Both explant and dispersed cell culture preparations of brain tissue provide a means to directly assess the functions of neuropeptide-containing brain cells isolated from the complex influences present in vivo. The validity of the approach depends on reproducibility of observations. In dispersed cell cultures, we find that cell responses, as determined by secretion of the peptide somatostatin, have remained relatively constant both quantitatively and qualitatively over numerous preparations. In addition, for pharmacologic studies on somatostatin secretion, data from several laboratories are in good agreement. The validity of the dispersed cell approach also depends on whether the pharmacologic and physiologic behavior of cells parallels that expected of excitable tissue. The variability of the explant cultures from culture dish to culture dish makes quantitative experiments, such as demonstrated with the dispersed cultures, difficult. On the other hand, explant cultures better maintain the integrity of the tissue components, so that interactions between neurons and glial cells could occur as in vivo. The long-term health and viability of neuropeptidergic cells in explant and dispersed culture make both preparations potentially useful models to examine central nervous system physiology. Future work with such preparations must eventually address the problems of culturing adult brain tissue, the precise nutrient and hormonal requirements of brain cells, so that undefined medium components, such as serum, can be eliminated from the culture environment, and the general question of whether observations made in vitro facilitate our understanding of intact brain physiology.