Retinoic acid (RA) is involved in the control of a variety of differentiation processes. In vitro experiments have clearly shown that RA blocks myoblast proliferation and induces their differentiation. We are studying RA function in myogenesis control during zebrafish development. To determine whether RA acts on muscle development in vivo, we altered RA signaling by treating zebrafish embryos either with exogenous RA or with retinaldehyde dehydrogenase inhibitors (DEAB or Citral), or with a pan-RA receptor antagonist, reagent BMS493. We have shown that RA is able to induce expression of early and late muscle differentiation factors (MyoD, myohenin, mef2c...). On the contrary, in DEABor BMS493-treated embryos expression of these myogenic genes was reduced. Analysis of expression patterns suggests that RA could participate to the establishment of a differentiation zone in the anterior presomitic mesoderm (psm). We have analysed “classical” muscle signalling pathways and showed that neither Hh/hedgehog nor Wnt/wingless pathway seems to be involved. We have analysed the expression of fgf8 gene and showed that fgf8 expression is controlled in a similar way as myogenic factors albeit at earlier stages and independently of an intermediate translation. Furthermore we showed that in ace mutants (the zebrafish mutant acerebellar (ace) is devoid of fgf8), RA activation of myogenic factors is blocked. Taken together these elements strongly suggest that Fgf8 mediates the differentiating effect of RA on myogenesis. We are currently investigating the involvement of other Fgfs and of other gene networks through a functional genomic systematic analysis.