The authors evaluated the characteristics of infants' spontaneous movements by using dynamical systems analysis. Participants were 6 healthy 1-month-old full-term newborn infants (3 males, 3 females). They used a triaxial accelerometer to measure limb acceleration in 3-dimensional space. Acceleration signals were recorded during 200 s from the right wrist when the infant was in an active alert state and lying supine (sampling rate 200 Hz). and was stored in the system's memory. Digitized data were transferred to a PC for subsequent processing with analysis software. The acceleration time series data were analyzed linearly and nonlinearly. Nonlinear time series analysis suggested that the infants' spontaneous movements are characterized by a nonlinear chaotic dynamics with 5 or 6 embedding dimensions. The production of infants'spontaneous movements involves chaotic dynamic systems that are capable of generating voluntary skill movements.