Production, Quality, and Low Temperature Incubation of Eggs of Atlantic Cod Gadus morhua and Haddock Melanogrammus aeglefinus in Captivity

Atlantic cod Gadus morhua and haddock Melanogrammus aeglefinus broodstock maintained under altered regimens of temperature and photoperiod spawned up to 8 mo per year. The cod broodstock produced viable embryos from October through June. The haddock broodstock produced viable embryos from December through May. Egg diameters were largest during the middle of the spawning season when water temperature was at a minimum, resulting in an inverse relationship between egg diameter and water temperature in both species. Egg quality was high, as evaluated by buoyancy, fertilization rate, regularity of early cleavage, and percent viable hatch. Low temperature incubation of cod and haddock eggs extended the embryonic period. Cod embryos tolerated a wider range of temperatures than haddock. High mortality (1 90%) was observed before hatching in haddock embryos incubated at 1 C. Atlantic cod embryos hatched at temperatures as low as —1 C, extending the embryonic period to 59 d. At 8 C Atlantic cod and haddock embryos hatched in 11–12 d. To determine if extending the embryo incubation time by using low temperatures had a detrimental effect, embryos were incubated through hatch at either 1 C or 6 C, and the larvae from both groups reared at 6 C. Growth and early survival of larvae were comparable in both treatments.     

Spawning dynamics of American shad (Alosa sapidissima) in the St. Lawrence River,Canada–USA

Maltais E, Daigle G, Colbeck G, Dodson JJ. Spawning dynamics of American shad (Alosa sapidissima) in the St. Lawrence River, Canada–USA.
Ecology of Freshwater Fish 2010: 19: 586–594. © 2010 John Wiley & Sons A/S Abstract – The most northerly population of American shad (Alosa sapidissima), located in the St. Lawrence River, is considered vulnerable because of low population abundance and limited spawning habitat located at the upstream extent of the population’s anadromous migration. Here, we aimed to establish the temporal and spatial extent of spawning based on a novel hatch‐date analysis of juveniles. Spawning activity lasted from early May to early July. We found that juveniles captured downstream during the summer hatched later in the year than those captured further upstream. As a result, younger juveniles were distributed somewhat further downstream. In addition, we found significant multimodality in hatch‐date distributions at midstream and downstream sampling stations. Together, these results provide evidence that the 2‐month spawning period involved numerous spawning events that progressed in a downstream direction as the season advanced, rather than being restricted to upstream sites over the spawning season.     

Stability in reproductive timing and habitat usage of Chinook salmon across six years of varying environmental conditions and abundance

The spatial and temporal distribution of spawning activity by autumn‐run Chinook salmon Oncorhynchus tshawytscha (Walbaum) was examined across multiple years. The study period included two years of extreme drought conditions when water temperatures in the spawning reach of the Stanislaus River were considered sub‐optimal for spawning and egg incubation. Despite varying levels of superimposition, redd counts and densities remained stable and positively associated with river location, indicating that superimposition may be driven by habitat preference rather than the absence of suitable spawning locations. Spawning occurred slightly later during drought years (6–10 days later compared with wetter years). This delay was attributable to deferred migration instead of deferred spawning, and the majority of redds were constructed at water temperatures exceeding the optimal temperature range. As a consequence, estimates of juvenile production during 2014 and 2015 were among the lowest on record. These findings may be related in part to the high hatchery contribution to the population, above‐average temperatures during spawning and incubation, and superimposition rates. Management recommendations include adequate cold‐water storage in the upstream reservoir, refined spawning habitat restoration techniques in the light of superimposition rates and, on a broader system scale, actions that reduce the amount of stray hatchery Chinook salmon.     

