Goal of this series of experiments with pot plants was to optimize some steps of the method for resistance screening of Musa cultivars to R similis. Radopholus similis per 100 g of roots was significantly higher in river sand than in local banana soil, however it followed the same tendency in both growing media. Reproductive index and final root weight decreased with increasing initial inoculation density. In a subsequent experiment with a broader initial inoculation range and two different P, similis populations, reproductive index, root and shoot weight decreased with increasing initial inoculation densities for both populations. When eight exposure times were evaluated, R similis per 100 g of roots showed a lag phase of about five weeks, then started to grow during six weeks, and stabilized afterwards. Inoculation of plants in a variable pot volume experiment resulted in a lower reproductive index in higher pot volumes. Finally, R similis per 100 g of roots was not different among Musa AAA cvs. Grand Name, Gros Michel, and Musa cv. AA Pisang Mas (P less than or equal to 0.25), but varied significantly between the first group and both Musa AAAA cv. FHIA-23 and Musa AAA cv. Yangambi Km5 (P less than or equal to 0.0001). In summary, the use of 1.8 liter pots, filled with sterilized local banana soil and initially inoculated with 0.28 R similis/ml of substrate and exposed to nematodes during 8 to 12 weeks, gave the most consistent results.