A comparative ecological study on the scyphozoans Aurelia aurita,Cyanea capillata and C. lamarckii in the Gullmar Fjord,western Sweden, 1982 to 1986

During the years 1982 to 1986, the life cycles and population dynamics of three scyphozoans, Aurelia aurita (L.), Cyanea capillata (L.) and C. lamarckii (Person and Lesueur), were studied in the Gullmar Fjord on the Swedish west coast. The settling of planulae, strobilation of scyphistomae and release of ephyrae were followed on ceramic settling plates in the laboratory and in the field. Weekly to bi-weekly hauls with Bongo nets were used to study the abundance of ephyrae and medusae. The results show great differences in the life cycles and ecology of the three species. A. aurita utilizes the best season for scyphistoma growth (August to September) and strobilates during the highest zooplankton abundance in October. C. capillata strobilates during the spring (March to May), and the abundance of C. capillata medusae is more dependent on immigration from the North Sea than A. aurita. C. lamarckii does not reproduce at all in the Gullmar Fjord and is totally dependent on immigration from the North Sea. The possibility of interspecific competition between A. aurita and C. capillata is discussed. A preliminary experiment showed that scyphistomae of A. aurita eat planula larvae of C. capillata during the autumn.     

Nucleic acid and protein levels in strobilating polyps of Chrysaora quinquecirrha and Aurelia aurita

Levels of DNA, RNA, and protein were measured in scyphistomae of the scyphozoans Aurelia aurita and Chrysaora quinquecirrha during strobilation. In synchronously developing populations of A. aurita, the amount of DNA per polyp increases 41 to 260% during strobilation, whereas RNA and protein remain relatively constant. The RNA/DNA ratio drops by about 50% during strobilation in both C. quinquecirrha and A. aurita, and protein/DNA decreases by 50% in A. aurita. Specific RNA reserves are probably not accumulated in advance of strobilation, since strobilation is blocked immediately in both species by 1 to 3 g/ml actinomycin D. The size and DNA content of polyps are usually increased by prolonged incubation at temperatures of 12° to 15°C, which facilitates strobilation; however, the RNA/DNA and protein/DNA ratios of such polyps are not significantly increased by the cold conditioning.Contribution No. 472 from the Virginia Institute of Marine Science.     

Ultrastructural observations on prey capture and digestion in the scyphomedusa Aurelia aurita

Aurelia aurita medusae are able to catch their prey with their entire body surface. Catch efficiency in medusae caught in Kiel harbour in May 1985 was found to be highest at the tentacles and lowest at the subumbrella. Surface structures of the medusa as well as the cnidom are described by SEM observations. Microbasic heterotrichous euryteles and atrichous isorhizas were found. Discharged nematocysts on the prey's skin indicate different functions of the two types. The villi in the gastral cavity show a characteristic morphological differentiation that consists of a ciliated distal and a basal area covered by vesicles. Four types of glandular cells were identified by TEM observations. Mucous cell types preferably occur in densely ciliated areas. The presence of serous cells is restricted to the basal region of the gastral villi and gastral cavity where the extracellular predigestion takes place. The time of food passage in young medusae of A. aurita decreases from 19 h at 4°C to 4 h at 22°C.     

Bimodal vertical distribution and diel migration of the copepods Metridia pacifica,M. okhotensis and Pleuromamma scutullata in the western North Pacific Ocean

