Physiologia Plantarum 130: 318–330. 2007 Copyright ª Physiologia Plantarum 2007, ISSN 0031-9317 REVIEW Recent advances in the biogeography and genecology of symbiotic Frankia and its host plants David R. Benson a and Jeffrey O. Dawson b, * a Department of Molecular and Cell Biology, University of Connecticut, Storrs, CT 06269, USA b Department of Natural Resources and Environmental Sciences, 1316 Plant Sciences Laboratory, University of Illinois at Urbana–Champaign, 1201 South Dorner Drive, Urbana, IL 61801, USA Correspondence *Corresponding author, e-mail: jdawson2@uiuc.edu Received 24 November 2006; revised 15 April 2007 doi: 10.1111/j.1399-3054.2007.00934.x Molecular phylogenetic approaches have begun to outline the origin, distribution and diversity of actinorhizal partners. Geographic isolation of Frankia and its host plants resulting from shifting continents and dispersal patterns have apparently led to the development of Frankia genotypes with differing affinities for host genera, even within the same plant family. Actinorhizal plant genera of widespread global distribution tend to nodulate readily even outside their native ranges. These taxa may maintain infective Frankia populations of considerable diversity on a broad scale. Arid environments seem to have distinctive actinorhizal partnerships, with smaller and more specific sets of Frankia symbionts. This has led to the hypothesis that some host families have taxa that are evolving towards narrow strain specificity, perhaps because of drier habitats where fewer Frankia strains would be able to survive. Harsh conditions such as water-saturated soils near lakes, swamps or bogs that are typically acidic and low in oxygen may similarly lessen the diversity of Frankia strains present in the soil, perhaps limiting the pool of frankiae available for infection locally and, at a larger scale, for natural selection of symbiotic partnerships with host plants. Recent molecular ecological studies have also provided examples of Frankia strain sorting by soil environment within higher order cluster groupings of Frankia host specificity. Future frontiers for ecological research on Frankia and actinorhizal plants include the soil ecosystem and the genome of Frankia and its hosts. Introduction By using polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) techniques on peeled root nodules, nodular strains of Frankia can be character- ized directly without having to first obtain isolates (Hahn et al. 1999). Thus, phylogenetic, ecological and distribu- tional features of specific Frankia strains can be described. Frankia nodules for analysis can be collected in the field or obtained from a soil sample using actinorhizal bait plants. Approaches employing molecular phylogeny have given new perspectives to the classification of actino- rhizal plants, providing important evolutionary and ecological insights (Fig. 1). The more than 20 actinorhizal genera in eight families of angiospermous plants were, until recently, thought to be taxonomically diverse and unrelated. However, actinorhizal plant families, together with all legumes and the rhizobially nodulated genus Parasponia, have been placed together in the rosid clade according to Soltis et al. (1995). This clade apparently comprises plant taxa with a predisposition to nodular symbiosis with diazotrophs. Distribution of infective Frankia Infective Frankia are known to occur on all continents except Antarctica, and on many islands (Figs 2 and 3), but Abbreviations – Mya, million years ago; Myr BP , million years before present; NU, nodulation units; PCR-RFLP, polymerase chain reaction-restriction fragment length polymorphism. 318 Physiol. Plant. 130, 2007