Fricative production in children and adults has been widely studied using acoustic measures, but little information exists on how children learn to control both laryngeal and oral gestures for (voiceless) fricative production. Here, we report oral airflow data from four 5-year-old children and four women producing laryngeal and oral fricatives /h s z/. Functional data analysis was used to normalize the signals, obtain average productions for each subject, and determine how temporal and amplitude variability were distributed over the consonant. Higher temporal variability in the 5-year-olds than the adults was observed, but at localized regions in the signals. The locations of maximum phasing variability and amplitude variability differed among subjects. For an individual subject, regions of high phasing variability were not necessarily regions of high amplitude variability. Finally, whereas adult women evidence different airflow patterns for /s/ and /z/, some of the children do not.