Abstract
The gut microbiota of fish larvae evolves fast towards a complex community. Both host and environment affect the development of the gut microbiota; however, the relative importance of both is poorly understood. Determining specific changes in gut microbial populations in response to a change in an environmental factor is very complicated. Interactions between factors are difficult to separate and any response could be masked due to high inter-individual variation even for individuals that share a common environment. In this study we characterized and quantified the spatio-temporal variation in the gut microbiota of tilapia larvae, reared in recirculating aquaculture systems (RAS) or active suspension tanks (AS). Our results showed that variation in gut microbiota between replicate tanks was not significantly higher than within tank variation, suggesting that there is no tank effect on water and gut microbiota. However, when individuals were reared in replicate RAS, gut microbiota differed significantly. The highest variation was observed between individuals reared in different types of system (RAS vs. AS). Our data suggest that under experimental conditions in which the roles of deterministic and stochastic factors have not been precisely determined, compositional replication of the microbial communities of an ecosystem is not predictable.