Morphometric and Morphomeristic Variations in Five Populations of Indigenous Celebes Goby Glossogobius celebius (Perciformes: Gobiidae) from Southern Luzon, Philippines
To elucidate the population structure and pattern of morphological divergence of the indigenous Celebes goby Glossogobius celebius (Valenciennes, 1837), morphometric and morphomeristic characters of this species were compared among populations collected from five study sites in Southern Luzon, Philippines. Univariate analysis of variance confirmed significant differences among populations for 20 out of 23 allometrically-transformed metric characters (P < 0.05), while nine meristic traits of five populations were comparatively homogeneous (P > 0.05). Intrapopulation variability was very low (6.81–7.21%) for each location. Multivariate analysis of variance determined a very high significant group variability (P < 0.05). Also, principal component analysis (PC1=36.60%; PC2=18.34%) and canonical variate analysis (CV1=44.59%; CV2=37.30%) identified the 2nd dorsal fin length, anal fin length, and the 2nd dorsal fin origin as the main significant contributors of population differentiation. Derived classification functions correctly classified an average of 68% individuals into their original groups. The canonical variate analysis plot exhibited high isolation of lacustrine populations (Lake Taal and Laguna de Bay) from riverine populations (Dampalit River, Alitaaw River and Pansipit River), with the latter having longer metric characters. Likewise, squared Euclidean distance similarity using metric data clearly produced two clusters representing both river and lake populations. This morphological divergence is suggested to be environmentally-induced (phenotypic plasticity) primarily by the differences in flow rate and temperature of the study sites.
Key Words: Celebes goby, Glossogobius celebius, Laguna de Bay, morphometric, riverine population
Abbreviations: CV – canonical variate, CVA – canonical variate analysis, DOST – Department of Science and Technology, PC – principal component, PCA – principal component analysis, UPGMA – unweighted pair group method with arithmetic mean
For more communications and information on this journal, please contact the Editor-in-Chief at TeleFax: +63 049 536 2379, E-mail: email@example.com, Website: http://www.pas-uplbca.edu.ph/