<p>Physical observations in the Beatrix Basin, Marlborough Sounds, New Zealand, revealed four distinct regions: a partially exposed northern region with moderate water currents, low rainfall, small catchment area and scant freshwater inflow; a central highly exposed region with strong currents, moderate rainfall, no land catchment area and no direct freshwater inflow; an eastern region with a high degree of exposure and strong water movement, moderate rainfall, extensive catchment and seasonally high freshwater inflow; a southern sheltered region with sluggish water movement, comparatively high rainfall, extensive catchment and the highest rainfall in the study area. Differences between the four regions in monthly mean temperature and salinity were slight and there was an indication of more mixing in the central region than elsewhere. No thermocline was observed and surface water temperature demonstrated a diurnal rhythm in summer. Mean winter temperature fell to 10 degrees C and rose to 17 degrees C in summer. Mean salinity ranged from 32.8 degrees C in winter to 34.8 degrees C in summer and no halocline was present. Salinity fell occasionally after heavy rain, but tidal currents soon dissipated the freshwater except in the southern region where it persisted for a short time. The reproductive cycles of Perna canaliculus and Mytilus edulis aoteanus, the occurrence of their larvae in plankton samples, their season of settlement and peaks in settlement activity, were monitored. Physical environmental conditions preceeding or accompanying spatfall were recorded. A broad temporal relationship was found between spawning adult mussels, pelagic larvae, environmental conditions, and settlement of Perna canaliculus in Elie Bay. It appears that P. canaliculus spawned in early summer and in autumn when the water temperature was 18 plus-minus 0.5 degrees C. However, it was not possible to predict spatfall since the duration of the pelagic phase was not known and settlement is a function of larval survival and dispersal, as well as the presence of a suitable substrate. Accurate spatfall prediction is required to facilitate timed immersion of spat-collecting ropes while ensuring "seasoning" and avoiding fouling. As it was not possible to predict settlement by relating it to other biological or physical phenomena, this study sought an alternative spat-collecting material not so susceptible to fouling but still attractive to mussels. Carbon black fibrillated polypropylene film proved to be such an alternative. The determination of available food for mussels within the study area was attempted using particulate organic nitrogen as the indicator. The problems encountered are discussed. The concentration of particulate organic nitrogen was 11.78 microgram at N/litre. The larvae of Mytilus edulis aoteanus and Aulacomya maoriana were reared in the laboratory to the late veliger stage. The most likely causes of their failure to metamorphose were a too high larval density, an excess of food organisms and the lack of a suitable substrate for settlement. It was not possible to induce spawning of Perna canaliculus under controlled conditions.</p>
Autecology of Shallow-Water Octocorallia from Mediterranean Rocky Substrata, I. The Banyuls Area