The study evaluated the influence of vegetation characteristics on nutrient loss in vegetation fallows in a part of the rainforest belt in Agoi-Ekpo, Cross River State. Participatory method was used to identify fallows of 3-year and abandoned farmland. In each identified fallow category, 10 plots for vegetation estimation was established, while two runoff plots of 10m x 4m was constructed and used for nutrient loss estimation. Data was gathered during the cropping season of March to November, 2012. Results showed that on the 3-year fallow only basal cover was retained by the model and significantly explained Ca loss (R 2 = 0.627, F = 3.655, p<0.01), Mg loss (R 2 = 0.57.2, F = 3.271, p<0.05) and potassium loss (R 2 =0.527, F = 2.980, p<0.05). On the abandoned farmland, the model retained only crown cover and it significantly explained OC loss (R 2 = 0.591, F = 3.402, p<0.01), TN (R 2 = 0.599, F = 3.456, p<0.01), Ca (R 2 = 0.674, F = 4.067, p<0.01), Mg (R 2 = 0.796, F = 5.75, p<0.01) and K (R 2 = 0.823, F = 6.090, p<0.01). The study showed that more nutrient element losses were recorded in the abandoned farmland. The study suggests that trees and shrubs should not be completely cut down on farmlands facilitate rapid vegetation restoration during land abandonment.