EFFECTS OF SALINITY ON AGGRESSIVE AND CANNIBALISTIC BEHAVIOR IN POST-LARVAL COMMON SNOOK Centropomus undecimalis

Common snook Centropomus undecimalis are considered a species of interest for intensive aquaculture in various regions of North, Central and South America. Despite recent advances in larval rearing protocols, to date, there is still no established large scale production of fingerlings for commercial on-growing or restocking of recreational fisheries. High mortalities observed during the early life stages remains one of the major bottlenecks in the successful culture of common snook. Cannibalism has long been suspected to be a significant contributor to mortality before they reach 5 to 10 cm. Hence, the present study aimed to investigate the effects of contrasting salinity gradients on dynamics of cannibalism, growth and survival in hatchery reared common snook.

The indoor experimental setup consisted of three separate systems. At 25 days post-hatch (DPH) common snook were randomly stocked into twelve 100 L fiberglass tanks (200 larvae per tank) and acclimated to salinities of 35, 15 and 5 g L^-1 (n = 4 replicates) over a 72 hour period. Fish were given a commercially available micro-particulate diet and fed to satiation three times daily. Photoperiod was maintained at 1000 lux, 12 h dark/12 h light (LED), temperature at 27 ± 1 °C, pH at 8 ± 0.5 and dissolved oxygen at 6 ± 1 mg L-¹. Standard length measurements were obtained from 10 larvae per tank (40 per treatment) at 25, 30, 40 and 50 DPH. Throughout the study (25-50 DPH), behavioral and swimming patterns of fish held in each tank were documented with the use of video recordings. Daily mortality was recorded and total survival assessed.

Upon completion of the rearing trials, results showed highly significant differences in mean total survival where  the 35 g L^-1 treatment has a survival rate of 37.5% but survival in the lower salinity treatments was 75.6% (15 g L-1) and 69.6% (5 g L^-1) (Table 1). Observed mortality severely underestimated mortality rates and direct evidence of cannibalism was infrequent. Analysis of videography revealed the incidence of cannibalism and aggressive behaviors occurred more frequently among fish maintained at 35 g L^-1 than those at 15 and 5 g L^-1. No significant differences in fish growth were observed among the treatments during the trial period (Table 1).