The issue of whether or not neurons die with age is of fundamental importance for understanding senescent decline in motor, sensory, and cognitive functions. Yet after over 100 years of attempts to quantify age-related neuron loss, this fundamental issue is "swirling in controversy" and the results of numerous studies addressing this issue "rife with confusion and seemingly contradictory data" (7). This is in large part due to deficiencies in the way we have thought about what constitutes evidence for neuron loss and deficiencies in the methods that have been available for counting neurons. Over the past several years, a new generation of stereological principles have been developed that have made it possible to identify and eliminate these deficiencies. Although there are now a number of review articles that briefly describe these and other new stereological principles (4,9,13,14,15,18,23), for the most part detailed descriptions of the new principles for counting neurons are found in specialized literature and are accompanied by new terms and concepts that have overwhelmed potential users and inhibited their adoption. In this essay, I describe the principles and concepts that are involved in the new stereological approach to neuron counting and discuss why the data obtained from their application represent improvements over those obtained with previously available techniques. A practical guide to the application of the principles to counting neurons or any other cell, organelle, or object present in sectioned material has been included as an appendix.