Evidence for division of labor in the social caterpillar Eucheira socialis (Lepidoptera: Pieridae)

The caterpillars of Eucheira socialis westwoodi cooperatively spin and maintain a hollow silken nest and an elaborate network of silken foraging trails on their host plant, madrone (Arbutus spp.: Ericaceae). Nests typically contain several hundred larvae. Two populations are known to harbor a sex ratio distorter. The primary sex ratio in these two populations for four generations has been exceedingly male biased (64–79% male). Lepidoptera larvae are easily sexed using external morphology, allowing us to uniquely mark male and female larvae and to assemble larval groups of particular sex ratios. We report here the results of experiments on sex-specific larval behavior and physiology and the effect of colony sex ratio on individual behavior. We found that male larvae spent more time spinning silk on the nest and less time feeding than female larvae. Males were the first to emerge from the nest and the first to venture out along trails to feed. Male-biased nests had a significantly greater amount of silk deposited on their surfaces than female-biased nests. In the field, male-biased nests produced heavier male and female pupae than female-biased nests. Male and female larvae in 75% male nests became active earlier than males and females in other sex ratio treatments. Received: 11 September 1998 / Received in revised form: 24 February 1999 / Accepted: 27 March 1999     

In situ swimming and settlement behaviour of larvae of an Indo-Pacific coral-reef fish, the coral trout Plectropomus leopardus (Pisces: Serranidae)

Late larvae of the serranid coral trout Plectropomus leopardus (Lacepède), captured in light traps, were released during the day both in open water and adjacent to two reefs, and their behaviour was observed by divers at Lizard Island, northern Great Barrier Reef. Coral trout larvae (n = 110) were present in light-trap catches from 18 November to 3 December 1997, including new moon (30 November). The swimming speed of larvae in open water or when swimming away from reefs was significantly greater (mean 17.9 cm s⁻¹) than the speed of larvae swimming towards or over reefs (mean 7.2 cm s⁻¹). Near reefs, larvae swam at average depths of 2.7 to 4.2 m, avoiding 0 to 2 m. In open water, swimming depth varied with location: larvae >1 km east of Lizard Island swam steeply downward to >20 m in 2 to 4 min; larvae >1 km west oscillated between 2.6 and 13 m; larvae 100 to 200 m east of Lizard Island oscillated between 0.8 and 15 m. Nearly all larvae swam directionally in open water and near reefs. In open water, the average swimming direction of all larvae was towards the island, and 80% (4 of 5) swam directionally (p < 0.05, Rayleigh's test). Larvae swam directionally over the reef while looking for settlement sites. The frequency of behaviours by larvae differed between two reefs of different exposure and morphology. Depending on site, 26 to 32% of larvae released adjacent to reefs swam to open water: of these, some initially swam towards or over the reef before swimming offshore. In some cases, offshore-swimming seemed to be due to the presence of predators, but usually no obvious cause was observed. Depending on the reef, 49 to 64% of the larvae settled. Non-predatory reef residents aggressively approached 19% of settlers. Between 5 and 17% of the larvae were eaten while approaching the reef or attempting to settle, primarily by lizardfishes but also by wrasses, groupers and snappers. A higher percentage of larvae settled in the second week of our study than in the first. Average time to settlement was short (138 s ± 33 SE), but some larvae took up to 15 min to settle. Average settlement depth was 7.5 to 9.9 m, and differed between locations. No settlement took place on reef flats or at depths <4.2 m. Larvae did not appear to be selective about settlement substrate, but settled most frequently on live and dead hard coral. Late-stage larvae of coral trout are capable swimmers with considerable control over speed, depth and direction. Habitat selection, avoidance of predators and settlement seem to rely on vision. Received: 7 July 1998 / Accepted: 26 January 1999     

