INTRODUCTION

Orientation is crucial for almost all organisms, from bacteria to mammals, for selecting habitat or feeding grounds, deciding up from down, or just keeping track of your position, or your neighbour's. The marine environment is an especially rich source of material, particularly for the study of migration, where many animals cover enormous distances, and appear do this with reference to very few sensory cues. The mechanisms underlying these processes, and the benefits they bring to the organism, are only now beginning to be understood.

The Haptophyta comprises a group of microalgae of particular importance in marine habitats, often occurring in ‘bloom’ concentrations, sometimes with devastating effects where the bloom is composed of species toxic to other forms of life. The most familiar species are the coccolithophorids, unicellular organisms encased in calcified scale-like structures, the coccoliths, which are readily preserved in marine sediments and have for a long time been important indicators in micropalaeontological studies. In the middle of this century it was recognized that there was a need to compile and standardize the terminology used in coccolith morphology (Braarud et al., 1955; Halldal & Markali, 1955). This approach was continued by several authors (e.g. Hay et al., 1966; Okada & McIntyre, 1977; Tappan, 1980; Perch-Nielsen, 1985) in published articles, and in the report from a Round Table session at the Rome 1970 Plankton Conference (Farinacci, 1971), which included terms from both fossil and extant taxa. Over the last two decades many new terms have been introduced as observations on coccolith morphology have improved through the use of the electron microscope, and recent glossaries covering various aspects of haptophyte terminology have been published by Heimdal (1993), Kleijne (1993) and Margulis et al. (1993).

Using depth-telemetering, sonic tags orally inserted into the stomachs of Gadus morhua (Pisces: Gadiformes) found over-wintering in Trinity Bay, Newfoundland, the movements of individual fish were observed as surface waters warmed in early spring. Physiological measurements (antifreeze protein levels in the blood) indicate that many cold-adapted, bay cod change their thermal regime at this time. Fish acclimatized to subzero water temperatures enter the newly-formed thermocline and become available to a cod trap fishery. Most sonically-tagged fish resided in 0–1°C waters along the shoreline. Tracking data confirmed indications from declining antifreeze protein levels that cold-adapted cod, having moved into shallow waters in early spring, do not return to deeper, subzero-temperature waters for any appreciable time. At night some cod swam pelagically near the surface. Fish moved at times in the same direction as the tidal current, but ground speeds were several times greater than current velocities. Nocturnal pelagic swimming was also observed during the summer when temperatures within the thermocline exceeded 10°C.

Vertical migration interacts with oscillatory tidal flows to produce horizontal transport. When the vertical migration period is an exact multiple of the tidal period then net uni-directional horizontal transport can occur. This happens most effectively in ‘selective tidal-stream transport’ for which vertical migration into the flow is specifically synchronized with the dominant (usually M2) tidal constituent. When migration and tide are not synchronized there is no net transport and, instead, a horizontal displacement oscillation of the organism takes place at the beat period between the vertical migration and tidal periods. The most common form of vertical migration is synchronized, not with the tide, but with the day-night cycle. Diel migration in M2-period tidal currents induces no net transport, but can produce excursions of several tens of kilometres over just a few days. Diel vertical migration can, however, interact with the sun-generated part of the semi-diurnal tide to produce net horizontal transport, the direction of which is controlled by the phase of the S2 tidal currents. The spatial distribution of phase implies that regions of horizontal convergence and divergence will result from diel migration. Diel migration in diurnal-period tides causes the preferred direction of transport to reverse at six-monthly intervals.

