A comparative study was carried out to compare the effect of caging mullet and tilapia in a shrimp polyculture system. In six shrimp tanks (three tanks for each fish species), either mullet, Mugil cephalus (CCT-SM), or tilapia, Oreochromis niloticus (CCT-ST), was stocked in cages. In three other tanks, mullets were allowed to roam freely in shrimp tanks (D-SM). White shrimp, Litopenaeus vannamei (0.50 g), was cultured as the predominant species were distributed randomly into nine fibreglass tanks (5 m3) at a density of 300 shrimp/tank, while fish (1.50 g) were stocked at the same density of 10% of the initial total shrimp biomass. The results showed that water quality parameters were not significantly different among treatments (p > .05), except for total suspended solids (TSSs). System performances based on parameters such as total weight gain (2,808.15 g/tank) and nutrient recovery were higher in D-SM treatment (39.80% for nitrogen and 27.40% for phosphorus) than in CCT-SM and CCT-ST treatments (p < .05). These system performance parameters were significantly affected by the mullet-holding strategy; however, they were not affected by fish species. The addition of mullet or tilapia in shrimp tanks did not affect shrimp growth differentially. Fish growth performances based on parameters such as final weight (98.43 g/fish) and DGR (1.29 g/day) were significantly higher in D-SM treatment and were significantly different among D-SM, CCT-SM and CCT-ST treatments (p < .05). It is concluded that in shrimp–fish polyculture with a stocking density of fish at 10% of the initial total shrimp biomass, tilapia is more effective than mullet, when caged. However, under free-roaming conditions, the use of mullet is more effective in terms of system performances relative to a system holding caged tilapia. |