Since their initial description in 1992, neurospheres have appeared in some aspect of more than a thousand published studies. Despite their ubiquitous presence in the scientific literature, there is little consensus regarding the fundamental defining characteristics of neurospheres; thus, there is little agreement about what, if anything, the neurosphere assay can tell us about the relative abundance or behavior of neural stem cells in vivo. In this review we will examine some of the common features of neurospheres, and ask if these features should be interpreted as a proxy for neural stem cells. In addition, we will discuss ways in which the neurosphere assay has been used to evaluate in vivo treatment/manipulation, and will suggest appropriate ways in which neurosphere data should be interpreted, vis-a-vis the neural stem cell. Finally, we will discuss a relatively new in vitro approach, the Neural-Colony Forming Cell Assay, which provides a more meaningful method of quantifying bona fide neural stem cells without conflating them with more growth-restricted progenitor cells.