Populations of the earthworm Aporrectodea caliginosa were analyzed on three sampling dates (maize emergence in spring, crop harvest in autumn and post-harvest, in winter) and two soil depths (0-15, 16-30 cm) in four tillage systems: 1) shallow tillage, 2) disk ploughing, 3) ripper decompaction (one pass) and 4) ripper decompaction (two passes). The soil was an irrigated clay loam in Central Mexico. The tillage systems were applied in winter during four consecutive years on experimental plots (400 m 2 area) cropped with maize. Tillage treatments did not cause differences in juvenile, adult and total earthworm individuals. In contrast, the sampling date had a significant effect (P < 0.05) on earthworm population: individuals were absent at maize emergence (April 2007), 135 m −2 at harvest (November 2007) and 45 m −2 at post-harvest date (January 2008). The recovery of earthworm population during the maize growing period was attributed to favourable conditions for earthworm growth consisting in abundant food (2-5 t ha −1 y −1 of maize stover) and high soil water content. Earthworm abundance showed differences caused by soil depth, individuals migrate to the deep layer after crop harvest. A very high number of burrows was observed (2200-2800 burrows m −2) at 10 cm of soil depth, in plots worked with the different tillage sequences in the three sampling dates.