-Individual Pinctada margaritifera molluscs were collected from the Takapoto atoll (Tuamotu Archipelago, French Polynesia) and used to produce ten first generation full-sib families in a hatchery system, following artificial breeding protocols. After three years of culture, these progenies were transferred to Rangiroa atoll (Tuamotu Archipelago, French Polynesia) and tested for their potential as graft donors. A large-scale grafting experiment of 1500 grafts was conducted, in which a single professional grafter used ten individual donor oysters from each of the ten families, grafting 15 recipient oysters from each donor. The recipient oysters were all obtained from wild spat collection in Ahe (Tuamotu Archipelago, French Polynesia). After 18 months of culture, 874 pearls were harvested. Highly significant donor family effects were found for nucleus retention, nacre thickness, nacre weight, pearl colour darkness and visually-perceived colour (bodycolor and overtone), pearl shape categories, surface defects and lustre, the last two of which are components of the Tahitian classification grade. No significant difference was recorded between the ten G1 families for the absence or presence of rings. The progenies could be ranked from "best" (i.e., the donor whose grafts produced the greatest number of grade A pearls) to the "worst". Some progenies had extreme characteristics: family B presented the greatest number of pearls with lustre (98%) and a high proportion of dark gray to black with green overtone pearls (70%). These results have important implications for the selective breeding of donor pearl oysters: it may be possible to reach a point where specific donor lines whose grafts produce pearls with specific quality traits could be identified and maintained as specific breeding lines.

No abstract

Ecological factors may influence the number of parasites encountered and, thus, parasite species richness. These factors include diet, gregarity, conspecific and total host density, habitat, body size, vagility, and migration. One means of examining the influence of these factors on parasite species richness is through a comparative analysis of the parasites of different, but related, host species. In contrast to most comparative studies of parasite species richness of fish, which have been conducted by using data from the literature, the present study uses data obtained by the investigators. Coral reef fishes vary widely in the above ecological factors and are frequently parasitized by a diverse array of parasites. We, therefore, chose to investigate how the above ecological factors influence parasite species richness in coral reef fishes. We investigated the endoparasite species richness of 21 species of butterfly fishes (Chaetodontidae) of New Caledonia. We mapped the diet characters on the existing butterfly fish phylogeny and found that omnivory appears to be ancestral. We also mapped the estimated endoparasite species richness, coded from low to high parasite species richness, on the existing butterfly fish phylogeny and found that low parasite species richness appears to be associated with the ancestral state of omnivory. Different dietary and social strategies appear to have evolved more than once, with the exception of obligate coralivory, which appears to have evolved only once. Finally, after controlling for phylogenetic relationships, we found that only the percentage of plankton in the diet and conspecific host density were positively correlated with endoparasite species richness.