The ¢sh were fed a tilapia diet (35% crude protein)
at a daily rate of 30% of their live weight (El-Sayed
2002), reduced to 20% beginning of the second
month, three times a day for 50 days. The ¢sh were
weighed at 10-day intervals, and their daily ration
was re-adjusted accordingly. At the termination of
the experiment, all the ¢sh in each tank were netted,
weighed collectively and their average ¢nal weight
was recorded.
A one-way analysis of variance (ANOVA) was conducted to test the e¡ect of water temperature on the
performance of Nile tilapia fry, using the computer
program SPSS (SPSS Version 11.0.0, 2003). Least signi¢-
cant di¡erence was used to compare means at
P o 0.05, as described by Gill (1981). Because all the
experimental groups were subjected to the same
water source and tank conditions, except water temperatures, it is assumed that the possible interaction
of treatment group with tank variation was minimal.