jar immediately transferred into an 18L bucket filled with 10 to 15L of ambient seawater. This dilution step was executed to reduce the stress of a new environment as previous research has shown copepods to be sensitive to sudden changes in their surrounding condition (Harris et al., 2000). The dilution also helps to alleviate the rapid warming effects of the sun. After transferring the copepods to the laboratory, 3 liters of the sample were added to 3.6L of GF/C filtered seawater. The copepods were stored at room temperature (~25°C) throughout the duration of the experiments.
Measuring Survival Rates
In the first experiment, testing the effect of algal food additions on copepod survival was completed with three algal species (Chaetoceros, Isochrysis, and Rhodomonas) run at six concentrations (Table 1). Triplicates for each algal concentration were run for two water conditions: filtered and non-filtered sea water (108 tests total). Tests were run in 20 ml glass vials. Prior to dilution of the test concentrations, the densities of the phytoplankton stock cultures were determined by duplicate counts with a hemacytometer.
Figure 1. Survival rates for Bestiolina simlis in response to varying volumes of A) Chaetoceros neogracille B) Isochrysis galbana and C) Rhodomonas sp. food source additions. Additions of Rhodomonas (C) show the highest survivorship with approximately 50% of the animals remaining after 5 days
Ten Bestiolina similis were sorted under a dissecting microscope and placed into each of the 108 vials. The vials were gently aerated for 3 minutes each day and stored at room temperature for the duration of the experiment. To eliminate the need for recounting and to prevent fouling, the vials were checked daily, and dead copepods were counted and removed. Survival rates were back calculated using the daily mortality counts.
In a second experiment, Rhodomonas was added as described in the methods for experiment 1. Each volume addition of Chaetoceros and Isochrysis was diluted so as to match the corresponding concentration of Rhodomonas. Constant phytoplankton densities were maintained throughout experiment 2. Phytoplankton densities in the experimental vials were checked daily and adjusted to treatment levels through the addition of phytoplankton from stock cultures. In this experiment, the percentage of copepods remaining after one week was determined. The daily survival rates as calculated in the previous experiment were not determined.
Nutrient Chemistry
Nutrient analyses were conducted on water samples taken from the vials following experiment termination (≤10% survival in the individual vials). Ammonia concentrations were measured colorimetrically with an indolphenol based test kit (Palintest). In the presence of chlorine, ammonia reacts with alkaline salicylate to form a green-blue indophenol complex which was read with a Yellow Springs Instrument (YSI) photometer at 640 nm.
The nitrite levels were determined using a colorimetric procedure with an iodide reagent system. Under mildly acidic conditions, nitrites catalyze the oxidation of iodide to iodine, producing a brown color. The color produced is proportional to the nitrite concentration measured with an YSI photometer at 490 nm.
Cultivation of copepods for egg production experiments
Two days prior to the collection and isolation of copepods, eight 2L plastic Nalgene® containers were filled with 1.5L of GF/C filtered seawater and sealed with Parafilm® to prevent water evaporation. The water was also gently aerated with use of bubblers (ceramic type medium pore; 37:12:12 mm). Duplicate cultures were set up for each of the three algal species (Chaetoceros, Isochrysis, and Rhodomonas) and cultures containing no added food were used as controls. The phytoplankton concentrations maintained in the 2L cultures and used in the quantification of egg production were determined from the concentrations that optimized survival rates. Chaetoceros was maintained at 2.4 105 cells ml-1, Isochrysis was maintained at 2.8 105 cells ml-1, and Rhodomonas was maintained at 5.0 104 cells ml-1. Throughout the experiment hemacytometer counts were taken daily and used to calculate the phytoplankton concentration (cells/mL). The desired phytoplankton concentrations were maintained by adding the appropriate volume of phytoplankton from stock cultures. 150 freshly collected adult Bestiolina were isolated and placed into each of the eight cultures, thus creating an initial copepod:volume ratio of 1 copepod per 10 ml. The cultures were incubated at room temperature for 24 hours prior to measuring egg production rates.
Measuring Egg Production Rates
Figure 2. Survival rates for Bestiolina similis in response to optimal phytoplankton concentrations (figure 1) for Chaetoceros, Isochrysis, and Rhodomonas. Vertical Bars represent standard deviation. Rhodomanas is shown to be a superior feed for achieving high survival rates in comparison to optimal concentrations of Chaetoceros and Isochrysis
Four copepods were isolated from each of the 8 2L cultures 24 hours after the first phytoplankton food addition and placed in 10 mL Petri dishes previously filled with 5 ml of GF/C filtered seawater and no food. At 12-hour intervals, 4 individuals from each of the 8 cultures were isolated for 32 additional copepods over a total of 72 hours. The eggs were counted and removed from each petri dish at periods of 12 and 24 hours. The total number of eggs produced by the females that were fed with the same food source was taken as an average rate of #eggs female-1day-1.
