Deep Time Project: A Comprehensive Phylogenetic Tree of Living and Fossil Angiosperms

Comments and questions: Dr. Doug Soltis

Last modified: 9/6/02

Copyright © FLMNH. This site is maintained by FLMNH











Back to Top















Back to Top


lower eudicot



Archaefructaceae is a basal angiosperm family proposed by Sun et al. (2002). The family consists of two fossil species, i.e., Archaefructus liaoningensis Sun, Dilcher, Zheng et Zhou (Sun et al., 1998) and A. sinensis Sun, Dilcher, Ji et Nixon (Sun et al., 2002). It is proposed that Archaefructaceae is a sister clade to all angiosperms when their characters are included in a combined 3 gene molecular and morphological analysis (see figure below).

Figure 1. Consensus cladogram of most parsimonious trees fro the analysis of 173 living taxa of seed plants, plus the fossil Archaefructus. Analyses included 1628 molecular characters and 17 to 108 morphological characters. Taxa with numbers in brackets indicate the number of species in that clade that were analyzed as separate terminals and are monophyletic in all trees but are no shown in this figure (Courtesy of David Dilcher).


James A. Doyle, Section of Evolution and Ecology, University of California, Davis, CA 95616;

Helena Eklund, Uppsala, Sweden;

Patrick S. Herendeen, Department of Biological Sciences, The George Washington University, 2023 G Street NW, Washington, DC 20052;

Extant Chloranthaceae are a small "magnoliid" family of four very distinct genera with extremely reduced flowers, but fossil dispersed pollen, leaves, and reproductive structures indicate they were one of the first abundant angiosperm groups. We have compiled a morphological cladistic data set of 131 characters from all parts of the plant, 38 living species of Chloranthaceae, 10 outgroups, and six fossil reproductive structures with in situ pollen scored for characters preserved in the fossils. We have analyzed this data set with and without fossils and obtained generic relationships consistent with higher-level molecular analyses. Addition of fossils had only minor effects on the arrangement of extant species, but most of the fossils had several equally parsimonious positions in the tree. Our results indicate that Barremian-Aptian female flowers with Asteropollis pollen are basal members or outgroups of the basal genus Hedyosmum, and three species of Late Cretaceous trilobed androecia are outgroups to or nested within Chloranthus. However, Cenomanian Couperites fruits with Clavatipollenites pollen, which has been compared with pollen of extant Ascarina but is plesiomorphic for the family, may represent either an extinct line nested within Chloranthaceae or an extinct sister taxon. We hope to combine this morphological data set with molecular sequences being amassed by Hongzhi Kong (Beijing), Susanne Renner (University of Missouri St. Louis), and Lennart Andersson (Gothenburg), with the help of Taylor Field (Berkeley), who is studying the ecophysiology of Chloranthaceae and other "basal" angiosperms, to obtain a better estimate of relationships within the genera and the position of the fossils, and to relate the results to early floral evolution, molecular estimates of divergence times, paleoecology, and biogeographic history.

Study of the Lower Eudicots (Ranunculales, Proteales Through to Core Eudicots)

Susana Magallon and Sara Hoot

  1. Synthesize a non-molecular data set for placeholder taxa within these groups. This data set is meant to be definitive and useful for paleobotanists and others interested in character evolution. When data is not available (or of dubious reliability) for a particular trait, we will do the necessary morphological, anatomical, etc. work. We will concentrate our attention on characters which are most likely to turn up in plant fossils or of prime interest in terms of character evolution.
  2. Complete sequencing for any placeholders that are not represented by the most commonly used genes (five or six?) employed to resolve angiosperm phylogeny.
  3. Test methodologies for incorporating fossil data into molecular/morphological data sets.
  4. Test the molecular clock hypothesis, estimate ages of lineage splitting using appropriate rate-constant or rate-heterogeneous methods, and compare the results with the available fossil record.
  5. Explore character evolution and reliability of morphological characters using the resulting phylogeny.

Deep Time Home | Projects | Meetings | News and Info | Goals | Links | Site Map | FLMNH