plant biodiversity. We advocate a modest investment in the period 2003–2008 for development of genomics resources in species outside the model, references, and their crop relatives. As an initial step, we recommend the survey of roughly 50 species with EST sequencing. ESTs remain the most rapid and cost-effective way to sample gene content and discover new genes. This effort will go a long way toward making gene discovery and comparative genomics possible.

To ensure that genomics investment in additional species builds effectively on existing resources and on the EST sequencing suggested above, we recommend that the evolutionary-genomics community pursue further the selection of a small number of key species (5–10) spanning critical evolutionary nodes in preparation for communitywide genomic investigation over the next 3–10 years. The species should be selected to provide a broad view of the evolutionary potential of genomes and a deep understanding of gene diversity and adaptation. Other federally funded projects—such as Deep Green (2002) and its successor, Deep Gene (2002)—have identified a phylogenetically diverse array of species for application of genomics. The NPGI should give priority to developing tools for species from this set or from among species closely related to them. In the context of evolutionary studies, the genomes of cyanobacteria, from which up to 20% of the genes in contemporary plants originated, and the eukaryotic algae also are legitimate objects of study in the NPGI. The specific approach taken to achieve the goals of evolutionary studies broadly is not immediately obvious and will require consensus building. Our key concern is that, for any given species or evolutionary question, there needs to be at least some minimum concentration of scientists ready to exploit genomics data.

Beyond that concern, criteria for choosing any evolutionary-genomics focal species should include:

• Distributed position in the phylogeny, with emphasis on early branches of the green plant phylogeny.

• Genome size, with emphasis on small genomes and simple ploidy.