Spontaneous prolongation of the luteal phase has been described in horses, but the underlying causes are still unclear. The present study investigated details of gonadotrophin and progestogen secretion in pregnant mares (n = 11) with or without experimentally reduced early postovulatory luteal function. From days 0 to 3 after ovulation, they were treated with the prostaglandin F2α (PGF2α) analogue cloprostenol or left untreated. After conceptus collection on day 34, they were assigned to the opposite treatment. Mares were affiliated to the group primary corpus luteum (n = 6) or diestrous corpus luteum (n = 5) depending on diestrous corpus luteum (CL) detection in the PGF pregnancy. For statistical comparisons, a p-value < 0.05 was significant. There was an effect of treatment (p < 0.01), but not of group on progestogen concentration. The concentration of LH was higher in PGF-treated than in untreated pregnancies (p < 0.05), but did not differ between groups. The FSH concentration did not differ between groups nor treatments. The total luteal tissue area was greater in mares with a diestrous ovulation during the PGF treatment pregnancy. Low progestogen concentration in the early postovulatory phase diminish the negative feedback on the hypothalamic-pituitary axis in early pregnancy and, thus, stimulate a luteal tissue response. Detection of secondary CL at the time of pregnancy examination in mares may reflect that early post-ovulatory progestogen concentrations were low.