Results
Survival Rates of Bestiolina similis - Effect of Algal Food
In all control group trials (no added food), there was high survival (> 80%) in the first 24 hours followed by a rapid decline in survival with > 90% mortality by the third day (diamonds, Figure 1). We obtained similar results for the survival rates of copepods in filtered and non-filtered seawater, suggesting that ambient food levels in surface waters were rapidly exhausted and thus were not sufficient to increase survival rates. However, greater variability in survival was observed in the animals maintained in bay water in comparison to GF/C filtered seawater. Survival rates as a function of Chaetoceros, Isochrysis, and Rhodomonas algal food sources are presented in Figure 1. All three phytoplankton species increased survival by at least one day in comparison to the controls.
There was no significant difference in the survival curves at different food concentrations for Chaetoceros neogracille. Most survived to the second day followed by a rapid decline (< 10% survival) by day 5 for all food additions (Figure 1A). At an algal density of 2.4105 cells ml-1, 50% of the copepods survived to approximately 3.2 days.
Figure 3. Average number of of Bestiolina similis alive after 7 days of being maintained in varying concentrations of Chaetoceros, Isochrysis, and Rhodomonas. Food concentrations and water volumes were kept constant in this experiment. Vertical Bars represent standard deviation.
In response to being supplied with Isochrysis galbana Tahitian strain, the copepods that were fed with the three highest additions (750, 1000, and 1500 μl), had the highest survival rates. However, 50% of the copepods survived to approximately 3.8 days despite the volume of the three additions (Figure 1B). Bestiolina similis that were fed with Rhodomonas sp also had the highest survival rates in response to the three highest food additions. In comparison to Chaetoceros and Isochrysis, however, Rhodomonas clearly yielded the best survival rates (Figures 1, 2). These were achieved by copepods fed at a phytoplankton concentration of 5.0 104 cells ml-1. At this feeding level, 50% of the copepods survived to 5.8 days (Figure 1C).
To verify that the mortality rates were not the result of a starvation effect due to a decline in ambient phytoplankton densities in the experimental trials, an experiment (experiment 2 of trials) was conducted in which all three phytoplankton species were added in equal quantities and algal concentrations were maintained thereafter throughout the experiment. In order to maintain equivalent phytoplankton concentrations, daily hemacytometer counts were taken, and microalgae from stock cultures were added to the 20-mL vials as needed. Survival rates after 7 days in the Chaetoceros and Isochrysis treatments were less than 10% for all algal densities and were therefore not significant (Figure 3). Rhodomonas sp. once again yielded the highest survival, with best survival observed at a food concentration of
4.6 104 cells ml-1 (Figure 3).
Figure 4. Average egg production rates in response to additions of A) no food B) Chaetoceros neogracille C) Isochrysis galbana Tahitian strain and D) Rhodomonas sp. as phytoplankton food source. Vertical bars represent standard deviation.
Nutrient Concentrations
The highest nitrite concentration found in the experimental vials at the termination of the experiment was .03 μM; the highest ammonia concentration obtained was 7 μM.
Egg Production Rates
Because of egg production depressions in the second 12 hours of the egg production experiments, only the eggs produced in the first 12 hours were used to determine daily egg production rates. The results for egg production for Bestiolina similis are shown in Figures 4 and 5. Figures 4A-D show the number of eggs produced in the initial 12 hours following both evening and morning isolations in response to no added food and to the three phytoplankton additions. Egg production rates differed with food source. In the controls, egg production was very low and ceased at day 3. This agrees with the low survival observed in the controls and suggests that Bestiolina similis in Kaneohe Bay had low energy reserves. Egg production was enhanced in all treatments with food additions and egg production rates remained fairly constant over the 3 days. Egg production varied with diet and was typically higher at night than during the day for Chaetoceros (Figure 4B) and Isochrysis (Figure 4C). At night, Bestiolina consistently produces an average of 30% and 45% more eggs in Chaetoceros and Isochrysis, respectively. Bestiolina similis fed on Rhodomonas, which gave the best survival rates, had low reproductive rates (Figures 4D and 5).
