An experiment were conducted to evaluate The effect of feeding frequencies and stocking size on growth and feed conversion ratio (FCR) of rainbow trout (O. mykiss) in brackish water for 102 days. Three stocking sizes (10, 25 and 40 g/fish) of rainbow trout and three feeding frequencies (feeding fish twice, three times and four times per day) were examined as the same time. Fish were fed at a restricted feeding program according to the rainbow trout standard table in the definite times during the light period. The results showed that in the group with 10 g initial weight increasing the feeding frequency, some growth factors like final weight, daily growth, and weight gain. But the other factors such as food conversion rate (FCR) and specific growth rate (SGR) decrease. It means in this group (10 g) two times feeding per day is not enough for an acceptable growth rate. In the two other groups with (initial weight 25 and 40 g) although by increasing the feeding frequency, some growth factors a little increase but these differences in 95% level (p،ـ0.05) are not significant. Therefore in rainbow trout culture with initial weight at least 25 g two times feeding per day was enough. The results of carcasses analysis of fish, showed that by increasing the feeding frequency there were no significantly difference at 95% level between the body compositions in all of treatments. With connivance of feeding frequencies, the best FCR was in the group with 10 g initial body weight and there was not any significant difference in SGR. So according to the results, it suggests that fish should be stocked with 10 g initial body weight.

Resumo: O objetivo deste trabalho foi avaliar a qualidade de água na produção de alevinos de tilápia-do-nilo alimentadas com diferentes níveis de proteína bruta (PB) em sistema experimental com bioflocos sem renovação de água. Foram utilizados 270 alevinos (6,31±0,40g), distribuídos aleatoriamente em 18 aquários de 200 L (15 animais/aquário), e que foram alimentados com três níveis 28%, 32% e 36% PB (tratamentos), durante 60 dias. Todas as rações experimentais foram formuladas para conter 3.200 kcal/kg de energia digestível. O delineamento experimental utilizado foi o inteiramente casualizado com três tratamentos e seis repetições. Os seguintes parâmetros de qualidade de água foram avaliados: temperatura, oxigênio dissolvido, pH, amônia, nitrito, nitrato, condutividade, alcalinidade, turbidez e volume de sólidos sedimentáveis. Os dados obtidos foram submetidos análise de variância (ANOVA), e complementadas com o teste de Tukey (p<0,05). Não foram encontradas diferenças (p>0,05) entre as médias dos parâmetros físicos e químicos da água. Por outro lado, para os valores de sólidos sedimentáveis houve diferença (p<0,05), sendo que os aquários onde os peixes foram alimentados com ração contendo 36% PB apresentaram maiores valores, quando comparado com o tratamento 28% PB, no entanto, eles não diferiram do tratamento 32% PB. O sistema bioflocos manteve os parâmetros físico-químicos da água em condições favoráveis para a produção de alevinos de tilápia-do-nilo sem renovação de água, independentemente dos três níveis utilizados de proteína das rações, resultando em melhor utilização do recurso hídrico e consequente diminuição da emissão de efluentes. Abstract: This study aimed to evaluate the water quality of Nile tilapia fingerling culture fed diets with different crude protein (CP) levels in an experimental biofloc system (BFT) with no water renovation. Fingerling (n=270, 6.31±0.40g) were distributed in 18 aquaria (200 L, 15 fish/aquarium) and fed with three levels 28%, 32% e 36% of CP (treatments), during 60 days. All experimental diets were formulated to be isoenergetic 3,200 kcal/kg of digestible energy. The experimental design was completely randomized with three treatments and six replicates. The following water quality parameters were evaluated: temperature, dissolved oxygen, pH, total ammonia, nitrite, nitrate, conductivity, alkalinity, turbidity and sedimentable solids. Experimental data were submitted to analysis of variance (ANOVA) and complemented with Tukey &#769;s test (p<0.05). There were no significant difference (p>0.05) among physical and chemical water parameters of different treatments. On the other hand, sedimentable solids showed statistical difference (p<0.05), and the aquaria where fish were fed with 36% CP showed the highest values compared to 28% CP; however, did not differ from 32% CP. The biofloc technology kept favorable water quality to Nile tilapia fingerling culture without water renovation independently of the three dietary protein levels, resulting in improved utilization of hydric resource, and consequent reduction of effluents emissions.