Spatial distribution of developmental egg ages within a herring Clupea harengus spawning ground

Herring Clupea harengus are demersal spawners that lay their eggs in contin uous, multilayer mats on coarse sediment. Although the reproductive biology of herring is extensively studied, little is known about the processes of egg deposition and development within a spawning ground. Here, the spatial distribution of developmental egg ages from a spawning ground (Ballantrae B ank, west of Scotland) is modelled using regression trees and generalised a dditive models. We refer to developmental, rather than absolute, ages to in dicate that ageing the eggs according to the apparent state of development in the samples may underestimate their true age. Egg development was found to be related to the position within the egg mat (horizontal and vertical), the sampling day and the egg density in the samples. Eggs near hatching (1 2 to 14 d old) were mainly deposited at the eastern and northeastern bounds of the mat, where eggs in the top layer exceeded the developmental age of eggs in the bottom layer. Newly laid eggs (1 to 2 d old) were found in the western parts of the ground, where substrate type was less suitable and egg densities were lowest. The results suggest that spawning progressed in a w esterly direction over a period of at least 2 wk, constrained by the availa bility of suitably coarse sediment. The presence of eggs in the bottom laye r with lower developmental age than in the top demonstrates that multiple l ayering causes developmental retardation nearest to the substrate. Developm ental retardation can therefore be seen as a contributing factor in the rep orted variability in hatching lengths of herring larvae.