An abscission agent (5-chloro-3-methyl-4-nitro-1H-pyrazole [CMNP]) at 300 mg·L–1 in a volume of 2810 L·ha–1 was applied to Valencia orange trees [Citrus sinensis (L.) Osb.] on 22 May 2004. At this time, immature and mature fruit were present on the tree simultaneously. Three days after application, fruit were mechanically harvested using a trunk-shake-and-catch system. The power to the shaker head was operated at full- or half-throttle (FT or HT, respectively), and the duration of trunk shaking was 2 seconds at FT or 4 seconds at FT and HT. Mature fruit removal percentage and number of immature fruit removed, and fruitlet weight and diameter were determined. Mature fruit removal percentage with 2 seconds at FT or 4 seconds at FT harvesting ±CMNP, or 4 seconds at HT + CMNP was not significantly different and ranged between 89% to 97%. Harvesting at 4 seconds HT without CMNP removed significantly less mature fruit than any treatment. CMNP did not affect immature fruit removal by the trunk shaker. Harvesting at 4 seconds at HT removed significantly less immature fruit than 2 seconds at FT or 4 seconds at FT. No significant difference in fruitlet weight or diameter was measured between any trunk shaker harvest operation and CMNP treatment. Trunk shaking frequency was estimated to be 4.8 and 8.0 Hz at HT and FT, respectively. Yield in 2005 was determined on the same trees used for harvest treatments in 2004. CMNP did not impact yield. No significant difference in yield was seen between the hand-picked control and 4 seconds at HT, whereas yield in the remaining treatments was lower. The results demonstrate that CMNP application combined with low frequency trunk shaker harvesting can achieve high percentage of mature fruit removal with no significant impact on return yield of the following crop.