Soft bedding increases the risk for death among prone infants. We compared the softness of beds and bedding and infant sleep position for infants sleeping alone and for those bed sharing. Study design: Questionnaires were used to record the bedding and sleep practices of 218 consecutive African American infants. Enrollment was prospective. Mechanical models were used in the homes of a subgroup to measure the softness of bedding and its propensity to cause rebreathing. Results were compared by using the Student t test, Mann-Whitney U test, and X2 analysis. Results: In a cross-sectional sample of infants, at 8.2 ± 3.3 weeks of age, 61% (133 of 218) had bed shared ≥1 of the previous 14 nights and 48.6% (106 of 218) had bed shared the night before. Breast-feeding rates were not different for bed sharers and those sleeping alone. The rates of maternal smoking for both groups were low (13.6% vs 11.8%). Comforters, pillows, and waterbeds were more commonly used beneath bed-sharing infants. Bed sharers were twice as likely to habitually be placed prone for sleep (18% vs 9%). In the subgroup studied in their homes (15 bed sharing, 19 alone), the shared beds were softer (P < .0001) and could cause more rebreathing (P = .007). Conclusions: Infants at increased risk for sudden infant death syndrome, by sociodemographic criteria, who also bed share are more likely to sleep prone and to use softer beds. These findings may explain part of the risk associated with bed sharing among US infants, a risk that appears to be independent of the effects of maternal smoking.