Diel changes in fine-scale vertical distributions of three calanoid copepods Metridia pacifica, M. okhotensis and Pleuromamma scutullata in the subarctic waters of the western North Pacific were examined. Sampling was carried out in June and August 1983, at two stations in Oyashio water using a Longhurst-Hardy Plankton Recorder (LHPR). Sampling, down to about 1 000 m, was repeated four to five times at intervals of several hours. Vertical resolution was 5 to 40 m. Copepods were concentrated in two strata, the surface (0 to 60 m) and the mesopelagic (200 to 300 m) layers, throughout the day at both stations. Younger M. pacifica (C III and C IV) were dominant in both strata. Although most female C V and adult females demonstrated diel vertical migration at 20 to 30 m h^-1, a significant number of females did not migrate upward but remained in the deep stratum at night. The same trend was evident in M. ohkotensis and P. scutullata. Foregut content observations indicated that feeding activities of the deep mode populations were as high as those of the surface mode, though food of deep individuals was different. Such a bimodal distribution may increase intraspecific diversity of copepod populations and is possibly why metridiid copepods dominate during late summer to winter in the relatively simple ecosystems of high latitudes.     

Feeding rates of the jellyfish Aurelia aurita on fish larvae

We quantified feeding rates of field caught Aurelia aurita feeding on yolk sac cod (Gadus morhua) larvae in a series of incubation experiments. A short-time (~1 h) functional response experiment with a wide range of prey concentrations (0.5–16 prey l⁻¹, initial concentration) revealed that ingestion rates increased linearly over this range, such that clearance rates were similar between the different prey concentrations. This suggests that A. aurita is capable of efficiently utilizing dense prey patches. This indication was further supported by a linear increase of prey captured by A. aurita during 2.5 h of feeding at extremely high prey concentration (>200 prey l⁻¹). Clearance rate in darkness scaled with jellyfish diameter to a power of ~1.7 for jellyfish 3.9–9.5 cm in diameter. The jellyfish did not alter their umbrella pulse frequency in response to presence of fish larvae. There were no significant differences between A. aurita feeding rates in light and darkness for yolk sac prey ages 0–7 days (at 7.5°C). Although prey vision and escape abilities of fish may develop rapidly during early larval ontogeny, these factors apparently have little impact on interactions with predators such as A. aurita during the yolk sac stage.     

Sun-compass migration by Aurelia aurita (Scyphozoa): population retention and reproduction in Saanich Inlet,British Columbia

The scyphomedusa Aurelia aurita in Saanich Inlet, a north-south oriented fjord in British Columbia, uses a sun compass to migrate in a southeasterly direction during the day. When the sky is overcast and at night, A. aurita orients randomly and is dispersed passively by gentle tidal currents. The net result is daily reaggregation of medusae into enormous swarms along the southeastern shore of the fjord. Observations of spawning A. aurita in these swarms suggest that sun-compass migration and aggregative behavior may have evolved to facilitate reproduction and to maintain the population within this fjord throughout the year.     

Allozyme and morphological evidence for a newly introduced species of Aurelia in San Francisco Bay,California

The jellyfish Aurelia aurita (Linnaeus, 1758) is usually considered to be a cosmopolitan species. Aurelia sp. medusae observed at Foster City, San Francisco Bay, California, USA, since 1988 are morphologically distinct from Aurelia sp. collected 200 km away in Monterey Bay, but are morphologically similar to aquarium-cultured Aurelia sp. from Japan. They differ consistently in radial canal morphology. In allozyme electrophoresis, significant differences at 12 of 14 polymorphic loci strongly suggest that Aurelia sp. from Foster City and Tokyo Bay belong to one species, while Aurelia sp. from Monterey Bay and Vancouver Island belong to a second species. We propose that Aurelia sp. at Foster City is a recent introduction, possibly from Japan via ships' ballast water. The identities and taxonomic affinities of the two Aurelia defined in this study, and their relationships with the Linnaen A. aurita described from the North Atlantic, will require genetic and morphological study of the currently recognized species A. aurita and A. limbata (Brandt, 1838) from several zoogeographical provinces.     

