The purpose of this study was to investigate whether endometrium with retarded development differs, functionally, from endometrium with normal 'in-phase' development. Precisely timed endometrial biopsies were obtained from 24 women suffering from unexplained infertility at 4, 7, 10 and 13 days following the luteinizing hormone (LH) surge. Frozen sections were labelled with an anti-placental protein (PP) 14 monoclonal antibody using an avidin-biotin peroxidase technique and semi-quantification of endometrial PP14 was performed using a Quantimet 970 image analyser. Serum PP14 and saliva progesterone were measured for each patient. Data were analysed using one- and two-way analysis of variance. Normal and retarded endometrium were identified in 16 (group I) and eight (group II) women respectively. Both groups demonstrated a significant increase of the area of precipitate measured for PP14 from day LH + 4 to LH + 13. However, two-way analysis of variance showed that endometrial PP14 was significantly (P < 0.05) lower in the retarded endometrium group at LH + 10 and LH + 13. Serum PP14 was also significantly lower (P < 0.01) in women with retarded endometrial development at LH + 13. Women with normal endometrial development had a significantly higher (P < 0.05) concentration of cumulative saliva progesterone from LH + 3 to LH + 5. This study indicates that there are functional differences between normal and retarded endometrium. These differences may adversely affect uterine receptivity during implantation and the early placentation stage.