This study used telemetry to determine spawning behavior and subtidal habitat use of horseshoe crabs Limulus polyphemus. We attached combined acoustic and radio transmitters to 12 gravid female horseshoe crabs at Ted Harvey Beach and 12 at North Bowers Beach (both on the western shore of Delaware Bay) over a 5-d period before peak spawning on the new moon. Horseshoe crabs were acoustically tracked and radio-tracked daily for 8 d during both high tides and during the incoming dominant (higher) high tide. All horseshoe crabs were relocated at least once, and 83% of females spawned from two to six times (x?? = 3.35, SE = 0.18). Of these females, 85% spawned on two to five consecutive nights (x?? = 3.31, SE = 0.59). Most (95%) females spawned on the beaches where they were initially tagged. Typically, the shoreline used by an individual for spawning ranged from 70 to 1,160 m (x?? = 351 m, SE = 38 m). Between spawning events, horseshoe crabs remained 50-715 m offshore (x?? = 299 m, SE = 57 m) from their established spawning beaches. Following the new moon, all but one (96%) moved out of range of our survey area, which extended approximately 1 km from the shoreline. Multistate mark-recapture models were used to estimate recapture probabilities and daily probabilities of spawning and departure from the vicinity of the spawning beaches. The probability of recapture by acoustic telemetry was high and estimated to be 0.95 (95% confidence interval, 0.73-0.99). Horseshoe crabs equipped with acoustic and radio transmitters have high rates of recapture, can be tracked continually, and can be relocated in both foreshore and inshore habitats.