Gene flow and local adaptation in a sunfish-salamander system

There is increasing evidence that populations may not be well adapted to their local environments, and as a result, recent interest has focused on understanding factors that constrain adaptive evolution. This study presents data suggesting gene flow may constrain the ability of larvae of the streamside salamander Ambystoma barbouri to avoid predation by fish via escape behavior and life history tactics. Streamside salamander larvae face conflicting selection pressures in different streams. Some streams are ephemeral, where larvae should be active to feed, grow, and reach metamorphosis before stream drying. Other streams contain predatory fish, where larvae should be generally inactive to avoid predation. Previous work has shown that streamside salamander larvae exhibit ineffective antipredator behavior by having inappropriately high activity levels with fish, resulting in high predation in laboratory and field experiments. This study investigated the possibility that gene flow from larvae in ephemeral habitats may reduce the escape performance of larvae from populations with fish and alter their life history characteristics to increase their susceptibility to fish predation. I assayed escape behavior (speed, acceleration, and duration of escape) and life history characteristics (hatching date, size, stage) associated with predator avoidance among laboratory-reared larvae from four populations. As predicted, two populations (one with fish and the other fishless and ephemeral) connected by gene flow were not significantly different in almost all assays. In contrast, larvae from an isolated population with fish had significantly stronger escape behaviors and delayed hatching than both an isolated population that lacked a history of fish co-occurrence and the population with fish but gene flow from a fishless population. These results support theory suggesting that gene flow can constrain adaptive evolution. Received: 22 February 1999 / Received in revised form: 4 April 1999 / Accepted: 26 April 1999     

Chemical defense of the ladybird beetle Epilachna paenulata

Summary. The defensive chemistry of the ladybird beetle Epilachna paenulata (Coleoptera: Coccinellidae) was characterized as a mixture of piperidine, homotropane and pyrrolidine alkaloids. Whole body extracts of adult beetles contain four major alkaloids: 1-(6-Methyl-2,3,4,5-tetrahydro-pyridin-2-yl)-propan-2-one; 1-(6-methyl-2-piperidyl)-propan-2-one; 9-aza-1-methyl-bicyclo[3.3.1]nonan-3-one and 1- (2′′- hydroxyethyl)-2-(12′-aminotridecyl)-pyrrolidine. Comparative studies of the defensive chemistry of eggs, larvae, pupae and adults showed both qualitative and quantitative differences in alkaloid composition among the four life stages, and also within adult age. Laboratory predation bioassays with wolf spiders showed that the adults are better protected than the larvae and pupae. Field tests showed the adult alkaloid extract to be deterrent to ants.     

Osmoregulation, nutritional effects and buoyancy of marine larval fish: a bioassay for assessing density changes during the earliest life-history stages

It has been hypothesized that marine fish larvae in the advanced stages of starvation would show increased density (ρ = mass volume⁻¹) from water loss due to osmoregulation failure. Changes in larval buoyancy are currently attributed to swim bladder regulation and protein synthesis or catabolism. Osmoregulation-related changes in density is an alternative mechanism, the importance of which remains untested in the laboratory and the influence of which on vertical distributions is unknown. We provide evidence that loss of osmotic control is a plausible mechanism for increased density of larval cod (Gadus morhua L.). Furthermore, our results show that this mechanism is not restricted to larvae in the advanced stages of starvation. “Relative” larval densities are estimated using a modified density gradient. We use a gravimetric method to separate the effects of nutrition from osmoregulation failure. We assessed the importance of sampling strata on estimates of larval density. Proportional sampling within three depth strata (stratified sample) produced the least biased method for determining the “average” density of a population of larvae in laboratory culture. Larvae sampled from the bottom third of the culture tank were significantly more dense then those sampled from the surface. This was true for larvae of all ages. The average change in density from hatching till death from starvation for larvae sampled in the surface stratum was nominal (Δρ = 5.0 × 10⁻⁴ g cm⁻³), while the change for those sampled from the bottom stratum was large (Δρ = 3.8 × 10⁻³ g cm⁻³). These large density differences suggest that larvae sampled from the bottom stratum were either osmotically stressed or were facultatively changing their density via regulatory pathways. Preliminary observations suggest that vitality is lower amongst those larvae which are sampled near the bottom. The small change in average density of larvae sampled from the surface stratum was due to starvation. The density differences we observed between “osmotically stressed” and “starving” larvae could readily have been misconstrued as differences in feeding and growth experienced by individual larvae. The potential bias of increased density from osmoregulation failure must be considered as a factor in experimental designs developed to assess the effect of fed and starved treatments on buoyancy for larvae of all ages. The simple bioassay we describe may prove useful both as a means of assessing larval condition and as a mechanism for evaluating factors affecting larval vertical distributions in the field. Received: 13 January 1997 / Accepted: 3 February 1997     

