Files in this item



application/pdfDanderson_Clark.pdf (49MB)
(no description provided)PDF


Title:A Phylogenetic Study of the Arracacia Clade (Apiaceae)
Author(s):Danderson, Clark A.
Director of Research:Downie, Stephen R.
Doctoral Committee Chair(s):Downie, Stephen R.
Doctoral Committee Member(s):Levin, Geoffrey A.; Molano-Flores, Brenda; Soto-Adames, Felipe N.; Whitfield, James B.
Department / Program:Plant Biology
Discipline:Plant Biology
Degree Granting Institution:University of Illinois at Urbana-Champaign
South America
rapid species radiation
"True" Rhodosciadium group
Abstract:The Arracacia clade (Apiaceae: Apioideae) contains a heterogeneous assemblage of 12 genera, comprising 111 species distributed in high montane temperate and sub-alpine (= páramos) habitats of meso- and South America. Recent molecular studies have indicated that some of the larger genera of the Arracacia clade (i.e., Arracacia, Prionosciadium, and Rhodosciadium) are polyphyletic. Given this evidence and historically that these taxa have been difficult to circumscribe, a thorough examination of the clade using molecular and morphological data is warranted. This dissertation seeks to address six major objectives: 1) elucidate the intergeneric and interspecific relationships of the Arracacia clade and ascertain the degree of polyphyly of its constituent genera; 2) examine the comparative utility of cpDNA and nrDNA loci at resolving phylogenetic relationships in the Arracacia clade; 3) delimit major clades within the Arracacia clade; 4) conduct a cladistic study of the morphological characters of the “True” Rhodosciadium group, a clade that contains some species of Arracacia, and many but not all of the species of Prionosciadium and Rhodosciadium (including the nomenclatural type of the latter genus, hence the name of the group); 5) examine character evolution within the “True” Rhodosciadium group, and determine whether there are any subclades that can be delimited unambiguously on the basis of morphology; and 6) produce a revision of the “True” Rhodosciadium group. Initially, cladistic analyses of nuclear ribosomal DNA internal transcribed spacer (nrDNA ITS) sequences were carried out on taxa of the Arracacia clade, the closely related South American páramo genera Cotopaxia, Niphogeton, and Perissocoeleum, and outgroups. To further resolve relationships within the clade, a chloroplast DNA (cpDNA) pilot study examining the utility of 20 non-coding regions (intergenic spacers and introns) was performed. Following this pilot study, cladistic analyses of the four most variable cpDNA regions (trnQ–5’ rps16, trnD–trnT, rpl32–trnL, and ndhA intron) and two nrDNA regions (ITS and ETS [= external transcribed spacer]) were conducted. From the combined analyses, groups were recognized and delimited for future studies. Cladistic analyses of molecular and morphological data were performed for the “True” Rhodosciadium group so as to determine the relationships within the group and the evolutionary pattern of the 23 morphological characters examined. Results of the phylogenetic analyses show that many of the genera of the Arracacia clade are indeed polyphyletic, particularly Arracacia, Prionosciadium, and Rhodosciadium, whereas Donnellsmithia and Myrrhidendron are each monophyletic. Although most relationships in the majority of the trees are poorly resolved and supported, nine groups are recognized on the basis of the combined nrDNA and cpDNA analyses. Low molecular sequence divergence, large polytomies, the frequency of polyploidy, and the presence of short internal and long terminal branches indicate that the Arracacia clade, as well as the genera Cotopaxia, Niphogeton, and Perissocoeleum, have undergone rapid species radiation. The relationships within the “True” Rhodosciadium group remain largely unresolved and/or poorly supported on the basis of nrDNA and cpDNA sequences, and there are no discernable morphological features separating this group from the other groups recognized in the Arracacia clade. The morphological characters examined were poor indicators of phylogenetic relationships, and mapping of the characters onto the molecular-derived trees revealed that none of the characters provided unique synapomorphies useful for delimiting recovered clades. With these problems in mind, a tentative treatment of the “True” Rhodosciadium group was produced. In conclusion, the above-mentioned studies and the resultant recognized subclades provide the necessary foundation for future studies needed to better understand the relationships of the taxa and to revise the genera of the Arracacia clade. More in-depth morphological and anatomical studies are required to determine what, if any, characters delimit the “True” Rhodosciadium clade and subclades within. Additional fast-evolving loci from the nuclear and chloroplast genomes should be investigated in an attempt to further resolve the relationships of the clade. For those clades that can be delimited on morphology, revisions should be pursued. Also, further studies on the Arracacia clade and the genera Cotopaxia, Niphogeton, and Perissocoeleum are needed to shed light on their biogeographic histories.
Issue Date:2011-05-25
Rights Information:Copyright 2011 Clark Danderson
Date Available in IDEALS:2013-05-26
Date Deposited:2011-05

This item appears in the following Collection(s)

Item Statistics