standard deviation were calculated. The exception to this was for 
the PCB analyses. Both sections of the core were combined and only 
a single analysis was conducted. 
Statistical tests were performed to determine if the 
concentrations of contaminants were different at each of the five 
disposal mounds compared to the Reference station concentrations 
(Tables 3-5 and 3-6) and also to see if there were significant 
differences between the top core section (0-2 cm) and bottom 
section concentrations (Table 3-7). For PCBs, only a single 
analysis was conducted for each station, therefore the PCB 
concentrations between stations could not be compared 
statistically. Because total carbon is a more informative measure 
of increased organic matter input compared to total hydrogen or 
total nitrogen, only this value was compared statistically among 
stations. Cadmium and nickel were below the analytical detection 
limits in most samples and were not tested. In most cases, the 
concentrations of contaminants were significantly (p<0.05, Mann- 
Whitney U-test) higher at the five test stations (NL-85, NL-I, NL- 
II, NL-III and NL-RELIC) than at the Reference station. This is 
true for both the core tops and bottoms (Tables 3-5 and 3-6). 
Zinc, chromium, and copper concentrations were elevated in the 
bottom sections of cores from each of the five disposal mounds 
compared to the Reference station concentrations. In the top core 
sections this was also true except that zinc was not higher in the 
NL-II core sections and chromium concentrations were not elevated 
in the NL-III sections. In the top core sections from the NL-RELIC 
station, all parameters except arsenic, total carbon, and oil and 
grease were significantly elevated and all but arsenic were 
elevated in the bottom sections of these cores. For the remainder 
of the parameters, no consistent trends in the data could be 
identified. Comparison of the top 2 cm core sections with the 2- 
10 cm sections did not reveal any consistent trend in elevated 
levels of contaminant. In only six cases (out of 42, Table 3-7) 
did a chemical concentration differ significantly between the top 
and bottom core sections. 
37215 Benthic Community Analysis 
The sieve residue from the samples collected at the 
Reference station in November 1985 contained thousands of 
ampeliscid amphipod tubes and little or no sand and gravel. The 
residue from the sample collected at the center of the NL-85 mound 
contained up to 1 liter of sand, gravel, and large shells. Five 
species were in common among the top ten abundant species at both 
stations: the amphipods Ampelisca and Unciola and the polychaetes 
Mediomastus, Owenia and Tharyx (Table 3-8). Taxa were also 
identified as being exclusively found at one station or the other 
(Table 3-9). The effects of grain size, feeding competition, 
depth, and prey availability most likely all played a role in this 
segregation. Both suspension feeders and deposit feeders occurred 
28 