Sublethal impairment of respiratory control in plaice (Pleuronectes platessa) larvae induced by UV-B radiation, determined using a novel biocybernetical approach

The increasing amount of UV-B radiation (280 to 320 nm) reaching the earth's surface could endanger transparent fish and zooplankton in upper water layers. Previous studies on marine fish eggs and larvae focused mainly on the effects of high, lethal doses of ultraviolet irradiation. Now a sensitive test for accumulated, sublethal impairments of physiological performance has been established. In developing fish larvae (Pleuronectes platessa L.), the respiratory control responses to hypoxia were evaluated by combining a biocybernetical approach and a novel optical test system. Before testing respiratory control behavior, the larvae were irradiated with light containing UV-B (test individuals) or free from it (control individuals). Differences in the physiological reactions of these larvae were detected by biocybernetical analysis (linear systems analysis) of respiratory control. The analysis was based on the evaluation of changes in ventilatory rate caused by sinusoidally modulated partial pressures of ambient oxygen. The ventilatory movements of the larvae were recorded, analyzed and evaluated by video microscopy and digital image processing. After several days of irradiation with moderate UV-B doses or intensities (8.6 kJ m⁻² d⁻¹, corresponding to an irradiance of 0.24 W m⁻² for 10 h), sublethal impairments of respiratory control could be detected in plaice larvae by this method; both the linearity of respiratory control itself and the parameters of linear control behavior were affected by UV-B. Concerning environmental implications, the results must still be interpreted cautiously, because natural irradiation conditions are more complex than the artificial regimen. Our biocybernetical approach will be useful for further studies on sublethal impairments of physiological functions caused by UV-B in fish and other transparent marine or limnic animal groups. In addition, the method could be used to test the sublethal impact of chemical contaminants. Received: 4 November 1997 / Accepted: 6 May 1998     

Mycosporine-like amino acids prevent UVB-induced abnormalities during early development of the green sea urchin Strongylocentrotus droebachiensis

Experiments were performed to determine how ultraviolet radiation (UVR) in the environmentally relevant range affects development of the sea urchin Strongylocentrotus droebachiensis (Müller) and whether mycosporine-like amino acids (MAAs), present in the early life stages, reduce UV-induced damage. Eggs, embryos, and larvae contained five MAAs having absorption maxima ranging from 320 to 334 nm. Eggs contained principally shinorine and porphyra-334, which absorb maximally at 334 nm and half-maximally at 312 and 348 nm, spanning much of the environmental range of biologically effective UVR. Concentrations of MAAs remained constant in unirradiated embryos through the gastrula stage, but decreased significantly in two-armed pluteus larvae. Daily exposure to combined photosynthetically active radiation (PAR, 400–700 nm) and UVR did not affect the concentration of MAAs in these embryos up to the two-armed pluteus stage. Prism larvae of sea urchins and the sand dollar Echinarachnius parma (Lamarck) did not accumulate shinorine from the surrounding seawater. Daily exposure of embryos to UVA (320–400 nm) and UVB (295–320 nm) radiation in the presence of PAR induced delays and abnormalities during development, and removing UVB eliminated this effect. Abnormalities in embryos included thickening of the blastoderm wall, filling of the blastocoel by abnormal cells, exogastrulation, and formation of abnormal spicules. The percentage of embryos that developed normally was lower in batches of embryos exposed to PAR + UVA + UVB, except in embryos from urchins maintained on MAA-rich diets. In all cases, the percentage of PAR + UVA + UVB-exposed embryos that developed normally was positively related to the concentration of MAAs in eggs from which the embryos developed. Thus, the MAAs found in S. droebachiensis embryos protect them against UVB-induced abnormalities during their development to at least the four-armed pluteus larval stage. Received: 8 May 2000 / Accepted: 29 September 2000     

