FISHERY BULLETIN: VOL. 78, NO. 4 



% SIMILARITY 

 25 50 75 



100 



SUMMER 



SPR^G- 

 SUMMER 



-WIMER 



— Anchoviella perfasciata 

 -Sardinella anchovia 

 -Mugil curema 

 — Membras martinica 

 — Anchoa mitchiili 

 — Eucinostomus sfi. 

 r—Anchoa lyolepis 

 *— Harenguia jaguana 

 — Trachinotus carolinus 

 — Anchoa hepsetus 

 — Menticirrhus littoralis 

 — Brevoortia patron us 

 — Lagodon rhomboides 

 — Leiostomus xanthurus 

 — Mugil cephalus 



Figure 3. — Similarity dendrogram for species by month, based 

 on presence or absence data, for fishes collected from the surf 

 zone of Horn Island, Miss. 



The third seasonal species group, B. patronus; 

 pinfish, Lagodon rhomboides; spot, Leiostomus 

 xanthurus; and striped mullet, Mugil cephalus, 

 was prevalent during winter or early spring, gen- 

 erally in water temperatures below 24.5° C, with 

 the first three occurring at temperatures as low as 

 11.5° C. Together these four species composed the 

 secondary abundance peak of March 1976 and 

 1977 (cf. Figure 2). 



Rough silverside, Membras martinica, and bay 

 anchovy, Anchoa mitchiili, did not fit within the 

 three seasonal categories. Anchoa mitchiili was 

 collected in greatest abundance during the spring 

 and in the fall, while M. martinica was most com- 

 mon in the spring, but only during 1975 and 1977. 

 Membras martinica was infrequent in 1976. 



Multiple regression analysis explained little of 

 the variation associated with fish abundance (i.e., 

 fish per seine haul). However, while tentative, the 

 analysis may indicate the relative importance 



of these variables in controlling fish occurrence 

 (Table 4). The dominant factor affecting the clu- 

 peiform fishes was tide level. Tide contributed 

 only 5.1% and 8.9% in the regression equations 

 for H. jaguana and A. hepsetus, and accounted for 

 19.2% of the model for A. lyolepis (P<0.05). The 

 remaining variables contributed little to the pre- 

 dictive ability of the regression equations, al- 

 though salinity composed 5.6% of the variance 

 model for A. hepsetus (P<0.05). Temperature was 

 the dominant parameter in the model for T. 

 carolinus (not significant) and Menticirrhus lit- 

 toralis (P<0.05). 



While not apparent from Table 4, our observa- 

 tions indicate that catch per effort (cf. Figure 2) 

 may coincide with wind direction. Wind patterns 

 in the study area undergo annual cycles in which 

 direction is primarily from the north during the 

 winter and from the south during the summer. 

 Fishes were collected in greatest numbers during 

 the summer when southerly winds predominated. 



Daily Activity Patterns 



The greatest number of fishes were present 

 within the surf zone during the early morning 



Table 4. — Stepwise multiple regression of the five major 

 environmental parameters contributing to the average number 

 offish collected per seine haul in the surf zone of Horn Island, 

 Miss. 



•Significant (P<0,05), 



916 



