Host-plant resistance is the best means to control the key disease of lentil-vascular wilt, caused by Fusarium oxysporum um Schlecht. emend. Snyder & Hansen f.sp. lentis Vasudeva and Srinivasan. Systematic screening for resistance to lentil wilt was initiated in the field in 1993, in a wilt-sick plot in North Syria, with a core collection of 577 germplasm accessions from 33 countries. A subset (88 accessions) of mostly resistant accessions was re-screened in 1994. The most resistant accessions came from Chile, Egypt, India, Iran and Romania. Variation among accessions in the temporal pattern of wilting was analyzed. The limited wilting in resistant accessions followed a linear model through time, whereas the pattern for susceptible accessions was better described with an exponential model. This temporal variation emphasizes the need for repeated scoring during screening for resistance to lentil vascular wilt to identify 'late-wilters'. To overcome spatial variation in inoculum density, the efficacy of using wilt scores from a systematically-repeated susceptible control in covariate analysis was tested. Covariance analysis significantly improved overall screening by 3% in 1993, but the improvement was non-significant in 1994. The results emphasize the relative uniformity of disease pressure in the wilt-sick plot and suggest that covariance analysis of a systematically arranged control will be of greater benefit in land which is less uniformly infected.