MINELLO ET AL.: PREDATION ON BROWN SHRIMP 



percentage of body weight eaten by the fish over 

 the experimental period, ranged from 6.1% in 

 clear/sand tanks to 24.8% in turbid/no sand tanks. 

 The feeding behavior of southern flounder (84-94 

 mm TL) on brown shrimp was also observed in 

 aquaria. These fish exhibited a variety of feeding 

 behaviors including active searching for prey on the 

 bottom and in the water column as described by 011a 

 et al. (1972) for summer flounder, Paralichthys den- 

 tatus. Generally, however, the fish remained motion- 

 less on the bottom and waited for potential prey to 

 come within striking distance before attacking. Fish 



Table 2.— Predation on brown shrimp by southern flounder. A) 

 Number of shrimp eaten per fish over the 1 2-h experimental period 

 for treatment combinations of turbidity and substratum. B) ANOVA 

 results using the number of shrimp eaten per fish as the obser- 

 vation. 



Turbid 



Clear 



in the family Bothidae have been classified as 

 primarily visual feeders by de Groot (1971). In our 

 observations, all stalking activity by southern 

 flounder was accompanied by active eye movements, 

 tracking potential prey, which suggested the 

 primary use of vision in prey detection. A study of 

 diel feeding periodicity, similar to that conducted 

 on red drum and Atlantic croaker by Minello and 

 Zimmerman (1983), however, indicated that south- 

 ern flounder could also feed at night even when 

 tanks were enclosed in black plastic to completely 

 eliminate light (unpubl. data). This finding suggests 

 that sensory mechanisms, in addition to vision, can 

 be used by these fish to detect prey. 



Pinfish 



In both experiments with pinfish, the largest num- 

 ber of shrimp were eaten in tanks with clear water 

 and without sand (Fig. 2). The ANOVA on the num- 

 ber of shrimp eaten in the first experiment (pinfish 

 I) indicated a significant interaction between turbid- 

 ity and substratum (Table 3). The substratum ap- 

 parently did not affect predation in tanks with turbid 

 water, but in clear water the presence of sand sig- 

 nificantly reduced predation rates (Fig. 2A). In a 

 similar manner, turbidity did not significantly affect 

 predation in tanks with sand substrata, but it did 

 reduce predation rates in tanks without sand. An 

 ANCOVA with incident light and an ANOVA using 

 the weight of shrimp eaten (Table 3) did not alter 



X 



iZ 



I- 

 < 



lU 

 Q. 



S 



cc 

 z 

 tn 



u. 

 o 



oc 



UJ 

 03 



S 

 Z 



2 - 



\ NO SAND 

 \ 

 \ 

 \ 

 \ 

 \ 

 \ 

 \ 

 \ 

 \ 



SAND 



\ 



^4 



V 



A. PINFISH I 



CLEAR 



TURBID 



CLEAR 



TURBID 



Figure 2.— Mean predation rates on brown shrimp by pinfish in treatment combinations of turbidity 

 and substratum. Vertical Hnes, representing one half of Tukey's oj (Steel and Torrie 1960) on either side 

 of the mean, can be used to compare means at the 0.05 significance level. 



63 