Chilled storage of white shrimp (Litopenaeus vannamei) spermatophores

Chilled storage of spermatophores from white shrimp (Litopenaeus vannamei) is needed to generate a consistent and reliable supply of spermatozoa for domestication purposes. The objective of this study was to develop a protocol for the chilled storage of white shrimp spermatophores and to evaluate bacterial propagation during such storage. In the first experiment, spermatophores were immersed in four extenders, mineral oil, Ringer's solution, phosphate buffer and 0.85% NaCl, and stored at low temperature (2-4 °C) for 35 days. Characteristics of preserved spermatophores changed the least and viable sperm was highest when spermatophores were stored in mineral oil. Spermatophores preserved with mineral oil appeared morphologically normal. Bacillus circulans, Staphylococcus hominis and S. lugdunensis, S. sciuri, S. xylosus and Micrococcus spp. were identified as the predominant bacteria during chilled storage, and total bacterial counts gradually increased during the experiment. A second experiment investigated the effect of antibiotic on chilled storage. Spermatophores were preserved in only mineral oil or mineral oil with 0.1% penicillin-streptomycin. These were evaluated for changes in external morphology of spermatophores, sperm viability and total bacteria count every week during a 35-day experimental period. Percentages of viable sperm (69.5 ± 3.9%) were significantly higher (P < 0.05) among spermatophores preserved in mineral oil with 0.1% antibiotic compared with those preserved only in mineral oil (57.7 ± 3.4%) over 35 days. The number of total bacteria in the treatment with mineral oil ranged between 28.3 ± 4.8 and 2416.7 ± 299.4 CFU/g, but in mineral oil containing antibiotic bacteria were undetectable. This study suggests that chilled storage of spermatophores is a feasible approach for the management and spawning of white shrimp broodstock.     

Towards development of large-scale hatchery cultivation of larvae and spat of the pearl oyster Pinctada mazatlanica in Mexico

The results of a series of pilot-scale runs with P. mazatlanica larvae from 2004 through 2006 are reported. Preliminary runs in 2004 and 2005 used broodstock collected in summer, when massive spawning of wild populations naturally occurs. However, results of larval development were very poor and failed to produce spat in both years. In 2006, ripe broodstock were still collected in summer, but also in the spring time, based on the hypothesis that the gonads in this season were in better reproductive condition that in summer. Three larval runs were conducted in 2006: two in spring and one in summer. Larvae growth and survival greatly increased in both spring runs, ending with two successful productions of spat (∼ 20 × 10³ and ∼ 100 × 10³ juveniles). The summer larval run in 2006 failed again to produce spat. Additionally, the first run of April 2006 refers to an experiment that evaluated two different larval culture conditions: constant temperature (27 °C) and low stocking density (3-4 larvae ml^− 1) versus variable temperature (24-28 °C) and high stocking density (8-9 larvae ml^− 1). The first trial significantly increased larval survival and growth, which in turn resulted in greater numbers of settled spat, in comparison of the second trial, where survival, growth, and settlement of spat were significantly lower. Also in 2006, the quality of seawater used at the hatchery was evaluated with microbiological and chemical tests. The implication of these tests, together with results from all experiments are analyzed and discussed in terms of the potential development of large-scale hatchery cultivation of P. mazatlanica larvae in Mexico.     

The influence of water temperature on spawning patterns and egg quality in the Atlantic halibut (Hippoglossus hippoglossus L.)

A three-year study was conducted to investigate the effects of water temperature on Atlantic halibut broodstock reproductive performance. Two groups of fish held under ambient photoperiod were established onto contrasting temperature regimes. The ‘chilled’ group were held at below 9 °C from late October and at approximately 6 °C from December until the end of the spawning period whilst the ‘ambient’ group received no temperature control. The temperature profiles for the ‘ambient’ group changed over the 3 years but were generally characterised by warmer conditions prior to spawning and an earlier temperature rise in the spring. Total egg production was higher in the ‘chilled’ group each year. Absolute fecundity was significantly reduced in the ‘ambient’ group compared to the ‘chilled’ group every year of the study (0.6 million eggs/female for ‘chilled’ group vs. 0.3 to 0.4 million eggs/female for ‘ambient’ group) and egg viability, in terms of fertilisation and hatch rate, was significantly impaired in the ‘ambient’ group in years 2 and 3 (mean fertilisation rate in the ‘ambient’ group was between 27.0% and 54.8%; vs. 63.4% to 77.4% in the ‘chilled’ group, mean hatch rate in the ‘ambient’ group was between 3.1% and 25.6% vs. 60.7% to 71.7% in the ‘chilled’ group). Eggs spawned at high temperatures, later in the season were generally of low viability. In the ‘ambient’ group the spawning season became progressively delayed during the study and average duration of spawning season over 3 years was shorter (between 23.5 to 26.3 days for the ‘ambient’ group vs. between 30.5 and 41.2 days for ‘chilled’ group). It is hypothesised that high temperature during the vitellogenesis period caused a delay in spawning and a reduction in quantity and quality of eggs and that this effect was exacerbated by high temperature during spawning.     

