The hypothesis tested was that increasing concentration of 17beta-estradiol (E(2)) subsequent to luteolysis stimulates the preovulatory surge of LH and that a decline in E(2) after the initial rise is not necessary to trigger the preovulatory surge of LH in the bovine female. Beef cows were synchronized to Day 16 of the estrous cycle. At Hour 0, all cows were ovariectomized and received one of four E(2) treatments: 1) luteal phase E(2) (LE; n=5), 2) increasing then decreasing E(2) (DE; n=5), 3) increasing and subsequent maintenance of high E(2) (IE; n=4), and 4) no E(2) (NE; n=3). Cows in the LE group received one E(2) implant at Hour 0 which provided low concentrations of E(2). Cows in the DE group received one E(2) implant at 0, 8, 16, 24, 32 and 40 hours; implants were subsequently removed at 8-hour intervals, thus mimicking the preovulatory rise and fall of E(2). Cows in the IE group were treated with the same regimen of E(2) implants as cows of the DE group, except that no E(2) implants were removed. Blood samples were collected at Hour 0 and at hourly intervals from Hour 2 through 80, for serum LH and E(2) quantification. The number of cows responding with a surge of LH was 0/3, 0/5, 4/5 and 3/4 for the NE, LE, DE and IE treatments, respectively. The proportion of cows responding with an LH surge was different (P<0.01) when data for cows in the NE and LE groups were pooled and compared with the pooled data of cows in the DE and IE groups. The mean time of the LH surge was not different (P>0.80) for cows responding with an LH surge (DE and IE treatments). Thus, increased levels of E(2) greater than luteal phase concentrations are needed to initiate preovulatory surges of LH, and it appears that concentrations of E(2) need to reach a certain level but do not need to decrease after this initial rise to stimulate a surge release of LH.