type associated with different acoustic reflectance patterns. 

 Visual observations were recorded for each sample describing the 

 physical and biological conditions and the apparent redox layer 

 thickness. Three polycarbonate plastic core liners (6.5 cm. ID) 

 were pushed into each sediment grab sample to a minimum depth of 10 

 cm and extracted; the three cores were combined into a polyethylene 

 bag and refrigerated for subsequent grain size analysis at the NED 

 laboratory. 



2.6 Benthic Community Sampling and Analysis 



Quantitative benthic samples were obtained with a 0.1 m 2 

 Smith-Mclntyre grab at 12 "transect" stations in and around the 

 historic Brenton Reef disposal site (Figure 2-4) . Mr. Sheldon 

 Pratt from URI participated in the field sampling and supervised 

 the preservation, sorting, and identification of organisms. 



One grab per station was placed in a five gallon bucket, 

 labelled, and brought to shore without treatment. In the 

 laboratory, organisms were relaxed by mixing 8% MgS0 4 into sandy 

 samples and holding silty samples overnight in stagnant conditions. 

 The material was sieved to 0.5 mm with filtered seawater and the 

 residue preserved in 10% formaldehyde with 0.2% rose bengal stain. 

 After fixation, samples were separated into a number of fractions 

 by sieving with 2.0, 1.0, and 0.5 mm screens both by successive 

 decantation in tap water and suspension in "Ludox" colloidal silica 

 solution (density 1.3 gm/ml) . Material retained on the 2.0 mm 

 screen was sorted in glass trays without magnification; all other 

 fractions were sorted with binocular microscopes at 7.5x and 6.6x. 



Samples from three stations (stations 3, 11, and 12) had 

 large numbers of individuals along with fine amphipod tube detritus 

 of similar density, requiring an excessive amount of time to sort 

 (50+ hours; approximate sorting time ranged between 4-6 hours for 

 an average sample) . Consequently, it was necessary to split these 

 samples. Nested sieves were used to remove organisms by "Ludox" 

 flotation prior to splitting the fine sieve fraction. This greatly 

 reduced the number of individuals subject to splitting error and 

 improved the accuracy of the final counts. For example, the ratios 

 of Aj_ agassizi and of all individuals in splits from sample 12 were 

 0.73:1 and 0.99:1, respectively. The numbers of A^ agassizi found 

 in a 1/4 split of the fine detritus fraction of this sample was 13. 

 The actual number counted in the coarser sieve fractions which were 

 not split in the sample was 2,685; the calculated total of 2,737 

 could not be more than a few percent different than the true count. 