Effects of temperature on the induced spawning of channel catfish and the production of channel × blue catfish hybrid fry

Hybrid catfish (channel catfish, Ictalurus punctatus, female × blue catfish, I. furcatus, male) can be obtained by induced spawning and artificial fertilization but with variable results. Temperature of the surrounding environment affects the rates of physiological processes in fish including time to ovulation following hormone injection and time post-ovulation when quality eggs can be obtained. Brood females were held at 24, 26, and 28 °C in 100 L aquaria and injected with LH-RHa at 20 μg/kg followed 12 h later with 100 μg/kg. Fish were monitored hourly recording the time of the first egg deposit. Some females were manually stripped soon after the first eggs were observed, and the remaining females were stripped 4-6 h after the first eggs were observed. Eggs were artificially fertilized with blue catfish sperm and incubated. The percent of females that ovulated were 52.9%, 82.4% and 95.5% at 24, 26, and 28 °C (P = 0.001) respectively. The majority of females that ovulated did so between 58 to 64 h at 24 °C, 48 to 52 h at 26 °C and 24 to 40 h at 28 °C for a degree hour response time of 1405 ± 117, 1141 ± 238 and 951 ± 261 respectively (P < 0.001). Differences in eggs/kg female, eggs/g of eggs, percent viable eggs, percent hatch and survival of sac fry to swimup resulted in an average of 384.4 ± 316.97 fry/kg female at 24 °C, 370 ± 219.2/kg at 26 °C and 1284 ± 1394.1/kg at 28 °C (P = 0.136). Egg quality varied with how soon eggs were stripped after the first egg was observed. For fish held at 26 and 28 °C. the combined effects of egg quality and fry survival resulted in an average 1081 ± 1483.9 fry/kg female for fish stripped within 2 h after the first eggs were released and 500 fry/kg female when stripped after 4 or more hours once the first eggs were released.     

Reproductive biology of the carp, Labeo bata (Ham.) from the River Kali,India

Spawning in Labeo bata occurs during July and August. Each individual spawns only once, with the onset of spawning directly dependent on the monsson season. Its success depends upon the access of gravid fish to a suitable spawning ground. The average fecundity is 192 785.     

Spawning, early development and first feeding in the gobiid fish Priolepis nocturna

The reproductive behavior, embryonic development and early larvae of Priolepis nocturna are described. Three pairs of P. nocturna began spawning 41 days after acquisition and maintained a 5–10 day spawning cycle lasting beyond several months. Spawning was initiated by the female who signaled her readiness to spawn by displaying to the male. Egg clutch size averaged 1578 ± 51.23 eggs and ranged from 268 to 3121. Egg length averaged 0.82 ± 0.01 mm total length (TL) and ranged from 0.75 to 0.90 mm. Egg width averaged 0.51 ± 0.51 mm total width (TW) and ranged from 0.49 to 0.52 mm. Fertilized eggs were ovoid in shape and attached to the ceiling of provided shelters via adhesive filaments at the proximal end. Hatching rates averaged 97.3 ± 0.51% and ranged from 91.9 to 99.8%. Larvae measuring 1.89 ± 0.04 mm TL hatched 121 ± 0.5 h post fertilization and did not rotate position prior to hatching. Skeletal elements of the chondrocranium were simplistic and dominated by the hyoid, hyomandibulosymplectic cartilage, ethmoid and Meckel's cartilage in first feeding larvae. No elements were added to the cranial architecture by 5 days post hatch (DPH) when larvae measured 2.05 ± 0.04 mm TL. First feeding larvae consumed only dinoflagellates and tintinnids suggesting that feeding was constrained by a poorly developed feeding mechanism. Embryology and larval development are described to 5 DPH.     

