Several scientific methods have been described in the identification of triploid fish. However, many of these methods are not applicable for routine management purposes due to their complexity and cost. In this study, the possibility of using morphological variation as a least cost and less complex method of distinguishing triploid and diploid African catfish Clarias gariepinus (Burchell, 1822) was examined. Triploid catfish were produced by cold shock of fertilized eggs in 5°C for 20 mins (at approximately 3 mins after fertilization). The fish were incubated, hatched and raised for 3 months. Ploidy levels of the fish were then ascertained by observing the erythrocyte shape. Triploid erythrocyte was ellipsoidal in shape while diploid was round. Morphological characterization was then carried out on 100 samples each of triploid and diploid African catfish. Although significant differences were observed in many parameters, the principal morphometric difference between triploid and diploid African catfish could not be clearly distinguished. It was therefore concluded that morphological characteristics is not ideal for discriminating triploids and diploids of African catfish. The used of erythrocyte characteristics still remains the cheapest and relatively effective method for triploid and diploid determination in African catfish.

This study was designed to examine the use of RAPD markers in discriminating triploid and diploid African catfish Clarias gariepinus (Burchell, 1822). Following a routine technique, triploidy was induced by cold shock and confirm by erythrocyte measurement in C. gariepinus. Thereafter, 80 RAPD markers were screened; out of which, three showed the highest percentage of polymorphism (i.e., OPB 16 = 71.43%; OPC 14 = 61.9%; OPD 12 = 75%). The results obtained showed genotype differences between triploid and diploid without overlapping. However, the development of a Sequence Characterized Amplified Region (SCAR) marker was not achievable because progenies of triploid and diploid C. gariepinus could not be differentiated based on a specific fragment. Consequently, the genetic distance showed high similarities for both treatments and the UPGMA-generated dendrogram could not separate the treatments into two distinct clusters. It was concluded that RAPD makers cannot be used to separate the ploidy status of fishes.