1 |
Molecular systematics of the South African gobiid fishes of the genus glossogobius gill, 1862 (teleostei : gobiidae)Maake, Pholoshi Abram January 2009 (has links)
Thesis (M.Sc. (Zoology)) --University of Limpopo, 2009 / The genus, Glossogobius Gill, 1862 belongs to the family Gobiidae, a large fish family represented by several thousand species. In southern African there are about 51 genera comprising about 107 species. It has been suggested that gobies underwent an extensive ecological radiation, with species
diversifying into marine, brackish and freshwater habitats. The genus Glossogobius has about 22 species with an Indo-west Pacific distribution. The overall aim of the current study was to examine the systematics of Glossogobius in southern Africa using molecular markers in an attempt to clarify
the geographic distribution and species status.
A total of 131 individuals of G. callidus were collected from rivers and estuaries throughout the species’ distribution in two southern African biogeographic regions (subtropical and warmtemperate) using six-meter seine nets, while individuals of G. tenuiformis were only collected in
KwaZulu-Natal region. Additional specimens and taxa were downloaded from Genbank and also loaned from the SAIAB DNA collection and the Art Gallery and Museum of the Northern Territory(NTM), Darwin Australia. This study employed mtDNA cytochrome b (cyt b) and the nuclear S7 ribosomal protein gene intron 1 (S7 intron 1) as genetic markers. In determining the phylogeographic
structure of the river goby general methods such as AMOVA, demography analyses, network of haplotypes and phylogenetic tree reconstruction were employed. Evolutionary relationships within the genus were resolved by employing parsimony (MP), Bayesian Inference (BI) and Maximum likelihood (ML) tree reconstruction analyses. Both MP and ML were computed in PAUP* while BI was computed using Mr Bayes.Initially, a large-scale assessment of the genetic structure and diversity of G. callidus was undertaken. This species has a wide distribution in most estuaries and coastal river systems along the
east coast of southern Africa, and in inland freshwater systems from the north-eastern parts. Overall,patterns of haplotype relatedness, high haplotypic diversity related to low nucleotide diversity, and
mismatch distribution analysis supported a population expansion hypothesis with continued restricted dispersal and gene flow. Two major historically isolated lineages (5% divergence, cyt b)were identified within this species, one comprising subtropical samples (KwaZulu-Natal lineage),and the other including the majority of the warm temperate localities and a few subtropical localities(EC, Eastern Cape lineage). Glossogobius callidus individuals collected from riverine localities within the EC lineage were geographically structured and support the general hypothesis of structure among freshwater species while the estuarine individuals were less structured. Although, the individuals from these two lineages were morphologically inseparable, the high genetic separation between the two conforms to the reports that G. callidus is a species complex. The results further revealed some evolutionary significant units (Nkanini/Kosi bay lineage, and Mozambique) in the KZN lineage, KwaZulu-Natal. However, there were other identified localities with restricted gene flow and some degree of reproductive isolation that were not evolutionary significant. Thus, the results indicate that G. callidus is a species with relatively strong population and phylogeographic
structure in southern Africa.The validity of the species of the genus Glossogobius in southern Africa (G. callidus and G. giurus
with its synonymy G. tenuiformis) was investigated by estimating the relationships among the Glossogobius taxa in southern Africa, using the same molecular markers. Based on a model selected under AIC in Modeltest, ML analysis of the dataset also recovered a phylogram which was similar to
MP and BI in recognising the major clades in many respects. All analyses recovered monophyly for all southern African species of the genus. Although the cyt b analyses included Glossogobius species from Australia, there was no evidence of a phylogeographic break between the two continents.
However, it appears that each southern African Glossogobius species has a sister species occurring in Australia according to the cyt b dataset. Glossogobius aureus was revealed as the probable sister species to G. tenuiformis while the African G. giuris group was monophyletic with the Australian G.
giuris. The two G. callidus groups that appear to represent two unique species (sp. 1 and sp. 2) were also sister taxa to the undescribed Glossogobius sp. from Australia. It could be postulated that this
inter-continental species relationships were a result of either allopatric or sympatric speciation events, to the ancestral taxa of each of these three sister groups which in turn split up during the Gondwana breakup. The inclusion of G. celebius (a freshwater and brackish species with an Asian
and Oceania distribution) in the S7 intron dataset did not affect the monophyly of southern African species. However, this species appears to be more basal in the genus than the southern African
species. The results further indicated that G. tenuiformis (previously a synonym of G. giuris) forms a separate clade that is distinct from the G. giuris species group. Although G. tenuiformis and G. giuris
formed a monophyletic group in all analyses involving the S7 intron 1, individuals from each species were monophyletic and strongly supported. The scanning electron micrographs of the cheek papillae of the two species further revealed morphological differences. The congruence between both morphology and molecular data indicated that G. tenuiformis is unequivocally distinct from G. giuris and it is therefore suggested that the taxonomic status of the G. giuris group be revised. / National Research Foundation, the Ernst and Ethel Eriksen Trust, the South African Institute for Aquitic Biodiversity (SAIAB), and the University of Limpopo
|
Page generated in 0.0197 seconds