Background: Whether the adult mammalian heart harbors cardiac stem cells for regeneration of cardiomyocytes is an important yet contentious topic in the field of cardiovascular regeneration. The putative myocyte stem cell populations recognized without specific cell markers, such as the cardiosphere-derived cells, or with markers such as Sca1(+), Bmi1(+), Isl1(+), or Abcg2(+) cardiac stem cells have been reported. Moreover, it remains unclear whether putative cardiac stem cells with unknown or unidentified markers exist and give rise to de novo cardiomyocytes in the adult heart. Methods: To address this question without relying on a particular stem cell marker, we developed a new genetic lineage tracing system to label all nonmyocyte populations that contain putative cardiac stem cells. Using dual lineage tracing system, we assessed whether nonmyocytes generated any new myocytes during embryonic development, during adult homeostasis, and after myocardial infarction. Skeletal muscle was also examined after injury for internal control of new myocyte generation from nonmyocytes. Results: By this stem cell marker-free and dual recombinases-mediated cell tracking approach, our fate mapping data show that new myocytes arise from nonmyocytes in the embryonic heart, but not in the adult heart during homeostasis or after myocardial infarction. As positive control, our lineage tracing system detected new myocytes derived from nonmyocytes in the skeletal muscle after injury. Conclusions: This study provides in vivo genetic evidence for nonmyocyte to myocyte conversion in embryonic but not adult heart, arguing again the myogenic potential of putative stem cell populations for cardiac regeneration in the adult stage. This study also provides a new genetic strategy to identify endogenous stem cells, if any, in other organ systems for tissue repair and regeneration.