Delphacodes kuscheli (Hemiptera: Delphacidae) is the vector of maize rough dwarf virus that affects maize production in the central region of Argentina. The spatial abundance pattern and body morphology of the insect vector were studied during the spring and summer of 1999 and 2000 in endemic and nonendemic areas affected by this virus. An estimation of insect density was obtained from high sticky traps (6 m high) placed at nine sampling sites along a 500-km transect that crossed the main maize production area in Argentina. Host patches were detected, and their areas were measured using Landsat 5 Thematic Mapper (TM) imagery from the region to be studied. Population phenology of the individuals collected in the sampling sites was compared using multivariate analysis of morphometric variables. The land use estimated from Landsat 5 TM images varied according to the different sites and years. Areas with similar proportion of winter pastures, winter cereals, and perennial pastures were grouped together by complete linkage cluster analysis. The mean abundance of the vector varied and did not show a consistent pattern. The highest abundances were not found in the traps placed in the endemic region. Multiple linear regression of the proportion of area (arcsine transformed) covered by annual pastures, perennial pastures, and winter crops versus mean abundance of D. kuscheli (expressed as insect/trap/d) were significant for both years (R² = 0.91 for 1999 and R² = 0.82 for 2000). The function that related insect abundance and host area was different each year; this suggested that other factors were affecting the “insect productivity” of a certain host area. There were clear morphometric differences among the individuals collected in the different sampling sites, although the sites were not so far away from one another (50 km). When comparing the phenotypes of the insects collected in different sites, 55 (1999) and 51% (2000) of the cases were correctly assigned to the sites from which the individuals were collected. These percentages increased to 74 and 62% if the individuals were grouped according to the cluster of land use obtained from Landsat 5 TM images. In all cases, Wilk’s lambdas were highly significant. The role of the distribution, abundance, and quality of the host plants is analyzed in this study.