FISHERY BULLETIN: VOL. 78, NO. 4 



MISSISSIPPI SOUND 



GULF of MEXICO 



10 KM 



Figure l.— Map of Horn Island, Jackson County, Miss., show- 

 ing the four sampling areas. The southern side of the island 

 represented the windward shore. 



bars which extend the length of the island. The 

 surf zone habitat is characterized by a sand sub- 

 strate, the absence of any rooted vegetation, and 

 sufficient wave activity to be categorized as a high 

 energy beach (Odum and Copeland 1974). 



We began sampling in April 1975 along the 

 southwestern edge of the island (Station 1), and 

 collections were made at about 7-wk intervals 

 until November 1975. From May 1976 to November 

 1977 we sampled four stations along the windward 

 shore of Horn Island (Figure 1) at about 5-wk 

 intervals (Table 1). We also sampled sheltered 



Table l. — Sampling dates for fish taken from the surf zone 

 habitat on the southern shore of Horn Island, Miss., between 

 April 1975 and November 1977. Each collection represents a set 

 of seine hauls at a specific location. 



beach areas adjacent to Stations 3 and 4 during the 

 summer of 1976. All of the above collections were 

 taken between 0900 and 1600 h c.s.t. (central 

 standard time). 



Every month between March and September 

 1976 (excluding August) we sampled either Sta- 

 tion 1, 3, or 4 over a 24-h period, taking samples at 

 about 4-h intervals. The choice of station was 

 based in part on the availability of a safe an- 

 chorage for our boat. In order to compare data 

 throughout the study, collections made between 

 1600 and 0900 h were not included in seasonal or 

 annual comparisons. 



Fishes were collected with a 3.2 mm Ace"* mesh 

 bag seine measuring 9.1 x 1.8 m. Hauls were made 

 perpendicular to the beach face beginning 16-18 m 

 offshore. The area sampled extended from the 

 swash zone to the midlongshore trough, and we 

 made an effort to take regular samples only in 

 areas directly exposed to surf. We continued sein- 

 ing at each location until no additional new 

 species were collected; usually 5-9 hauls sufficed. 

 Each collection at each location was thus com- 

 prised of a successive number of seine hauls. 

 Fishes collected from all seine hauls at a single 

 station were pooled for analysis. Catch-per-effort 

 data from all stations were pooled to provide 

 monthly means. The study included 613 seine 

 hauls. 



Species similarity by months was analyzed by 

 the unpaired group arithmetic average clustering 

 (UPGMA) method (Sneath and Sokal 1973). Only 

 the 15 most abundant species, which were col- 

 lected in at least 15% of the locations sampled, 

 were analyzed. Pair similarity based on species 

 presence or absence (Odum 1971) was determined 

 by: 



S =2C/A +B 



where C = number of species common to samples 

 a and b, 

 A = number of species in sample a, 

 B - number of species in sample b. 



We used stepwise multiple regression to define 

 the dominant factors associated with the abun- 

 dance (fish per seine haul) of the five most fre- 

 quently occurring species. Environmental para- 

 meters selected as independent variables were 



■'Reference to trade ngimes does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



912 



