Zooplankton samples were collected during daylight at 5 stations, spaced 18 km apart between the lagoon and the Coral Sea in the central region of the Great Barrier Reef. Stations 1 (39 m depth) and 2 (60 m depth) were located on the mid-shelf, while stations 3 (64 m depth) and 4 (75 m depth) were located on the outer-shelf and station 5 (300 m depth) was located in the Coral Sea.Stations were sampled at total of 5 times at two-weekly intervals between 9 January and 21 March 1983 using a 1 m diameter plankton net of 236 µm mesh. At each station, three replicate, stepped-oblique tows were made from 10 m depth to the surface and from approximately 1 m above the bottom to 10 m above the bottom. At Station 5, the bottom was taken as 120 m. Each tow was of 5 min duration and tow speed was approximately 1 m/s.Larval fish samples were collected during daylight from Station 5, Station 4, Station 2 and a nearshore station, 1 km from Pandora reef (15 m depth), on 5 trips at two-weekly intervals between 24 January and 31 March 1983. Fish were collected with a 2m Tucker trawl. At each station, three replicate stepped-oblique tows were made from approximately 1 m above the bottom to the surface. Each tow was of 30 min duration and tow speed was about 1 m/s.Zooplankton samples were split using a Folsom splitter and half the sample was oven-dried at 60°C for 24 hours to calculate total dry weight. The other half was further split as far as 1/64 to produce a sub-sample of about 1000 to 2000 copepods. The total count of copepods in the sample was used to determine copepod density. Relative abundances of species in each sample were estimated from complete counts of the first 200 copepods in each sub-sample. A total of 1000 individuals in the sub-sample were also examined to identify rare species. If a species was found in the first 1000, but not the first 200 individuals, it was scored with a nominal density of 0.1/m³.Fish larvae were sorted from the other plankton collected in the Tucker trawl. To counter size-selection of larvae by the different fore and aft meshes of the net, the extracted larvae were then gently washed on a 2 mm sieve. Larvae retained on the sieve were identified to family and counted. Taxa were loosely grouped into "reef' and "non-reef' forms, based on predominant distributions of adults.