Patterns of embryo mortality in a demersally spawning coral reef fish and the role of predatory fishes

Citation
Mj. Emslie et Gp. Jones, Patterns of embryo mortality in a demersally spawning coral reef fish and the role of predatory fishes, ENV BIOL F, 60(4), 2001, pp. 363-373
Citations number
57
Categorie Soggetti
Aquatic Sciences
Journal title
ENVIRONMENTAL BIOLOGY OF FISHES
ISSN journal
03781909 → ACNP
Volume
60
Issue
4
Year of publication
2001
Pages
363 - 373
Database
ISI
SICI code
0378-1909(2001)60:4<363:POEMIA>2.0.ZU;2-O
Abstract
The biological significance of embryo mortality in demersally spawning cora l reef fishes is poorly understood. Here we describe patterns of variation in embryo mortality in Pomacentrus amboinensis (Pomacentridae) at Lizard Is land (Great Barrier Reef). The aim was to determine whether numbers of embr yos hatched substantially differed from egg production, and if so, identify whether predatory fishes were a source of embryo mortality. Spawning succe ss (number of eggs laid), embryo mortality (proportion of embryos that died prior to hatching) and number of embryos hatching were estimated from dail y maps of clutches laid on artificial surfaces (PVC tiles) defended by nest ing males. Patterns of variation in eggs laid, embryo mortality and numbers of embryos hatched were examined at three spatial scales: (1) among widely -spaced locations around the island; (2) between adjacent reef slope and pa tch reef habitats occupied by P. amboinensis at a single location; and (3) among different males within these two habitats. The embryo mortality was e xtremely high, with a mean of 25.9 +/- 6.2% (S.E.) for 4 locations examined in 1994 and a mean of 69.2 +/- 2.9% for two habitats surveyed in 1995. The re were no significant differences in embryo mortality among locations or h abitats in either year. This meant that spatial patterns in the number of e mbryos hatching reflected differences in the number of eggs laid on tiles. Embryo mortality was extremely variable on the scale of individual territor ies, with embryo mortality commonly ranging from 0% to 100%. Much of the mo rtality could be attributed to diurnal predatory fishes, especially the wra sse Thalassoma lunare. However, variation in predator densities did not exp lain spatial patterns in embryo mortality rates. Both solitary and group pr edatory behaviour was observed, with groups often causing 100% embryo morta lity. The level of embryo mortality observed suggests that predation prior to hatching may have a substantial effect on the reproductive output of pop ulations of this demersal-nesting fish.