Search

Rapid, ancient radiations pose one of the most difficult challenges for phylogenetic estimation. We used DNA sequence data of 9,006 aligned base pairs from five genes (chloroplast atpB, matK, rbcL, and 18S and 26S nrDNA) to elucidate relationships among major lineages of Saxifragales (angiosperms, eudicots). These relationships were poorly supported in previous studies, apparently because the lineages originated in rapid succession. Using an array of methods that explicitly incorporate assumptions about evolutionary process (weighted maximum parsimony, maximum likelihood, LogDet/paralinear transformed distances), we show that the initial diversification of Saxifragales was indeed rapid. We suggest that the poor resolution of our best phylogenetic estimate is not due to violations of assumptions or to combining data partitions having conflicting histories or processes. We show that estimated branch lengths during the initial diversification are exceedingly short, and we estimate that acquiring sufficient sequence data to resolve these relationships would require an extraordinary effort (approximately 10(7) bp), assuming a linear increase in branch support with branch length. However, our simulation of much larger data sets containing a distribution of phylogenetic signal similar to that of the five sampled gene sequences suggests a limit to achievable branch support. Using statistical tests of differences in the likelihoods of topologies, we evaluated whether the initial radiation of Saxifragales involved the simultaneous origin of major lineages. Our results are consistent with predictions that resolving the branching order of rapid, ancient radiations requires sampling characters that evolved rapidly at the time of the radiation but have since experienced a slower evolutionary rate.

Chrysosplenium (Saxifragaceae) consists of 57 species widely distributed in temperate and arctic regions of the Northern Hemisphere, with two species restricted to the southern part of South America. Species relationships within the genus are highly problematic. The genus has traditionally been divided into two groups, sometimes recognized as sections (Oppositifolia and Alternifolia), based on leaf arrangement, or, alternatively, into 17 series. Based on morphological features, Hara suggested that the genus originated in South America and then subsequently migrated to the Northern Hemisphere. We conducted phylogenetic analyses of DNA sequences of the chloroplast gene matK for species of Chrysosplenium to elucidate relationships, test Hara's biogeographic hypothesis for the genus, and examine chromosomal and gynoecial diversification. These analyses revealed that both sections Oppositifolia and Alternifolia are monophyletic and form two large sister clades. Hence, leaf arrangement is a good indicator of relationships within this genus. Hara's series Pilosa and Macrostemon are each also monophyletic; however, series Oppositifolia, Alternifolia, and Nepalensia are clearly not monophyletic. MacClade reconstructions suggest that the genus arose in Eastern Asia, rather than in South America, with several independent migration events from Asia to the New World. In one well-defined subclade, species from eastern and western North America form a discrete clade, with Old World species as their sister group, suggesting that the eastern and western North American taxa diverged following migration to that continent. The South American species forms a clade with species from eastern Asia; this disjunction may be the result of ancient long-distance dispersal. Character mapping demonstrated that gynoecial diversification is dynamic, with reversals from inferior to half-inferior ovaries, as well as to ovaries that appear superior. Chromosomal evolution also appears to be labile with several independent origins of n = 12 (from an original number of n = 11) and multiple episodes of aneuploidy.

The phylogenetic relationships of some angiosperm families have remained enigmatic despite broad phylogenetic analyses of rbcL sequences. One example is the aquatic family Podostemaceae, the relationships of which have long been controversial because of major morphological modifications associated with their aquatic habit. Podostemaceae have variously been associated with Piperaceae, Nepenthaceae, Polygonaceae, Caryophyllaceae, Scrophulariaceae, Rosaceae, Crassulaceae, and Saxifragaceae. Two recent analyses of rbcL sequences suggest a possible sister-group relationship of Podostemaceae to Crassulaceae (Saxifragales). However, the branch leading to Podostemaceae was long, and use of different outgroups resulted in alternative placements. We explored the phylogenetic relationships of Podostemaceae using 18S rDNA sequences and a combined rbcL + 18S rDNA matrix representing over 250 angiosperms. In analyses based on 18S rDNA data, Podostemaceae are not characterized by a long branch; the family consistently appears as part of a Malpighiales clade that also includes Malpighiaceae, Turneraceae, Passifloraceae, Salicaceae, Euphorbiaceae, Violaceae, Linaceae, Chrysobalanaceae, Trigoniaceae, Humiriaceae, and Ochnaceae. Phylogenetic analyses based on a combined 18S rDNA + rbcL data set (223 ingroup taxa) with basal angiosperms as the outgroup also suggest that Podostemaceae are part of a Malpighiales clade. These searches swapped to completion, and the shortest trees showed enhanced resolution and increased internal support compared to those based on 18S rDNA or rbcL alone. However, when Gnetales are used as the outgroup, Podostemaceae appear with members of the nitrogen fixing clade (e.g., Elaeagnaceae, Ulmaceae, Rhamnaceae, Cannabaceae, Moraceae, and Urticaceae). None of the relationships suggested here for Podostemaceae receives strong bootstrap support. Our analyses indicate that Podostemaceae are not closely allied with Crassulaceae or with other members of the Saxifragales clade; their closest relatives, although still uncertain, appear to lie elsewhere in the rosids., (Copyright 1999 Academic Press.)