1. Intracellular recordings were made from 78 phrenic motoneurons (PM) in anesthetized, paralyzed, artificially ventilated cats that were slightly hypercapnic. 2. Three subpopulations of PM (types A, B, and A/B) were identified on the basis of their membrane potential trajectories during expiration (E). Type A cells exhibited wholly linear trajectories. These were rapidly hyperpolarized at the onset of E followed by a slow ramp of increasing hyperpolarization observed in 51 of 59 type A cells. Types B (13 cells) and A/B (6 cells) had nonlinear trajectories in E. Type B cells approached their end-expiratory potential levels more slowly. 3. Measurements of axonal conduction velocity, expiratory phase input resistance, initial depolarization rate, and initial spike onset during inspiration revealed that type B cells had significantly slower axonal conduction velocities, higher input resistances, greater initial depolarization rates, and earlier initial spike onsets than type A cells. The properties of type A/B were intermediate between the other cell types. These results support the hypothesis that the PM pool is not homogeneous. 4. Active E-phase inhibition of all types of PM was directly demonstrated by reversal of the increasing hyperpolarizing wave to a depolarizing wave with hyperpolarizing current injection using a bridge circuit. Thus hyperpolarization of PM during E is not merely due to a central disfacilitation. 5. During hyperpolarizing current injection the inspiratory phase membrane potential trajectory of all PM became a ramp depolarization similar to that seen during control conditions in type A cells. These results support the conclusion that all cells within the PM pool are receiving a similar central excitatory synaptic input during inspiration. The rapid initial depolarization of type B and their concomitant early spike onset is a consequence in part of a rebound excitation from their expiratory phase inhibition as well as a higher input resistance, probably due to a smaller cell size. 6. Expiratory related neural activity was recorded within the phrenic motor nucleus. It is suggested that these expiratory related neural elements, based on the temporal pattern of their activity, may be responsible for the active inhibition of PM.