FISHERY BULLETIN: VOL. 82. NO. 1 



Santa Monica 

 Bay 



Redondo Beach 



Los 

 Angel es 



100 500 



I 1 1 1 1 1 I 



meters 



FIGURE 1.— Study area at King Harbor, Redondo Beach, Calif. 



water immediately surrounding the plume), or at the 

 base (the substrate surrounding the discharge). 



The abundance of fishes at the Point has been 

 documented since 1974 (Stephens and Zerba 1981), 

 and work continued at this area during the same time 

 period the discharge was examined. Two divers 

 equipped with slates and depth gauges swam in one 

 direction along the rock breakwater at a fixed depth 

 for 5 min, counting all fish seen 1.5 m above and 

 below them and within sight to either side. Transects 

 were run at depths of 1.5, 4.5, 7.5, and 10.5 m, with 

 replicates at each depth. 



In order to determine the nature of the feeding 

 habits of blacksmith at the discharge versus those 

 feeding at the Point, utilization of food items based 

 on stomach analysis was examined for each area. 

 General availability of food was estimated by sam- 

 pling plankton at both sites. 



Stomach analysis closely followed methods em- 

 ployed by Ellison et al. (1979). Fish were collected 

 from each study site by scuba divers using pole 

 spears. During fish collection, a temperature profile 

 was taken using a temperature probe coupled to a 

 telethermometer (Yellow Springs Instruments Co., 

 Model 43 ID 5 ). After capture the fish were placed on 



ice. The body wall was cut open and the stomach 

 injected with a 207c Formalin solution. The fish were 

 then preserved in a 1 0% Formalin solution for at least 

 48 h, rinsed in running water for 2 h, and placed in 

 70% isopropyl alcohol. 



Within 2 wk from date of capture, fish stomachs 

 were removed and placed in vials of 70% isopropyl 

 alcohol. At this time the standard length, wet weight, 

 and sex of each fish were noted. Each stomach was 

 then blotted dry (with special care taken to remove 

 the internal fluid) and weighed, food items dissected 

 out, and the empty stomach weighed again. Stomach 

 fullness was estimated using a scale from (empty) to 

 5 (full). 



Individual prey items were separated into the 

 lowest identifiable taxa and counted, and the percent 

 of the total volume estimated. In most cases, only 

 whole organisms or whole organism indicators were 

 counted. In prey items which were not eaten whole 

 (i.e., algae and ectoprocts), only the percent volume 

 was estimated. 



In 1979-80, 73 fish were collected at the discharge 

 area from 13 sampling days during a 15-mo period. 

 Four sampling days were in the afternoon (1430- 

 1830 h) and 10 were in the morning (0830-1100 h). A 

 total of 35 blacksmith were collected from the Point 

 area before noon (1000-1130 h). 



During the study period, 28 plankton samples from 

 the discharge plume and 13 plankton samples from 

 the Point were collected. The mean rank order abun- 

 dance of prey items from each site was determined 

 for comparison with blacksmith stomach contents. 



Observations comparing different prey items from 

 two locations were tested using contingency table 

 analysis, the G-test (Crow 1982), and Kendall's coef- 

 ficient of rank correlation. When only one variable 

 (fish weight, stomach fullness etc.) was tested be- 

 tween two locations, a two-sample f-test was used, 

 assuming separate variances. Values of the Index of 

 Relative Importance (IRI) were calculated for con- 

 sumed prey from the sum of the percent number and 

 the percent volume, multiplied by the frequency of 

 occurrence (Foe) (Pinkas et al. 1971). 



Dietary overlap between blaeksmth from the Point 

 and discharge was examined using the formula of 

 Schoener (1970): 



a 



1-0.5 ( Z \Px r Py, 



'Reference to trade names does not imply endoresement by the 

 National Marine Fisheries Service, NOAA. 



where n is the number of food categories, x, is the 

 average percentage of estimated volume that food 

 category ;' contributed to species at location x, andy, 

 is the average percentage of estimated volume that 

 food category i contributed to species at location y. 



200 