Average daily egg production rates determined from the initial 12-hour period following evening isolation in addition to the 12 hours following morning isolation are shown in Figure 5. Highest egg production rates of 23.2 eggs fem-1 d-1 (maximum of 30 eggs fem-1 d-1) were obtained from the copepods that were fed with Isochrysis, with the second highest average egg productions of 17.5 eggs fem-1 d-1 observed in Bestiolina similis fed on Chaetoceros.
Discussion
Figure 5. Egg production rates (eggs female-1 day-1) for Bestiolina similis in response to different phytoplankton. Initial concentrations in the cultures were maintained throughout the experiment. The highest egg rate was 23.3 and was obtained from the copepods fed on Isochrysis. Vertical bars represent standard deviation.
The focus of this project was to compare the rates of mortality and egg production in response to various phytoplankton food source additions (i.e. Isochrysis, Rhodomonas, and Chaetoceros) in order to determine the biological parameters that optimize animal longevity and reproductive performance. As part of a larger project, these data will be used to optimize conditions for Bestiolina similis for large scale and continuous production of nauplii.
Similar to other small tropical and sub-tropical calanoid copepods, Bestiolina similis was characterized by short survival times under conditions of no and low food. These animals appear to maintain few reserves, and thus require constant feeding. For large scale cultivation, it means that phytoplankton densities in the cultures need to be checked regularly and adjusted to maintain optimal conditions.
Phytoplankton additions improved survival rates, with best survival observed at phytoplankton concentrations between 104 to 105 cells ml-1. Optimal survival was achieved with the addition of Rhodomonas sp. at a concentration of 3 to 5 104 cells ml-1. Although water quality may have contributed to mortality rates in these experiments, nitrite and ammonia concentrations were well within the tolerance level of Bestiolina similis (Clauberg, 2004). It is likely that the small size of the culture vessels may have caused a significant increase in mortality.
Although best adult survival was achieved with Rhodomonas, egg production rates were very low on this food. This is in contrast to results from McKinnon et al. (2003) that shoed highest egg production rates on ad libitum diets of Rhodomonas (25 eggs fem-1 d-1) and Heterocapsa niei (48 eggs fem-1 d-1). We obtained good reproductive output on Isochrysis and Chaetoceros diets with egg production rates of 23 and 17 eggs fem-1 d-1, respectively. These rates exceeded those reported by McKinnon et al. (2003) on these phytoplankton species (Isochrysis: 5 eggs fem-1 d-1; Chaetoceros: 10 eggs fem-1 d-1). Although there may have differences in the algal strains used in these studies, it may also reflect differences between individuals of Bestiolina similis from 2 very different populations (Hawaii vs. Australia). This emphasizes the need for determining optimal culturing conditions for each combination of copepod and algal strains being considered for a particular aquaculture operation
Egg production rates were 4-fold higher than the 3 to 6 eggs fem-1 d-1 reported by Clauberg (2004) for Bestiolina similis fed on Isochrysis galbana in long term cultures. This difference may be primarily due to differences in feeding conditions. Food levels in the experiments were 5 times higher than in Clauberg's culture (5 104 vs. 1 104 cell ml-1).
Table 1. Twelve 20-ml glass vials each containing 10 Bestiolina similes and varying food volumes with added filtered (GF/C) and non-filtered seawater
In an efficient culturing system, it would be necessary to optimize both survival and reproductive yields. According to this study, the survival rates were optimized with Rhodomonas sp. and the egg production was optimized with Isochrysis galbana. This is consistent with results from Clauberg (2004), which found that the best long term cultivation of B. similis from Kaneohe Bay was obtained on Isochrysis galbana and Rhodomonas sp. with copepod densities reaching 0.5 to 1 per ml with approximately 50% of the population as nauplii (Clauberg, 2004). However, better results might be obtained by designing a mixed diet for Bestiolina. McKinnon et al. (2003) optimized culturing conditions by using mixed phytoplankton with a density ratio of 2:1:1:1 for Isochrysis /Rhodomonas /Tetraselmis /Heterocapsa, respectively. Such an approach may further improve long-term cultivation of B. similis.
Information from this study can be used to develop an improved automated copepod culturing systems. Future challenges include the development of a system that allows the recirculation of water to improve water quality and an efficient method for separating eggs and nauplii from the culture while maintaining a healthy and reproductive adult population. At densities of 0.25 to 0.5 adults ml-1 (Clauberg, 2004) and maximum reproductive rates of 20 eggs fem-1 d-1, a 10 L tank of Bestiolina similis could produce 25000 to 50000 eggs d-1 assuming an adult sex ratio of 1:1.
