New Trends in Plant Systematics

Abstract

Hybridization is thought to be an important phenom enon in angiosperm evolution, and it has been suggested that a majority of all plant species may be derived from past hybridization events (e.g., Stebbins, 1959; Raven, 1976; Grant, 1981; Arnold, 1997). In addition, there is an increasing interest in the reconstruction of reticulate pat terns (e.g., Linder & Rieseberg, 2004), with increased emphasis on the need to explore multiple independent markers to investigate the origin of putative hybrid species (e.g., Hamzeh & Dayanandan, 2004; Koontz & al., 2004 and other examples listed in Table 1). Several reviews have recently been published on the process of hybridization itself or on issues indirectly related to it: Hegarty & Hiscock (2005) and Zhou & al. (2005) present an overview of molecular techniques as well as criteria for distinguishing hybrid speciation; Gross & Rieseberg (2005) evaluate the ecological genetics of homoploid hybrid speciation, and Seehausen (2004) reviewed the possible role of hybridization in adaptive radiation. Mallet (2005) presented several hybrid examples in plants and animals to discuss the evolutionary signifi cance of hybridization. Also, many studies have been published in the past several years on the incidence and role of (allo)polyploidy in evolution (e.g., in Soltis & Soltis, 1993, 2000; Ramsey & Schemske, 1998; Otto & Whitton, 2000; Crawford & Mort, 2003; Soltis & al., 2004). In addition to species-level hybridization, other (genome-level or molecular) evolutionary processes such as recombination, gene conversion or horizontal gene transfer can confound the phylogenetic signal in the data to such an extent that it may become non-treelike, and phylogenetic methods are not appropriate for analysis. It is best to check prior to phylogenetic analysis whether this applies, and if so, then use network methods to rep resent it (Bryant & Moulton, 2004). Nevertheless, the pages of botanical systematic jour nals are still remarkably devoid of examples of reticulate patterns, and plant species-level relationships are pre dominantly depicted as trees. The question can be asked whether this is because there are no suitable tools avail able for detection or whether the problem is merely ignored. In this paper we will explore the current practice of dealing with hybrid terminals in published phyloge netic studies, briefly describe a selection of network-pro ducing methods currently available, as well as discuss future possibilities of reconstructing reticulate patterns in angiosperm evolution. Many recently published studies report the occur rence of plant hybrids in several plant genera and fami lies, both at the polyploid and homoploid level (e.g., Spartina [Poaceae] in Ainouche & al, 2004; Actinidia [Actinidiaceae] in Chat & al, 2004; Glycine [Fabaceae] in Doyle & al, 2004; Phoenix [Arecaceae] in Gonzalez Perez & al, 2004; Pleione [Orchidaceae] in Gravendeel & al, 2004; Gagea [Liliaceae] in Peterson & al, 2004). Ellstrand & al. (1996) surveyed frequency and taxonom ic distribution of spontaneous hybridization in vascular plants in five major floras. They concluded that most hybrids are concentrated in particular families such as Poaceae, Cyperaceae and Rosaceae, and several genera within these families account for most of the hybrid species encountered (Ellstrand & al, 1996). Some life history characteristics seemed to be associated with hybridizing taxa such as perennial habit, asexual repro ductive modes, and outcrossing breeding system (Ellstrand & al, 1996; Rieseberg, 1997; Wisseman & Ritz, 2005). However, it is unclear whether the observed uneven distribution is due to intrinsic (biological) differ ences of the lineages involved (such as breeding system or ecological preferences), or to extrinsic factors such as extreme habitat or ecological transitions (Rieseberg, 1997; Gross & Rieseberg, 2005), or distribution pattern (the extent of sympatry with other species). Also, sam pling bias could be a factor with the number of reports on hybrids influenced by the systematic attention given to