Abstract: This study was carried out to monitor the suitability of Clarias gariepinus as a predator on Tilapia (Oreochromis niloticus) populations. In order to (a) recommend a stocking ratio for Tilapia-clarias prey-predation relationship, (b) determine the size/level at which the prey is most vulnerable to predation and (c) to determine whether sudden cohabitation with or without feeding/sparing feeding will trigger predation. Four different treatments of (1) 10 Clarias juveniles/100 Tilapia fingerlings fed to satiation (2) 10 Clarias juveniles/100 Tilapia fingerlings fed sparingly (3) 50 Clarias fingerlings/50 Tilapia fingerlings fed sparingly (4) 15 Clarias fingerlings/90 Tilapia fingerlings fed to satiation and (5) 15 Clarias fingerlings/90 Tilapia fingerlings fed sparingly were monitored over a period of 12 weeks. Predation occurred only in Treatments 1 and 2 where the predator was bigger than the prey and the highest level of predation of 71% Tilapia fingerlings occurred in Treatment 2 compared to 26% in Treatment 1. The predation level was highest during the first 6 weeks of the study due to increasing size of the prey. Hence growth performance from 7th-12th week depended solely on the feed fed for Treatments 1 and 2. However, since virtually no predation was recorded for Treatments 3 (0%), 4 (0%) and 5 (2.22%) growth performance for these 3 treatments from weeks 1-12 was dependent solely on the feed fed whether fed to satiation or not. The Analysis of Variance test (ANOVA) to test significance in the predation level showed significant differences (p<0.05) among the treatments (at least for treatments 1 and 2) and Duncan`s Multiple range tests puts the order of predation as Treatment 2 (71a)>1(26b)>3(0c)>4 (0cd) and > 5 (0cde), but there were no significant differences (p>0.05) among 3, 4 and 5 where the prey and predator were of similar sizes. Treatment 1 recorded the best growth performance in both species of fish with respect to mean weight gain, food conversion ratio and increase in total length. The order of growth performance is Treatment 1>2>4>3>5. The food conversion ratio among treatments 2 and 1 (where predation occurred) showed that it is higher in Treatment 1 where the level of predation was lowest. Analysis of Variance (ANOVA) shows growth rate of prey and predator were significant at (p<0.05), respectively. The correlation coefficients `r` between the size of the prey and predator in all the treatments were all positive viz: Treatment 1 (1.00), 2(0.96), 3(0.88), 4(0.95) and 5(0.81), this implying the size of the predator increased as the pray size increases. The sudden cohabitation of Clarias gariepinus and Tilapia (Oreochromis niloticus) with or without feeding did not trigger predation because the initial number stocked in Treatments 4 and 5 (each specie of fish raised separately and the two brought together at the end of 2 weeks) were recovered at the end of the 12 weeks experiment. The best level of growth performance and fairly balanced survival/predation level is achievable with Treatment 1 which gave the highest overall weekly mean weight gain and 90/74% Clarias/tilapia survival rate closely followed by Treatment 2 with 100/29% Clarias/tilapia survival rate. However if the intention of Clarias/tilapia polyculture is not to embrace predation, but to maintain at least about 100% survival of both species solely dependent on the feed fed Treatments 4>3>5 is recommended in that order.