Restraint system effectiveness for rear seat adult (16 years or older) car occupants is estimated by applying the double pair comparison method to Fatal Accident Reporting System (FARS) data for 1975 through 1985. As this data set contains information on fatal crashes only, the results apply exclusively to fatalities, and should not be generalized to other levels of injury. Rear seat occupants coded as using any restraint system are assumed to be using the lap belt only. Occupants in all four outboard seating positions (that is, driver and right front passenger, right and left rear passengers) serve as "other" occupants. Disaggregating the "other" occupant by restraint use generates six estimates of restraint system effectiveness for each of the two rear outboard positions. Insufficient data precluded estimating effectiveness for the center rear (or center front) positions, and also use of these occupants as "other" occupants. Average restraint system effectiveness for the two outboard rear seating positions is estimated as (18 +/- 9)%, where the error limit indicates one standard error. These estimates suggest that there is a 39 in 40 chance that rear lap belts reduce fatality likelihood, but a less than 1 in 10 chance that the reduction exceeds 30%.