Morphology,fluid motion and predation by the scyphomedusa Aurelia aurita

Although medusan predators play demonstrably important roles in a variety of marine ecosystems, the mechanics of prey capture and, hence, prey selection, have remained poorly defined. A review of the literature describing the commonly studied medusa Aurelia aurita (Linnaeus 1758) reveals no distinct patterns of prey selectivity and suggests that A. aurita is a generalist and feeds unselectively upon available zooplankton. We examined the mechanics of prey capture by A. aurita using video methods to record body and fluid motions. Medusae were collected between February and June in 1990 and 1991 from Woods Hole, Massachusetts and Narragansett Bay, Rhode Island, USA. Tentaculate A. aurita create fluid motions during swimming which entrain prey and bring them into contact with tentacles. We suggest that this mechanism dominates prey selection by A. aurita. In this case, we predict that medusae of a specific diameter will positively select prey with escape speeds slower than the flow velocities at their bell margins. Negatively selected prey escape faster than the medusan flow velocity draws them to capture surfaces. Faster prey will be captured by larger medusac because flow field velocity is a function of bell diameter. On the basis of prey escape velocities and flow field velocities of A. aurita with diameters of 0.8 to 7.1 cm, we predict that A. aurita will select zooplankton such as barnacle nauplii and some slow swimming hydromedusae, while faster copepods will be negatively selected.     

Vertical distribution and mortality of <Emphasis Type="Italic">Calanus finmarchicus</Emphasis> during overwintering in oceanic waters southwest of Iceland

The seasonal vertical distribution and the predatory regime encountered by Calanus finmarchicus were studied along a transect across the Reykjanes Ridge in the oceanic area southwest of Iceland from data collected during four cruises between November 1996 and June 1997. The mortality for the overwintering period was estimated using linear regressions of density estimates from November 1996 to April 1997. In addition, we also estimated the mortality of the oldest population stages (C4 and older) in April and June by applying the vertical table method. During winter (November/December–January/February), the animals mainly resided at a depth of ∼300–1,500 m in the water of Atlantic origin. Ascent to upper layers took place mainly during March and April, and continued until May. During all cruises, continuous deep-scattering layers were observed, mainly within the range of 400–500 m to 700–800 m depth. Based on sampling with a Harstad pelagic trawl in April, the scattering was mainly ascribed to jellyfish (mainly Periphylla periphylla), small mesopelagic oceanic fishes (several species but Benthosema glaciale and Maurolicus muelleri were most abundant), euphausiids (mainly Meganyctiphanes norvegica) and shrimps (mainly Sergestes arcticus). These species may represent a predatory threat to overwintering C. finmarchicus. From November to April, daily per capita mortality rates were estimated to be (mean ± 95% CL) 0.004 ± 0.0028 for the total data set, and 0.004 ± 0.0033 and 0.004 ± 0.0023 (day^-1) for the Iceland Basin and Irminger Basin, respectively. Mortality rates were higher later in life (mean ± 95% CL) for C5/females (0.13 ± 0.044) and C5/males (0.19 ± 0.051) than for C4/C5 (0.00 ± 0.035) when averaged over all samples taken in April and June 1997. We discuss how the observed distribution and mortality rates of overwintering C. finmarchicus might be related to predatory regime.     

Laboratory study of predation by Aurelia aurita on larvae of cod,flounder, plaice and herring: development and vulnerability to capture

Capture success of the medusa Aurelia aurita preying on various developmental stages of fish larvae was measured together with larval reactivity and escape speed after being stung. These experiments were conducted in the spring of 1983 with A. aurita medusae collected from Loch Etive, Scotland and laboratory-reared larvae of Gadus morhua L., Platichthys flesus L., Pleuronectes platessa L. and Clupea harengus L. Capture success of the medusae increased with medusa size, but decreased with advancing larval development. Smaller species of larvae were more vulnerable to capture. Larval reactivity to encounters with medusae increased with advancing development, and larger species of larvae were more reactive to encounters. Larval escape swimming speeds also increased with advancing larval development and size. These results indicate that earlier stages of larvae within a species and smaller species of larvae at a given stage are more vulnerable to predation by medusae since they are less reactive to encounters. Apparently they are more susceptible to the effects of neurotoxins. Predation rates on different developmental stages of herring larvae are documented and compared with rates predicted by a predation model. Predictions fell within the range of observed predation rates, but tended to overestimate rates by larger medusae feeding on larger herring larvae. This indicates the possibility of predator satiation and/or behavioural avoidance.     

