The aim of the study was to evaluate the main cutting force for two species of tropical Mozambican wood and to develop predictive models. Cutting these hardwoods is difficult. Determination of cutting parameters is required to optimize cutting processes, machines and tools in the cutting operations. This determination would enable the forestry and wood sector to achieve higher financial results. Samples of a lesser-known wood species Pseudolachnostylis maprounaefolia (ntholo) and a well-known wood species Swartzia madagascariensis (ironwood) were machined in a test apparatus. A standard single saw tooth mounted on a piezoelectric load cell was used to evaluate the main cutting force. Data were captured using an A/D converter integrated with National Instruments LabVIEW software. The measured signals were recorded at a sampling frequency of 25 kHz. The experimental set-up used response surface methodology for developing predictive models. The experimental clearly determined the relationship between the main cutting force and edge radius, wood density, rake angle, chip thickness, moisture content (MC) and cutting direction (CD). Among the studied variables, chip thickness and CD had the highest effect on the main cutting force level while wood density, MC and rake angle had the lowest effect.