The effect of initial weight composition and heterogeneity on growth and behaviour of African catfish (Clarias gariepinus)

It is generally accepted that disrupting social hierarchy by size grading allow the smaller to grow better. However, many studies have shown a negative or no effect at all. This raises the question whether other factors besides social hierarchy are responsible for the increase in weight variation over time. Despite the studies made on the consequences of size sorting on growth performance, little is known about the underlying behavioural causes/consequences of the observed effects. Therefore, the main objective of this study is to investigate the inof initial weight composition and heterogeneity on performance and behaviour of catby comparing 3 groups with homogeneous weight (light, middle-weight and heavy) with a control group of heterogeneous (mixed) Two thousand with an average weight of 8.2 1.2 g were fed restrictively at 20 g.kg-0.8.d-1. After 8 weeks were individually weighed and manually selected. Four treatments were established: homogeneous light (HL, 66.5 - 94.9g), homogeneous middle (HM, 125.4 - 154.9g), homogeneous heavy (HH, 184.5 - 214.1g) and a heterogeneous group containing one third of each group mentioned above (HET, 65.2 - 214.3g). Each treatment contained three replicates with 36 each. Selected were tagged to examine individual growth trajectories. For 27 days were fed ad libitum. Fish behaviour was studied using video cameras on day 2, 4, 8, 11, 15, 18 and 26. Five behaviours elements were monitored (swimming, resting, air breathing, waiting-in- feeding-area and speed of eating). Relative growth rate, feeding level and feed conversion ration did not differ signibetween all the treatments and when similar-weight were compared in different treatments. However, behaviour differed signibetween the treatments (Table 1). These results suggest that social hierarchy is not playing a major role in explaining differences in growth performance of African catHowever, feeding motivation seems to be an important factor, with heavier exhibiting a higher feeding motivation by swimming more, spending more time on the feeding area and eating their meal faster than lighter Table 1. Behaviour of African catas affected by the experimental treatments (HL = homogeneous light; HM = homogeneous middle; HH = homogeneous heavy; HET = heterogeneous) before and during feeding. Data are shown as the mean SD (n = 3 replicates/treatment, each replicate is the mean of day 2,4,8,11,15,18 and 26) abc Means within a row lacking a common superscript differ signi(P