The long term influence of individual biological rhythms on group feed demand behaviour was investigated in European sea bass (Dicentrarchus labrax) held under controlled environmental conditions with an on-demand feeding system. The experiment was realized over 219 days with 190 fish distributed on 4 tanks. Sea bass had a mean body mass comprised between 139 g to 183 g. The number of feed demand acts by each individual was calculated daily, and the population could thus be partitioned into three categories (high-, low- and zero-triggering fish). The duration of the period that an individual held high-triggering status could vary, but was 63 ± 16 days on average. The transition period between two highest-triggering fish in one tank was on average 4 ± 4 days. The group feeding rhythm followed the same pattern of feed demand rhythm as the highest-triggering individual fish. When the highest-triggering fish was nocturnal, the totality of feed demand in the group was realized during the night with one peak at 22:00, corresponding to dusk under experimental conditions. When the highest-triggering fish was diurnal, the majority of feed demand in the group was realized during the light period with one peak at 06:00, corresponding to dawn, and/or another at 12:00. This study therefore highlights that sea bass group feeding behaviour is not the sum of individual feed demand behaviours, but is directed by the rhythm and behaviour of a few high-triggering fish. The regular changes of high-triggering fish in the group proved that it was not the identity of these particular fish that was most important for the group, but their role as a feed demand leader.

The capability of turbots to use self-feeders was studied using 3 groups of 100 juveniles. Feeding activity reached a plateau after 20 days. Almost all feeding activity was observed during the photophase, and the acrophase was associated with dawn. The proportion of nocturnal demands was the highest in the replicate with the lowest feed efficiency, suggesting that feed waste was mainly occurring during nocturnal feeding activity. During the last 30 days of the experiment, demands were only rewarded during two periods of 2.5 h.day−1. Feed efficiency was improved and was homogeneous between the three tanks, but the voluntary feed intake was reduced. It is concluded that turbots are able to use self-feeders but they do not adapt quickly, and they are not able to compensate a reduction of feed access by increasing demand activity during the access period.

Les Siluroidei sont connus pour avoir, en milieu naturel, une activité trophique essentiellement nocturne. Ceci a été confirmé au cours d'expériences réalisées en conditions d'accès constant à une source de nourriture, ou encore à l'aide de distributeurs d'aliments à la demande. Cependant, en conditions limitantes en température ou en oxygène, l'activité trophique de certains Siluroidei devient dépendante de ces deux facteurs, et n'est plus synchronisée par l'alternance jour/nuit. Le rythme nycthéméral d'activité trophique des espèces étudiées présenterait donc une acrophase préférentiellement nocturne, mais posséderait une certaine plasticité. L'existence d'un contrôle d'origine endogène de ce rythme est aussi discuté. Rares sont les poissons Siluroidei qui refusent de s'alimenter lors de distributions diurnes d'aliments. Il convient d'évaluer les conséquences que pourraient avoir sur la croissance des horaires d'alimentation non synchrones avec les rythmes alimentaires. L'effet de l'heure des repas sur les performances de croissance montrent une tendance générale à une croissance supérieure chez les animaux nourris de nuit par rapport aux animaux nourris de jour. Cependant, le mode de distribution choisi (en repas ou en continu) semble avoir plus d'influence que l'heure de distribution sur les performances de croissance. Certains auteurs ont aussi montré une augmentation de l'adiposité des poissons lorsqu'ils n'étaient pas nourris en phase avec leur rythme alimentaire.

Hoplosterum littorale (Hancock, 1828) shows a marked circadian rhythm of food demand. Feeding activity is mainly nocturnal, with two peaks, and is synchronized with the diel light cycle. We tested the effect of infradian (period >28 h) light/dark cycles on this rhythm in fish with demand-feeders: fish subjected to a 13.5 L/22.5 D light/dark cycle for 8 days and a 25.5 L/10.5 D cycle for a further 8 days showed a strictly nocturnal feeding activity, a response that only passively reflected the cyclic alternation of light and dark.