Individual variation in growth of African catfish Clarias gariepinus: a search for explanatory factors

Samenvatting

Among farmed animals, fish exhibit the largest individual variation in growth, yet most of the studies reporting data on growth do not take individual variation into account. Usually a mean value is considered and although the variation around the mean is also mentioned, it is generally viewed as a sort of "statistical noise". The importance of individual variation in growth should not be underestimated since it has important consequences for water quality, aggression, stress levels, farm management, selection programmes, etc. Among the factors responsible for growth variation, social hierarchy is often considered as the most important. Social hierarchies may induce behavioural inhibition and stress on subordinate fish, affecting their feed intake, feed efficiency and as a consequence reducing their growth. However, for most fish species there is no unambiguous proof that individual differences in feed intake, feed efficiency and growth result from social hierarchies. Some studies suggest that inherent (genetic) factors may also cause the variation in growth.The general aim of this study was to understand the underlying factors responsible for the individual variation in growth of African catfish Clarias gariepinus . The following factors were investigated: 1) if individual variation in growth is mainly a consequence of social hierarchies, 2) the contribution of individual differences in feed intake and feed efficiency to the individual differences in growth and 3) the contribution of feeding behaviour and stress response in explaining individual differences in feed efficiency.Chapters 2 and 3 investigated the importance of social hierarchy as an explanatory factor for the individual variation in growth of African catfish. The growth performance, behaviour (feeding behaviour, aggression levels) and stress response between groups of low-, medium- and heavy- weight fish were compared. Chapter 2 showed that low-weight fish do not exhibit increased growth rates in the absence of heavier fish. Apparently, the growth differences were not induced by social hierarchies where the larger fish suppress the growth of smaller fish. Instead, this study suggests that feeding behaviour is a crucial factor. Heavier fish exhibit feeding behaviours that may give advantage when feed is limited, such as being more active swimmers, spending more time at the feeding areas and eating their meal faster than low-weight fish. These differences in behaviour may result in growth variation, as found in this study.Chapter 3 showed that the aggression and stress levels did not increase in heterogeneous (weight) groups as compared with homogeneous (weight) groups. Furthermore, low-weight fish did not exhibit a higher number of skin lesions and higher stress levels when reared in heterogeneous groups as compared with low-weight fish reared in homogeneous groups. These results suggested that low-weight fish were not behaving as subordinates and heavy fish as dominants.To further investigate the importance of inherent differences in growth variation, a set of experiments were designed using individually housed fish. Housing fish individually enabled the study of individual differences in the absence of social interactions and to measure individual feed intake accurately. This raised the question whether the results obtained from housing fish individually could be representative of a group housing situation. Chapter 4 compared the growth performance, feeding behaviour and stress response of isolated and non-isolated fish. This study suggested that in African catfish feed intake is stimulated by the presence of conspecifics resulting in higher feed intake and growth rates. However, isolation per se seems not to act as a stressor in the short term or to affect the stress response, probably because periods of isolation are part of the African catfish lifestyle. In addition, Chapter 5 compared the growth of fish housed individually and afterwards in a group. The average growth of individually housed fish was lower than fish in group housing. However, slow and fast growing fish under individual housing remained slow and fast growing fish, respectively, under group housing. This suggests that the different growth rates observed when fish are housed individually are a characteristic of the individual and not simply a consequence of isolation.Chapters 5 to 7 used individually housed fish to supply experimental data on inherent factors responsible for individual variation in growth. Chapter 5 quantified individual differences in performance traits and feeding behaviour and focused on the repeatability of such individual differences when fish were fed ad libitum. Fish exhibited pronounced individual variation in growth (CV = 52.8 %), in feed intake (34.3 %) and in total feeding time (>100 %). The repeatability estimates were 0.55 for growth, 0.70 for feed intake, 0.49 for feed efficiency and 0.81 for total feeding time. These high repeatability estimates suggested that individual differences in growth, feed intake/efficiency and feeding behaviour are consistent over time and therefore probably inherent. Individual differences in growth were explained mainly by individual differences in feed intake (~85 %). Individual differences in feeding behaviour were shown to be related to feed efficiency, measured as residual feed intake (i.e., the difference between actual feed intake and that predicted from mean observed requirements for growth and maintenance). With increasing total feeding time, the maintenance requirements also increased suggesting that slow eaters have higher maintenance costs. Chapter 6 tested whether individual differences in feeding behaviour explained the differences in growth rate by affecting feed efficiency, using restrictively fed fish. This study showed that despite the low variation in initial body weight (6.5 %) and in cumulative feed consumption (7.5 %) over the experimental period, catfish exhibited high variation in final body weight (18.1 %), specific growth rate (17.2 %) and feed conversion ratio (27.9 %), suggesting that individual variation in growth/feed efficiency is important in determining growth. This individual variation may be related to individual differences in protein/fat deposition since faster growing fish deposited more protein and less fat than slower growing fish. Pronounced individual differences in feeding behaviour (reaction towards feed and time spent eating) were also observed and correlated to individual differences in growth/feed efficiency. Fast eaters were the fast growers.Chapter 7 presented two experiments to investigate individual differences in basal and post-stress levels of glucose, lactate and cortisol and their relation to individual differences in feed efficiency. There was a pronounced individual variation in both basal and post-stress levels of plasma glucose, lactate and cortisol. Basal levels of glucose, lactate and cortisol did not contribute significantly to explain differences in feed efficiency. However, glucose levels obtained after a stress test (netting) could explain differences in feed efficiency by 1.3 % in experiment 1 and 5.9 % in experiment 2. In experiment 2, the cortisol levels obtained after the stress test also explained part of the differences in feed efficiency (8.7 %). Apparently, high stress responders are less efficient fish. The stress response probably adds to differences in maintenance costs, thereby affecting the feed efficiency.The findings of this thesis are discussed and the main conclusions are presented in Chapter 8 . The importance of social hierarchy in explaining individual growth variation should be considered species-dependent. In addition, social hierarchy should not be accepted a priori as the major cause of individual growth variation without previous investigation. The results of this thesis suggested that in African catfish the individual variation in growth is not the result of marked dominance-subordinance relationships. Instead, genetic-based differences in feed intake, feed efficiency, feeding behaviour and stress response seem to play a role in explaining growth variation in African catfish. In practical terms, one may question the use of grading in this species as grading is done under the assumption that it disrupts an existing social hierarchy. Furthermore, the results of this thesis called for the development of selection programmes in African catfish. Selecting for feed efficiency (residual feed intake) is a promising direction to pursue. The most efficient fish (low residual feed intake) were shown to be fast eaters and low stress responders which may be advantageous under aquaculture conditions.It was also concluded that individual differences in feed intake and feed efficiency (residual feed intake) contributed ~85 and ~15 %, respectively, to the individual differences in the growth of African catfish. Individual differences in feeding behaviour (total feeding time) and stress response (plasma glucose and cortisol after an acute stress) contributed to explain variation in residual feed intake (maintenance requirements) up to 8.7 %.Despite the results obtained in this thesis, our understanding of the causes of growth variation in African catfish is far from being complete. The challenge is to find the mechanisms responsible for this variation and how they are related to the individual differences in behaviour and stress response found in this thesis