An ongoing research program to document the development of the sucking response in low- and high-risk newborns is described. Goals are (a) to facilitate early oral feedings, and (b) ultimately to determine whether varying levels of self-regulatory mother-young interaction, uninterrupted by birth, differentially affect life span development. In pilot research, finger sucking opportunities were given twice daily to 2 female and 8 male critically ill premature newborns. A clinical scoring system (range 0--12) measured quality of the sucking response beginning as early as 1 hour of life. The sucking response was present in all 10 newborns. The mean sucking score and standard error of the mean on the first day of life were 6.0 +/- 0.8. Sucking scores did not correlate with birth weight or gestational age, but correlated positively with pH (r = +0.52, p less than .01) and negatively with pCO2 (r = 0.47, p less than .05). Sucking scores generally increased with age and with closely time-related sucking opportunities. Sucking scores were negatively correlated with serum bilirubin levels. Sucking opportunities seemed to facilitate neuromuscular coordination, alert activity, alert inactivity, and deep sleep. A description is given also of a portable electronic suckometer and research nipple developed to quantitatively measure the sucking response of low- and high-risk newborns. The pilot research with this instrument is summarized, as are 3 current studies, 2 of which begin at birth. Sucking is a major component of mother-newborn interaction. Perhaps the isolated transitional newborn is in a nonphysiologic state.