Changes in cortisol and behavioral responses were examined longitudinally in 83 infants (39 girls, 44 boys) tested at their well-baby exams with inoculations at 2, 4, 6, and 15 months (72 infants completed all testing). Another sample of 2-, 4-, and 6-month-olds (n = 18 per age) received mock exams without inoculations to determine early developmental changes to the exam procedures. Behavioral distress was coded every 30 sec during the exam, a 5-min inoculation period, and a 20-min recovery period. Salivary cortisol was obtained upon arrival at the clinic and 25 min after the beginning of the inoculation period. To the exam-inoculation procedures, cortisol responses were high at 2 months, decreased significantly between 2 and 4 months, remained comparable between 4 and 6 months, and then declined again between 6 and 15 months. Between 6 and 15 months there was also a decrease in pretest cortisol. By 15 months, significant increases in cortisol from pre- to posttest were no longer observed for most infants. To the exam-only procedures, cortisol responses decreased between 2 and 4 months, and by 4 months most infants failed to show pre- to posttest increases in cortisol. Behavioral distress decreased between 2 and 6 months, but increased again at 15 months. While crying and cortisol were modestly correlated during the 2- to 6-month exam-inoculation procedures, at 15 months no significant correlations were obtained. Behavioral and hormonal reactions thus followed different ontogenetic paths and may provide different information about infant functioning. There was some evidence that the emergence of the circadian rhythm in cortisol might be related to the early decrease in cortisol response.