| Abstract | Although tilapias are one of the well-studied groups of fishes in aquaculture and have been the subject of several significant symposium volumes in the last decade, poor broodstock productivity due to low fecundity and asynchronous spawning remains one of the most significant constraints to commercial production and its future expansion. Therefore, the objective of this study was to better understand several aspects of seed production, focusing on influence of maternal age and size, fecund type, feeding level, stocking density and water flow rate on fecundity, spawning frequency and egg quality of Nile tilapia, Oreochromis niloticus and finally to suggest a seed production strategy with a goal to reduce problems associated with the current seed production. A series of four experiments were conducted, one experiment conducted in hapas suspended in a pond and the other three conducted in concrete tanks in re-circulating system at the Asian Institute of Technology (AIT), Thailand using individually tagged Thai-Chitralada strain Nile tilapia. Fecundity (eggs female), weight (g), and egg quality were monitored by removing eggs from incubating females every four, five or seven days over the experimental periods. In an attempt to better understand and characterize the variability in female reproduction, tagged mature females (n = 68) from a single population were monitored in hapa-in-pond system over 12-month experimental period. Spawns female⁻¹, eggs female⁻¹ day⁻¹, eggs kg female⁻¹ day⁻¹, days spawn⁻¹ and weight (g) at first spawning of individual female were determined from the regular weekly sampling of eggs spawn⁻¹. For analysis the females were grouped into two categories, high spawning (HSF) and low spawning frequency (LSF) classes based on their spawning frequency record. Moreover, nested under these two classes were two groups each based on brood size, i.e., high frequency-large size (HL), high frequency-small size (HS), low frequency-large size (LL) and low frequency-small size (LS). There was no difference in eggs spawn⁻¹ among all females. The HSF females had five and three fold higher eggs female⁻¹ day⁻¹ and eggs kg female⁻¹ day⁻¹, respectively than the LSF females. Eggs female⁻¹ day⁻¹ and spawns female⁻¹ remained high among HSF group while low among LSF group throughout the 12-month experimental period. This suggests individual female spawning activity is consistent within a population under a common environment. Inter-spawn-interval increased with age in all four groups, and days spawn⁻¹ was shorter by 130% in HSF females compared with days spawn⁻¹ in LSF. The HSF group also spawned more successively ( ≥ 3 spawns without interruption) while LSF group of females had fewer successive spawns ( ≤ 2). Body weight of fish 95 ± 3.2 to 223 ± 24.8 g had no influence on the number of eggs produced. In the second experiment, effects of two levels of feeding (1 or 4 % BW day⁻¹), stocking density (3 or 10 /m²) and water flow rate (0.06 ± 0.00 or 0.35 ± 0.04 L sec⁻¹) on fecundity, relative fecundity, spawning frequency and egg quality of adult (125 days old) females reared in recirculating concrete tanks for 120 days. Higher relative fecundity (312 ± 36 eggs kg female⁻¹ day⁻¹) was observed at low levels of feeding (1% BW day⁻¹), stocking density (3 female /m²) and water flow rate (0.06 ± 000 L sec⁻¹) treatments. Under these conditions relative fecundity was 2 to 4 fold higher than the lower eggs kg female⁻¹) day⁻¹) found at high levels of feeding (4% BW day⁻¹), stocking density (10 water flow rate (0.35 ± 0.04 L see⁻¹) treatments.High feeding level (4% BW day⁻¹) increased growth (g day⁻¹) by 35% and eggs spawn⁻¹ by 18% while it had no influence on eggs kg female⁻¹ day⁻¹ and spawn female⁻¹. High stocking density (10 female /m²) lowered eggs spawn⁻¹ by 19%, the eggs kg female⁻¹ day⁻¹ by 52% and spawn female⁻¹ by 40% while increasing eggs m⁻² day⁻¹ by 30% and did not affect growth (g day⁻¹). A higher water-flow rate (0.35 ± 0.04 L sec⁻¹) increased growth (g day⁻¹) by 33% but decreased spawn female⁻¹ by 32%, eggs k female⁻¹ day⁻¹ by 51%, eggs m⁻² day⁻¹ by 30% and it had no effect on eggs spawn⁻¹. The treatment variables did not affect percent egg fertilization and hatchability, crude lipid and protein contents, and egg diameter (mm). However egg weight (mg) was affected by all variables. Feed level interacted with water flow rate in improving growth while stocking density interacted with water flow rate in lowering eggs kg female⁻¹ day⁻¹ and spawn female⁻¹. This study suggested that lower feeding level (1% BW day⁻¹), lower stocking density (3 female /m²) and lower water-flow rates (0.06 ± 0.00 L see⁻¹) is a productive management strategy to