Intensive studies of plants such as Arabidopsis and maize have helped determine how genes work to make plants look the way they do. At the same time, systematists have been able to reconstruct the evolutionary history of plants.

We connect these two areas by studying the genetic and developmental changes that make species appear different from one another. We focus on the grass family, which contains maize and rice. Maize has a long tradition of genetic studies and rice has a complete genome sequence. The data from these two species form the basis of our comparative studies.

With funding from the National Science Foundation, we are currently using two approaches to investigate the genetic basis of evolutionary change.

1. A focused study of a domesticated species, foxtail millet (Setaria italica), and its wild ancestor, green millet (Setaria viridis). The two species differ in the architecture of the plant and the shape of the inflorescence. Our colleagues (Mike Gale, Katrien Devos) at the John Innes Institute in Norwich, England crossed the two species and produced a genetic map. We are using their mapping population to identify quantitative trait loci (QTL) that control the architectural and inflorescence differences.

2. A broad investigation of candidate genes known to affect the way inflorescences look. This is part of a multi-investigator grant to Sarah Hake (UC-Berkeley). In this project we are investigating the sequence and expression of important developmental genes (including Knotted1, tasselseed2, indeterminate spikelet, zmm14) in eight cultivated species (rice, maize, oats, barley, common millet, foxtail millet, pearl millet, sorghum). We anticipate two distinct patterns of evolution among these genes. a) Genes that are evolving neutrally and have conserved expression patterns among the cereals. We conclude that their function has not changed in evolutionary time. This is useful for agronomists, because it allows them to extrapolate from maize to all the other cereals. b) Genes that appear to be under selection and have different expression patterns in different species. We conclude that these genes have changed function in evolutionary time and may have contributed to the formation of species differences. Most of the genes we have studied so far appear to be in the first, conserved category.

The QTL approach provides a direct answer to questions about differences between two species; however, the results apply only to the species pair under investigation. The candidate gene approach permits broad generalizations, but is much more indirect. The combination of the two provides a unique picture of patterns of genetic and phenotypic diversification.