FISHERY BULLETIN: VOL. 85. NO. 1 



well documented (^^''ickham and Minkler 1975; 

 Lakshmi et al. 1976; Minello and Zimmerman 1983), 

 and this species generally remains beneath the sur- 

 face of the substratum throughout the daylight 

 hours, emerging to forage at night. 



The objective of this search vi^as to determine 

 whether turbid water and a suitable substratum for 

 burrowing affect predation rates on juvenile brown 

 shrimp. Experiments were conducted in the labora- 

 tory, and predatory fish were southern flounder, 

 Paralichthys lethostigma Jordan and Gilbert, pin- 

 fish, Lagodon rhomboides (Linnaeus), and Atlantic 

 croaker, Micropogonias undulatus (Linnaeus). The 

 effect of turbidity on burrowing by brown shrimp 

 was also examined. 



METHODS AND MATERIALS 

 Predation Experiments 



Collection and Handling of 

 Experimental Animals 



Fish were collected with trawls and seines from 

 Galveston Bay, TX, and held in clear-water tanks 

 without a sand substratum. They were fed live 

 shrimp daily and starved for 24 h before an experi- 

 ment. Total lengths of fish were measured after each 

 experiment, and specimens from a subsample in 

 holding tanks were weighed and measured. A 

 length-weight relationship was calculated and used 

 to estimate weights of experimental fish. 



Shrimp were collected by trawling 2 to 3 d before 

 each experiment. They were fed daily with pelleted 

 shrimp food but not fed during experiments. 

 Measurements of total length (tip of rostrum to tip 

 of telson) were made on all shrimp placed into 

 experimental tanks and all shrimp removed 

 after an experiment. A length-weight relationship 

 was calculated for each experiment from sub- 



samples of shrimp and used to estimate individual 

 weights. 



Experimental Tanks 



Experiments were conducted in fiberglass tanks 

 (1.75 m X 5.8 m X 0.5 m) located in a building with 

 a white translucent roof which allowed the use of 

 natural photoperiods. Each tank was divided in half 

 by a wall of 1.5 mm mesh fiberglass forming two 

 compartments (1.75 m x 2.9 m) of 5.07 m^ bottom 

 area. A 5 cm layer of washed beach sand (well sorted 

 with a graphic mean grain size of 2.95 "t*; analyzed 

 according to Folk 1980) was placed in four tanks. 

 In four other tanks, approximately 1 mm of sand 

 was used to reduce the contrast between prey and 

 the bottom of the tank. Tanks were filled to a depth 

 of 26 cm with seawater (24-26%o) pumped from the 

 beachfront off Galveston Island. During experi- 

 ments, water temperatures varied among tanks by 

 only 0.5°C, and diurnal ranges are listed in Table 1. 



Pulverized kaolinite was used to make the water 

 turbid in four tanks (two with sand bottoms and two 

 without sand). Particle size analysis (Folk 1980) in- 

 dicated that the kaolinite was poorly sorted with a 

 graphic mean grain size of 8.82 <t>. A clay slurry was 

 introduced into tanks through a 19 L settling bucket 

 with an outlet hose (5 mm ID) located 5 cm from 

 the bottom. This settling bucket served to remove 

 some of the heavier particles and flocculated aggre- 

 gates from the clay suspension. Each tank contained 

 a small submersible pump (252 L/minute capacity) 

 connected to a discharge pipe which extended along 

 the length of the tank and sprayed water over the 

 surface. This pump together with 12 airstones/tank 

 provided some vertical mixing which helped keep 

 clay particles suspended. 



Turbidity, light, and temperature were measured 

 at 2-h intervals during each experiment. Turbidity 



Table 1.— Design and conditions for predator-prey experiments. 



^Number of predators per compartment. 



^Number of replicate compartments used per treatment combination 



^Average initial and final turbidity in turbid tanks over experimental period. 



'Average light levels measured in clear tanks over the first 5 h of the experimental period (n 



16). 



60 



