1. Single- and multiunit recordings were obtained from neurons in the auditory cortex of the echolocating bat Myotis lucifugus, while trains of stimuli designed to simulate the bat's frequency-modulated (FM) orientation pulse and its returning echo were delivered. It was found that many neurons in the cortex responded selectively to pulse-echo pairs only if the time delay separating the artificial pulse and the echo was within a certain range. This response property is called "delay-dependent facilitation." Since echolating bats are known to utilize echo-delay information for the determination of target distance, it is postulated that these neurons are involved in the process of distance perception. 2. Two types of delay-sensitive neurons were characterized on the basis of their response patterns. P-type units had short maximum response delays, narrow delay response functions, and response latencies for pulse-echo pairs that were similar to their response latencies for single loud FM pulses. E-type units had longer maximum response delays, wide delay response functions, and pulse-echo pair response latencies that were time-locked to the echo. Another important difference between these two classes was that changes in the amplitude of the artificial echo caused systematic changes in the delay response of E-type units but not of P-type units. 3. The sharpness and stability of the delay response functions of P-type units suggested that they may encode target distance by responding at discrete echo delays. In contrast, delay tuning may not be an unambiguous determinant of echo delay in E-type units. Here, the most consistent and reliable response parameter for echo delay is the time at which the responses occurred. This suggested that echo delay could be encoded by the temporal pattern of responses in E-type units in relation to the responses evoked by the outgoing orientation cry. The different range of delay of delay sensitivity of P-type and E-type units indicates that these two mechanisms could be operating at different ranges of target distance. 4. P-type and E-type responses may not be due to different populations of neurons but to different response properties of the same population under different conditions. Evidence for this proposition was obtained by showing that in some recordings, decreases in the amplitude of the artificial pulse caused a switch in response from a long best delay, E-type response to a short best delay, P-type response. This suggested that the delay sensitivity of cortical neurons could be under the bat's control based on the intensity of its pulse emissions.