This study identified the appropriate time to visually score rice germplasm in order to select materials having the lowest yield reduction under iron toxicity. The experiment took place at Korhogo, 9° 22' N 5° 31' W, in northern C te d'Ivoire, during the wet season (May-Aug 1995) using a randomized complete block design with three replications on an irrigated Ultisol lowland that contained 343 ppm Fe in soil solution at the beginning of the season. Of 28 transplanted, Suakoko 8 and Bouak 189 were tolerant and susceptible checks, respectively, and two O. glaberrima were land races. Each plot of 40 hills had a planted border using the check varieties. The experiment was repeated simultaneously on a nearby control field which was free of iron toxicity.