weights of young stages has been overcome to a 

 certain degree with the advent of experimental 

 laboratory programs at a few research facilities 

 during the last 10 yr. 



The experimental larval fish program at the 

 Northeast Fisheries Center Narragansett 

 Laboratory, National Mai'ine Fisheries Service, 

 NOAA, has been studying grow th, metabolic, and 

 trophodynamic factors for a number of important 

 commercial and sport species, and it is the object of 

 this report to present larval length-weight rela- 

 tions for seven species including Atlantic cod, 

 Gadus ruorhua: haddock, Melanogrammus 

 aeglefinun: scup, Stcnotonuis chrysops: Atlantic 

 herring, Cliipea harengiis; winter flounder, 

 Pseudoplci/ronectes omericanus\ summer floun- 

 der, Paralicbthys dentatus\ and yellowtail floun- 

 der, Limanda ferruginea . The larval length- 

 weight relations presented here are previously 

 unreported in the literature for six of the seven 

 species with the exception of the Atlantic herring, 

 which is included because it represents the only 

 data available for western North Atlantic stocks. 



Materials and Methods 



All larvae were obtained from experimental 

 spawning of adults in the laboratory and reared by 

 techniques reported by Smigielski ( 1975a, b) and 

 Laurence ( 1975). The length-weight data were col- 

 lected coincident with a variety of experimental 

 studies on larval growth, survival, metabolism, 

 and feeding reported by Laurence (1974, 1977, 

 1978, and as yet unpublished). 



In all cases the data were collected from larvae 

 reared at prey concentrations in the range of 0.5 to 

 3.0 organisms ml. Concentrations of 0.5 and above 

 have been shown to be adequate for normal grow th 

 in the studies cited above. Rearing temperatures 

 were optimum for growth and survival or within a 

 3°C nonlethal range about the optimum depend- 

 ing upon the experiment from which the data were 

 taken. Optimum temperatures determined in 

 laboratory studies for rearing the seven species 

 were 7°C for cod and haddock, 8°C for winter 

 flounder, 10°C for herring and yellowtail flounder, 

 16°C for summer flounder, and 18°C for scup. 



Length measurements were taken from the tip 

 of the snout to the end of the notochord in the 

 preflexion stage. During flexion of the notochord 

 measurements were taken to a line vertically per- 

 pendicular to the tip of the notochord until the 

 hypural bones became prominent or exceeded the 



line vertically perpendicular to the notochord tip. 

 At this time, a standard length measurement to 

 the posterior end of the hypural plate was re- 

 corded. Since the original experiments were not 

 designed for developmental anatomy purposes, 

 the different flexion stages were not recorded coin- 

 cident with the length and weights. 



Lengths were recorded to the nearest 0.1 mm 

 with a filar ocular micrometer. Dry weights were 

 determined after rinsing larvae in distilled water, 

 pipeting onto a glass Petri dish, and drying to a 

 constant weight at 60°-90°C for 24 h. Individual 

 dry weights were recorded to the nearest 0.1 /xg on 

 a gram electrobalance. 



All measurements were made on post yolk-sac 

 larvae that were freshly sacrificed and unpre- 

 served. The data points for each species represent 

 lengths and weights for individual larvae except 

 for winter flounder and haddock. The data for 

 these two species are the means of lengths and 

 weights for samples of 10-25 larvae collected on a 

 weekly basis during different experiments. The 

 experimental procedures precluded the matching 

 of individual lengths with weights for these two 

 species. 



Regression equations and associated parame- 

 ters were calculated as geometric mean, func- 

 tional regressions using log base 10 transformed 

 data according to the methods of Ricker (1973) 

 rather than using the previously standard predic- 

 tive, regression techniques. Ricker demonstrated 

 the advantages of using functional rather than 

 predictive regression calculations to reduce bias in 

 length-weight conversions where the populations 

 of measurements are typically open ended, where 

 only a portion of the length and weight distribu- 

 tions are represented, and where the variability 

 may be more inherent in the biological material 

 itself rather than the means of measuring length 

 and weight. 



Results and Discussion 



The exponential relation between length and 

 weight for all seven species are presented in 

 linearized form by logarithmic transformation in 

 Figures 1-7. The larvae studied in this research 

 are from different taxonomic families (Clupeidae, 

 Gadidae, Sparidae, Bothidae, and Pleuronec- 

 tidae), represent different adult life styles (pelagic 

 and demersal), develop in a range of different 

 temperatures, and demonstrate different patterns 

 of metamorphosis from larval to juvenile stages. 



891 



