The nutritional ecology of the ctenophore Bolinopsis vitrea: comparisons with Mnemiopsis mccradyi from the same region

Bolinopsis vitrea is a warm water lobate ctenophore which doesnot overlap in its distribution with Mnemiopsis mccradyi incontiguous waters. We examined its feeding ecology on a seriesof cruises. B. virrea ingested increasingly more prey at higherfood concentrations (2–100 prey l^–1) but feedingeffort (clearance rate) decreased with increasing food availability.On a dry weight basis, smaller tentaculate Bolinopsis ingestedseveral times more than larger lobates, but based on carbonweight, specific ingestion was fairly uniform over the entiresize range investigated (6–60 mm total length). Bolinopsiscollected during the daytime in the Bahamas rarely had morethan three prey items in their guts. These results and laboratorymeasurements of digestion times (av. = 1.9 h) allowed computationof daily rations, which could not account for the metabolicrequirement as measured on the same cruises. Results of feedingexperiments, however, implied that prey densities in excessof 11^–1 were sufficient to sustain a growing populationof Bolinopsis. Prey concentrations about an order of magnitudehigher were required for M. mccradyi based on similar experiments.These results were in general agreement with observed densitiesand distributions of ctenophores and their zooplankton preyin the Bahamas and coastal South Florida.     

Vertical distribution, life cycle and production of the chaetognath Sagitta crassa in Tokyo Bay, Japan

The daytime vertical distribution of Sagiita crassa in TokyoBay was examined from February 13, 1988 to February 20, 1989.High densities of larger-size chaetognaths were found near thesea bottom, whereas the smaller animals tended to inhabit theupper layers. This feature of distribution is discussed in relationto the distribution of their main food organisms, e g. Pseudodiaptomusmarinus, Acartia omoru, Centropages abdomialis and Oithona davisae.The two periods of replacement of two morphs were confirmedby the variation only in mean body length of this chaetognath,unlike the previous authors who made additional morphologicalobservations. It was hypothesized that S.crassa has at leastfive generations Two generations, including mostly the largerforms, had higher growth rates than the generation consistingmainly of the small form. Yearly respiration of S.crassa was8.2 g C m^–Abstract. Yearly production of this animal wasestimated to be 3.8 g C m^–. A feeding estimate revealedthat chaetognaths require a prey production of 13.1 g C m^–year^–1. The impact of this chaetognath on the prey populationin Tokyo Bay and the propriety of an estimated value of annualproduction of S crassa is discussed.     

Vertical distribution of pelagic chaetognaths and feeding of Sagitta enflata in the Central Equatorial Pacific

The vertical distribution and feeding of pelagic chaetognathsat 5°S, 160°W in the Central Equatorial Pacific wereinvestigated using a series of 0–500 m vertical haulswith a VMPS net over a 24 h period between 6 and 7 October 1990.The total number of individuals per haul was between 370 and688. Fourteen species in four genera were found at this station.The most abundant species was Sagitta enflata which comprised32.4–61.1% of the individuals collected from the 0–500m layer. Mesopelagic species made up 9.3–15.1% of thetotal number of individuals. Sagitta enflata and Pterosagittadraco were found in the upper part of the thermocline both byday and at night. The fraction of the population containingfood items (FCF) of S.enflata in the 0–50 m layer variedbetween 4.8 and 12.5% (mean 10.8%) and feeding activity washighest between sunrise and noon. The percentages of Copepoda,Foraminifera, crustacean larvae, Chaetognatha, Pteropoda, Ostracoda,fish and unidentified material in the gut of S.enflata were51.9,6.7,3.8,2.9,1.9,1.9 and 30.9%, respectively. Sagitta enflataconsumed food organisms which were mainly between 0.5 and 1.0mm in length. The daily feeding rate of S.enflata was 1.81 preyper individual, which was equivalent to 8.06 mg C m^–2day^–1. This corresponded to     

Trophodynamics of the scyphomedusae Aurelia aurita. Predation rate in relation to abundance, size and type of prey organism