Variability in larval development of the tropical reef fish Pomacentrus amboinensis (Pomacentridae): the parental legacy

The effects of food availability, female size, and social interactions on the quality of Pomacentrus amboinensis larvae at hatching were examined using two field-based experiments. In Experiment 1, food availability and female size significantly influenced size, eye diameter and levels of yolk reserves of larvae at hatching. Small females (47 to 52 mm standard length, SL) whose diets were not supplemented, produced the longest larvae (3.0 ± 0.01 mm total length, TL) with the least yolk reserves (50.1 ± 1.04 μm²). Irrespective of female size, those that received additional food produced larvae with the largest yolk-sacs (large females: 87.60 ± 1.53 μm²; small females: 80.14 ± 1.24 μm²). In Experiment 2, interactions with conspecifics had a greater affect on the somatic development of larvae at hatching than food availability. Increased social interactions resulted in larvae that were ⋍3% longer, with 2% greater head depth, than larvae from females that spawned in isolation on the experimental reefs. Fed females produced larvae with ⋍20% more yolk than larvae from females whose diets were not supplemented. All three factors (food availability, female size, and intensity of social interactions) tested within these experiments vary spatially and temporally among reefs. There is the potential, therefore, for larvae at the onset of the planktonic stage to vary in quality, level of development, and probability of survival. Received: 12 August 1996 / Accepted: 26 August 1996     

Poecilogony in an estuarine opisthobranch: planktotrophy, lecithotrophy, and mixed clutches in a population of the ascoglossan Alderia modesta

A San Diego population of the opisthobranch mollusc Alderia modesta (Lovén, 1844) exhibits poecilogony, the presence of two development modes within a single species. In spring, half of the adults spawned masses containing ∼300 eggs with a mean diameter of 68 μm. After 3 d, these egg masses hatched planktotrophic veligers with a maximum shell dimension of 116 μm. The remaining adults spawned masses containing ∼30 eggs with a mean diameter of 105 μm. These egg masses hatched after 5 to 6 d, releasing lecithotrophic larvae with a maximum shell dimension of 186 μm. About 1% of field-collected adults produced mixed clutches containing a continuum of larval sizes, spanning the size extremes of planktotrophy and lecithotrophy and hatching larvae with a mean maximum shell dimension of 152 μm. Adults producing planktotrophic and lecithotrophic larvae were interfertile, and no hybrid breakdown was observed through the F3 generation. When starved, adults which previously produced only lecithotrophic larvae switched to producing planktotrophic larvae or mixed clutches with both planktotrophic and lecithotrophic larvae. Sequence-polymorphisms from a portion of the mitochondrial cytochrome c oxidase I gene support the conclusion that the two reproductive morphs represent a single species. Most of the lecithotrophic larvae and a small percentage of the larvae from mixed clutches were metamorphically competent within 3 d of hatching. A. modesta is the only molluscan species as yet known to have both planktotrophic and pelagic lecithotrophic development within a single natural population. Received: 14 August 1997 / Accepted: 11 April 1998     

Halogenated metabolites in two marine polychaetes and their planktotrophic and lecithotrophic larvae