The chromosomes of Platynereis dumerilii (Audouin & Milne-Edwards) are described here for the first time. A modal chromosome number of 2n=28 was recorded, based on counts conducted on metaphase spreads prepared from 24-h-old larvae. The karyotype comprises seven pairs, each of relatively large median (arm ratio, p/q=l.00–0.59) and submedian (arm ratio, 0.59–0.33) chromosomes. Attempts were also made to band the chromosomes using C-banding and silver staining methods. C-band-positive regions were localized on four chromosome pairs (three median, one submedian). A further two chromosome pairs (both median) were observed to have terminal nucleolar organizer regions (NORs). To our knowledge, this is the first time that these banding methods have been successfully applied to any polychaete species. These results are discussed in relation to the karyotypic variation within the class Polychaeta, and the family Nereidae in particular. A stable karyotype consisting of a moderate number of large and morphologically well-differentiated chromosomes, coupled with the ease of culture under laboratory conditions and short generation time, suggests that P. dumerilii is a potentially suitable model for evaluating marine contaminants for genotoxic activity.

The seasonal development of the microbial food web in eastern McMurdo Sound, Antarctica, was investigated during and immediately after the 1990–1991 bloom of Phaeocystis sp. (Prymnesiophyceae: Prymnesiales). From 23 November to 7 December, which was before the appearance of macroscopic colonies of Phaeocystis, both phytoplankton and Protozoa were low in abundance. During the Phaeocystis bloom (~10 December to 7 January), phytoplankton biomass was high and was dominated by colonial and singlecelled Phaeocystis, but other phytoplankton taxa, including diatoms and photosynthetic dinoflagellates, co-occurred. Heterotrophic nanoplankton and protozoan microzooplankton increased dramatically in biomass during the bloom. Non-thecate heterotrophic dinoflagellates were the most abundant microzooplankters. There was no evidence for inhibition of the microbial food web by Phaeocystis. By the post-bloom period, macroscopic colonies were rare and phytoplankton biomass had declined markedly. Microzooplankton biomass continued to rise until the end of the sampling period (23 January). Microfaecal pellets (median size range, 10–30 μm) were abundant during the bloom and post-bloom periods. In the post-bloom period, the heterotrophic protistan assemblage became very diverse, with numerous trophic linkages within the microbial food web. The abundance and diversity of the heterotrophic protist assemblage suggests that there was little control of protists by metazoans and that the microbial food web, consisting of bacteria, algae and Protozoa, was poorly coupled to metazoan zooplankton.

The reproductive biology of 1124 male (165–760 mm TL) and 1476 female (175–898 mm TL) black dogfish, Centroscyllium fabricii, was examined. The sharks were caught off western Greenland with bottom otter trawls. Size at maturity was about 550 mm TL in males and 650 mm TL in females. Ovarian ova did not continue to develop during gestation. The sex ratio of embryos was 1:1. Litter size ranged from 4 to 40, with a mean of 16–4. The total number of mature ova, fertilized ova, or embryos tended to increase with size of the females. Development of black dogfish embryos is dependent solely on yolk reserves. Near-term embryos (152–192 mm TL) had completely absorbed their external yolk sacs. The smallest free-living specimens collected in this study were 165 mm TL for males and 175 mm TL for females. The black dogfish does not seem to have a well-defined breeding season. This species displayed a well-defined pattern of depth segregation by size. The sex ratios in five shallow depth strata (500–599, 600–699, 700–799,800–899, and 900–999 m) were 1:1, but those of three deeper depth strata (1000–1099,1100–1199, and 1200–1299 m) contained a significantly greater number of females than males. Four abnormal hermaphrodites possessed gonads containing both testicular and ovarian tis-sues. Two specimens had normal female reproductive organs (e.g. ovaries, oviducts, nidamental glands, and uteri) as well as claspers.

