Volume 86 AnnalsNumber 1 of the1999 Missouri Botanical GardenPHYLOGENY OF BASAL Sara B. Hoot,2 Susana Magalldn,: andEUDICOTS BASED ON THREE Peter R. Crane'MOLECULAR DATA SETS:atpB, rbcL, AND 18SNUCLEAR RIBOSOMAL DNASEQUENCES'ABSTRACT Phylogenetic analyses were conducted for 73 genera of "lower" eudicots (Ranunculidae and "lower" Haamamelididae).magnoliid outgroups, and appropriate representatives for higher taxa within the "higher" eudicot clade (e.g., Rosidae,Dillenidae, Asteridae) based on sequences of three genes: the two chloroplast genes atpB and rbcL and nuclear ribo-somal 18S DNA. Based on the partition homogeneity test, the three data sets were relatively congruent (P > 0.13).The data were analyzed using heuristic parsimony searches and bootstrap analyses in three ways: individually. the twochloroplast sequences combined, and ail three sequences combined. Both ingroup and outgroup sampling were variedto test the stability of the tree topology. The trees resulting from a combination of the chloroplast data and ail threedata sets had the best resolution and the strongest branch support. The following higher taxonomic groups wererecognized with high bootstrap values (> 90%): Eudicots (including Nelumbo), Ranunculidae (including Euptelea),Papaverales, "core" ranunculids, a clade consisting of "lower" hamamelids and "core" eudicots, "core" eudicots (in-cluding caryophyllids, asterids, and rosids), Dilleniaceae, caryophyllids (including Simmondsia), and asterids. All ran-unculid families, including Circaeasteraceae s.l. (including Kingdonia) and Lardizabalaceae s.l. (including Sargento-doxa), formed well-supported monophyletic groups. Other well-supported eudicot clades were Platanus/Proteaceae,Buxaceae/Didymeles, Trochodendraceae/Tetracentraceae, and a group with poor internal resolution that included generain Hamamelidaceae, various rosids, and Paeonia. Morphology (especially floral features) and other characteristics aredescriled in some detail for well-supported clades determined by the molecular data. ' This work was supported in part by grants from the National Science Foundation (DEB-9596011 to SBH and DEB-9020237 to PRC). We thank M. Grayum, D. Nickrent. and an anonymous reviewer for helpful comments on themanuscript; we also thank 1). Swofford for access to test versions of PAUP* 4.0. We are grateful to the following forproviding us with leaf material, )NA. or unpublished sequences: M. Chase, Royal Botanic Gardens, Kew; M. Dillonand A. Douglas, Field Museum; E. van Jaarsveld, Kirstenbosch Botanical Garden: K. Kron, Wake Forest University;T. Lammers, Field Museum: R. McBeath. Royal Botanic Garden, Edinibrgh; M. Maunder and T. Upson. Royal BotanicGardens, Kew; D. Nickrent, Southern Illinois University; R. Olmstead, University of Washington; H.-N. Qin, ChineseAcademy of Sciences, Beijing; Y.-L. Qiu, University of Zurich; and A. Reznicek, University of Michigan. We thank G.Zurawski for rbcL sequencing primers. SDepartment of Biological Sciences, University of Wisconsin-Milwaukee, P.O. Box 413, Milwaukee, Wisconsin53201, U.S.A. SThe University of Chicago, 5734 S. Ellis Ave., Chicago, Illinois 60637, U.S.A. 4 Department of Geology, The Field Museum, Roosevelt Road at Lake Shore Drive, Chicago, Illinois 60605, U.S.A. ANN. MISSOURI BOT. GARD. 86: 1-32. 1999.
BioStor
