Two important trends are recognizable to increase the efficiency in crop breeding and production by manipulation of the reproductive systems: i)exploitation of novel genetic variability after hybridization of remotely related plant species and ii) introduction of apomictic seed formation.

Valuable features can be transferred from wild species into cultivated plants by interspecific hybridization and subsequent backcrosses. That was exemplified in forage grasses by the introgression of resistance to Puccinia coronata from Festuca species into Lolium species. Allopolyploid hybrids between important forage grasses (L. multiflorum x Dactylis glomerata. F. pratensis x D. glomerata and F. rubra x L. perenne) have been produced for the first time. Even crosses between species of the subfamilies of Pooideae and Panicoideae were successful and could be used for production of haploid plants via genome elimination in wheat or for the generation of plants having the oat genome and one to eight chromosomes of Zea mays, Penniisetum americanum or Tripsacum dactyloides. One aim of such crosses is the transfer of apomixis from T. dactyloides into cereals.

The main advantage of apomixis is that it permits the fixation of hybrid vigour and the production of true breeding hybrid varieties. We have developed an efficient technique to screen for apomixis in Pooideae and proposed a new breeding scheme for the apomictic crop Poa pratensis based on selected sexual plants. A new approach is in progress to isolate the gene(s) for apomixis by means of differential display using segregating progenies of crosses between sexual and apomictic plants. Additionally, the so-called `Salmon system` of wheat is used to analyse the initiation of autonomous embryo development at the cellular and molecular level as well as to induce complete apomixis within the parthenogenetic `Salmon` lines.