Levels of norethisterone and medroxyprogesterone acetate were measured in serum and milk of women receiving the injectable contraceptive formulations Norigest and DepoProvera, respectively, throughout complete injection intervals. In 5 of 10 women receiving Norigest, serum norethisterone levels were undetectable by 8 weeks after injection and only 2 women had detectable levels of norethisterone in milk at this time. In contrast, 8 of 10 women receiving DepoProvera had detectable levels of medroxyprogesterone acetate in both serum and milk 12 weeks after injection. The ratio of the milk:serum concentrations of norethisterone varied from 0.12 to 0.92 (mean 0.34) and for medroxyprogesterone acetate from 0.12 to 2.60 (mean 0.88). It is unlikely that these differences between the two formulations are due entirely to differences between the binding of norethisterone and medroxyprogesterone acetate to serum proteins. The area under the curve of serum steroid concentrations plotted against time was only 50% higher for women injected with DepoPovera than for those injected with Norigest but the area under the curve for milk values was 400 times higher. Assuming the infant ingests 600 ml daily, the daily intake of steroids in the first week after injection would be 0.5 to 2.4 micrograms for norethisterone and 1 to 13 micrograms for medroxyprogesterone acetate. By 8 weeks after injection, the amount of norethisterone ingested would be small but that of medroxyprogesterone acetate would still be significant.