Seasonal changes in feeding, gonad development and lipid stores in Calanusfinmarchicus and C. hyperboreus from Malangen, northern Norway

 Seasonal dynamics of feeding activity, oil sac volume, gonad development, sex ratio and spawning periods in the two sibling species Calanusfinmarchicus and C. hyperboreus, the key zooplankton copepod organisms throughout the northern Atlantic waters, were studied simultaneously in Malangen, northern Norway, during 1992. We were also tracing differences in surface- and deep-dwelling components of these populations in terms of oil sac volume and gonad development during the time period when the G₁ is preparing for a subsequent generation (G₂) or hibernation. The main difference in the life cycle strategies of these species is the earlier maturation and spawning of C. hyperboreus. No feeding activity in either of the two species was found in February, but both commenced feeding in March, prior to the spring phytoplankton peak. The larger copepod, C. hyperboreus, had a more intensive energy deposition than C. finmarchicus. The period of active feeding was much shorter for the former species, only from March through July in copepodite stages CIV and CV, and even less in females – March and April. Basically, a similar pattern of seasonal changes in gonad length and lipids was observed in the two species. In June, oil sacs in the surface- and deep-dwelling specimens were about equal, during the rest of the year, lipids in the deep CVs exceeded those in the surface. We propose that as copepods accumulated sufficient lipid reserves, they started to descend, while others, containing less fat, stayed in the upper layers feeding. The mean length of the gonads in the surface-dwelling copepods was consistently less than in their deep counterparts from October to February, so that gonad development at the expense of accumulated reserves during resting stage was confirmed. C. finmarchicus males were found in considerable numbers only in February and March, and were only occasionally found in the upper layers (0–100 m), while adult male C. hyperboreus were present from October to March, but were never found in the surface layers. The differences in life cycle timing among the two species are discussed in relation to tradeoffs with regard to foraging strategies, generation numbers, bioenergetics and predator avoidance. Received: 31 March 1999 / Accepted: 23 November 1999     

Life history of the nematode Theristus anoxybioticus from sublittoral muddy sediment at methane seepages in the northern Kattegat,Denmark

Field and laboratory studies from 1989 to 1994 of the nematode Theristus (Penzancia) anoxybioticus Jensen, 1995 inhabiting muddy sediment at methane seepages in the northern Kattegat, Denmark, have revealed a series of biological features of adults and juveniles not observed before in free-living marine nematodes. Reproductive adults were present in April and May only and inhabited the uppermost cm of sediment, while juveniles inhabited deep anoxic and sulphidic sediment layers during the remaining months of the year. This points to a downwards migration of newly hatched juveniles and an upwards migration during adulthood. Gut content analyses showed that adults feed mainly on benthic diatoms by swallowing whole cells while juveniles probably feed on bacteria; this differentiation in food clearly relates to the main bulk of potential food items in their respective microhabitats. The generation time is estimated to be one year, with a juvenile life span of about nine months, which is four to eight times longer than that found in related aerobic species of similar body size. Adults survived in oxic water but 50% died within 15 h in hypoxic water (5 to 6% atm sat), and all died within 12 h in anoxic water. Juveniles on the other hand died within 0.5 h in oxic water but survived anoxic water for at least 15 d. These results suggest that juveniles of T. anoxybioticus are adapted to an anaerobic life style.     

