Tribute to Gwilym P. Lewis

The young Gwilym Lewis first joined the Royal Botanic Gardens, Kew, as part of the horticultural staff in 1974. There, his enthusiasm for botany was quickly noticed and he was soon recruited and, subsequently, promoted as part of the team of scientists in the herbarium, later going on to become a senior scientist in integrated monography at Kew. This drive, energy and dedication to botany, which were noted at the age of 22, have never left him. Gwilym has been a key contributor to research at Kew Gardens for nearly 45 years. These four decades have seen monumental changes in legume taxonomy, systematics and evolution, and Gwilym has been an important instigator and contributor to this movement both scientifically and as head of the legume section atKew, and indeed inmanyways, as leader of the legume systematics international research community as a whole in recent years. In the four decades sinceGwilymstarted his career, our vision of legume relationships has profoundly changed. In the early 1980s, the main legume lineages were thought to have evolved from broad common ancestors, exemplified by simple-flowered extratropical woody Caesalpinioideae (Polhill et al. 1981), and were usually portrayed as explicit ancestor-descendant relationships among informal groups of genera. Later in the 1980s, cladistics began to be adopted by legume researchers, the first phylogenetic analyses were presented in Advances in Legume Systematics (ALS) Part 3 (e.g. Crisp and Weston 1987; Lavin 1987; Zandee and Geesink 1987), and Jeff Doyle (1987) described the new molecular-biology tools and data and their tremendous potential for resolving evolutionary relationships. These approaches were fully embraced by the legume systematics community, and, in 1995 (ALS 7), 16 new phylogeneticstudiesacross thefamily,mostusingmoleculardata, were published. In the 1990s, the plastid rbcL gene was being sequenced to look at higher-level legume relationships (Doyle et al. 1997), in 2001, the first full legume plastomes were sequenced, and, in 2008, the full nuclear genome of Lotus japonicus appeared. By the mid-2000s, the phylogenetic framework and time frame of higher-level legume relationships were broadly established (Wojciechowski et al. 2004; Lavin et al. 2005; Bruneau et al. 2008), and, in 2019, we have now amassed DNA-sequence data for more than 5500 legume species, including nuclear genomes (~25 species) and several hundred plastid genomes (e.g. Egan and Vatanparast 2019), as well as produced a much more comprehensively sampled phylogeny of the family as a whole (Legume Phylogeny Working Group 2017). We are now seeing unprecedented use of these phylogenies to test hypotheses on Gwilym Peter Lewis