Antineoplastic treatments for cancer and other non-malignant disorders can result in long-term or permanent male infertility by ablating spermatogonial stem cells (SSCs). SSC transplantation using testicular tissue harvested before a sterilizing treatment is a promising approach for restoring male fertility in these cases, but a lack of exclusive biomarkers to unequivocally identify prepubertal SSCs limits their therapeutic potential. To address this, we performed single-cell RNA-seq on testis cells from immature baboons and macaques and compared these cells with published data from prepubertal human testis cells and functionally-defined mouse SSCs. While we found discrete groups of human spermatogonia, baboon and rhesus spermatogonia appeared less heterogenous. A cross-species analysis revealed cell types analogous to human SSCs in baboon and rhesus germ cells, but a comparison with mouse SSCs revealed significant differences with primate SSCs. Primate-specific SSC genes were enriched for components and regulators of the actin cytoskeleton and participate in cell-adhesion, which may explain why the culture conditions for rodent SSCs are not appropriate for primate SSCs. Furthermore, correlating the molecular definitions of human SSC, progenitor and differentiating spermatogonia with the histological definitions of Adark/Apale spermatogonia indicates that both SSCs and progenitor spermatogonia are Adark, while Apale spermatogonia appear biased towards differentiation. These results resolve the molecular identity of prepubertal human SSCs, define novel pathways that could be leveraged for advancing their selection and propagation in vitro, and confirm that the human SSC pool resides entirely within Adark spermatogonia.