Respiratory-distress syndrome (RDS) in the newborn is a major cause of neonatal mortality and morbidity. Although prematurity is the most-important risk factor for RDS, the syndrome does not develop in many premature infants. The main cause of RDS is a deficiency of pulmonary surfactant, which consists of phospholipids and specific proteins. The genes underlying susceptibility to RDS are insufficiently known. The candidate-gene approach was used to study the association between the surfactant protein A (SP-A) gene locus and RDS in the genetically homogeneous Finnish population. In the present study, 88 infants with RDS and 88 control infants that were matched for degree of prematurity, prenatal glucocorticoid therapy, and sex were analyzed for SP-A genotypes. We show that certain SP-A1 alleles (6A2 and 6A3) and an SP-A1/SP-A2 haplotype (6A2/1A0) were associated with RDS. The 6A2 allele was overrepresented and the 6A3 allele was underrepresented in infants with RDS. These associations were particularly strong among small premature infants born at gestational age <32 wk. In infants protected from RDS (those that had no RDS, despite extreme prematurity and lack of glucocorticoid therapy), compared with infants that had RDS develop despite having received glucocorticoid therapy, the frequencies of 6A2 (.22 vs.71), 6A3 (.72 vs.17), 6A2/1A0 (.17 vs.68), 6A3/1A1 (.39 vs.10), and 6A3/1A2 (.28 vs.06) in the two groups, respectively, were strikingly different. According to the results of conditional logistic-regression analysis, diseases associated with premature birth did not explain the association between the odds of a particular homozygous SP-A1 genotype (6A2/6A2 and 6A3/6A3) and RDS. In the population evaluated in the present study, SP-B intron 4 variant frequencies were low and had no detectable association with RDS. We conclude that the SP-A gene locus is an important determinant for predisposition to RDS in premature infants.