The fish assemblages of rocky intertidal and subtidal (<3 m depth) habitats of the south-western Portuguese coast were sampled between May 1992 and November 1993, using traps, angling and hand-netting, to investigate their structure and seasonal dynamics. The main purpose of the study was to describe the patterns of prey availability for marine-feeding otters (Lutra lutra L., Mammalia) in south-western Portugal, and to compare these with other areas in Europe. A total of 475 fish of 14 species was captured over 150 d of trapping (788 trap-days), and 1575 fish of 15 species were captured over 24 d of angling (76.3h). Hand-netting was carried out occasionally, to supplement the observations from the other methods, and only 96 fish of seven species were captured. The three methods yielded different patterns of the fish assemblage: giant gobies (Gobius cobitis Pallas, 33.7%) and rock gobies (G. paganellus L., 22.9%) were the most frequently trapped fish; corkwing wrasse (Symphodus melons (L.), 81.2%) were the usual species caught by angling; shannies (Lipophrys pholis (L.), 42.7%) and Montagu's blennies (Coryphoblennius galerita (L.), 28.1%) were the most frequently netted species. Conger eels (Conger conger (L.)), giant gobies and shannies showed an uneven distribution along the coast, occurring most frequently in schist shelves, rather than in areas of large sandstone boulders. There was a large seasonal variation, with the catches of most species peaking in winter and early spring, including the majority of those most important in the diet of otters: conger eels, shore rocklings (Gaidropsarus mediterraneus L.), rock gobies and corkwing wrasse. This was also the period when the individuals of several species were heaviest.

This paper describes the random amplification of polymorphic DNA markers (RAPDs) in Lasaea rubra (Erycinidae: Bivalvia). Present evidence suggests that L. rubra is an asexual species; however, the exact mode of clonal reproduction in this species is still a matter of debate. In this preliminary study, four of the primers used generated polymorphic RAPDs. One primer was able to distinguish between individuals from the same or different crevice population. This same primer also resolved a single band difference between otherwise identical RAPD patterns of a parent and its offspring. No familial differences have been detected in several previous studies using allozyme electrophoresis. This paper suggests that many polymorphic markers could be obtained with this species using the RAPD technique. Population genetic analysis of L. rubra has long been hampered by a dearth of polymorphic markers due to its small size. These findings suggest that this technique has the potential to further the study of population genetics in this asexual species.

Proximate aspects of diel vertical migration in the freshwater and marine environment are compared using data from the literature. Examples of migrations in both environments are presented, from which it is concluded that relative changes in light intensity before sunrise and after sunset are primary causes of migrations. Experiments have shown that photoreactive behaviour is enhanced in the presence of predators but inhibited by shortage of food. These factors are called secondary causal factors. A hierarchy of causal factors is proposed. In lakes fish exudates suffice but in marine biotopes like bays, it is possible that fish have to be actually present for enhancement to take effect. To what extent the presented stimulus-response mechanism holds for mesopelagic animals in oceans is discussed on the basis of vertical distributions of euphausiids.

Collecting water from rock pools just above rising tides, usually in winter and spring, yielded 1412 Lipophrys pholis, 255 Pholis gunnellus and 29 Coryphoblennius galerita, bearing respectively 172, 8 and 1 Oceanobdella blennii, 744 Taurulus bubalis bearing six Oceanobdella microstoma and 14 Sanguinothus pinnarum, and a Myoxocephalus scorpius with a single O. microstoma; fishes of 23 other species totalled 1243, but none bore leeches externally. Oceanobdella blennii was found mainly in February and March (some in gill chambers), at temperatures generally below 8°C, with lower numbers (all external) in April and May and none later. Infestation was highest in Northumberland. Leeches on P. gunnellus were smaller than those on L. pholis, which they would not attack. They were confined to Northumberland and Scotland, where L. pholis was scarcer. The southern limit of O. blennii seemed to be in north Cornwall, but O. microstoma, though more stenothermal, extends from the Arctic to south Devon, its hosts living mostly subtidally. It is hidden under the chin of Cottidae, whereas S. pinnarum is easily seen on the fins. Abundance in south-west Britain of the ‘cleaner fish’ Crenilabrus melops may explain why S. pinnarum, though common in Scotland and tolerant of summer temperatures, is scarce in Anglesey and not found further south.

