We evaluated several techniques for their ability to record membrane potential changes with voltage-sensitive dyes introduced into CNS neurons in the brain slice preparation. Using a probe designed for intracellular application, JPW1114, we found that iontophoresis or pressure pulses could not push the lipophilic dye through electrodes whose resistance was sufficiently high to produce good electrical recordings in cerebellar Purkinje neurons. However, properly selected patch electrodes could introduce the dye into the cell and still give good electrical records. Using this technique we recorded depolarizing and hyperpolarizing transients and climbing fiber responses using either a single photodiode or a fast, cooled CCD camera. While these results are promising, there are still problems due to the slow diffusion of the dye in the dendrites and a low sensitivity which requires signal averaging to acquire traces with a good signal to noise ratio.