Ephyra larvae and small medusae (1.7–95 mm diameter, 0.01–350mg ash-free dry wt, AFDW) of the scyphozoan jellyfish Aureliaaurita were used in predation experiments with phytoplankton(the flagellate Isochrysis galbana, 4 µm diameter, {smalltilde}6 x 10^–6 µg AFDW cell^–1), ciliates (theoligotrich Strombidium sulcatum, 28 µm diameter, {smalltilde}2 x 10^–3 µg AFDW), rotifers (Synchaeta sp.,0.5 µg AFDW individual^–1) and mixed zooplankton(mainly copepods and cladocerans, 2.1–3.1 µg AFDWindividual^–1). Phytoplankton in natural concentrations(50–200 µg C I^–1) were not utilized by largemedusae (44–95 mm diameter). Ciliates in concentrationsfrom 0.5 to 50 individuals ml^"1 were consumed by ephyra larvaeand small medusae (3–14 mm diameter) at a maximum predationrate of 171 prey day^–1, corresponding to a daily rationof 0.42%. The rotifer Synchaeta sp., offered in concentrationsof 100–600 prey I^–1, resulted in daily rations ofephyra larvae (2–5 mm diameter) between 1 and 13%. Mixedzooplankton allowed the highest daily rations, usually in therange 5–40%. Large medusae (>45 mm diameter) consumedbetween 2000 and 3500 prey organisms day^"1 in prey concentrationsexceeding 100 I^–1. Predation rate and daily ration werepositively correlated with prey abundance. Seen over a broadsize spectrum, the daily ration decreased with increased medusasize. The daily rations observed in high abundance of mixedzooplankton suggest a potential ‘scope for growth’that exceeds the growth rate observed in field populations,and this, in turn, suggests that the natural populations areusually food limited. The predicted predation rate at averageprey concentrations that are characteristic of neritic environmentscannot explain the maximum growth rates observed in field populations.It is therefore suggested that exploitation of patches of preyin high abundance is an important component in the trophodynamicsof this species. ¹Present address: University of Bergen, Department of MarineBiology, N-5065 Blomsterdalen, Norway     

Daily feeding rhythm and rations of Pacific salmon of the genus Oncorhynchus in the period of marine prespawning migrations

The daily feeding rhythm and rations of the humpback salmon Oncorhynchus gorbuscha, the sockeyed salmon O. nerka, and the chum salmon O. keta during marine prespawning migrations is investigated with consideration of materials collected at daily stations in waters off eastern Kamchatka in June–July 1999 and 2001 (from catches of drift nets). The bulk of humpback salmon and sockeyed salmon food consists of euphausiids, hyperiids, large copepods, pteropods, and fish juveniles. In the food of chum salmon, pteropods dominated. In a 24-h period, salmon manifest a pronounced evening peak of stomach fullness, while at night feeding discontinues. Samples collected in the morning consisted of fish who had just started feeding after the night pause. In addition to nightly period of rest, all species manifested a daytime decrease in foraging activity, though less pronounced than in the night. The daily rhythm of Pacific salmons’s feeding depends on the vertical migrations of their food items (representatives of sound-scattering layers). During the marine feeding period, the most intensive feeding is recorded in the humpback salmon and chum salmon. The daily ration of the humpback salmon is lower than that of the chum salmon but includes animals of higher food value. Due to a high digestion rate in combination with a large stomach volume, the chum salmon can consume a large quantity of low-calorific food in a short time. The daily ration of the sockeyed salmon is considerably lower than that of other salmon species.     

Diel periodicity of Subantarctic copepods: relationships between vertical migration, gut fullness and gut evacuation rate

A zooplankton community in the Polar Frontal Zone north of SouthGeorgia was sampled for 5 days in February 1994. Feeding ofvarious copepodite stages of six copepod species was assessedby a series of gut fluorescence/gut evacuation experiments.Feeding periodicity was compared to vertical distribution andmigration patterns, as revealed by Longhurst Hardy PlanktonRecorder (LHPR) and ring net catches. Despite chlorophyll alevels reaching only 0.8 mg m^–3 and daily carbon rationsbased on phytoplankton intake being low, feeding of all thecopepods was restricted mainly to the 8 h night-time perid.During the daytime, the epipelagic community was verticallydispersed within the top 100 m. At night, upward migration bymost species led to a convergence of almost all zooplanktersin the upper half of the surface mixed layer. However, large-amplitudevertical migrations which crossed the thermocline were performedonly by Metridia lucens and Pleuromamma robusta. Although feedingby both migrants and non-migrants was mainly at night, therewas no diel signal in gut evacuation rate. The gut evacuationrates of the nine copepod species and stages differed significantly(5-fold) and were negatively related to the extent of theirdiel vertical migration. The long gut-passage times of the migratoryspecies, M.lucens and P.robusta, would have allowed them timeto defaecate some of the food eaten near the surface at depth,contributing to an active carbon transfer out of the mixed layer.However, their scarcity at this site meant that their grazingcomprised only -<1.4% of total copepod ingestion, so theircombined gut flux is likely to have been negligible (-0.4 mgcarbon day^–1).     

