We examined the roles of fibroblast growth factor (FGF)-2 and FGF-8 in the migration of mesencephalic mouse neural crest cells. Our in vitro migration assay has shown that FGF-2 (basic FGF) and FGF-8 have chemotactic activity for these cells. Chemotaxis was inhibited by anti-FGF-2 and anti-FGF-8 neutralizing antibodies. In addition, anti-FGF-2 blocked neural crest cell migration in cranial organ cultures. This observation suggests that FGF-2 functions as a chemoattractant in migration of mesencephalic neural crest cells in vivo. In organ culture, the antagonist of FGF binding to a low-affinity fibroblast growth factor receptor (FGFR) heparan sulfate, inositolhexakisphosphate (InsP6), inhibited migration as well. Mesencephalic neural crest cells had high-affinity FGFRs, in particular FGFR-1 and FGFR-3. Thus, the chemotactic activities of FGF-2 can be mediated by the low-affinity FGFR alone or by a combination of low- and high-affinity FGFRs (FGFR-1, FGFR-3, or both). Moreover, differential localization of FGF-2 was found at the mesencephalic axial level of intact embryos during neural crest cell migration. FGF-2 protein expression was predominant in the target regions, in particular the mandibular mesenchyme, that are colonized by mesencephalic neural crest cells. This characteristic distribution supports the notion that FGF-2 acts as a chemoattractant in the mouse embryo that directs mesencephalic neural crest cell migration. Whereas FGF-8 showed chemotactic activity in vitro, neural crest cell dispersion was observed in explants that had been treated with anti-FGF-8 neutralizing antibodies. This result suggests that FGF-8 may not be a chemoattractant in vivo. However, the distribution of neural crest cells in explants treated with anti-FGF-8 differed from that in control explants or in intact embryos. Extreme FGF-2 distribution was observed in the mandibular arch and FGF-8 is expressed in the epithelium. FGF-8 may play a role in mesencephalic neural crest cell migration, and its role may be concerned with the differential localization of FGF-2. To establish this notion, we performed immunohistochemical examination of FGF-2 distribution in explants treated with FGF-8 and analysis of FGF-2 gene expression levels by reverse transcriptase-polymerase chain reaction by using RNA from explants. The data indicate that FGF-2 is distributed throughout the mesenchyme in FGF-8-treated explants and that expression of FGF-2 is promoted by FGF-8. Therefore, we conclude that the expression of FGF-8 in the mandibular arch epithelium is a prerequisite for the differential localization of FGF-2 and that the FGF-2 distribution pattern is essential for chemotaxis of mesencephalic neural crest cell migration.