Potential dietary effects on the fatty acid composition of the common jellyfish Aurelia aurita

Fatty acid composition of the natural and aquarium-reared common jellyfish Aurelia aurita was investigated. Fatty acid composition of the aquarium-reared A. aurita clearly reflected that of the diet, brine shrimp (Artemia). In the same way, fatty acid composition of the natural A. aurita was assumed to reflect those of natural diets. Samples of natural A. aurita were collected from April 1995 to September 1995 in the Seto Inland Sea, Japan, and their fatty acids were analyzed by gas chromatography and mass spectrometry. Variation of fatty acid compositions was seasonal rather than dependent on body size. Two major seasonal groups were divided by the cluster analysis of the A. aurita fatty acid composition: the April–June and the August–September clusters. The April–June cluster was characterized by high contents of the (n − 3)-fatty acids of diatom origin, accumulated via the grazing food chain. By contrast, the August–September cluster was characterized by an increase in (n − 6)-fatty acids of macroalgal origin, probably transferred via the detritus food chain. These results suggest that the diet of natural A. aurita may shift between the diatom-based food chain and the detritus-based food chain. Received: 12 April 2000 / Accepted: 1 December 2000     

Distribution and abundance of moon jellyfish (Aurelia aurita) and its zooplankton food in the Black Sea

The distribution of moon jellyfish (Aurelia aurita Linnaeus, 1758) in the Black Sea was determined from plankton samples collected above the anoxic zone (maximum depth 200 m) in the summer, winter and spring during 1991–1995. Distribution was patchy. Average biomass ranged from 98 to 380 g m⁻², and abundance varied from 2 to 14 individuals m⁻². Biomass and abundance peaked in late spring and summer. The distribution was correlated with hydrographic features in the Black Sea, with higher concentrations occurring at the peripheries of anticyclonic eddies. Centers of the two main cyclonic gyres generally had a low biomass of A. aurita. From July 1992 to March 1995, the populations were largely concentrated in offshore regions. A. aurita were confined to the upper part of the mixed layer. Smaller A. aurita (≤1 cm) were present in early spring (March), and individuals reached maximum size in the summer. Release of the epyhrae occurred in spring on the northwestern shelf of the sea when the seawater temperature was 11–12 °C. Microscopic analysis of stomach contents showed that copepods and mollusks form their main diet. Received: 3 September 2000 / Accepted: 29 September 2000     

Invading <Emphasis Type="Italic">Aurelia aurita</Emphasis> has established scyphistoma populations in the Caspian Sea

The Caspian Sea has no endemic Scyphozoa. In 1999, a mass accumulation of Aurelia medusae was recorded, indicating that sometime earlier, jellyfish had invaded the basin, but since then no scyphozoans have been reported in the Caspian. In the fall of 2008, we found scyphistomae (scyphoid polyps) during a cruise to the eastern Middle Caspian. The scyphistomae were numerous (100–10,000 ind. m⁻²) and occupied a depth range of 30–73 m. Genetic data (18S rDNA, ITS-1 and COI) showed that the scyphistomae belonged to the species A. aurita. The current genetic data set is insufficient to determine the source region(s) of the invasive A. aurita. It remains unclear why no moon jellies have been recorded in the Caspian in the last 10 years. Because swarming scyphomedusae are often pests, the presence of scyphistomae should be considered as a warning of a possible outbreak of A. aurita medusae in the Caspian.     

Sexual reproduction of the scyphomedusa Aurelia aurita in relation to temperature and variable food supply