Acknowledgments
Support for this work was provided through NSF grant 0243600 and the University of Hawaii Sea Grant College Program. I would like to acknowledge Dr. Petra Lenz, Dr. Michael Cooney, and Ben Clauberg for their advice and expertise in this investigation. Phytoplankton cultures were kindly provided by Brian Nedved and Piera Sun. I also would like to thank Hawaii Institute of Marine Biology for the use of their boats and facilities for the copepod collections.
References
Clauberg B. (2004). Biology of short term copepod egg preservation and long term cultivation of the copepod Bestiolina similis. Report completed for credit, Fachhochschule Mannheim University of Applied Sciences: 1-33.
Doi M, et al. (1997). Preliminary investigation of feeding performance of larvae of early red-spotted grouper, Epinephelus coioides, reared with mixed zooplankton. Hydrobiology. 259-358.
Harris RP, et al. (2000). Zooplankton Methodology Manual. Academic Press:76-8.
Hernández Molejón OG and L Alvarez-Lajonchére. (2003). Culture experiments with Oithona oculata Farran, 1913 (Copepoda: Cyclopoida), and its advantages as food source for marine fish larvae. Aquaculture. 219:471-483.
Kittredge JS, et al. (1974). Chemical signals in the sea: marine allelochemicals and evolition. Fishery Bulletin, National Oceanic and Atmospheric Administration of the United States. 72:1-11.
Kraul S, et al. (1992). Evaluation of live feeds for larval and postlarval mahimahi Coryphaena hipparus. J World Aquaculture. 23:299-307.
Kuhlmann D, et al. (1983). Rearing of turbot larvae (Scophthalmus maximus L.) on cultured food organisms and postmetamophosis growth on natural and artificial food. Aquaculture. 23:183-96.
McKinnon AD, et al. (2003). The potential of tropical paracalanoid copepods as live feeds in aquaculture. Aquaculture. 223:89-106.
Lee CS. (2003). Biotechnology advances in finfish hatchery production: a review. Aquaculture 227:439-58.
Mauchline J. (1998). Advances in marine biology,biology of calanoid copepods. Acedemic press, San Diego:5-349.
McMichel RH and KM Peten. (1989). Early life history of spotted seatrout, Cynoscion nebulosus (Pisces: Sciaenidae), in Tampa Bay, Florida. Estuaries. 12(2):98-110.
Nellen W. (1985). Live animal food for larval rearing in aquaculture. Non-Artemia organisms. In: Bilio, H. Rosenthal and C.J Sinderman (eds.). Realism in Aquaculture. Proceedings of the World Aquaculture Society, Venice 1981.
Ogle JT, et al. (2000). Culture of red snapper. Global Aquaculture. 3(5):23-7.
Ogle JT, et al. (2002). Culture of the copepod Acartia tonsa utilizing various artificial feeds. Gulf and Caribbean Fisheries Institute. 53:234-40.
Ostrowski AC and CQ Laidley. (2001). Application of marine foodfish techniques in marine ornamental aquaculture: Reproduction and larval first feeding. Aquarium Sciences and Conservation. 3:191-294.
Payne MF and RJ Rippingale. (2000). Evaluation of diets for culture of the calanoid copepod Gladioferens imparipes. Aquaculture. 187:85-96.
Payne MF et al. (2001). Cultured copepods as food for West Australian dhufish (Glaucosoma hebraicum) and pink snapper (Pagrus auratus) larvae. Aquaculture. 194:137-50.
Payne MF and RJ Rippingale. (2001). Effects of salinity, cold storage and enrichment on the calanoid copepod Gladioferens imparipes. Aquaculture. 201:251-62.
Payne MF and RJ Rippingale. (2001). Intensive cultivation of the calanoid copepod Gladioferens imparipes Aquaculture. 201:329-42.
Schipp GR, et al. (1999). A method for hatchery culture of tropical calanoid copepods, Acartia spp. Aquaculture. 174:81-8.
Strøttrup JG, et al. (1986). The cultivation of Acartia tonsa Dana for use as a live food source for marine fish larvae. Aquaculture. 52:87-96.
Strøttrup JG and NH Norsker. (1997). Production and use of copepods in marine fish larviculture. Aquaculture. 155:231-47.
Sukgeun J and E Houde. (2003). Spatial and Temporal variabilities of pelagic fish community structure and distribution in Chesapeake Bay, USA. Estuarine Coastal and Shelf Science, 58(2):335-51.
Watanabe T, et al. (1983). Nutritional values of live organisms used in Japan for mass propagation of fish: a review. Aquaculture. 34:115-43.
Watanabe T. (1993). Importance of docosahexanoic acid in marine larval fish. J World Aquaculture Soc. 24:152-61.