We previously found that long-term potentiation (LTP) in the dentate gyrus (DG) is attenuated by lesion of the basolateral amygdala (BLA), but it remained unclear whether or not there is neural connection between the BLA and the DG. In the present study, we tried to provide physiological evidence that the DG receives neural inputs from the BLA. Single-pulse electrical stimulation of the BLA evoked two distinct components of field potentials in the ipsilateral DG: the P1 component with 26-ms peak latency was elicited by lower intensity of BLA stimulation, whereas the P2 component with 14-ms peak latency was elicited at higher stimulus intensity. The PI response (1) was evoked when the stimulating electrode was positioned within the BLA, (2) showed the laminar profile similar to the perforant path (PP)-evoked response and (3) exhibited strong paired-pulse facilitation. On the other hand, the P2 wave (1) was evoked even with the stimulating electrode outside the BLA, (2) did not reverse its polarity at any location in the DG and (3) showed only slight facilitation by paired-pulse stimulation. These data indicate that the P1 component represents synaptic responses of DG granule cells to neural inputs from the BLA. Furthermore, the BLA-evoked DG field potentials were not affected by local injection of tetracaine into the PP and displayed LTP independently of the PP-evoked response, suggesting that neural inputs from the BLA are not mediated by the PP.