The oxygen diffusion rate (ODR) technique measures the flux of oxygen to a platinum microelectrode, similar to oxygen movement to roots. Instead of merely determining mean ODR, the ODR can be measured along a transect. The spatial pattern can then be examined and related to patterns of aerated and anoxic zones in the soil. The purpose of this study was to examine the within-sample variability of ODR as a tool to characterize soil variability at the small scale, autocorrelation, and binary analysis of the two-dimensional plane. Undisturbed samples (640 cm3) were collected from pots in a greenhouse study, in which the soil had been pulverized, and from plots in a field trial under minimum tillage. After equilibrating the samples at 100 mm negative soil water head, horizontal ODR transects were measured in the center plane of each sample. The ODR values varied considerably within the samples, and mean ODR values were larger for the field samples (55 μg m−2s−1) than for the pulverized greenhouse samples (26 μg m−2s−1). The autocorrelation lengths were much greater for the disturbed greenhouse samples (53 mm) than for the field samples (5 mm). Variances may not have been stationary for the homogeneous greenhouse samples. The fraction of ODR reading >33 μg m−2s−1was 0.19 for the greenhouse samples and 0.61 for the field samples. For the same ODR cutoff, the ratio of cluster size to sample area was 0.03 for the greenhouse samples and 0.06 for the field samples. In summary, the two-dimensional analysis provided more insight than the mean ODR values alone. The ODR transect technique may provide useful information on aeration patterns in soil that will complement more detailed image analysis approaches that have a finer pixel scale.