Cold Spring Harbor Symposia on Quantitative Biology, Volume LXVII.© 2002 Cold Spring Harbor Laboratory Press 0-87969-678-8/02. 57 and Trasler 1991, 1992) and that lack of migration resulted in cardiovascular defects. Some early studies indicated abnormal migratory behavior of cardiac neural crest cells derived from Splotch embryos (Moase and Trasler 1990), and later studies appeared to confirm these results (Conway et al. 1997). However, subsequent reanalysis of these latter studies suggests that the migratory population dependent on Pax3 is actually a pool of hypoglossal and hypaxial muscle progenitors rather than cardiac neural crest (Epstein et al. 2000). We reexamined neural crest migratory behavior in Splotch embryos using a transgenic mouse line to aid in the identification of neural crest cells. A portion of the connexin 43 (Cx43) gene upstream regulatory region was used to drive expression of lacZ and was found to direct expression to neural crest cells in developing embryos (Waldo et al. 1999). We crossed Cx43-lacZ mice with Splotch mice and examined neural crest patterning at E12. 5 in wild-type and homozygous Splotch embryos. Interestingly, we found significant numbers of labeled cells that had migrated throughout the branchial arches, encased the aortic arch arteries, and invaded the outflow tract of the heart in both wild-type and mutant embryos (Epstein et al. 2000). The precise patterning of neural crest derivatives in the heart was not identical since the outflow tract septum was poorly formed or absent in mutant embryos, and there appeared qualitatively to be fewer labeled cells at the most distal zones of migration in Splotch embryos. In addition, in some embryos, labeled cells were not as tightly clustered in the region of the endocardial cushion in mutant embryos when compared to wild type. Nevertheless, the unequivocal findings from these studies were that neural crest migration is not entirely dependent on Pax3 function, and the Splotch cardiac phenotype including persistent truncus arteriosus is not due to absence of neural crest cells in the heart. Rather, the defect is more subtle, stochastic, or related to postmigratory neural crest function.