Influence of hormonal induction and broodstock feeding on longfin yellowtail Seriola rivoliana maturation,spawning quality and egg biochemical composition

To optimize broodstock management for Seriola rivoliana, a survey over 5 years was performed to evaluate the effects of successive hormonal inductions with GnRHa or dietary regime on spawning quality. Running males and females with an oocyte diameter > 500 μm were injected with GnRHa. The spawning quality was compared among consecutive years and different moments along the spawning season. Besides, three different feeding regimes were tested. Spawning quality parameters were not significantly affected by time along the spawning season, whereas they were clearly influenced by diet. Particularly, broodfish fed the mackerel regime (M) showed a significantly higher number of eggs than other treatments. Besides, dietary protein content significantly improved broodstock fecundity of S. rivoliana. Moreover, hatching rates were also higher in broodfish fed mackerel and the dietary docosahexaenoic acid (DHA) levels significantly increased this parameter. Despite polar lipids were not significantly affected, neutral lipids fatty acid composition of S. rivoliana eggs showed higher 18:2n‐6 in eggs, whereas ARA content was not affected. The results of this study denoted the high fecundity and good spawning quality of S. rivoliana broodfish and suggested the importance of high protein, energy and DHA levels in broodstock diets.     

Growth and gonadal development in diploid and triploid turbot (Scophthalmus maximus)

This study determined the effect of triploidy on the survival, growth and gonadal development of turbot from 6 to 48 months of age. From 6 to 24 months of age (first sexual maturity), survival was similar in both ploidies (P > 0.05). From 24 to 48 months of age, after the first sexual maturity, survival was 91.9% in diploids and 100% in triploids, which did not exhibit the post-spawning-associated mortality. Growth was similar for both ploidies during the first year of life. After that, triploids grew significantly (P < 0.05) more that diploids, with more marked differences after each spawning season. From 24 to 48 months, the average weight difference between both ploidies was 11.4 ± 1.9%, ranging from 4.3 to 23.0%. At 47 months of age, the biomass of triploids was 10.3% greater in total weight and 14.3% greater in eviscerated weight. Gonads of triploid males were similar to that of diploids, whereas in triploid females, they were significantly smaller and rudimentary. A histological analysis carried out at 47 months of age showed complete sterility of triploids in both sexes. Sex ratio was 1 male (M):0.6 female (F), for diploids, significantly (P < 0.05) different from 1:1, and 1 M:3.3 F for triploids, significantly (P < 0.05) different from 1:1 and from the diploids. Since females grow more than males, culture of triploids benefited from the high female ratio, which helped to reduce size dispersion. In addition, their sterility allowed better performance by avoiding the reduction in growth that takes place during the spawning periods. Together, these observations indicate that triploidy induction can be an interesting option for turbot aquaculture, especially for the production of large-size fish of more than 2 years of age.     

Serotonin stimulates ovarian maturation and spawning in the black tiger shrimp Penaeus monodon

Serotonin (5-hydroxytryptamine, 5HT) has been reported to induce ovarian maturation and spawning in the crayfish Procambarus clarkii and white Pacific shrimp Litopenaeus vannamei. The aim of this study was to explore the role of exogenous 5HT on the reproductive performance of the black tiger shrimp Penaeus monodon. 5HT solution was injected into domesticated P. monodon broodstock at 50 μg/g body weight and ovarian maturation and spawning were recorded. The presence of 5HT in the ovary and oviduct of P. monodon was also studied by immunohistochemistry and its levels in the ovary by enzyme link immunoabsorbance assay (ELISA). The 5HT-injected P. monodon developed ovarian maturation and spawning rate at the level comparable to that of unilateral eyestalk-ablated shrimp. Hatching rate and the amount of nauplii produced per spawner were also significantly higher in the 5HT-injected shrimp, compared to the eyestalk-ablated shrimp. 5HT-positive reactions were found in the follicular cells of pre-vitellogenic oocytes, in the cytoplasm of early vitellogenic oocytes and on the cell membrane and cytoplasm of late vitellogenic oocytes. 5HT in the ovary was present at 3.53 ± 0.26 ng/mg protein level in previtellogenic stage and increased to 17.03 ± 0.57 ng/mg protein level in the mature stage of the ovary. The results suggest a significant role of 5HT, possibly directly on the ovary and oviduct, on the reproductive function of female P. monodon.     

Induced spawning and artificial incubation of Tilapia

Methods for induced spawning and artificial incubation of Tilapia nilotica and T. aurea are described. These methods were required for the production, in aquaria, of F₁ hybrids between the two species. Spawning families were established in long aquaria from which sexually ripe individuals were selected for stripping by their “on heat” display coloration. Injections of pituitary hormones were not required. The fertilized eggs were incubated in a low speed “shaker”. At correct temperatures Tilapia fry may be produced by this artificial process all year round.     

