Gene expression in luteinizing hormone-releasing hormone (LHRH) neurons was analyzed during the periovulatory period to (1) characterize temporal patterns of LHRH gene expression and their relationship(s) to gonadotropin surges, and (2) determine if any such changes are uniform or dissimilar at different rostrocaudal levels of the basal forebrain. The number of neurons expressing mRNA for the decapeptide, and the relative degree of expression per cell were analyzed using in situ hybridization and quantitative image analysis. Rats were killed at 1800 hr on metestrus (Met), 0800 hr, 1200 hr, 1800 hr, and 2200 hr on proestrus (Pro), or 0200 hr, 0800 hr, and 1800 hr on estrus (E; n = 5-6 rats/group). All sections were processed for LHRH mRNA in a single in situ hybridization assay. Sections were atlas matched and divided into four rostrocaudal groups for analysis: vertical limb of the diagonal band of Broca (DBB), rostral preoptic area/organum vasculosum of the lamina terminalis (rPOA/OVLT), medial preoptic area (mPOA), and suprachiasmatic/anterior hypothalamic area (SCN/AHA). Plasma LH and FSH levels from all animals were analyzed by RIA. The labeling intensity per cell was similar among all time points at all four rostrocaudal levels. The number of cells expressing LHRH mRNA, however, varied as a function of time of death during the estrous cycle, and this temporal pattern varied among the four anatomical regions. At the level of the mPOA, the number of cells was highest at 1200 hr on Pro, and then declined and remained low throughout the morning of E. At the level of the rPOA/OVLT, the greatest number of LHRH neurons was noted later in Pro, at 1800 hr, dropping rapidly to lowest numbers at 2200 hr. No significant changes in LHRH cell number occurred at the DBB or SCN/AHA levels. At all anatomical levels, the secondary surge of FSH was unaccompanied by any change in the number of neurons expressing LHRH mRNA. These data demonstrate that (1) the number of detectable LHRH mRNA-expressing cells fluctuates during the periovulatory period and (2) peak numbers of LHRH-expressing cells are attained in the mPOA before the onset of the LH surge and before peak LHRH cell numbers are seen at more rostral levels. A model is proposed in which gene expression in this subpopulation of LHRH neurons may be activated by preovulatory estrogen secretion and acutely reduced following the proestrous surge of progesterone.