Sixty rice accessions were assessed in the field for resistance to Aspavia armigera F. attack under natural and artificial infestations. Out of these, 30 were lowland and 19 upland varieties derived from the International Institute for Tropical Agriculture (IITA) breeding programmes, while the others were obtained from other local and international organizations. From the upland varieties, the highest population density of A. armigera was recorded on CAN 4143 (p < 0.05), while the highest percentages of unfilled grains were obtained on ITA 132 (33.1%) and CAN 4143 (32%), and were both rated the most susceptible to A. armigera. CAN 6656, on which the lowest value for unfilled grains (4.7%) was found was rated resistant. Among the lowland varieties, TOX 3118-2-E2-2-1-2, on which was recorded the highest population density of A. armigera and highest percentage of unfilled grains (p < 0.05) was considered susceptible, while ITA 230, ITA 308, ITA 123, TOX 3100-32-2-1-3-5, TOX 3107-39-1-2-1-3, TOX 3027-43-1-E3-1-1, TOX 3441-7-1-1-1, TOX 3226-5-2-2-2, TOX 3716-15-1-1 and TOX 3561-56-2-3-2 were considered resistant. However, no-choice tests revealed that five from each of ten upland and lowland varieties initially rated as resistant were severely damaged. The other five upland ITA 315, IRAT 169, ITA 321, FAROX 41 and M 55 with consistent expression of resistance which also had low number of stylet sheath: 11, 17, 23, 24.3, and 29.3, respectively, were therefore rated resistant. Similarly, lowland ITA 230, TOX 3561-56-2-3-2, TOX 3100-2-1-3-5, TOX 3226-5-2-2-2 and TOX 3107-39-1-2-1-3 with low number of stylet sheath were rated resistant as well.