The effects of food availability and temperature on sexual maturation and female reproductive output of the scyphomedusa Aurelia aurita was examined in two populations from the contrasting environments of Southampton Water and Horsea Lake, England. Trends in oogenesis and subsequent reproductive output differed markedly between the two populations. In Southampton Water, the onset of sexual maturation occurred earliest in the larger medusae, but eventually all females became ripe, the smallest being 45 mm bell diameter (BD). The decrease in minimum size at maturity was correlated with increasing temperature. In A. aurita from Horsea Lake, size at maturity varied on a seasonal basis, with the smallest ripe female being only 19 to 20 mm BD. There were spring and autumn periods of sexual maturation in this population. During the autumn period, it is likely that food limitation was playing a more critical role in determining medusa size, with decreasing temperature indirectly affecting A. aurita by limiting primary and secondary production. In similar-sized ripe medusae, fecundity was greater in Southampton Water, but the planula larvae produced were significantly smaller than those in Horsea Lake. It is suggested that in Horsea Lake, the quality of the larvae are greater in terms of biochemical content to ensure survival of the few gametes produced (i.e. K-strategy). Comparison of the reproductive effort of the two A. aurita populations revealed that medusae from Southampton Water, which experience greater food availability, are able to direct more energy to reproduction than Horsea Lake medusae. In the latter, A. aurita medusae appear to partition the available food resources into either somatic growth (and therefore increased future fecundity) when food is abundant, or reproductive growth when food is scarce. Received: 24 June 1997 / Accepted: 23 March 1998     

Prey selection by the scyphomedusan predator Aurelia aurita

We describe feeding behavior of Aurelia aurita (Linnaeus) using gut content analyses of field-collected specimens and a mesocosm experiment. The field studies were conducted in Narragansett Bay, Rhode Island, USA from March to April 1988, and the mesocosm studies were done at the Marine Ecosystems Research Laboratory at the University of Rhode Island. Patterns of prey selection changed with medusa diameter. Smaller medusae (12 mm diameter) consumed mostly hydromedusan prey whereas larger medusae (up to 30 mm diameter) ingested greater numbers of copepod prey. While larger medusae did feed on copepods, their diet also contained more barnacle nauplii and hydromedusae than expected from the relative abundances of these prey types in plankton samples. A marginal flow mechanism of feeding by A. aurita provided an explanation for the patterns of prey selection we observed in medusae of different sizes and among widely divergent prey types. Our data indicated that large prey, with escape speeds slower than the marginal flow velocities around the bell margins of A. aurita, made up a substantial fraction of the daily ration when they were available. Such prey species may be more important to nutrition than the more abundant copepods and microzooplankton. Successful development of young medusae may depend upon an adequate supply of slowly escaping prey.     

Spatio-temporal changes in diversity and community structure of planktonic copepods in Sagami Bay, Japan

Seasonal changes in diversity and community structure of planktonic copepods at a shelf site in Sagami Bay, Japan was studied in relation to cross-shelf interaction of species components. Seasonal mesozooplankton samples were collected from the shelf station (St. M) of the north-west part of Sagami Bay from 1995 to 1997. Vertical multi-layered samples were collected near the center of Sagami Bay (St. P) in June 1996. A total 185 copepod species were identified from the two stations. We observed a clear seasonal succession in calanoid diversity and community structure at St. M from a simple shelf water community (>11 species) during spring blooming periods to highly diverse and mixed communities (ca 20–30 species) of shelf water species coupled with various Kuroshio Current species during late summer to autumn. Cluster and non-metric multidimensional scaling ordination analyses showed two distinct calanoid community groups. One group, which included samples of St. M and the surface layer of St. P, consisted of shelf water species, such as Calanus sinicus, Ctenocalanus vanus, Paracalanus spp., and Kuroshio species, such as, Canthocalanus pauper, Scolecithrix danae, etc. The other cluster was restricted to the samples collected from mid and deep layers at St. P, which consisted of meso- and bathypelagic species and Oyashio species (cold-current species, such as Neocalanus cristatus, Pseudocalanus spp., Eucalanus bungii and Metridia pacifica). In the mid and deep layers at St. P, the population of dormant copepodid stage V (CV) of Eucalanus californicus and C. sinicus were dominant. The deep CV population of C. sinicus might be ecologically discriminated from the surface and shelf water population due to their larger body length and dormant life cycle. E. californicus was also collected at the shelf site during each spring bloom period, whereas the population might descend into the mid- and deep-layers of the central bay before summer. Our results suggest that the seasonal fluctuation of community structure in the shelf water was controlled by both physical (Kuroshio Current) and biological factors, i.e., spring bloom and ontogenetic vertical migration of E. californicus. In particular, transport and diffusion processes of Kuroshio Current in Sagami Bay played a key role in controlling the shelf water calanoid community.     

