The Journal of Comparative Neurology recently published a Commentary concerning the appropriate use of methods for the quantification of structures in biological samples (Coggeshall and Lekan, 1996). The article addresses a critical issue for scientists who wish to pursue a wide range of quantitative neurobiological studies and who must choose among a variety of methods (and variations of methods) that are available. The stereological approach promoted by Coggeshall and Lekan is indeed one of the viable alternatives. As with all methods, it is based on certain assumptions and is not always the most appropriate to apply. Coggeshall and Lekan, however, take a more dogmatic stance in their article; they characterize all nonstereological methods as ‘‘assumption-based’’and, hence, inappropriate for serious studies. With this approach, their Commentary misses a crucial point: All methods of quantification have underlying assumptions and principles. Each must be applied critically, and each has its own set of strengths and weaknesses that must be carefully considered in relation to the problem to be solved, to the type of tissue available, and to the level of accuracy needed, before it is used or its results are interpreted.