Abstract: To quantify the ear shape of corn (Zea mays L.) and to understand the effects of growth stress on the ear shape are of importance in corn crop science. By means of digital image analysis on whole ear RGB images, and using the definition of extent (E), volumetric odds (V), longitudinal centroid (L),　transverse centroid (T), and sphericity (S) to represent the similarity to rectangle, volumetric equality, longitudinal symmetry, transverse symmetry, and similarity to sphere, respectively, the ear shape data from the field experiment with 2 corn cultivars, 2 contrast growth conditions, and 2 defoliation treatments in a split plot design were analyzed by cluster analysis, canonical variate analysis, and variance analysis. It was found that within the first 2 weeks after silking, the ear shape varied obviously, namely, the E increased from 0.72 to 0.78, the S from 0.40 to 0.48, while the L decreased from 0.56 to 0.51, the T decreased from 0.02 to 0.01, and the V decreased from 0.25 to 0.21. Afterwards, all the 5 shape traits tended to be stable. Growth stress (high planting density and zero fertilizer application) shifted the E by -4.5% and the V by +17.7% at maturing stage, which, combined with the variation of ear length and width, contributed 87%-97% of grain yield change. Throughout the ear growth duration, the ear shape traits L, V, and S were more sensitive to growth stress, being considered as the potential quantitative indicators of stress.

Key words: model., corn (Zea mays L.), ear growth, environmental stress, shape, image process, soil water change, summer maize, land surface-air temperature difference