This study investigated the occurrence and ontogenetic changes of halogenated secondary metabolites in planktotrophic and lecithotrophic larvae and adults of two common, infaunal polychaetes, Streblospio benedicti (Spionidae) and Capitella sp. I (Capitellidae), with different life-history traits. S. benedicti contains at least 11 chlorinated and brominated hydrocarbons (alkyl halides) while Capitella sp. I contains 3 brominated aromatic compounds. These halogenated metabolites are potential defense compounds benefiting both larvae and adults. We hypothesized that: (1) planktotrophic larvae contain halogenated metabolites because they are not protected by adult defenses, (2) quantitative and qualitative variation of planktotrophic larval halogenated metabolites parallels that of adults, and (3) brooded lecithotrophic larvae initiate the production of halogenated metabolites only after metamorphosis. To address these hypotheses, volatile halogenated compounds from polychaete extracts were separated using capillary gas chromatography and identified and quantified using mass spectrometry with selected ion monitoring. All four life stages (pre- and post-release larvae, new recruits, adults) of both S. benedicti and Capitella sp. I contained the halogenated metabolites previously identified from adults. This is the first report of halocompounds identified and quantified in polychaete larvae. Allocation of potential defense compounds to offspring varied as a function of species, feeding type and developmental stage. Pre-release larvae of S. benedicti with planktotrophic development contained the lowest concentration of total halogenated metabolites (1.75 ± 0.65 ng mm⁻³), post-release and new recruits contained intermediate concentrations (8.29 ± 1.72 and 4.73 ± 2.63 ng mm⁻³, respectively), and planktotrophic adults contained significantly greater amounts (28.9 ± 9.7 ng mm⁻³). This pattern of increasing concentrations with increasing stage of development suggests synthesis of metabolites during development. Lecithotrophic S. benedicti post-release larvae contained the greatest concentrations of halometabolites (71.1 ± 10.6 ng mm⁻³) of all S. benedicti life stages and developmental types examined, while the amount was significantly lower in new recruits (34.0 ± 15.4 ng mm⁻³). This pattern is consistent with a previously proposed hypothesis suggesting a strategy of reducing potential autotoxicity during developmental transitions. Pre-release lecithotrophic larvae of Capitella sp. I contained the highest concentration of total halogenated metabolites (1150 ± 681 ng mm⁻³), whereas the adults contained significantly lower total amounts (126 ± 68 ng mm⁻³). All concentrations of these haloaromatics are above those known to deter predation in previously conducted laboratory and field trials. As a means of conferring higher larval survivorship, lecithotrophic females of both species examined may be expending more energy on chemical defenses than their planktotrophic counterparts by supplying their lecithotrophic embryos with more of these compounds, their precursors, or with energy for their synthesis. This strategy appears common among marine lecithotrophic larval forms. Received: 14 July 1999 / Accepted: 20 January 2000     

Food-particle size and selection by bivalve larvae in a temperate embayment

Epifluorescence microscopy was used to analyze the stomach contents of bivalve larvae collected in the Baie des Chaleurs (western Gulf of St. Lawrence, Canada) in order to document food-particle sizes, compare feeding among taxa, and compare the diet with the in situ phytoplankton community. Stomach contents were mainly composed of small autotrophic flagellates (<5 μm) and cyanobacteria (<2 μm), reflecting the microbial food web which characterizes these waters. More than half (55%) of all veligers examined contained algal cells of 5 to 15 μm, whereas only 3% had cells of 15 to 25 μm. Differences in the size ranges of ingested algal cells among similar-sized larvae of different species suggests that veligers actively selected food particles. Among the smallest veligers (185 to 260 μm), scallops (Placopecten magellicanus) and mussels (Mytilus edulis) ingested more <5 μm and 5 to 15 μm algae than clams (Mya arenaria). Among larger veligers (261 to 405 μm), clams contained significantly more <5 μm cells than mussels, whereas mussels contained significantly more 5 to 15 μm algae than clams. Algal cells of 15 to 25 μm were preferentially ingested by mussel veligers. Feeding also differed between different-sized veligers within taxa, i.e. the smallest clam veligers ingested fewer of 5 to 15 μm algae than the larger size classes. Mussel veligers ingested significantly more 15 to 25 μm and fewer <5 μm cells as their size increased. The dominance of ultraplankton in the nearshore waters of Baie des Chaleurs and in the stomach contents suggests that veliger larvae may be an important export path for carbon produced by small phytoplankton. Received: 17 July 1996 / Accepted: 20 September 1996     

Comparison of somatic growth and otolith increment growth in laboratory-reared larvae of Pacific saury, Cololabis saira, under different temperature conditions