Acetylcholinesterase (AChE) activity, a marker of neural differentiation, was histo-chemically localized in embryos and larvae of Mytilus galloprovincialis Lamarck and M. edulis L. (Mollusca: Bivalvia). The results show that: (1) the first AChE-active cells develop as two bilaterally symmetrical, postero-dorsal pioneer sensory neurons and their supporting superficial cells; (2) the pathways of their pioneering longitudinal axons are the same as those of two bilaterally symmetrical nerve cords which differentiate later in a postero-anterior direction; (3) in the conchostoma larva the posterior neurons and their associated superficial AChE-active cells give rise to two ciliary sensory-like organs which are the earliest rudiments of pedal ganglia and primary byssus glands; (4) three pairs of posttrochal gangliar rudiments, most probably visceral, parietal and pleural ganglia, develop on the same longitudinal nerve cords in the trochophore larva; (5) starting from the veliger stage, the presumptive foot rotates on the ventral side and migrates forwards, so that this gangliar chain is bent and the viscero-parietal ganglia become localized dorso-posteriorly to the pedal ganglia; (6) the cerebral ganglia differentiate together with the apical organ and the nerve network of the velum after the first pedal nerve rudiments have been detected.

Atlantic mackerel (Scomber scombrus L.) were sampled by pelagic trawl over the Porcupine Bank to the west of Ireland between 52°N and 54°N at peak spawning time, between 27 May and 10 June 1992. Distribution of juveniles (age classes 1 and 2) was inversely related to depth, comprising over 90% of hauls close to the coast of Ireland and 20% of the catch on the Porcupine Bank. Egg production increased westwards and was high along the continental shelf edge. Adult spawning fish were found throughout the shelf area and beyond, into waters over 3000 m deep. The results indicate that mackerel spawn over deep water some distance from the shelf edge and may exploit oceanic food resources from deep water.

Late stage juveniles and adults of Callinectes sapidus in Chesapeake Bay, USA, and Maja squinado off the Ria de Arousa, Spain, were compared for ontogenetic changes in movement patterns (speed, distance, orientation) and habitat selection (depth, substrate) using ultra-sonic telemetry and published information. After settling in submerged grass beds in the lower Bay, 20-mm juvenile C. sapidus disperse long distances into low salinity sub-estuaries to feed and grow to maturity in two years. Within the Rhode River sub-estuary, juvenile C. sapidus moved with a mean speed of 12 m h1 in nearshore shallows (1·1 m); whereas adults averaged 24 m h·1 in the deeper (2·9 m) channel areas and moved freely in and out of the main estuary. Individuals of both life stages exhibited a pattern of slow meandering (juveniles, 2 m h1, adults 10 m h·1) within a limited area, alternating with faster, directionally-oriented movement (both stages >50 m h·1) between meandering sites. Juvenile and adult males over winter in deeper water nearby, while inseminated females migrate long distances into high salinity areas to incubate the eggs. Callinectes sapidus completes the migration cycle only once per 2·5-y generation. Maja squinado settles on rocks in shallow kelp forests in the coastal zone, where they grow to maturity in 2 y. Juveniles moved slowly (0·5 m h·1) while meandering without directional orientation on shallow (4 m) small patch reefs during summer. After the pubertal moult in summer, adults also meandered slowly (1 m h·1) mostly on rocks at slightly greater depth (7 m). In late summer and autumn, newly mature and older adults moved with directional orientation into deeper (10·40 m) water for the winter, until migrating back to the shallows for the summer; whereas juveniles remained inshore on rocks for the winter. Adult M. squinado live several years after puberty and complete the seasonal migratory cycle several times during their lives.

A single-season mark-recapture experiment was carried out on queen scallops, Aequipecten opercularis (L.) (Bivalvia: Pectinidae) to assess the suitability of the technique for estimating population sizes and mortality rates of exploited populations. Tag-loss and tagging-induced mortality were 7.7% and 11.9% respectively. Non-reporting was negligible.

In a single fishing season (June-December 1989) 25.4% of releases were recaptured alive. Mean density of commercial-sized queens was estimated as 0.47 m−2 over the fishing ground. Instantaneous rates of total (Z), fishing (F) and-natural (M) mortality, calculated from the monthly decline of recaptures during the fishing season, were 0.;41,0.21 and 0.20 per month respectively. The calculated M is very high but includes non-yield or indirect fishing mortality. Analysis of the age-frequency distribution of queens from an adjacent, largely unexploited, area indicated an M of only 0.036 per month (0.43 y−1). There was some indication of an increase in M in older year classes, associated with senescence.