Biochemical composition and quality of captive-spawned cobia Rachycentron canadum eggs

Interest in the commercial production of cobia Rachycentron canadum continues to rise as additional insight is gained into the hardy and fast growing nature of this species. However, research regarding the biochemical composition of captive-spawned eggs and egg and larval quality remains scarce. Such data is essential as a common bottleneck to production is a steady supply of fingerlings for grow-out. This study quantified the biochemical composition and quality of cobia eggs produced over 2 spawning seasons by broodstock on a traditional ‘trash fish’ diet which is commonly fed to tank spawning cobia. Throughout the study, batch fecundity, proportion of floating eggs and percent hatch averaged > 1 million eggs,  0.8 and 70%, respectively. Batch fecundity was significantly higher during the second spawning season as a result of the increased size of the females which weighed 18/22 kg and 22/26 kg at the beginning of each season. A positive correlation was found between the proportion of floating eggs and hatch rate for both spawning seasons. No correlations were found between egg composition (total lipid (30.0 ± 1.1% dry wt), protein (25.4 ± 2.2% dry wt), carbohydrate (2.4 ± 0.3% dry wt), vitamin E (10.2 ± 0.6 μg/g wet wt) or dry weight (119.1 ± 5.5 μg/egg)) and egg quality (proportion of floating eggs, hatch rate, larval growth and larval survival). Further, no differences in egg composition were noted between seasons or over the course of each season. The fatty acid composition of cobia eggs varied between seasons possibly due to changes in the quality of frozen feed (fish, shrimp, squid) given to the broodstock. The only correlation between the fatty acid profile and egg quality was a decrease in the proportion of floating eggs as the total amount of n-3 highly unsaturated fatty acids increased. No relationship between egg quality and amino acid content was noted with the most prominent amino acids being glutamate, leucine, alanine, proline, lysine and aspartate nor were any differences detected between spawning seasons.     

Spawning patterns provide further evidence for multiple stocks of sardine (Sardinops sagax) off eastern Australia

Previous studies have suggested that sardine, Sardinops sagax, off eastern Australia spawns across its entire range when habitat conditions are suitable. However, recent studies have suggested that separate sub‐populations and spawning groups may occur in the region. Spawning patterns off eastern Australia were investigated using data collected during nine ichthyoplankton surveys conducted between 1997–2015, and adult reproductive data obtained from ad hoc commercial sampling off New South Wales (NSW). The egg surveys covered the known distribution of sardine off eastern Australia and included year‐round sampling in the northern and southern parts of this range. Egg distributions and analysis of gonadosomatic indices identified two spatio‐temporally separate spawning groups; one occurring off southern Queensland to northern NSW during late winter and early spring, and a smaller group off eastern Tasmania to southern NSW during summer. Most eggs were collected from waters 50–90 m deep, with sea surface temperatures of 18–23°C. Additive modelling indicated depth was the most significant factor driving selection of spawning habitat, followed by the interaction of month and latitude. Low egg densities were recorded in waters between 34–37°S, despite conditions within the ranges suitable for spawning. The presence of two spawning groups of sardine off eastern Australia supports recent findings that two sub‐populations occur in the region. Findings of this study will help to optimise the spatio‐temporal extent of future egg surveys and further confirm the need to coordinate future management of each sub‐population among relevant jurisdictions.     

Spawning aggregation behaviour in the Creole perch,Percichthys trucha (Percichthyidae): A target species for conservation

1. Habitat fragmentation and loss threaten freshwater biodiversity worldwide. Habitats that are essential as fish spawning and nursery sites are critical, and must be identified for conservation purposes. Littoral zones within Patagonian lakes, especially shallow vegetated areas, represent important areas for the Creole perch, Percichthys trucha.
2. Spawning behaviour has been little studied, so it is not known whether these movements are performed in groups. A particular area of a deep lake in northern Patagonia was studied to identify the environmental conditions under which spawning occurs, and to analyse spawning aggregation behaviour in relation to the lunar cycle.
3. To identify spawning sites, spawning occurrence time, and the environmental parameters that determine it, fish capture data were analysed. Underwater cameras were also set up at three depth strata in the shallow zone to detect spawning aggregation behaviour. Aggregation was scored as the greatest number of individuals present in a single frame, and the number present per depth stratum. The relationship between the lunar cycle and Creole perch abundance by depth was then determined.
4. This is the first visual record of a spawning bed and the spawning aggregation behaviour of Creole perch in Patagonian lakes. The abundance of mature Creole perch in the shallow strata during the austral spring suggests that vegetated areas constitute spawning sites. Spawning aggregation takes place at a shallow depth 2 days before the full moon.
5. As the reproductive behaviours of Creole perch are important phenomena in the life history of the species, the results of this study have important implications for conservation and management.

