The motion resistances of 660 mm pneumatic and rigid bicycle wheels of the same rim diameter were measured experimentally using the developed tractor-towed single non-lug narrow wheel motion resistance test rig for traction studies. The motion resistances measured were taken to be the towing forces determined in real time using Mecmesin Basic Force Gauge (BFG 2500). The test variables included two test surfaces [tilled and wet (mud) surfaces], the dynamic load and the towing velocity. The tyre inflation pressure of 414 kPa was chosen to make the surface synonymous with that of the rigid wheel. Motion resistance ratios of the two wheels were determined empirically and through semi-empirical approach. The motion resistances of the rigid wheel were found to be greater than those of the pneumatic wheel for both surfaces. Consequently, the motion resistance ratios of the rigid wheel were greater than those obtained from the pneumatic wheel. Analysis of variance showed that there were significant differences between the means of the motion resistance measured on the test surfaces, as well as between the two wheels and their interactions with the test surfaces. The motion resistance ratio exhibited a linear relationship with the towing velocity, while the relationship with the dynamic load was quadratic. However, such a relationship is either direct or inverse with the respective variables. The motion resistance ratio models for the pneumatic and rigid wheels showed that on different test conditions of the dynamic loads and the towing velocities, the relationships between the motion resistance ratio and the dynamic load, and motion resistance with dynamic load were also different.