improve current tilapia hatchery seed production, however, optimal water flow-related stocking density needs further investigation. The effect of maternal age on fecundity, spawning interval and egg quality was examined for four age groups (4, 9, 16 and 24-month) in recirculating circular concrete tanks. Fecundity (eggs spawn⁻¹), body weight (g), and egg quality were monitored by removing eggs from incubating females every four days over a 168 day experimental period. Eggs spawn⁻¹ doubled between 4-month age and 24-month age females. While fecundity correlated with fish age (r = 0.794, d.f. = 96, P < 0.05), weight was not (r = 0.050, d.f. = 106, P < 0.05). Young females had more spawns female⁻¹ and produced 1.8 and 2.2 times more eggs than 16 and 24-month age females, respectively. Egg production declined along with a drop in spawning frequency after 18 months age. Large numbers of good quality hatchling were achieved from females of 6 to 18 months old. This study also evaluated egg production strategy based on size at early maturation and spawning frequency. Individually tagged 125 day-old females were first separated into large size (LS; 56.3 ± 1.2 g) and small size (SS; 28.9 ± 0.6 g) groups. The first part of the experiment monitored individual female's eggs spawn⁻¹ and growth rate (g day⁻¹) within a population in communal tanks for 60 days. In the second part of the experiment the females both large and small size groups were separated into four groups (i.e., SF ≥ 3, SF = 2, SF = 1, SF = 0) respectively, based on their spawning frequency records. Eggs spawn⁻¹ and growth rate (g day⁻¹) were monitored for 100 days after stocking. In the control treatment mixed size (39.6 ± 4.2 g) females were also monitored for 160 days. The spawning activity and egg production of individual females observed within the population remained unchanged over 100 days postseparation based on spawning frequency. The variation in body weight (g) observed at 125-day sampling disappeared over the experimental period. No differences were observed in mean fecundity (850 ± 16.5 eggs spawn⁻¹), egg diameter (1.87 ± 0.0 mm), fertilization (95 ± 0.4%) and hatchability (84 f 0.5%) across treatments. Small size SF ≥ 3 females had 167% and 82% higher mean eggs kg female⁻¹ day⁻¹ than the control and large size SF ≥ 3 females, respectively. This rate is also 35% and 47% higher than the highest reported for tilapia with different species and broodstock management strategies, respectively. The separation by size and spawning frequency also yielded a 96% active spawning breeder population compared with 63 ± 3.7% active spawning breeder non-selected population. Overall the strategy improved eggs kg female⁻¹ day⁻¹ of the broodstock population by 167%, allowing a consistent increase in egg production of 56 ± 6.9% and reduced number of breeders required to be maintained by a hatchery by 63%, thereby significantly increasing egg production. Higher and lower levels of feeding, stocking densities and water flow rates variations did not influence egg diameter (mm), percent crude protein and lipid contents, and fertilization and hatchability. There was no difference in egg diameter (mm), percent crude protein, lipid contents, fertilization and hatchability among same age but different size and fecundity females even under different culture conditions. Egg diameter (mm) increased with maternal age. Egg weight (mg) differed even within same age, size and fecund groups and was not correlated with maternal age (r = 0.294, d.f. = 108, P > 0.05). Females 6 to 18 month-age had uniform egg diameter (mm) and weight (mg). Differences in egg weight (mg) were large in comparison in old females. For example, there was 96% difference between lightest and heaviest eggs in 24 month age females. The lack of uniformity in egg weight (mg) and diameter (mm) in older females could lead to variations in fertilization and hatching that could also possibly increase the size variation in yolk-sac fry. Overall this study demonstrated that tilapia hatchery productivity in recirculating concrete tanks could be improved by selecting productive/fecund brood females and using females 6 month to 18 month age and regulating ration size at 1% BW day⁻¹, stocking 3 female /m² with 0.06 ± 0.00 L sec⁻¹ water-flow rate. This study also identified four distinct groups of females within a population and found that selection based on small size at early maturation and frequent spawning produced more eggs than large size at early maturation and less frequently spawning females. These series of experiments provide tilapia hatchery operation valuable information in improving production without a significant amount of additional input |