In autumn 1986, all the F2 progenies from the 21 monosomic F1 and from the disomic F1 cross between « Courtot » and « L22 » were sown in a complete randomized block with two replications. Every plot contained 20 plants sown on a line at a low density. All the trials were inoculated twice (at boot and heading stages) with spores of Leptovphaeria nodorum Muller (Septoria nodorum Berk.) strain n° 6, a wheat isolate. The following characters were recorded for each plant : two elements of partial resistance - incubation time (L.P.I.) and rate of spread of leaf necrosis (P. max) ; and two elements of tolerance - disease intensity on the ear (I.E.) and thousand kernel weight of diseased plants (P.M.G.M.). The earliness of heading time and stem length were also noted. Table 1 shows the average results obtained for all the F2 monosomic lines and for lines L22. As in the F1 chromosomes 3A, 2B and 5B of line L22 gave leaf resistance increased relative to that of F2 Di (table 2) ; also, as in the F2, chromosome 6D give more leaf susceptibility. Such chromosomes are supposed to bear gene (s) as recessive alleles. On the other hand, chromosomes ineffective in F1 were found to confer resistance in the F2 (i.e. 5A, 7A and 7B for L.P.I., 3B for P. max) or susceptibility (2A and 1B for L.P.I, 7D for P. max). Such chromosomes are supposed to bear dominant genes. More difficult to explain is the change from susceptible to resistant (5D for L.P.I., 6A for P. max). That could be due to the climatic conditions of 1987 or to some genetic interactions. A smaller relation was found between L.P.I. and P. max (r = 0.423) than in F1 (tables 2 and 4). No relations (table 3) were found between the foliar parameters and I.E. or P.M.G.M., nor between I.E. and P.M.G.M. Also no relation was observed with length of stem. Finally the partial resistance to Leptosphaeria nodorum of line L22 was polygenic, and even line L22 bore some genes for susceptibility. For breeding purposes, chromosomes 3A, 2B and SB could be useful, but, for breeding hybrid wheat, dominant genes are better. A method is needed to associate partial resistance and tolerance in two independent ways.