first with the 16 Phthalate ester standards. The identified 

 sample Phthalate ester was further confirmed by matching with its 

 known standard mass spectrum. Because the extraction procedure 

 was designed for analyzing PCBs, the presence of Phthalate esters 

 in the mussel extracts could only be considered as a qualitative 

 study. Any quantitative determination of this trace organic 

 compound would require the use of a specific extraction 

 procedure. 



Histopatholoaical Studies . To aid fast penetration of the 

 neutral- buffered formalin into the soft tissues of the mussels, 

 one valve of each animal was cracked with a sharp blow using the 

 handle of a knife, and the shell liquor drained. The mussel was 

 immersed in the fixative so that air was not trapped inside the 

 mantle cavity. To standardize the section, cross sections of the 

 mussels were cut just anterior to the foot. The sections were 

 further processed and stained with hematoxylin and eosin using 

 the standard histological procedure by the Histology Laboratory 

 of the Department of Pathobiology, University of Connecticut, 

 Storrs, Connecticut. 



Finished histological preparations were examined with an 

 Olympus VANOX microscope at magnifications of 4X, 10X and 40X. 

 Each specimen was critically scrutinized for stages of gonadal 

 development, staining characteristics of the Leydig tissue, 

 tissue integrity of the gill, kidney tubules, and intestinal 

 epithelium, as well as the degree of leucocytic infiltrations. 

 In addition, the prevalence of parasitic infections by 

 trematodes, pea crabs, and sporozoans were also recorded. 



Statistical Analyses of the Data . Prior to conducting any 

 statistical procedure, the data set of trace metal and PCB 

 concentrations was tested for normality using the procedure of 

 Shapiro and Wilk (1965) . If necessary, data sets were normalized 

 by transformations (log [x] , In [x] or ![x]). Statistical 

 analyses were performed using an IBM 3081 computer and the 

 software for analysis of variance (ANOVA) outlined in SAS User's 

 Guide: Statistics (SAS Inst. Inc., 1982). For the one-way 

 ANOVA, the data were classified by station, while for the two-way 

 ANOVA, the data were categorized by station (spatial) and 

 sampling period (temporal), i.e. before and during disposal. In 

 the present study, samples were collected before and during the 

 disposal period; post disposal samples were not collected because 

 sampling was not continued beyond June 1985. If the null 

 hypothesis was rejected, Tukey's multiple range test (Sokal and 

 Rohlf, 1969) was applied to discern which set(s) was different. 



The frequency of the occurrence of a given parameter, e.g., 

 gonadal development obtained from histopathological studies, was 

 tested using the G-statistics (Sokal and Rohlf, 1969) . This 

 procedure tests the null hypothesis that the stages of gonad 

 development or any other parameter, are independent of stations 