The mark-recapture methodology used here is suitable for this species, but single-release experiments place too much reliance on commercial catch-statistics.

Age, growth and maturation of Loligo vulgaris were studied by examination of growth increments within statoliths of 294 specimens (mantle length, ML, ranging from 31 to 498 mm) caught on the west Saharan shelf between 1985 and 1988. Maximum age was 335 d (290 mm ML) for females, and 396 d (498 mm ML) for males. Growth rates varied considerably among individuals with a greater range in males. Sexual dimorphism in length was apparent after about 210 d. Males and females diverged considerably in weight, with males reaching a greater weight after about 180 d. Growth in length between 124 and 396 d was best described by a power function, while growth in weight was best described by the Gompertz function. Males started maturing at 180–210 d and mature males ranged in age between 250 and 396 d; while females started maturing at 240–270 d and mature females ranged between 285 and 335 d. Loligo vulgaris hatched throughout the year with two distinct peaks; in winter (December - early March) and summer (June-July). The life cycle of L. vulgaris populations on the west Saharan shelf lasts ~1 y, with large males (>450 mm ML) living slightly longer.

A dense assemblage of the brachiopod Magellania fragilis was sampled by trawl and underwater photography during the expedition ANT IX/3 (1991) of RV ‘Polarstern’ on the shelf of the Lazarev Sea, Antarctica. Mean abundance and biomass estimates for M. fragilis were 26.15 individuals m2 and 1.13 g AFDM m2, respectively. Growth bands visible on the shell were interpreted as annual growth marks caused by the strong seasonality of food input to the benthos and were treated as size-at-age data. The von Bertalanffy growth function Lt (mm) = 51.67 (1 - e0020 (t +1.326))3.828 described these data best. The annual somatic P/B ratio was very low, 0.046 y1, and annual production amounted to 0.052 g AFDM m2 y1 at this particular site. These results indicate that M. fragilis is a comparatively slow-growing species with very low annual productivity.

The shrimp, Crangon crangon (L.) (Crustacea: Crangonidae), is a significant predator of the smallest sizes of plaice, Pleuronectes platessa L. (Teleostei: Pleuronectidae), during and immediately after the fish settle on sandy beaches when predation rate is strongly dependent on the size of both the predator and the prey. Laboratory experiments showed that this size-dependency is caused principally by the superior escape capabilities of larger fish once captured rather than differences in the ability of different sizes of shrimps to capture their prey. Fish that escape after capture are often wounded and some of these wounds may subsequently be fatal. Many shrimps capture and eat fish that are larger than their stomach volume resulting in long handling times and low prey profitabilities. For all sizes of shrimps used (36–65 mm total length) prey profitability (mg prey ingested min−1) increases with decreasing fish length.

INTRODUCTION

Individuals of the Antarctic krill Euphausia superba Dana assemble in spring into dense formations (estimated 20,000 m3) which swim for long distances and at considerable speed (estimated 20 cm s−1) to search for patches of phytoplankton (Hamner, 1984). Recruitment to schools and maintenance of a defined position in a travelling formation, a remarkable social behaviour, requires some kind of communication system between individuals.

Vision, as a sophisticated image-processing system, habituating strongly with time, appears inappropriate for the task of continuously monitoring position in the formation. This is not in contradiction of the fact that the prominent eyes of Euphausia orient the shrimp with respect to the axis of light coming from above, and direct the photophores 180° away from light (Land, 1980). The function of the photophores is only partially explained (e.g. predator avoidance by counter-shading, Grinnell et al., 1988), their role in formation swimming being doubtful because they are reportedly active only during dawn and dusk. Of course the eyes help Euphausia to assemble in schools and to evade predators (Strand & Hamner, 1990) as well as fishing nets (Everson & Bone, 1986a,b), but a simple mechanical reflex seems more suitable to control individual position in the formation.