Submitted to: International Conference on the Comparative Biology of the Monocotyledons
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: January 14, 2009
Publication Date: September 2, 2010
Citation: Meerow, A.W. 2010. Convergence or reticulation? Mosaic evolution in the canalized American Amaryllidaceae, in Seberg, O., Petersen, G., Barfod, A.S. and Davis, J.I. (eds.) Diversity, phylogeny and evolution in the monocotyledons, pp. 145-168. Aarhus University Press, Aarhus. Interpretive Summary: Canalization is defined as the suppression of phenotypic variation, or, in the context of molecular evolution, genetic buffering that has evolved under natural selection in order to stabilize the phenotype. Very little is understood on the processes behind canalization, even in today’s genomic era. Canalization seems to be powerful force in the Amaryllidaceae from which only a few lineages have escaped, and this has often in the past led to misdiagnosis of phylogenetic relationships. The American Amaryllidaceae resolves into two well-supported clades with plastid and nrDNA ITS sequences – the primarily diploid hippeastroid clade (x =11), within which there is limited morphological diversity, and the Andean tetraploid-derived clade (x = 23), which exhibits greater phenotypic diversity. Moreover, several patterns of morphological diversity in the Andean clade manifest in a homoplastic mosaic at different taxonomic levels in various subclades. Both major clades consist of a small tribe (2-4 genera) that is sister to a larger clade, but only in the hippeastroid group do two genera, Rhodophiala and Zephyranthes, appear polyphyletic based on ITS sequences. The mosaic patterns of phenotypic diversity are discussed in the context of genome duplication, possible reticulation, and biogeographic factors.
Technical Abstract: Using the American members of the amaryllis family as a model, the parallel evolution of similar characteristics is discussed against the framework a family trees generated by analysis of DNA sequences. It is hypothesized that the amaryllis family is under genetic constraints that force developmental responses to selection pressures to proceed similarly in different lineages. Possible genes involved in these processes are suggested, and the possibility of hybridization between early lineages of the family in the Americas is discussed.