The aim of this study was to clarify changes in sex steroids at the final menstrual period (FMP). We have shown previously that estradiol (E2) declines substantially in the 4-year period around the FMP, but hypothesize that testosterone (T) declines modestly and adrenal Δ5 androgens dehydroepiandrosterone (DHEA) and androstenediol (Adiol) remain unchanged. Methods: Liquid chromatography tandem mass spectrometry (LC-MS/MS) and immunoassay was used in approximately annual samples collected before and following FMP in 1490 women. We estimated time-related changes in each log-transformed androgen using piecewise linear mixed modeling, with knots (slope changes) at FMP-2 yrs and FMP+2 yrs as seen for E2. These models then were re-estimated for subgroups with different time courses identified using group-based trajectory modeling. Results: In the full sample, T was generally stable, although time course varied by subgroup, with a significant decrease of 5%/year in T in [FMP-2yrs, FMP+2yrs] only in the lowest T women. For DHEA and Adiol, declines were similar across all 3 time segments and across subgroups. Mean circulating androgen concentration declined modestly (P> 0.05) from five years before to five years following FMP. However, when stratified only the lowest 7% of circulating T declined significantly (p< 0.05) in the four years surrounding FMP when mean circulating E2 declined. This trajectory divergence of the lower circulating T suggests a different, non-adrenal source that is decreased at FMP which may be useful in clarifying ovarian versus adrenal testosterone production during the post-menopause. Paired results from samples collected before and following FMP in the same subjects indicate mean circulating E2 is less than 5% of mean circulating T suggesting that a relatively large portion of circulating E2 may be largely a result of peripheral conversion of adrenal androgens. Longitudinal LC-MS/MS analyses of circulating E2 and T indicate that the principal change in sex steroid influence at menopause is largely a decrease and dampening of ovarian and not adrenal steroid production.