Caryophyllaceae is a principally holarctic family including around 2200 species often classified into the three subfamilies Alsinoideae, Caryophylloideae, and Paronychioideae. Complex and possibly homoplasious morphological characters within the family make taxa difficult to delimit and diagnose. To explore part of the morphological evolution within the family, we investigated the phylogeny of the Caryophyllaceae by means of analyzing plastid and nuclear sequence data with parsimony and Bayesian methods. We describe a mode of tracing a stable phylogenetic signal in ITS sequences, and a significant common signal is shared with the plastid data. Parsimony and Bayesian analyses yield some differences in tree resolution. None of the subfamilies appear monophyletic, but the monophyly of the Caryophylloideae is not contradicted. Alsinoideae are paraphyletic, with Arenaria subg. Eremogone and Minuartia subg. Spergella more closely related to the Caryophylloideae. There is strong support for the inclusion of Spergula-Spergularia in an Alsinoideae-Caryophylloideae clade. Putative synapomorphies for these groupings are twice as many stamens as number of sepals and a caryophyllad-type of embryogeny. Paronychioideae form a basal grade, where tribe Corrigioleae are sister to the rest of the family. Free styles and capsules with simple teeth are possibly plesiomorphic for the family.

We present a cladistic analysis of 53 taxa from Asteraceae using 72 characters of mainly morphological data. Resuhs reveal subfamily Cichorioideae is paraphyletic. The Mutisieae constitute a basal grade that contains monophyletic groups with Mutisia and Cardueae sensu lato. The Arctotideae, Liabeae, Vernonieae, and Lactuceae form a clade together with subfamily Asteroideae. Our results show some inconsistencies with those from cladistic analyses ol molecular data.The relationships within the Asteraceae, dis-cladistic analysis of the family, noting strong morcussed since Cassini and Lessing in the early nine-phological support for this dichotomy. Bremer s teenth century, have been scrutinized using cla-(1987) analysis further supported the monophyly distic methods (Bremer, 1987; Jansen & Palmer, of the subfamily Asteroideae, and the subfamily 1987 Jansen et aL, 1990). Many results Cichorioideae emerged as a paraphyletic assemof cladistic analyses were inconsistent with the in-blage. The analyses of Jansen et al. ( 1990) yieldea trafamilial systems proposed by other authors (e.g., quite different topologies. The only monophyletic Carlquist, 1976;Cronquist, 1977 Cronquist, , 1981 Jeffrey, groupings established in all trees, both Wagner and 1978;Wagenitz, 1976). Jansen & Palmer (1987) demonstrated that Mutisieae-Barnadesiinae (Ca-Dollo analyses, are the subfamily Asteroideae, the Vernonieae as the sister group to the Liabeae, an brera, 1961, 1977) form a sister group to the rest the Eupatorieae as a subclade within a paraphyletic of the family, a finding that was corroborated by Heliantheae sensu lato. Dollo analysis always supan apomorphic cpDNA inversion in the latter group, ported a monophyletic Cichorioideae, but the Wag-Bremer (1987) also came to this conclusion in his ner analysis resulted in a paraphyletic Cichorioi-1* I* ^ This study was supported by a grant for Compositae taxonomy and Asteraceae phylogeny from the bwe i Natural Science Research Council. We thank Arne Anderberg for discussions of complex cladistic analyses and o letting us examine material that he had on loan from K. We also thank one anonymous reviewer, Michael J. Donog •

No abstract

No abstract

No abstract