The influence of food concentration and feeding rate on the gut residence time of Daphnia

The gut residence time of Daphnia pulicaria feeding on threedifferent concentrations of Cryptomonas erosa was measured byobserving the passage of small plastic beads through the intestine.Passage times ranged from 4 to 106 min (mean, 27 min), theydecreased with increasing food concentration, and were negativelycorrelated with rates of mandible movement.     

Gut evacuation of walleye pollock larvae in response to feeding conditions

Gut residence times of first-feeding walleye pollock larvae were measured at 6°C in continuous and discontinuous feeding regimes. Larvae fed tagged copepod prey evacuated their guts more quickly in continuous feeding as compared to the discontinuous feeding treatment. The mean gut residence time was estimated to be 5.0 h for larvae feeding continuously. Gut evacuation by larvae fed tagged prey and then isolated without food (discontinuous treatment) was slower and more variable, with an estimated gut residence time exceeding 8.0 h. Field and laboratory observations suggest that the larval fish gut may be modeled as an intermittent plug-flow reactor (PFR) in response to diel feeding patterns. The cyclical nature of gut dynamics has implications for gut content analyses and the estimation of daily food rations.     

Aspects of feeding,including estimates of gut residence time,in three mytilid species (Bivalvia,Mollusca) at two contrasting sites in the Cape Peninsula,South Africa

Individuals of three mytilid species (Choromytilus meridionalis; Perna perna; Aulacomya ater) from two sites characterised by different qualities of ration available to these suspension feeders showed different rates in some components of the physiological energy budget. These differences included higher feeding (=clearance) and respiration rates, but lower absorption efficiencies, in individuals from the site with the higher quality ration. A novel technique was employed to estimate the residence time of food particles in the digestive gland. Individuals showing higher feeding rates had shorter residence times than those feeding more slowly and a significant positive correlation was demonstrated between residence time and absorption efficiency. These relationships, together with an exponential increase in rates of respiratory heat loss with an increase in ingested ration, are suggested to provide these animals with a physiological flexibility to compensate for reduced food quality in a way consistent with some theoretical predictions. Such compensations are made more effective if the total gut capacity can also change in response to the quality of the ration.     

Growth dynamics of a ctenophore (Mnemiopsis) in relation to variable food supply. I. Carbon biomass, feeding, egg production, growth and assimilation efficiency

This paper contains new experimental data on the growth dynamicsof a lobate coastal ctenophore, Mnenmiopsis mccradyi, whichadd significantly to our understanding of the nutritional ecologyof ctenophores and their role as opportunistic predators. Theseexperimental observations were necessary to refine the dynamiccarbon budget presented as a simulation model in another report.The ratio of carbon biomass to dry weight may vary several-folddepending on the nutritional state and size from >12% inwell-fed larvae to <1% in starved adults. Feeding effort(clearance rate) is higher for previously starved animals, fallingsharply within a few hours after re-exposure to food. Directvisual observations of feeding activity of animals confirmedthis. Assimilation efficiency can be high (72%) in these animalsbut if they continue to feed at high food concentrations, incomingfood displaces material which is only partially digested andassimilation efficiency decreases substantially. Except at verylow food concentrations, growth efficiency ranges between 20and 45%. Mnemiopsis, begins to produce eggs at a size much lessthan its maximum. Egg production is very sensitive to food supply,and somatic growth is favored over egg production at low fooddensities. Even though several thousand eggs may be producedover a few days, they represent <2% day^–1 of the carbonbiomass of the ctenophore and several-fold less than respiratorycarbon.     

