Mestranol [3-Methoxy-19-nor-17α-pregna-1,3,5 (10)-trien-20-yn-17-ol (C21H2602)] was tested at 2 percent (active) as a repellent for protecting Douglas-fir (Pseudotsuga menziesii) seed from deer mice (Peromyscus maniculatus). In 5-day laboratory bioassays, deer mice consumed 61 to 66 percent fewer mestranol-treated seeds than control seeds; these results were about equal to those with a standard 0.5 percent (active) endrin seed treatment. Deer mice showed a progressive aversion to the mestranol seed treatment from 24 percent to 76 percent in 5 days. Thereafter, with minimal reinforcement, avoidance was maintained at 90 to 99 percent for 6 months. In six field trials in Washington, Oregon, and California, areas seeded with 2 percent mestranol-treated Douglas-fir seed yielded 1.6 to 5.9 times more germinants than areas seeded with control seed. In three of these areas, endrin seed treatments were included; they yielded 1.2 to 3.4 times more germinants than the mestranol treatment and 1.9 to 17.3 times more germinants than the control seed. Although the endrin treatments yielded higher numbers of germinants, the mestranol treatments in these tests generally resulted in acceptable numbers of germinants for first-year stocking. Mestranol's nontoxic, nonpersistent properties plus the aversion shown by deer mice to mestranol in our tests makes it a leading candidate as a Douglas-fir seed protectant in western United States.