Two spring barley composites, one based on eight West-European two-rowed cultivars (A) and the other, the predominantly six-rowed composite XXI, based on several thousands of barley cultivars (B), formed the starting point of a recurrent selection procedure . The aim was to study how effective a repeated process of mild selection against susceptibility followed by recombination of the remaining material was in accumulating partial resistance in four host-pathogen situations .i) Only partial resistance is present and the pathogen population is defined (a given race) .ii) Partial and major gene resistance both occur and the pathogen population is defined . iii) Only partial resistance is present and the pathogen population is not defined ; a mixture of races that varies over the years . iv) Partial and major gene resistance are both present and the pathogen population is not defined .The variation in partial resistance to barley leaf rust was large in both populations . Population A carried no effective major resistance genes, population B possibly a few at low frequency . The variation in partial resistance to powdery mildew was moderate in A and possibly large in B . As far as effective major resistance genes is concerned A did not carry any, although some recombinations of defeated genes might have been partially effective, while B seemed to carry many .During the selection procedure the populations were always exposed to race 1-2-1 of barley leaf rust and to the mixture of powdery mildew races that was naturally present .The selection procedure consisted of three cycles of recurrent selection . In the initial heterogeneous populations (SO) single plant selection was applied followed by line selection the next year . In both populations 12 lines were selected that were intercrossed in all directions within the two populations . Again single plant (S3) and line selection were exercised and 12 lines selected . The 12 A-lines were intercrossed in all directions with the 12 B-lines and the recombined population again exposed to single plant (S6) and line selection (S7) . The selection in this population was done within two-rowed entries (A*) and within six-rowed entries (B*) .The selection pressure was mild . In each selection stage about 30% plants or lines most affected by barley leaf rust and some 30% plants or lines most affected by powdery mildew were removed . Among the remaining plants or lines (ca . 45%) a selection for other useful agronomic characteristics was applied .The response to selection was measured in four evaluation trials . Both single plant and line selection contributed to the progress in resistance in both populations to both pathogens .
262The gain in partial resistance to barley leaf rust was the same in the two populations . The average amount of sporulating leaf tissue in the S7 was about twenty times less than that in the SO . The best S7 lines showed a sixty fold decrease compared with the SO . Corrected for the levelling effect of interplot interference, very strong with barley leaf ru...