Numerical taxonomic study of Ipomoea species in Southwestern Nigeria.

Olorungbeja, John Adewumi (2013)

xiv, 79p

Thesis

This study investigated the characters of Ipomoea species with a view to improving the taxonomy of the genus in Nigeria, delimit the species into their traditional sections and provide a better understanding of the affinities among species. Specimens of Ipomoea species were collected from different sites in Southwestern Nigeria. Twenty- three Ipomoea species were collected from natural population in Southwestern Nigeria and thirteen additional species were herbarium specimens. Eighty one (81) characters were examined on each specimen. The basic data matrixes of 81 × 36 were prepared by coding for presence or absence of the attributes of characters involved. Principal Component Analysis (PCA) was carried out on the taxonomic data using Palaeontological Statistics. A combination of two- state and multistate coding were employed. Single Linkage Cluster Analysis (SLCA) was performed for the estimation of resemblance among the groups of Operational Taxonomic Unity (OTU) and the affinities of the different OTUs were determined. The Single Linkage Cluster Analysis revealed interspecific relationships within the genus Ipomoea. At lower similarity coefficients below 58%, there was no separation of species. Several clusters or phenons were formed at higher similarity coefficients value based on the affinities of the species. There was correlation between the SLCA and the PCA of the Ipomoea species. At lower % variance, the level of similarity was so high that Ipomoea could hardly be separated leading to the overlapping and clustering of the species. The PCA disclosed the behavior of the Ipomoea species at 7.58, 13.2 and 35.10% variance. There was radial distribution of the Ipomoea species at lower % variance of 7.58% and 13.27% which were on components 3 and 2 respectively. When the power of variance was reduced radial distribution took over to strengthen the monophyly hypothesis of Ipomoea species. The study concluded that Ipomoea species could be delimited into their traditional sections using the species characters.

Collections: