1. Large neurons from layer V of cat sensorimotor cortex (Betz cells) were studied to determine the influence of the anomalous rectifier current (IAR) on slow afterhyperpolarizations (AHPs). The neurons were examined using intracellular recording and single-microelectrode voltage clamp in an in vitro brain slice preparation. 2. A faster medium-duration AHP (mAHP) and slower AHP (sAHP) followed repetitive firing (22, 23). The amplitude of the mAHP often increased or remained constant during membrane potential hyperpolarization. The membrane potential trajectory resulting solely from IAR activation was similar to the mAHP. 3. Postrepetitive firing voltage clamp was used to measure directly slowly decaying K+ currents (IK) and IAR at different membrane potentials. IK exhibited both a fast and slow decay. The time constants of the fast decay of IK and IAR activation were similar. IAR increased with hyperpolarization or raised extracellular K+ concentration [( K+]o), whereas both the fast and slow components of IK reversed or nulled near -100 mV and behaved as pure K+ currents in response to raised [K+]o. 4. To determine the precise contribution of IK and IAR to the AHP waveform, theoretical AHPs were computed using a quantitative model based on voltage-clamp measurements. The calculated AHPs were qualitatively similar to measured AHPs. The amplitude of the mAHP showed little change with hyperpolarization because of the increasing dominance of IAR at more negative membrane potentials. The sAHP was little affected by IAR activation. 5. Several model parameters subject to biological variation among Betz cells were varied in the calculations to determine their importance in the AHP waveform. With IK parameters held constant, the amplitude and time course of the mAHP depended on resting potential, membrane time constant, the kinetics of the anomalous rectifier conductance (GAR), and the maximum value of GAR. IAR activation could result in a biphasic AHP even when the fast decay of IK was omitted from the calculations. 6. A wider variation of model parameters revealed behavior that may be relevant to other neurons. Certain values of membrane or IAR activation time constants resulted in a monophasic AHP even when the fast decay of IK was present. The decay of a biphasic AHP could reflect either the onset of IAR or the fast decay of IK, depending on the relative value of their time constants. Procedures are outlined to discriminate between these possibilities using current clamp methods.(ABSTRACT TRUNCATED AT 400 WORDS)