wore floated from the sediments onto a O.") nun mesh sieve and the float and 

 residue preserved separately in 70% ethyl alcohol . Organisms were removed 

 under dissecting microscopes, sorted into major groups (annelids, 

 arthropods, molluscs, and others), identified to the lowest possible taxon 

 and counted. Organisms not sampled adequately by our methods because of 

 their small size e.g. nematodes, ostracods, copepods, and f oraminif era) 

 were not removed from the samples. Riomass of each of the major phyla was 

 estimated using two samples of five pooled replicates. Samples were dried 

 at 80°C for 24 h, weighed (to the nearest O.I mg) , combusted at 500°C for A 

 b, and reweighed ; ash-free biomass was then expressed as the difference 

 between dry-weight and combusted weight. Prior to drying, m.olluscs were 

 decalcified in O.IN HCL to eliminate carbonates. 



Sediment analyses were performed on a 3.5 cm (i.d.) x 5 cm deep cere 

 taken at each station. The dry sieving method and the method of moments 

 technique was used to calculate the arithmetic mean phi (Folk 1974) , which 

 was then converted to mean particle size in millimeters. Organic content 

 of the sediments was estimated, after combusting dried sediment at 500°C 

 for 24 h, as the difference between dry-weight and combusted weight. 



DATA ANALYSI S 



Since the number of replicates differed during the study, quarterly 

 values were converted to number /core (0.0078 sq. m) ; annual numbers of 

 species and individuals (by phylum) were expressed as percents. 



Species Diversity 



Seasonal species diversity for each station was estimated using the 

 Shannon information index (H'), calculated as: 



S n. n. 



H' = >: log 



i=l N - N ("ielou 1977) 



