Fluorescent dyes were used in the present study to map three rostrocaudal levels of the primitive streak of mouse embryos at 7.5 days of gestation. Injections at the most rostral level, the 1/4 streak level, principally labeled cells of the paraxial mesoderm, including mesenchyme in the head, and somites and segmental plate mesoderm in the trunk. Injections at the intermediate level, the 1/2 streak level, principally labeled cells of the lateral plate mesoderm of the trunk. Injections at the most caudal level, the 3/4 streak level, principally labeled cells of the extraembryonic mesoderm at the periphery (i.e., caudolateral portion) of the embryo adjacent to the developing amnion, yolk sac, and allantois. However, injections placed at all three primitive-streak levels provided some labeled cells to the areas of the ingressed mesoderm derived from immediately adjacent levels, suggesting that prospective mesodermal areas are not fully spatially segregated from one another at this stage. These data combined with those from a companion study in which the structure of the node was defined and its derivatives were mapped (Sulik et al.  Dev. Dyn. 201:260—278) allowed us to construct a prospective fate map of the mouse primitive streak at 7.5 days of gestation. This map revealed that mediolateral subdivisions of the ingressed mesoderm were arrayed in rostrocaudal sequence within the primitive streak, such that more rostral levels of the streak contained prospective medial cells (e.g., notochord and paraxial mesoderm), intermediate levels contained prospective intermediate cells (e.g., intermediate mesoderm or nephrotome and lateral plate mesoderm), and more caudal levels contained prospective lateromost cells (e.g., extraembryonic mesoderm). Prospective mesodermal cells, with the exception of those of the notochord, ingressed bilaterally from the primitive streak, contributing to the mesoderm on both the right and left sides of the axis. Our prospective fate map of the mouse primitive streak at 7.5 days of gestation, when compared to a map of the avian primitive streak at a comparable stage of development (Schoenwolf et al.  Dev. Dyn. 193:235—248), demonstrates that considerable congruity exists between the locations of mesoderm precursor cells in birds and mammals. Thus, similar morphogenetic movements underlie mesoderm formation during avian and mammalian gastrulation. Moreover, prospective mesodermal locations within the mouse primitive streak correspond to recently described patterns of early gene expression (Sasaki and Hogan  Development 118:47—59; Takada et al.  Genes Dev. 8:174—189; Rossant, personal communication). © 1994 Wiley-Liss, Inc.