1. Intracellular recordings were made from 43 spinothalamic (STT) neurons in the lumbosacral region of the spinal cord in anesthetized macaque monkeys. The antidromic responses of these neurons to electrical stimulation of the ventral posterior lateral (VPL) nucleus of the thalamus were examined, and orthodromic responses to electrical stimulation of the sural nerve or to mechanical stimulation of hindlimb skin were recorded to study the electrophysiological features of these neurons and their responses to afferent inputs. 2. The resting membrane potential of the neurons ranged from -26 to -70 mV and the antidromic latency from 2.3 to 9.1 ms. Three of the neurons were located in lamina I and were recorded so briefly that only antidromic and spontaneous activity could be studied. The rest of the neurons were located in laminae III-V and were of the wide-dynamic-range (WDR) type. 3. The antidromic action potentials recorded in the somas of STT neurons typically showed a fast rising phase and a short initial segment-somadendritic (IS-SD) delay. After repetitive antidromic stimulation, a progressive elongation of the IS-SD delay, widening of the spike, and failure of the SD spike were observed. 4. The afterpotential of the antidromic action potential consisted of a fast afterhyperpolarization (AHP(f)) and sometimes a delayed depolarization (DD) and a slow afterhyperpolarization (AHP(s)). The amplitude and the duration of the AHP(s) were progressively increased when longer trains of stimuli were used. When the membrane potential was hyperpolarized, the amplitude of the AHP(s) decreased, suggesting an involvement of K+ and/or Cl- ions. However, the AHP(s) completely disappeared when the strength of stimulation was adjusted to a level just below the threshold for the axon, suggesting that it was unlikely that recurrent inhibition contributed to the AHP(s). 5. The background activity of 32 STT neurons was analyzed. The membrane potential at which spikes were triggered in these neurons was around -42 mV. The width and the rise time of the spontaneous spikes were shorter than those of antidromic action potentials, although the maximum rate of rise was similar. The heights of the spontaneous spikes were slightly shorter than those of antidromic action potentials. 6. Three types of background activity have been observed. One type had a very low average spontaneous rate with a bursting firing pattern, consisting of a few spikes superimposed on a depolarization. This type of activity was seen mostly in lamina I neurons. The second type of activity had moderate to high spontaneous rate with a fairly constant interval between spikes. The slowly rising membrane potential between each pair of spikes resembled a pacemaker potential. The third type of activity was a mixture of the previous two. 7. The background firing rates of neurons recorded both extracellularly and intracellularly were not significantly different. Linear correlation analysis showed that the resting membrane potential recorded intracellularly was highly correlated with the height and the width of the antidromic action potential, but was not correlated with the background firing rate of STT neurons recorded intracellularly. 8. Electrical stimulation of the sural nerve evoked depolarizations and discharges in 35 STT neurons tested. Judged by the threshold and the latency of the evoked activity, three components could be distinguished corresponding to A-beta-, A-delta-, and C-afferent inputs. These depolarizations were all excitatory postsynaptic potentials (EPSPs), because the amplitude of the evoked depolarization was increased (or decreased) when the membrane potential was artifically hyperpolarized (or depolarized). 9. The threshold for EPSPs evoked by A-delta-input (A-delta-EPSPs) was about 5 times the threshold (5T) for the EPSPS evoked by A-beta-input (A-beta-EPSPs). In most cases, the A-delta-EPSPs could be separated from the A-beta-EPSPs by a small dip in the membrane potential and a discontinuation of spike discharges at that time. Furthermore, the membrane conductance increased more during the A-beta-EPSPs than during the A-delta-EPSPs, judging by the amplitudes of the spikes superimposed on the EPSPs. The latency of the potentials evoked by stimulation of the sural nerve and recorded in the dorsal horn indicated that at least some of the A-beta- and A-delta-EPSPs were evoked by monosynaptic input. 10. In 28 of 35 neurons, EPSPs evoked by C-afferent input (C-EPSPs) were recorded after single shock stimulation of the sural nerve. The threshold for the C-EPSPs was approximately 100T, and stronger stimuli evoked larger responses. 11. Repetitive stimulation (2 Hz) of the sural nerve was used to study the C-EPSPs in 13 STT neurons. Some neurons displayed a very synchronized C-EPSP pattern with a reliable latency, whereas some neurons had an unsynchronized or a mixed response pattern. Both incremental increases ("wind-up") or decreases ("wind-down") of the C-EPSP were observed during repetitive stimulation. In the case of wind-up of the C-EPSP, there was concurrently a wind-down of the A-afferent-evoked response in some neurons and a wind-up in other neurons. During repetitive stimulation, the membrane potential of the neurons recorded did not show any significant change. 12. Hyperpolarizations followed the excitation of some STT neurons by A- or C-afferent inputs. The amplitude of these hyperpolarizations was smaller when the membrane potential was hyperpolarized, suggesting that these potentials were inhibitory post-synaptic potentials (IPSPs). On the basis of the latency and other observations, the IPSPs were more likely to be evoked by small afferent (A-delta- and C-fibers) inputs than by A-beta-afferents. 13. Mechanical stimulation has been used to observe the pattern of EPSPs underlying evoked discharges in 3 STT neurons. Nociceptive stimuli, including pinching and squeezing, evoked larger EPSPs and a higher rate of discharge than did innocuous stimuli, such as brushing.