An AFLP data set comprising 95 accessions from 20 species of Lactuca s.l. (sensu lato) and related genera was generated using the primer combinations E35/M48 and E35/M49. In phenetic analyses of a data subset, clustering with UPGMA based on Jaccard's similarity coefficient resulted in the highest cophenetic correlation, and the results were comparable to those of a principal coordinates analysis. In analyses of the total data set, phenetic and cladistic analyses showed similar tree topologies for the well-supported parts of the trees. The validity of cladistic analysis of AFLP data is discussed. The results do not support a distinction among the serriola-like species L. sativa, L. serriola, L. dregeana, and L. altaica, which is in line with previous results. Therefore, we postulate that these species are conspecific. The serriola-like species L. aculeata occupies a clearly separate position, making it an ideal outgroup for studies of the closest relatives of L. sativa. The subsect. Lactuca as a group is well supported by our data, but the positions of L. saligna and L. virosa relative to the serriola-like species remain unclear. The close relationship between the sect. Mulgedium species L. tatarica and L. sibirica is corroborated by the present AFLP results and by additional crossability data.

We have investigated the use of AFLP technology as a tool for the high throughput enrichment of Radiation Hybrid (RH) maps. The 3000 rad TM112 bovine RH panel was assayed with 37 EcoRI/TaqI AFLP primer combinations. The number of selective nucleotides used during PCR was increased to seven, to reduce the complexity of the AFLP profile and minimise the overlap between hamster and bovine bands co-amplified from hybrid cell clones. Seven-hundred-forty-seven bovine AFLP bands were amplified that could be distinguished following electrophoresis. Repeatability was tested within and between laboratories on independent template preparations and an error rate of 1.3% found. Two-point linkage analysis clustered 428 AFLP fragments in 39 linkage groups of at least 4 markers. Multi-point maps were constructed for 5 sample linkage groups. The study demonstrated that the AFLP approach could be used to rapidly screen for the most informative clones during panel construction and to increase the number of markers on RH maps, which could be useful for joining linkage groups formed by other markers. The use of AFLP markers as anchor points between existing RH maps and other physical maps, such as BAC contigs, is also discussed.