The relationship between somatic growth and incremental growth of otoliths of Pacific saury, Cololabis saira (Brevoort), larvae under different temperature conditions was studied in the laboratory for three age groups (0 to 9, 10 to 20 and 20 to 30 d posthatch). Larvae were incubated from hatching to 9 d at 24, 20, and 16 °C. Further, larvae initially reared at an ambient temperature of 21.7 °C were transferred to experimental temperatures of 22, 18, and 14 °C on Day 10 and reared to Day 20 and similarly from Day 20 and reared to Day 30 posthatch. Growth trajectories of larvae sampled at the end of the three experiments were back-calculated using the biological intercept method and compared to the measured values 0 and 5 d after the start of each experiment. Back-calculated knob length at the different temperatures indicated no significant difference to the measured knob lengths except for the cases at 20 °C from hatching to 9-d-old larvae and at 14 °C from 20- to 30-d-old larvae. The close correlation between somatic and otolith growth shown in this study indicated the feasibility of estimating the growth history of Pacific saury larvae using otolith readings. Received: 14 April 1999 / Accepted: 27 October 1999     

Biological weighting of ultraviolet (280–400 nm) induced mortality in marine zooplankton and fish. I. Atlantic cod (Gadus morhua) eggs

In the Gulf of St. Lawrence, Canada, productivity-determining biophysical interactions occur in the upper 0 to 30 m of the water column. The eggs and larvae of several commercially important marine invertebrates and fishes (e.g. Gadus morhua L.) are found in this layer. Measurements of the diffuse attenuation coefficients for ultraviolet-B radiation (280 to 320 nm, UV-B) at various locations in this geographic region indicated maximum 10% depths (the depth to which 10% of the surface energy penetrates at a given wavelength) of 3 to 4 m at a wavelength of 310 nm. This represents a significant percentage of the summer mixed-layer water column: organisms residing in this layer are exposed to UV-B radiation. Laboratory experiments using a Xenon-arc-lamp based solar simulator revealed that cod embryos exposed to UV-B exhibited high wavelength-dependent mortality. The strongest effects occurred under exposures to wavelengths below 312 nm. This susceptibility was also dependent upon developmental stage; mortality was particularly high during gastrulation. At the shorter wavelengths (<305 nm) UV-B-induced mortality was strongly dose-dependent, and not significantly influenced by dose-rate. The biological weighting function (BWF) derived for UV-B-induced mortality in cod eggs is similar to that reported for naked DNA – suggesting that the mortality is a direct result of DNA damage. There was no evidence of a detrimental effect of ultraviolet-A radiation (320 to 400 nm). Calculations based upon the BWF indicate that, under current noon surface irradiance, 50% of cod eggs located at or very near (within 10 cm) the ocean surface will be dead after 42 h of exposure. Under solar spectral irradiance simulating a 20% decrease in ozone layer thickness, this time drops to 32 h. These are first-order estimates based upon surface irradiance taken at a time of day during which the values would be maximal. Nonetheless, they illustrate the relative changes in UV-B impacts that will result from ozone layer depletions expected over the coming decades. It is also important to point out that variability in cloud cover, water quality, and vertical distribution and displacement of cod eggs and larvae within the mixed layer, can all have a greater effect on the flux of UV-B radiation to which fish eggs are exposed than will ozone layer depletion at these latitudes. Received: 2 March 1998 / Accepted: 18 December 1998     

Effect of Ca2+ on survival and development of metamorphosing bonefish (Albula sp.) leptocephali