Seasonal development of scyphozoan medusae and the predatory impact of Aurelia aurita on the zooplankton community in the Bornholm Basin (central Baltic Sea)

The annual cycle of abundance and distribution of the scyphozoan medusae Aurelia aurita and Cyanea capillata was studied in the Bornholm Basin (central Baltic Sea) in 2002. Seasonal changes in prey composition and predatory impact were investigated by analyzing stomach contents. A. aurita occurred from July to November, with a maximum mean abundance of 2.3 ind. per 100 m³ in August, whereas C. capillata was caught in much smaller numbers from July to September. No ephyrae of either species were found; therefore, advection of medusae from the western Baltic Sea is assumed. From July to October, ~80% of A. aurita medusae was distributed in the upper 20 m above the thermocline, whereas C. capillata occurred only in the halocline below 45 m. A. aurita did not migrate vertically and fed mainly on the most abundant cladoceran species Bosmina coregoni maritima. Further prey organisms were the cladocerans Evadne nordmanni and Podon spp., mollusk larvae and copepods. Copepod nauplii and copepodite stages I–III were not eaten by the medusae, neither were fish eggs and larvae used as prey. Based on mean medusa and zooplankton abundance from the upper 20 m, the predatory impact was very low. In August, when mean abundance of A. aurita was highest, only 0.1% of the copepod and 0.5% of the cladoceran standing stock were eaten per day. However, in regions with higher medusa or lower zooplankton abundance, up to 7.9% of the cladoceran standing stock was consumed per day. Hence, A. aurita did not regulate the zooplankton community in the Bornholm Basin, and fish larvae did not suffer from competition with and predation by the medusae.Communicated by O. Kinne, Oldendorf/Luhe     

Spatial and temporal variations in the population size structure of three lanternfishes (Myctophidae) off Oregon,USA

Size-frequency distributions were determined for 3 common lantern-fishes (Stenobrachius leucopsarus, Diaphus theta, and Tarletonbeania crenularis) off Oregon in the summer. The fishes were caught mainly in sound-scattering layers by a large pelagic trawl with 5 opening-closing nets. Changes in depth distribution and diel vertical migration with growth were evident for all 3 species. The size of S. leucopsarus increased markedly with depth both at 0 to 90 m at night and 250 to 500 m during the day. Larger D. theta were also found deeper during the day (between 250 and 450 m), but neither D. theta nor T. crenularis demonstrated size segregation in the upper 90 m at night. Large D. theta and small T. crenularis did not appear to migrate into surface waters at night. Age-Group O (15 to 20 mm) S. leucopsarus were most abundant in deep water (400 to 480 m) in the daytime and did not migrate into near-surface waters at night. Age-Group I (30 to 40 mm) S. leucopsarus were common at about 300 m by day and within the upper 30 m at night. Age-Group II–III (50 to 60 mm) apparently followed the evening ascent of Age-Group I fish and most resided at 75 to 90 m at night, beneath Age-Group I fish. Age-Group III+fish (70 to 80 mm) were associated with Age-Group O at 400 to 480 m by day and usually did not migrate above 200 m at night. The size structure of S. leucopsarus differed among the nets of a single tow at one depth, or between two tows that fished the same depths on successive nights, indicating horizontal patchiness in age structure. D. theta demonstrated low within-tow variability in size composition which indicated a spatially more uniform age structure on a scale of kilometers. The size structures of these 3 lanternfishes were different in the same area and the same season during two different years, suggesting variable survival of year classes or horizontal patchiness of age composition in the area sampled.