|
Leaf Outline Evaluation in Selected Philippine Hoya R. Br. Species (Apocynaceae) Using Elliptic Fourier Analysis |
|---|---|
| รหัสดีโอไอ | |
| Creator | Jess H. Jumawan |
| Title | Leaf Outline Evaluation in Selected Philippine Hoya R. Br. Species (Apocynaceae) Using Elliptic Fourier Analysis |
| Contributor | Inocencio E. Buot, Jr |
| Publisher | National Science Museum, Thailand |
| Publication Year | 2564 |
| Journal Title | The Thailand Natural History Museum Journal |
| Journal Vol. | 1 |
| Journal No. | 15 |
| Page no. | 21-30 |
| Keyword | elliptic fourier analysis, principal component analysis, Hoya species, cluster analysis |
| URL Website | http://www.thnhmjournal.com/ |
| Website title | THE THAILAND NATURAL HISTORY MUSEUM JOURNAL |
| ISSN | 1686-770X |
| Abstract | Variations in leaf shape contours were analyzed using elliptic fourier descriptors (EFDs) of three Hoya species namely: Hoya buotii, Hoya coriacea, and Hoya halconensis. The species were grown under greenhouse conditions and exhibited resemblances which could become source of taxonomic confusion as reproductive parts were needed for species discrimination. Thirty leaf samples were photographed, converted to binary images, and subjected to elliptic fourier analysis (EFA). The extracted EFDs were quantitatively analyzed by PCA. The results utilized four effective principal component (PC) scores with 84.93% total variation. The largest variation contributed by PC1 (58.43% variation) was due to laminar size. Significant differences contributed by PC1 and PC3 (Anova P0.05) were suggested to be environmentally influenced. The three species depicted similarity at roundness of base, tapering towards tip (PC2) and roundness along leaf margin (PC4) were possible sources of taxonomic confusion. The dendrogram from cluster analysis revealed leaf outlines similarity between H. coriacea and H. buotii at about 50% Euclidean index. The leaf outline variations could be intricate by human recognition and small quantitative variations detected by EFA provided an effective platform in characterizing leaf shapes. |