Feeding of freshwater filter-feeders at very low food concentrations: Poor evidence for "threshold feeding" and "optimal foraging" in Daphnia longispina and Eudiaptomus gracilis

The feeding behaviour of two freshwater zooplankton species,the calanoid copepod Eudiaptomus gracilis and the cladoceranDaphnia longispina, at extremely low food concentrations hasbeen studied in order to test whether the animals react as predictedby the models of optimal foraging. Three different sized algae were offered in concentrations downto 1µg C/litre at 7°C during winter and 19°C duringsummer. Ingestion rates were measured by use of a flow-through¹⁴C-technique In contrast to the findings of several authors working withmarine copepods a threshold concentration where the animalsceased filtering could be detected only in one experiment byregression analysis (Daphnia, 19°C, food: Stichococcus).A depression of the filtering rate at low concentrations wasobserved only in Daphnia at 19°C with Stichococcus and Scenedesmusas food. It is supposed that in this case the reduction of thefiltering rate is not a direct response of the filtrator, butis caused by exhaustion due to the experimental conditions. As freshwater zooplankton is very rarely exposed to food abundancesas low as in the marine environment, there seems to be no selectionpressure favouring the evolution of an "optimal forager" behaviour. ⁽¹⁾Supported by Deutsche Forschungsgemeinschaft     

Chlorophyll a conversion and gut passage time for the pelagic tunicate Oikopleura vanhoeffeni (Appendicularia)

We determined the chlorophyll (Chl) a conversion efficiency,defined as the conversion of Chi a into pheopigments and non-fluorescentproducts, for the cold-water appendicularian Oikopleura vanhoeffeniusing ⁶⁸Ge as conservative tracer. In both laboratory experimentsand in freshly collected animals, there was enough undegradedChi a present in the gut to provide an index of feeding activityfor individual animals. In laboratory experiments with diatoms,Chi a conversion efficiency was predictable with an averageof 79%. The mean gut passage time (0.8 h) determined using cornstarch and diatoms as markers was not influenced by food concentrationor animal size over a trunk length range of 1.8–5 mm.When experimentally determined Chi a conversion efficiency andgut passage times were applied to O.vanhoeffeni from Logy Bay(Newfoundland), the estimated clearance rates were in closeagreement with previously published values using several differenttechniques. We therefore suggest that the modified gut pigmenttechnique is a useful tool to assess in situ ingestion ratesof O.vanhoeffeni.     

Growth dynamics of a ctenophore (Mnemiopsis) in relation to variable food supply. II. Carbon budgets and growth model

Our goal was to test our understanding of ingestion, assimilationefficiency and metabolism for Mnemiopsis mccradyi by formulatingand validating a simulation model of growth under differentconditions of food availability. The model was based on a carbonbudget approach using formulations derived from empirical results,including how each process was affected by food availabilityand ctenophore size. An experimentally measured carbon budgetfor pulsed food availability indicated that, relative to totalingestion, growth was high (17–48%), respiration plusorganic release was relatively low (24–48%) and little(<10%) of the ingested carbon was unaccounted for. New laboratoryinvestigations of feeding and assimilation efficiency were necessaryto refine the formulations so that model predictions comparedfavorably with a variety of laboratory measurements of growth,and growth efficiency, as well as the complete experimentallymeasured carbon budget. The refined model predicted a high ratioof growth to metabolism (>2) and a high gross growth efficiency(>30%) for smaller ctenophores at high food concentrations(>20 prey l^–1). Both growth rates and growth efficiencieswere predicted to decrease for larger ctenophores. Model predictionswere generally consistent with experimental results, includinginvestigations using pulsed food availability to simulate environmentalpatchiness. Although the model underpredicted ctenophore growthin some experiments at low food densities, the model predictionof a minimum prey concentration of about 8 l^–1 (24 µgC l^–1) for sustaining a ctenophore population of reproductivesize agreed with field observations.     

Cannibalistic feeding behavior of the brackish-water copepod Sinocalanus tenellus

Cannibalistic feeding behavior of the brackish-water copepodSinocalanus tenellus was examined in the laboratory using CI-II,CIII-IV and CVI female as predators and NI-II, NIII-IV, NV-VIand CI-II as prey. In each prey-predator combination, the ingestionrate increased with increasing prey density to an asymptoticvalue. Cannibalism took place even when phytoplankton was availableas an alternative food supply. Based on a daily ration, theoptimal prey stages for CVI females, CIII-IV and CI-II are NI-VI,NI-IV and NI-II respectively. Under average, natural prey density(10 nauplii l^–1), S tenellus can achieve only a smallfraction (max 9%) of the daily minimum food requirement by cannibalisticfeeding. However, the impact of cannibalism on naupliar survivorshipcan be significant. When adult females occur at a density of10 l^–1, the mortality due to cannibalism attains 99.2%during the naupliar stages.     

