Thesis (Ph. D.)--City University of Hong Kong, 2005. / "Submitted to Department of Biology and Chemistry in partial fulfillment of the requirements for the degree of Doctor of Philosophy." Includes bibliographical references (leaves 233-269).

This study reports on investigation of ways of improving the breeding programme for growth-related traits in common carp in Vietnam. The base population was synthesized following a single pair mating scheme from six carp stocks: (1) 2nd generation of family selection; (2) Hungarian 6th generation of mass selection; (3) Hungarian scaled carp; (4) Indonesian yellow 6th generation of mass selection; (5) Indonesian yellow carp; and (6) Vietnamese 6th generation of mass selection. The next two selected generations were produced using a partial factorial mating scheme, with each family being split and reared using communal early rearing (CER) or separate early rearing (SER) methods. The second generation (G2) was produced from selected fish from the CER G1 group. The total number of selection, control and reference families was 135 in the G1 and 101 in the G2 respectively. The control and reference (Hungarian P33 line) families were produced by single pair mating (reference families with the G2 only). Seven microsatellite loci were used for parentage assignment in the CER groups: 96.8% of the offspring (1284 individuals) and 96.2% offspring (1341 individuals) were unambiguously assigned to 113 families (selection, control) in the G1 and 99 families (selection, control and reference) in the G2 generations, respectively. Restricted maximum likelihood in the individual model was used to estimate phenotypic and genetic parameters. In CER, the estimated heritability values of common carp were from 0.20 ± 0.04 to 0.29 ± 0.05 for both weight and length at final harvest, indicating substantial additive genetic variation for selection on growth-related traits. The overall obtained maternal and common environmental effects were consistently close to zero. The average of direct response to selection for body weight was 15.0% per generation. In SER, the number of families in the G1 and G2 were 135 (selection and control) and 101 (selection, control and reference), respectively. The heritability estimates were from 0.20 ± 0.07 to 0.31 ± 0.08 at final measurement. Common environmental (full-sib family) effect were all lower at tagging and slightly higher at last measurement, ranging from 0.05 to 0.22. The response in each generation of selection as the difference between the selection and control lines was 8.1% on average for weight at final harvest, lower than under CER. The high genetic correlations of growth-related traits between the third (one year old, mature) and second (7 months old) measurements could allow selection to be based on the earlier assessment, reducing handling stress close to spawning. The benefits of using microsatellite markers to ascertain parentage, achieve greater growth rate (close to farming systems), shorten time to maturity and selection, and the overall relative merits of using CER v’s SER in this genetic improvement programme are discussed.