     

Seasonal variations in muscle growth dynamics and selected quality attributes in Atlantic halibut (Hippoglossus hippoglossus L.) fed dietary lipids containing soybean and/or herring oil under different rearing regimes

Atlantic halibut (Hippoglossus hippoglossus), 1630 ± 510 g at the start of the experiment were reared in tanks for 1 year and subjected to three different environmental regimes; Continuous light and Constant temperature (CC), Natural light and Constant temperature (NC) or Natural light and Ambient temperature (NA). Two diets were fed, where 50% of the dietary lipids were either herring oil or soybean oil supplemented. The fish were sampled in May, August and November the first year, and February and May the following year. The CC groups grew better than the NC and NA groups from May to November, and as a result had a significantly (p < 0.05) higher weight by the end of the experiment. This difference did however not affect muscle fibre size distribution, or the chosen muscle quality attributes texture, fat content, water or fat loss of the fillet. These quality attributes were influenced by season and body weight, whereas diet had no significant effect. The NA group had lower fat content of the fillet than the other environmental regimes in August, November and May_end, and fat content increased with increasing fish weight. Water and fat losses were lowest in November and February. The diet influenced the fatty acid profile of the musculature; the triacylglycerol fraction reflecting the dietary treatment while the polar fraction was less influenced by diet. The hardness of the fillet was not influenced by the environmental regimes, but by season, and hardness was at its lowest in August, and increased towards the following spring. The muscle fibre distribution varied only with season, and neither treatment nor diet had any effect on the muscle growth dynamics. Recruitment of new fibres seemed mainly to occur during winter, signified by maximum occurrence of fibres < 500 μm² by the end of the experiment in May.     

Broodstock spawning and larviculture of whiting (Merlangius merlangus L.) reared in captivity

In recent years, an increasing interest in the farming of marine ‘white’ fish, such as cod, has been observed in the aquaculture industry. One species being considered for aquaculture development is whiting (Merlangius merlangus L.). With little data being available on the domestication of this species, our study aimed to characterize spawning behaviour and rhythmicity in captive broodstock and provide original data on egg incubation, embryonic development and larval rearing. The results observed through video recording showed that spawning mainly occurred at night with active courtship behaviour. Daily egg batches were collected from two separate spawning seasons, with an average fertilization success declining from 77% to 53%. Time to hatch was inversely related to the incubation temperature and occurred after 78 degree‐days (9 days at 8 °C), with emerging larvae ranging in size from 3.3 to 3.46 mm total length and hatch success ranging from 48.3% to 99.9%. In comparison with a clear water environment, a green water regime during the first days of larval rearing significantly improved larval growth and survival. These results are the first to describe spawning activity, early development and larval performances in whiting and are promising for the potential future development of whiting aquaculture.     

OPTIMAL SALINITY-TEMPERATURE COMBINATIONS FOR THE EARLY LIFE STAGES OF GILTHEAD BREAM,Sparus auratus L.

High larval mortalities during rearing of gilthead bream, Sparus auratus L., led to experiments on the influence of salinity and temperature on eggs and yolk-sac larvae. Test salinities ranged from 5 to 70 ppt for eggs and from 15 to 45 ppt for larvae; experimental temperatures were 18–20°C for eggs and 18, 23 and 26°C for larvae. Spawning conditions were 18–20°C and 33–35 ppt salinity; the yolk-sac larvae were chosen from hatches obtained under similar conditions (18°C and 35 ppt salinity). For eggs the optimum survival range was found to be 30–50 ppt at 18°C and 15–60 ppt at 23°C, while that for yolk-sac larvae was 15–25 ppt at all three temperatures. Choosing normal development (no dorsal curvature) as the decisive criterion, the optimum salinity range for egg incubation was reduced to 30–40 ppt at 18°C and to 35–45 ppt at 23°C, while that for the yolk-sac stage remained 15–25 ppt at all test temperatures. Egg incubation was most successful at salinity-temperature combinations close to those during spawning, whereas salinity had to be reduced by at least 10 ppt for yolk-sac larvae.