Bonefish (Albula sp.) larvae (leptocephali) from the Gulf of California complete metamorphosis in ˜10 d in natural seawater (35‰S; Ca²⁺ conc = 10.5 mM). The increase in ossification that occurs near the end of the non-feeding metamorphic period, in addition to the ability of larvae to complete metamorphosis in dilute seawater (8‰ S) prompted the present study, where the effects of varying the external calcium ion concentration, [Ca²⁺]_e, of artificial seawater (ASW) on the survival, development and internal (whole-body) calcium ion content, (Ca²⁺)_i, of unfed metamorphosing larvae were investigated. Early-metamorphosing larvae placed in␣ASW, where [Ca²⁺]_e = 10.1 mM, survived for up to 10 d and developed normally without exogenous nutrients. In shorter-term experiments (4 to 5 d), no differences in survival were found for larvae in ASW with [Ca²⁺]_e rang-ing from 1.5 to 10.1 mM. However, in Ca²⁺-free ASW, most larvae died within 27 h and no larvae survived more than 42 h; the median lethal time (LT₅₀), and its 95% confidence limits, were 14.5 (10.0 to 20.9) h. High mortality (81% after 20 h) also occurred in 1.0 mM Ca²⁺ ASW, but 2 of 16 larvae tested survived for 96 h. The 96 h median tolerance limit (TL_M), corrected for control mortality, was 1.2 mM Ca²⁺. In natural seawater, larval (Ca²⁺)_i remained relatively constant ( = 0.419 mg larva⁻¹)␣in early- and intermediate-metamorphosing larvae, and then increased to a mean value of 0.739 mg larva⁻¹ in advanced larvae, indicating that Ca²⁺ was␣taken up from the medium at this stage; the increase in (Ca²⁺)_i corresponded to the period of ossification of the vertebral column. Internal (whole-body) magnesium ion content (Mg²⁺)_i showed no significant change during metamorphosis ( = 0.089 mg larva⁻¹). No significant differences in (Ca²⁺)_i were found in advanced larvae in natural seawater and those in ASW, with [Ca²⁺]_e ranging from 2.0 to 10.1 mM. However, clearing and staining revealed that ossification of the vertebral column had not yet occurred in advanced larvae from 2.0 to 10.1 mM Ca²⁺ ASW. Also, low [Ca²⁺]_e (1.0 to 2.0 mM) usually produced deformed larvae that swam erratically, at times showing “whirling” behavior. Received: 21 August 1996 / Accepted: 26 August 1996     

Microgeographic genetic differentiation in a colonial ascidian (Botryllus schlosseri) population

Limited gene flow via the restricted dispersal of larvae and gametes is expected to result in the genetic differentiation of populations of clonal invertebrates on small spatial scales. However, occasional dispersal events over greater distances may generate sufficient gene flow to maintain genetic homogeneity. We applied a spatial autocorrelation approach that does not require a priori definitions of subdivision boundaries to examine genetic differentiation within a continuous population of the colonial ascidian Botryllus schlosseri (Pallas) at two allozyme and five polychromatism loci. Colonies were sampled in July 1992, on a 12 by 18 m grid superimposed on a shallow subtidal (1 to 3 m) population in the Damariscotta River estuary in Maine, USA. Low but significant levels of positive autocorrelation were detected over very small spatial scales (<5 m), with negative autocorrelation occurring on larger scales (>8 m). This pattern indicates significant genetic differentiation over distances of 8 to 21 m, and is consistent with genetic drift and inbreeding creating small scale genetic structure. Received: 18 October 1999 / Accepted: 11 July 2000     

Effect of development rate on the swimming, escape responses, and morphology of yolk-sac stage larval American plaice, Hippoglossoides platessoides

To examine the impact of development rate on swimming performance, escape response, and morphology, yolk-sac larvae of American plaice (Hippoglossoides platessoides, Fabricius) were reared at two temperatures (5 and 10 °C). Videomicroscopy and silhouette collimation videography were used to examine swimming, escape behaviour, and morphology (standard length, finfold area, and yolk-sac area) of individual larvae. Larvae were examined from 0 d post hatch (dph) to 14 dph for the 5 °C treatment group and from 0 to 6 dph for the 10 °C treatment group (3 August to 17 August 1996). Since larvae were not fed, yolk-sac reserves were essentially exhausted by 14 and 6 dph for the 5 and 10 °C treatment groups, respectively. To control for the effect of testing temperature on behaviour, larvae from each temperature treatment were tested at both 5 and 10 °C. Testing temperature had an effect on some swimming parameters but not on escape response. Swimming performance, escape response, and morphology varied with age, while only morphology and escape response varied with development rate. Morphology and swimming performance, and morphology and escape response were found to be correlated as determined by canonical correlation. This study suggests that both types of swimming behaviours should be examined when developing models of the impacts of predation on the early life history of larval fish. Received: 13 September 1999 / Accepted: 21 June 2000     

Effect of temperature on viability and axial muscle development in embryos and yolk sac larvae of the Northeast Arctic cod (Gadus morhua)

