The tree of life: metaphor, model, and heuristic device.

The tree of life (ToL) is used variously as metaphor, model, and research tool to explore life’s evolution and genealogical relationships. It has broad appeal for multiple scientific communities and the public as a descriptive term covering life’s diversity. However, as biologists identify more entities and events to describe—including species, organisms, cells, genomes, gene families, extra-chromosomal genetic elements, endosymbioses, hybridizations, recombination types, and lateral gene transfer (LGT) events—the ToL strains under the weight of multiple uses and expectations. The multiple uses span levels of biological organization, from genes to clades, and they extend to all life forms despite their disparate mechanisms of reproduction and inheritance. Furthermore, some biologists consider the purpose of the ToL to be communicating patterns of relationship, whereas others consider evolutionary process to be a primary focus (e.g., Woese 1998). Use of both metaphors (figures of speech) and models (diagrams) comparing evolution to branching trees go back at least to Darwin (1859). Current multiple uses of the ToL, despite their differences, share reliance on phylogenetic methods. The ToL works well as a strictly bifurcating tree in the absence of reticulate evolution, which results from hybridization, lineage merger, and LGT. However, for organisms where inheritance of traits is often reticulate, as with LGT in Bacteria and Archaea, the evolutionary histories of organismal lineages and their molecular traits can differ dramatically. As a result, identifiable histories for organismal lineages can fade or disappear entirely for lack of characters sharing that same history, complicating recognition, and tracking of species or other taxa. Similarly, the ToL applied to life’s earliest evolution and origins is more specifically about molecular sequence lineages and their precursors than organismal lineages. For these reasons, the ToL has been criticized as disproven, obsolete, or restricted, applying to some life forms but not to others (e.g., Doolittle 1999; Bapteste et al. 2009; Puigbo et al. 2009). In this article, I argue for a pluralistic view of the ToL, not as a falsifiable hypothesis about the nature of the branching patterns, but as a broad metaphor for life’s common descent, and as an ongoing effort to model genealogical relationships, such as reticulations, for molecular and organismal lineages. The ToL has also become an important heuristic device. This stems from its use in studying evolutionary process. I seek to show the heuristic value of ToL studies by discussing the mutually informing nature of phylogenetic analysis and evolutionary biology, including the impact of ToL studies on core concepts of species, genes, and homology. I suggest that both instrumentalist and realist philosophical approaches to systematics are applied in estimating the ToL. The expanding number of distinctive entities whose common descent with modification can be traced via phylogenetics (genes, genomes, cells, organisms, populations, and clades) requires a pluralistic approach if the ToL is to be useful in developing and integrating new knowledge about evolution. Although the overwhelming majority of the ToL’s details are lost to time,much remains to be learned.