Problems in using Robertsonian rearrangements in determining monophyly: examples from the genera <i>Tatera </i>and <i>Gerbilluru</i><i>s</i>

Abstract

Chromosomal banding data on three species of Tatera from Kenya significantly alter the previous hypothesis of relationships between and within the genera Tatera and Gerbillurus based on cladistic analyses and the rule of parsimony (Qumsiyeh, 1986b). Of the many possible hypothetical relationships, the most parsimonious tree showed three homoplasies and allowed the genus Gerbillurus to be paraphyletic. The alternative trees, depicting larger number of homoplasies but with homoplasies restricted to fusion or fission events, were compatible with the morphological data in supporting the monophyly of Gerbillurus. To choose between the hypothesis based on the most parsimonious chromosomal tree and those supported by both morphology and slightly less parsimonious chromosomal trees, we performed an electrophoretic study on this group of gerbils. The conclusions are that the genus Gerbillurus is monophyletic and represents a branch that is closely related to the T. robusta group of Taterillini. The study documents that fissions and fusions must have occurred frequently and that in some cases the same fusions were acquired in two independent lineages in numbers exceeding those that are predicted by strict parsimony. The results raise questions about the validity of systematic conclusions based solely on fusion/fission data and utilizing the parsimony criterion.

Systematics and phylogeny of West African gerbils of the genus <i>Gerbilliscus</i> (Muridae: Gerbillinae) inferred from comparative G- and C-banding chromosomal analyses

“…The monophyly of Gerbilliscus and Gerbillurus (here represented by G. tytonis , in addition to G. paeba and G. vallinus already studied by Qumsiyeh et al, 1987) shown by this analysis is not surprising as it was suggested by earlier chromosomal (Qumsiyeh, 1986;, morphological (Pavlinov, 2001;Pavlinov et al, 1990) and molecular (Chevret, 1994;Chevret and Dobigny, 2005;Colangelo et al, 2007) studies. In fact, even the close relationship between these two genera was first suggested on the basis of comparative banding analysis (Qumsiyeh et al, 1987).…”

“…kempi (Codjia et al, 1994;Colangelo et al, 2001). Similarly, G. robustus in the Central African Republic (Matthey and Petter, 1970) and G. kempi in Guinea (Gautun et al, 1986) were both characterized by 2n = 46, whereas other authors attributed a 2n = 36 karyotype to the former species (Qumsiyeh et al, 1987;Granjon and Dobigny, 2003;Colangelo et al, 2005) and 2n = 48 karyotype to the latter (Codjia et al, 1994;Colangelo et al, 2001). Not surprisingly, this has resulted in inaccuracies in recent taxonomic lists of the genus Carleton, 1993, 2006), where the inclusion of certain species is clearly questionable from a chromosomal point of view.…”

“…While the East and South African Gerbilliscus were repeatedly the subject of chromosome banding (Qumsiyeh, 1986;Qumsiyeh et al, 1987;Qumsiyeh and Schlitter, 1991;Colangelo et al, 2005), biochemical (Qumsiyeh et al, 1987;Qumsiyeh and Chesser, 1988) and molecular (Colangelo et al, 2005) studies, the West and central African taxa have remained poorly documented: two taxa, G. guineae and G. cf . nigrita were chromosomally studied by R-banding (Benazzou et al, 1984) and only one, G. kempi , by using both G-and C-banding .…”

“…Another pertinent question highlighted in earlier studies is the relationship between the genera Gerbilliscus and Gerbillurus . Qumsiyeh (1986) and Qumsiyeh et al (1987) demonstrated that on chromosomal grounds two Gerbillurus species (G. paeba and G. vallinus) were evolutionarily very close to East African Gerbilliscus . This issue has recently been addressed by molecular data that have quite clearly shown that Gerbillurus should be included in Gerbilliscus (Chevret and Dobigny, 2005;Colangelo et al, 2007).…”

Recent radiation in West African Taterillus (Rodentia, Gerbillinae): the concerted role of chromosome and climatic changes

“…It is well supported by the CI, RI, RC and g1 statistics and by bootstrap and decay index values. In addition, the majority of the chromosome changes inferred correspond to Robertsonian changes, something which is commonly observed in mammals including gerbilline rodents (Qumsiyeh et al, 1987). More importantly, the cladistic method used here implies that character changes are not polarized a priori, thus allowing both fusions and fissions to be inferred (reviewed in Dobigny et al, 2004a).…”