Diel variations in gut pigment content, diel vertical migration and estimates of grazing impact for copepods in the southern Benguela upwelling region in October 1987

The gut fluorescence technique was used to estimate ingestionand filtration rates of the adult female copepods Paracalanusparvus, Cenlropages brachiatus and Calanus austrails, and copepoditestages 3, 4 and 5 of C.australis in the southern Benguela upwellingregion. During the study period chlorophyll concentrations withinthe upper 20 m of the water column were high, 5 µg I^–1in mid-shelf waters and 15–30 µg I^–1 in innershelf waters. Copepod gut pigment content was low and constantduring the day then increased sharply during the first 2 h aftersunset. Gut pigment content was 2–6 times higher duringthe night compared with daytime values. Small non-migratingcopepods (Paracalanus parvus) showed the smallest diel differencein gut pigment content and large migrating copepods (Centropagesbrachiatus and Calanus australis) the largest difference. Eggproduction rates were 20 and 50% of maximum at the mid-shelfand inner shelf stations respectively, suggesting food-limitation.Comparison of ingestion rates calculated from egg productiondata with ingestion rates calculated from gut pigment data suggestedthat the copepods were feeding omnivorously at the inner shelfstations but herbivorously at the mid-shelf stations. Assumingthat all of the phytoplankton was available as food, the nearshorecopepod assemblage grazed {small tilde}1% of the standing cropeach day, and the mid-shelf assemblage grazed 5% day^–1.Because of errors and uncertainties associated with the gutfluorescence technique, the feeding impact could be underestimatedby 2–4-fold. We discuss several approaches which couldlead to more precise estimates of feeding rates. ³Present address: Marine Sciences, SUNY, Stony Brook, NY, 11794-5000,USA     

Interannual variability in a predator-prey interaction: climate, chaetognaths and copepods in the southeastern Bering Sea

The interaction of the chaetognath Sagitta elegans with thecopepod community of the southeast Bering Sea middle shelf wasexamined in relation to environmental conditions during 1995–1999.Predation impact was estimated for 2 years, 1995 and 1997, usinggut content analysis, experimentally derived digestion time(DT) and abundances of chaetognaths and prey. Pseudocalanusconcentrations correlated with water temperature and Calanusmarshallae with sea ice extent. Sagitta elegans were less abundantbut individuals were larger in 1995, when C. marshallae predominated,compared to 1997, when Pseudocalanus and Acartia were the primaryprey. Predation by S. elegans removed <1% standing stockday^–1 of Pseudocalanus or C. marshallae in 1995 and 1.7to 2.3% of Pseudocalanus in 1997. The percent of the copepodcommunity biomass required by chaetognaths was estimated tobe <1% in 1995 compared with 8–12% in 1997. Calanusmarshallae may be more vulnerable than Pseudocalanus to cumulativepredation effects because of its reproductive strategy. Theeffect of chaetognath predation on the copepod community dependson which copepod species is predominant and its susceptibilityto cumulative predation effects, as well as on daily predationimpact, both of which varied between years with different climaticconditions.     

Feeding of nauplius stages of Eudiaptomus gracilis on mixed plastic beads

Eudiaptomus gracilis makes up 30–40 and 80–90% ofthe zooplankton in Lake Balaton during the summer and the winterrespectively. More than half of the species population consistsof nauplii We studied feeding and size selectivity of naupliiin suspensions which contained polystyrol latex beads in a concentrationdose to the natural seston. Guts of NI nauplii were free ofboth beads and remnants of natural food Of NII–NVI nauplii,67–87% took in beads. Older animals consumed more andlarger particles. The maximum diameter of ingested beads reached29 µm On an average, NII nauplii took in 128 µm³of beads in 10 min, whereas older animals consumed 615–900µm³. The clearance rate remained below 0 01 µl h^–1NII nauplii strongly preferred 1 2 µm particles Oldernauplii did not show any preference or selected only slightlyfor the smallest particles. Nauplii rejected 4–11 µmbeads. In some cases a weak positive selection could be observedtoward 12 µm or larger beads.     

Population biomass, feeding, respiration and growth rates, and carbon budget of the scyphomedusa Aurelia aurita in the Inland Sea of Japan

We investigated the seasonal occurrence, wet : dry : carbon: nitrogen weight ratios, population biomass, gastric pouchcontents, and rates of feeding, growth and respiration of thescyphomedusa Aurelia aurita in the central part of the InlandSea of Japan. Aurelia aurita medusae began to appear in January/Februaryas ephyrae, reached annual maximum body size in July/August,and disappeared, presumably due to death, by November. Initialslow growth in early spring was followed by a period of exponentialgrowth (mean growth rate: 0.069 d^–1) between April andJuly. In the Ondo Strait, which is characterized by strong tidalmixing, the A. aurita population (mean carbon biomass: 66.0mg C m^–3) overwhelmingly dominated the zooplankton-communitybiomass (mean biomass of micro- and mesozooplankton: 23.7 mgC m^–3) between May and early August The gastric contentanalysis revealed that A. aurita ate almost all micro- and mesozooplankters,of which small copepods were most important. On the basis ofdigestion time for small copepods (60 min) and their abundancein the gastric pouch of field-collected A. aurita, we determinedthe weight specific feeding rates and clearance rates. The formerincreases linearly with increasing copepod abundance, but thelatter was relatively constant irrespective of the food supply.We also measured the respiration rates of A. aurita and expressedthem as functions of body weight and temperature. These physio-ecologicalparameters enabled us to construct the carbon budget of theA. aurita population typical of early summer in the Ondo Strait.Predicted population-feeding rate (6.07 mg C m^–3 d^–1)was higher than the population-food requirement for both metabolismand growth (4.55 mg C m^–3 d^–1), indicating thatfood supply was sufficient to sustain the observed growth rate.This feeding rate was equivalent to 26% of micro- and mesozooplanktonbiomass, a significant impact on zooplankton.     

Feeding behaviour of the rotifer Ascomorpha ovalis: functional response, handling time and exploitation of individual Ceratium cells

The planktonic rotifer Ascomorpha ovalis feeds on large dinoflagellates(e.g. Ceratium sp., Peridinium sp.) and is able to extract theircell contents by means of its virgate mastax. This paper presentsthe results of experiments on the feeding behaviour of laboratory-culturedAscomorpha with Cerarium furcoides as food algae. Ascomorphaare three times larger than their prey Ceratium (by volume),but with regard to total length, their prey was even 20% larger.Ascomorpha showed a hyperbolic functional response curve witha plateau of the feeding rate at 8 Ceratium cells animal^–1dar^–1 when concentrations of Ceratium were >100 cellsml^–1. The mean handling time (time for capturing and extractingone Ceratium cell) was 3 min. The shape of the functional responsewas better described by a curvilinear model than by a rectilinearmodel. However, handling times cannot be responsible for this,since they were too short to set limits on ingestion rates.At low food concentrations, encounter rates with prey seemedto limit the feeding rates of Ascomorpha, whereas at mediumto high food concentrations, Satiation effects (lower attackrates) seemed to set limits on the feeding rates. Ascomorphashowed a significant decrease in the exploitation of singleCeratium cells at high prey concentrations. This decrease couldbe explained by a saturation effect in which the partly filledguts of Ascomorpha did not permit the total extraction of thecontents of a Ceratium cell.     

Distribution of euphausiids in the Kuroshio front and warm water tongue with special reference to the surface aggregation of Euphausia pacifica

The distribution of euphausiids and other macrozooplankton wasinvestigated in and around the Kuroshio front formed in theKa.shima-nada Sea. Japan. during the summer of 1993. Zooplanktonwere dominated by copepods, chaetognaths and euphausiids. andtheir biomass was significantly higher at the frontal stations.Eighteen species belonging to six genera of euphausiids werecollected. The species composition and community structure ofeuphausiids changed drastically with the ‘areas’corresponding to the hydrological conditions. In the area underthe influence of the Kuroshio. most euphausiids were warm-waterspecies. In contrast. euphausiids found in the frontal areawere those usually predominating in subarctic or cold Oyashiowaters. In the area of the warm water tongue. where warm Kuroshiowater lay above cold Oyashio water, the species compositionof euphausiids represented a mixture of both warm- and cold-waterspecies. Surface aggregations of Euphausia pacifica were observedin the frontal area during the night. The biomass and densityof the aggregations ranged between 90 and 136 mg C m^–3and 164 and 238 md. m^–3 respectively. These aggregationswere mostly made up of immature individuals of < 12 mm. Passivetransportation by convergent flow at the front seems to explainthe observed surface aggregations of E.pacifica.