Discussion

​

  Usually, reef fishes feed on algae, coral, open water or deposited nutrients in rubble, depending on species. Then, the fish immediately swims to a defecation site and deposits a faecal pellet. Finally, the animal return grazing to a feeding spot. From Figure 6 we can assess the positive relationship between traveling distances "feeding site -> defecation site -> feeding site", that is, both estimations d1 and d2 are likewise valid when it comes to compare the overall fishes' motility.

Throughout our investigation, we found that some families or species tend to feed far away from the defecation zone. The pattern that seems to emerge from Figure 1 is that parrotfishes travel the most distance from feeding ground before defecation. Concerning the species-level identification, only the species Colotomus viridescens does not display this pattern. However, the species Chlorurus sordidus and Scarus frenatus are consistent with the model, with Scarus frenatus travelling the most distance between the defecating site and the next feeding site (Figure 3). Most of the observed individuals feed on algae (Figure 2) thanks to their specific teeth allowing to scratch the algae of the surface of the coral (Reef Biosearch 2013). Then, they swim few meters away to different defecation spots as open water (Figure 2).

​

  The parrotfish family use a specific mechanism to eat that release fine algal in the water that surgeonfish family enjoy to eat (Reef Biosearch 2013). Indeed, most observed surgeonfish individuals graze on algae (Figure 5) and then defecate in open water (Figure 2). In Figure 1, we can see that the distance of surgeonfishes to their defecation site is small (0 to 2 meters), suggesting that they do not make use of a LDS and maybe even supply their feeding ground with fertilizing wastes, although this hypothesis remains speculative. Nevertheless, some individuals from Naso elegans and Acanthurus sohal species travel more than 2 meters whereas Zebrasoma xanthurum and Naso hexacanthus species never go up to 2 meters (Figure 3).

It seems however certain that the least mobile fish family, butterflyfishes, take no action at all regarding the disposition of its wastes. In Figure 2, it clearly appears that on whichever ground they feed (in most cases in open water), they also defecate, and the pattern seems remarkably consistent especially after noticing the diagonal line between feeding and defecating sites.

  The feeding-defecating system difference inter or intra-families might be the result of territorial defence. Bellwood (1995) found that Chlorus gibbus swim up to twenty meters away from a feeding spot to defecate whereas Chlorus sordidus stay closer to the feeding area. The difference in defecation behaviour between these two species may be explained by the fact that a territorial species (Acanthurus lineatus) seems to tolerate Chlorus gibbus to graze in its territory whether the individuals defecate away from the area. Indeed, many animals show their territory by their defecation (Nisbet 1983). In our case, the difference intra-family could be explain also by the fact that others territorial surgeonfishes were present as Acanthurus sohal (Alwany 2005).

​

A second possibility to this behavior to defecate few meters away could be an adaptation by positive selection to their feeding behaviour. Indeed, the process used by the parrotfish family to feed is knew to contribute to the bioerosion (Bellwood 1995). Thus, the parrotfish do not defecate in its grazing area to avoid acidic damage to coral structures and thus the loss of its diet. However, when compared to other fish families, the travelled distance between coral feeding and defecating is much shorter and this might be explained by the preference of butterflyfishes and surgeonfishes to feed on algae and suspended nutrients rather than coral, thus lacking the need for such specific behavioral adaptation.

​

  Many studies shown that the fact that herbivores deposit faeces at specific LDS might be a strategy resulting from behavioral selection most likely to avoid gastro-intestinal parasites’ re-infection (Hutchings et al 1999 & Mehlhorn 2008). This adaptive sanitary function has been also well studied in birds, where the parents deposit their faces far away from the nest and consume or carry away the faecal sac of their offspring to avoid the exposure of nestlings to pathogens (Quan et al 2015). We could easily imagine the same adaptation as strategy to lower the risk of parasitic re-infection in fishes as they are fish’s pathogens. It would be interesting to analysis the difference in parasitic rate between these three families to support this assumption.

​

​

Conclusion

​

  It is not clear simply from our brief observations whether any fish family or species do actually systematically use LDS, but a few behavioral patterns nonetheless still emerge. Parrotfishes seem to avoid defecating on top of their favorite feeding ground, corals, by depleting only after making large deviations towards open water areas, roughly 5 to 7 meters on average away from the reef. Surgeonfishes, who feed mostly on algae, do travel small to medium distances from the feeding ground and before defecation, but it is unclear whether this happens because of the baseline motility arising from the constant search for food or if it emerges from behavioral adaptation (for instance, promoting algal growth). A few observations hinted the use by surgeonfishes of specific LDS under rocky structures, but not enough to make assumptions nor statements. Butterflyfishes do not make use of LDS as they mostly remain very close to the foraging ground when defecating, this is further supported by the fact that they mostly feed on suspended nutrients on rubble or open water (no implied impact of waste on foraging ground) and exhibit very low traveling distances, averaging from 0 to 1 meter.

  This specific feeding-defecating behaviour have an important impact in reef ecology. And understanding the pattern of each species will allow us to understand the role of each species in the ecology of the reef. In our study, we shown that the species Chlorurus sordidus and Scarus frenatus have less an effect on the rate of the reef sedimentation than Colotomus viridescens, as they defecate far away from the defecating zones. Although, our sample size is too small to be significant. Nevertheless, it gives us a direction to the understanding of the reef ecology.

​

​

​