The objective of this study was to characterize normal temporal-spatial patterns during the Reach & Grasp Cycle and to identify upper limb motor deficits in children with cerebral palsy (CP). The Reach & Grasp Cycle encompasses six sequential tasks: reach, grasp cylinder, transport to self (T(1)), transport back to table (T(2)), release cylinder, and return to initial position. Three-dimensional motion data were recorded from 25 typically developing children (11 males, 14 females; ages 5-18 years) and 12 children with hemiplegic CP (2 males, 10 females; ages 5-17 years). Within-day and between-day coefficients of variation for the control group ranged from 0 to 0.19, indicating good repeatability of all parameters. The mean duration of the Cycle for children with CP was nearly twice as long as controls, 9.5 +/- 4.3 s versus 5.1 +/- 1.2 s (U = 37.0, P = .002), partly due to prolonged grasp and release durations. Peak hand velocity occurred at approximately 40% of each phase and was greater during the transport (T(1), T(2)) than non-transport phases (reach, return) in controls (P < .001). Index of curvature was lower during transport versus non-transport phases for all children. Children with CP demonstrated an increased index of curvature during reach (U = 46.0, P = .0074) and an increased total number of movement units (U = 16.5, P < .0001) compared to controls, indicating less efficient and less smooth movements. Total duration of the Reach & Grasp Cycle (rho = .957, P < .0001), index of curvature during reach and T(1) (rho = .873,P = .0002 and rho = .778, P = .0028), and total number of movement units (rho = .907, P < .0001) correlated strongly with MACS score. The consistent normative data and the substantial differences between children with CP and controls reflect utility of the Reach & Grasp Cycle for quantitative evaluation of upper limb motor deficits. (C) 2010 Elsevier B.V. All rights reserved.