Cod (Gadus morhua L.) eggs may develop and hatch within temperatures of −1.5 to 12 °C, but little is known about the effects of very low temperatures on larval characteristics. Eggs of the Northeast Arctic cod (Gadus morhua) were incubated at 1, 5 or 8 °C from Day 1 after fertilisation until hatching, and transferred to 5 °C after hatching. Histological samples of the axial musculature were taken at hatching and 5 d after hatching, and the data on muscle cellularity from these samples were related to survival and hatching, size, developmental data and viability of the yolk sac larvae. All larvae hatched at the same developmental stage. Incubation of eggs at 1 °C produced shorter larvae with a larger yolk sac and more, small deep fibres at hatching than larvae from eggs incubated at 5 or 8 °C. The larval size difference was still present 5 d after hatching, a time at which the larvae from 1 °C-incubated eggs were less developed and less resistant to an acute viability stress test (65 ppt salinity). Although there were no differences between temperature groups in number and size of muscle fibres 5 d after hatching, the deep fibres of the 1 °C-group contained less myofibrils than the two other groups. The phenotype of the larvae at hatching was thus affected within these incubation temperatures. Although all groups were transferred to the same temperature after hatching, the lowest egg incubation temperature (1 °C) still had a negative effect 5 d after hatching, as these larvae were both smaller, less resistant to stress and had less functional muscles at the time of first feeding. Our conclusion is therefore that 1 °C is close to, or below, the lower thermal tolerance limit for normal functional development of Northeast Arctic cod. The results are discussed in relation to larval viability and recruitment of this species in the wild. Received: 4 February 1998 / Accepted: 10 July 1998     

Effect of amino acids on the swimming activity of newly hatched turbot larvae (Scophthalmus maximus)

The swimming behaviour of newly hatched turbot (Scophthalmus maximus L.) larvae was observed in artificial seawater (ASW) and in solutions of 21 l-amino acids at a concentration of 10⁻⁵  M. The behaviour of 20 larvae was analysed in each solution. Each larva was observed for 1 min. Individual movements were recorded on video and analysed using a computer-assisted program. The larvae swam in convoluted, randomised three-dimensional paths, rested and started swimming again. There were large variations in the swimming behaviour of turbot larvae during ontogeny. In ASW the mean frequency of trajectories longer than a body length of 4 mm larva⁻¹ min⁻¹ increased from 1.2 at Day 1, to 10 at Day 4. Analysing the data (Dunnett's method) revealed that the frequency of swimming trajectories increased in the presence of glycine, histidine and glutamine, and decreased in the presence of proline. The total distance swum increased for glycine but decreased for proline. The threshold concentration for glycine detected by turbot larvae was 10⁻⁵  M. The straightness index did not change in the presence of the amino acids. The possible role of these changes in behaviour is discussed. Received: 12 June 1997 / Accepted: 13 January 1998     

Asymptotic growth trajectories of larval sardine (Sardinops melanostictus ) in the coastal waters off western Japan

Growth trajectories of individual larvae of Japanese sardine, Sardinops melanostictus, caught in the coastal waters off western Japan were back-calculated from the first feeding stage up to date of capture (approximate size of 20 to 35 mm total length; TL) based on individually determined allometric relationships between otolith daily ring radii and fish total lengths. The larvae in January-, February-, and March-hatched cohorts in the coastal waters grew faster and more uniformly than those in the oceanic waters offshore of the Kuroshio current. Growth trajectories of the three hatch-month cohorts were similar and could be expressed by the Gompertz model. The inflection points of the growth curves were reached at 9 to 11 d after hatching, when larvae were 10.8 to 11.8 mm TL. Maximum growth rates at these points were 0.80 to 0.85 mm d⁻¹. Growth rates gradually declined after the inflection points, and larval TLs converged into the infinite length of 29 to 32 mm, the sizes at which metamorphosis from larvae to juveniles is initiated. This asymptotic growth pattern in the larval stage resulted in the narrow ranges in TLs in spite of the wide range of ages of the larvae caught by boat seiners in the coastal waters. Slow growth and therefore long duration of the metamorphosing stage could be influential in determining the cumulative total mortality in the early life stages of the Japanese sardine. Received: 14 July 1996 / Accepted: 20 August 1996