which was anticipated because of targeting ORFs, we constructed a genetic map using the RI lines. The map constructed in the 86 RI lines of collard×cauliflower consisted of 130 SRAP and 120 AFLP markers, plus the GLS-ALK gene. All AFLP markers were dominant, whereas approximately 20% of the polymorphic SRAPs (26 markers) segregated as co-dominant markers. The AFLP as well as SRAP markers were distributed fairly evenly in nine major linkage groups covering 2,165 cM (Table 1, Fig. 2). No major differences of genome coverage were observed for the two marker techniques. Therefore, similar to AFLPs, the wide genome coverage of SRAP markers and their high reproducibility results in the construction of genetic maps with optimal marker distribution. As expected, each co-dominant marker pair (each band was scored independently, Fig. 3) was located in the map at almost the same position, at distances between 2 to 10 cM. They failed to fall exactly at the same
location because, being scored separately, their genetic distance theoretically matched the level of residual heterozygosity (1/25) which exists in this RI population after five rounds of selfing. The relative high frequency of co-dominant SRAP markers is another important advantage of this technique over AFLP markers.