Fast-spiking, parvalbumin-expressing basket cells (BCs) play a key role in the function of neuronal networks, such as feedforward inhibition, feedback inhibition, and oscillatory activity. In the dentate gyrus, BCs are activated by excitatory synapses emerging from perforant path, commissural / associational, and mossy fiber inputs. How synaptic events are integrated in the apical and basal dendrites of BCs to generate a unified action potential output, however, has remained unknown. To assess dendritic integration in BCs quantitatively, we developed detailed passive cable models of these neurons. Passive voltage responses to short and long current pulses (0.5 or 2 ms, 100–500 pA; 500 ms, 20–40 pA) were recorded from BCs in hippocampal slices from 17- to 19-day-old rats at 32–34°C in the presence of tetrodotoxin, blockers of excitatory and inhibitory synaptic currents, and (in a subset of experiments) the I_h channel blocker ZD7288. BCs were filled with biocytin during recording, visualized by 3,3'-diaminobenzidine after fixation, and identified based on the axonal arborization in the granule cell layer. Soma, dendrites, and the entire axonal arborization were reconstructed with a Neurolucida system. In eight fully reconstructed BCs, the mean surface areas of basal dendrites, apical dendrites, and axon were 3260, 9597, and 39034 mm², respectively. In a subset of five BCs, a dual somatic recording configuration was used to minimize series resistance artifacts. Uniform cable parameters R_m, C_m, and R_i were obtained by direct fitting of the experimental voltage transients with the model, using the simulation platform NEURON 6.1. On average, we obtained R_m = 18.5 8.9 kW cm², C_m = 1.0 0.1 mF cm^-2, and R_i = 122 16 W cm (n = 5). To assess nonuniformity of cable parameters, we further analyzed a subset of three BCs in which somato-dendritic recordings had been obtained (distance of dendritic recording sites 120, 180, and 220 mm from the soma). Our analysis suggests significant nonuniformity of R_m, which is highest in the axon, intermediate in the distal dendrites, and lowest in the proximal dendrites. Thus, in comparison to hippocampal pyramidal neurons (Golding et al., 2005, J. Physiol. 568:69), BCs show a reversed distance dependence of R_m along the dendritic axis. Simulation of synaptic events in BC cable models revealed that (1) summation of excitatory postsynaptic potentials is most effective if distal inputs are activated within a narrow time window and are distributed maximally, (2) the extensive axonal arbor of BCs has little influence on dendritic integration, and (3) shunting inhibition on proximal dendrites significantly reduces excitatory synaptic events generated on distal dendrites if the inhibitory synapse is "on path".