Morphological and Genetic Identification, and Population Structure of Kelah (Tor Tambroides)

Tor tambroides, locally known as kelah, is one of the most sought after of the freshwater fishes in the country, both as food as well as a highly priced game fish. The identification of this fish into various groups and names is largely based on external features and colours. Thus, this study was undertaken to determine morphological differences, genetic variation and genetic relationships among three populations of Tor tambroides. Samples of kelah were taken from three locations, namely Pahang, Negeri Sembilan and Kelantan. A total of sixty three samples were used with weights ranging from 3.1 to 1800 g, total lengths ranging from 5.5 to 50.7 cm and standard lengths ranging from 4.2 to 40.0 cm. Morphological studies based on conventional and truss morphometric and meristic methods were applied on each sample of fish. The non-quantitative morphology of kelah from the three populations did not show any distinct differences. However, the presence of a median lobe and large scales with dark vertical bands especially on the scales above the lateral line are the most distinctive features of kelah that can differentiate this fish based on phenotypic observations. The range of gill rakers and vertebrae numbers were 17 to 23 and 30 to 34, respectively. The number of branchiostegals rays was three. The meristic characters showed greater differences among the Tor tambroides populations than the morphometric characters. The mean number of caudal fin rays, upper lateral line scales, lower lateral line scales and lateral line scales showed significant differences (P<0.05) among populations. The four meristic characters summarized by the canonical discriminant function analysis plots together with both the hierarchical clustering analyses based on five conventional morphometric and four meristic characters, respectively showed similarities in shape, fin or scale counts of kelah from the three populations. For Randomly Amplified Polymorphic DNA (RAPD) markers, a total of 20 short arbitrary primers were screened but only twelve primers were used. The values of the genetic distances generated using 226 RAPD markers ranged from 0.7548 to 0.7643. In addition, thirty eight microsatellite markers developed for Mystus nemurus were used in this study. Only five primer pairs which amplified bands at the expected allele size regions were used for characterizing this species. The observed heterozygosity values were higher than the expected heterozygosity values for the bands amplified by primer pairs MnSC4-3B, MnLR2-1-52A and MnRmC3-1 across the three populations but the bands amplified by primer pairs MnSC4-1A and MnLR2-1-17B showed lower observed heterozygosity values than the expected heterozygosity values. The mean FIS value across the three populations was negative, indicating no deficit in heterozygosity. The mean value of FST or the fixation index was low indicating no gene was fixed within populations relative to the total population and the populations were still high in gene variation. The high value of Nm suggested high gene flow among the three populations. Both the (χ2) chi-square and the (G2) likelihood ratio tests significant differences (P<0.05), indicating deviations from Hardy-Weinberg equilibrium in most loci except for one locus (MnSC4-1A) in the Negeri Sembilan population and two loci (MnLR2-1-52A and MnSC4-1A) in the Kelantan population. The genetic distance values generated by the microsatellite markers ranged from 0.1053 to 0.1960. The UPGMA dendrograms constructed from the genetic distances based on both the RAPD and the microsatellite markers showed similar clustering patterns in that the Negeri Sembilan and Kelantan populations shared a similar cluster while the Pahang population was away by itself.

Preview PDF 600404_fs_2006_39_abstrak_je__dh_pdf_.pdf Download (134kB)

Actions (login required)

View Item