The movement of Meloidogyne incognita race 3 and Rotylenchulus reniformis through sand in response to natural fluctuations of vertical soil temperature gradients were compared within acrylic tubes 15 cm in diameter and 15 cm long. Tubes were buried between pairs of rectangular heat transfer coils within insulated boxes filled with moist sand. Multiple-endpoint ramp-and-soak process controllers were employed to control the temperatures of coils in a fashion that closely mimicked temperature patterns measured previously 5-25 cm deep in a cotton field in eastern Texas. Both species were pre-adapted for 24-36 hours within vials buried at the center of the boxes, then co-injected into 63 equal sections and extracting nematodes by direct flotation or Baermann funnel 24 and 48 hours after injection. In simulations of sunny and rainy periods during the summer, net movement of R. reniformis was away from the thermal surface regardless of tube orientation relative to gravity; M. incognita moved the opposite direction. In other tubes, 4 cm in diameter and 20 cm long, both species were pre-adapted at five depths and co-injected at the same depths as adapted at five points during the 24 hour cycle. The initial direction of movement was determined by cutting sand cylinders 1 1/2 hours after injection into 25 equal slices, such that each injection point was in the center of five slices. In this experiment, movement of R. reniformis away from the thermal surface occurred during the daytime, when the thermal surface was hot, and during the night, when the thermal surface was cool relative to the average temperature deep in the soil; M. incognita again moved the opposite direction. Results are consistent with the interpretation that under the conditions studied, R. reniform moves toward and M. incognita moves away from the temperature to which last exposed.