Sir.—The recent editorial by Drs Fulginiti and Ray1 concerning body surface cultures in the newborn infant highlighted their very limited value in predicting the cause of sepsis in neonates. In the United Kingdom, too, the usefulness of superficial specimens is being questioned,2 but it will almost certainly require several further studies demonstrating the futility of the exercise before an end to such practice is seen.
It may be important to divide neonatal infection into early-onset sepsis (EOS), ie, occurring within 48 hours of birth, and late-onset sepsis (LOS), ie, occurring after 48 hours. In EOS, the organisms responsible for the majority of infections are acquired either from the mother's lower genital tract or transplacentally.3 Although initial colonization of the neonate is considered essential in the pathogenesis of sepsis, the rapidity with which some bacteria (eg, group B streptococci) can produce fulminant infection means that processing