Vol. 72: 213-223. 1991 MARINE ECOLOGY PROGRESS SERIES Mar. Ecol. Prog. Ser. l Published June 4 Feeding, growth and survival of juvenile summer flounder Paralichthys dentatus: experimental analysis of the effects of temperature and salinity Kirk D. Malloy, Timothy E. Targett University of Delaware, College of Marine Studies, Lewes, Delaware 19958, USA ABSTRACT. Laboratory experiments were conducted on juvenile summer flounder Paralichthys dentatus (41 to 80 mm total length) to determine low temperature tolerance (2 to 3 "C) at 10, 20, and 30 %salinity, and to measure feeding rate, assimilation efficiency, growth rate and growth efficiency at 2, 6, 10. 14, 18 "C and 10, 20 and 30 '7' salinity. There was 100 % sunival at temperatures above 3 'C, suggesting that juvenile summer flounder are able to survive most winter water temperatures encoun- tered in north/central Mid-Atlantic Bight (MAB) estuaries. Mortality was 42 % after 16 d at 2 to 3 'C, and was highest in fish < 50 mm TL (1 g). Mean specific growth rates were not significantly different between 2 and 10 "C (mean = 0.14 % d-'), and these rates were not significantly different from zero. Mean growth rate increased to 2.4 "0 d-' at 14 'C and 3.8 O/O d-' at 18 OC. Ad libitum feeding rate showed a similar relationship to temperature. Mean assimilation efficiency (60.1 %) was not affected by temperature. Mean growth efficiency (K,) was significantly lower at 6 "C (-23.1 "h) than at 14 and 18 "C (18.4 and 22.1 %, respectively), and was highly variable. Salinity had no significant effect on any parameters measured, suggesting that factors other than salinity are controlling spatial distributions. Mortality resulting from acute exposure to low temperature probably occurs during one 2 to 4 wk period each winter Recruitment success from north/central MAB estuaries may be lower in years with late winter cold periods (i.e. March vs December) due to increased numbers of fish being exposed to lethal low temperatures. Additional mortahty probably results from low growth rates caused by sub-optimal temperatures (i.e. < 10 "C) throughout the spring. The annual contribution of new recruits from northern estuanes appears to be dependent on winter temperature regime, particularly on the magnitude and timing of temperature minima. INTRODUCTION Recruitment variability in fishes is considered to be a major cause of fluctuations in adult population size (Hunter 1976, Sissenwine 1984, Houde 1987). Small changes in mortality rates of early stages may have large effects on the subsequent size of adult popula- tions (May 1974, Houde 1989), and factors influencing the relatively high mortality of early life history stages have the greatest effect on recruitment success (Houde 1987). Although much research has been done on pre- dation and starvation in early larval stages (see Houde 1987), there is increasing evidence that mortality of late-larval/early-juvenile stages is an important con- tributor to year-class strength (Hunter 1976, Smith 1985, Folkvord & Hunter 1986, Bailey & Houde 1989). Since vulnerability to predation is size-dependent (Folkvord & Hunter 1986, Van der Veer & Bergman O Inter-Research/Printed in Germany 1987, Bailey & Houde 1989), slow growth during the critical larval-early juvenile period, resulting from sub- optimal environmental conditions such as low tempera- ture, may extend the period during which young fish are vulnerable to predation (Houde 1987). Temperature is a dominant abiotic factor controlling feeding and growth (Brett 1979), and conditions at the extremes of an organism's tolerance range exert the greatest limitations on scope for growth (Warren & Davis 1967, Weatherley & Gill 1987). Temperatures outside an optimum range have deleterious effects on early growth in many species (Holt et al. 1981, Houde & Taniguchi 1981, Taniguchi 1981, 1982, Cech & Mitchell 1984, Lee et al. 1984, McMullen & Middaugh 1985). Low temperatures experienced by larvae/juveniles in species spawning in winter can reduce ingestion rates, and, although maintenance requirements and activity may be lower, often result in reduced or negative