ALBERT

Notes: The age and growth of three widely distributed tropical clupeids, Amblygaster sirm, Herklotsichthys quadrimaculatus and Spratelloides delicatulus at four sites in Kiribati were examined by otolith increment counts. Amblygaster sirm grew to 230 mm s.l. and 130 g and lived for up to 2 years at all sites. Herklotsichthys quadrimaculatus grew to 110 mm and 15 g and lived up to 1 year; white S. delicatulus grew to 66 mm and 2.5 g and lived for 4-5 months. There was little difference in the rate of growth in length of any of the species at the sites, except for S. delicatulus at one site. All species grew rapidly in length at the expense of growth in weight during the early life-stages when mortality is highest, hence reducing the time they spend in this life-stage. The relationship between weight and age varied between the sites for H. quadrimaculatus and S. delicatulus, but these differences narrowed as fish size increased. Herklotsichthys quadrimaculatus also grew more rapidly in weight during the first 3 months than the other species. Growth comparisons between the three species showed that all grew in length at similar rates for the first 2 months. Spratelloides delicatulus became sexually mature at 2 months of age and subsequently grew more slowly than the other species. After 4 months A. sirm grew faster than H. quadrimaculatus. The rates of growth in length and weight of these species from Kiribati were compared with those from other regions of the Indo-Pacific. There were significant differences: A. sirm grew to a larger size in Kiribati than elsewhere; H. quadrimaculatus grew at a similar rate at all sites during the first 2 months, but then the growth rate of Kiribati fish declined and they reached a smaller asymptotic size than elsewhere; S. delicatulus from Kiribati grew in length at similar rates to Solomon Islands fish, but were significantly lighter in weight at the same age. The relative growth performance of these short-lived tropical species was compared to that of temperate clupeids. All species showed similar growth performance despite the slower growth rates of temperate species, which suggests that the clupeoids' phylogenetic conservatism in some life history traits extends to growth. Regional and local differences in growth rates of some tropical clupeids are as large as those found among most clupeid species.

Notes: Sprafelloides grarilis (Temminck & Schlegel), S. delirarulus (Bennett) and S. lewisi (Wongratana) live generally less than 4 months. They lived more than 5 months at only three of 10 sites sampled. Growth varied seasonally in S. grocilis(Temminck & Schlegel) at two sites in the Solomon Islands and in S. clelicululus (Bennett) at Helix Reef, near Townsville, Australia. This variation in the growth rate of S. grucilis could not be related to specific environmental patterns, but that of S. delicurulus at Townsville might be temperature-related. Instantaneous growth rates for all species were 0.4-1.7 mm day -1 at an age of I month and the range of variation in growth rate was similar for all species. These rates declined to 0.1 mm day -1 at 3 months of age, which is the rate reported for other short-lived clupeoids at the same age. The growth rates of fish from coral atoll and coastal lagoons were similar but barrier reef fish grew significantly faster. These results indicate that Sprarelloides have an extremely flexible growth pattern and that biological variation within a site can be as great as variation between sites.

Notes: The diets and prey selection of six species of tuna baitfish (the engraulids Stolephorus devisi and S. heterolobus, three dussumierids of genus Spratelloides, and the apogonid Archamiazosterophora. were examined at two heavily fished and one unfished site in the Solomon Islands. All species ate almost exclusively nektonic zooplankton, especially crustaceans. Calanoid copepods were the most importantpreyofallspeciesexcept A. zosterophora, whichateagreater biomassoflarvalfish. Baitfishcould be divided into two groups according to prey taxa: the apogonid A. zosterophora and the engraulids S. devisi and S. heterolobus ate similar prey in similar proportions; the three dussumierid species (Spratelloides) ate smaller zooplankton and a greater proportion of calanoids than the other species.The size of prey eaten differed greatly between baitfish species. A. zosterophora ate larger prey than other species. The two Stolephorus species ate similar-size prey, which were larger than the prey of the three species of Spratelloides. Among the Spratelloides, S. delicatulus ate smaller prey than the other two species. All species differed in the size of calanoids, carids and brachyurans they ate. In all cases, prey differed more by the minimum dimension (width) than by volume. There were significant relationships between fish length and prey width and volume, except for Archamia. However, in terms ofvolume, the prey of Stolephorus increased with fish size at a much greater rate than in Spratelloides species.Baitfish diets were compared to the zooplankton available. Prey selection by the three Spratelloides species and two Sfolephorus species was density-dependent for all common prey types, although Stolephorus positively selected hyperiid amphipods and brachyurans. Archamia showed strong negative selection for calanoids and selected larger zooplankters, including fish larvae and carids, avoiding smaller, more common, prey.Overall baitfish density and zooplankton density did not correlate at either of the heavily fished sites. Nor were there consistent differences for any species in diet or changes in prey selection between the heavily fished and unfished sites. This suggests that, in these lagoons, commercial baitfish catches are not directly influenced by the composition or density of prey.