Edge effect is an inherent problem when using live-trapping grids to census animals in populations. The relationship between grid size and home range size, shape, and dispersion pattern was investigated by a mathematical model and computer simulation. A simple relationship between overestimate of population density and grid size to home range size was found. To minimize the effect of home range size, shape, and dispersion pattern on census results, grid sizes at least 16 times larger than the average home range size should be used.