NOTE Franks et al.: Stomach contents of juvenile Rachycentron canadum 



379 



and were demonstrably important in the overall diet, 

 exhibiting the highest IRI (1,636.8) and %IRI (49.5) 

 of all food items. We observed that anchovies occurred 

 in 36% of group-1 juveniles, whereas only 14% of the 

 stomachs of group-2 juveniles contained anchovies. 

 Because anchovies were rarely encountered (0. 39(F) in 

 the diet of northern Gulf adult cobia (Meyer and Franks, 

 1996), our findings may suggest a possible ontogenetic 

 shift in feeding between juveniles^ and adults. 



Trachypenaeus similis was a key component of the 

 juvenile cobia diet on the basis of numeric and volu- 

 metric percentages, and our data show that shrimp, 

 including shrimp remains, occurred in more than 

 419? of the stomachs, exceeded only by fish in per- 

 cent frequency occurrence of prey. We found that crus- 

 taceans, primarily shrimp, occurred at a significantly 

 higher frequency in the stomachs of smaller juve- 

 niles (84%) than in larger juveniles (36%). In con- 

 trast, Meyer and Franks ( 1996) reported that crus- 

 taceans, primarily portunid crabs, were the primary 

 foods of adult cobia. based on percent occurrence, and 

 essentially dominated the adult diet. 



Loligo plei was one of the most important identi- 

 fied prey consumed by juvenile cobia. This species of 

 squid is abundant in the northern Gulf of Mexico, 

 particularly in shallower Gulf waters during spring 



and fall of the year (Hixon, 1980). Knapp (1951) re- 

 ported the percent occurrence for Loligo in the diet 

 of cobia as 17% . We report a similar percentage ( 18%) 

 for Loligo and a percent occurrence of 21% for all 

 loliginids combined. Meyer and Franks (1996) re- 

 ported the percent occurrence of loliginids as -12%. 



Evaluation of diel feeding periodicity based on 

 stomach fullness showed that juvenile cobia feed 

 primarily during daylight. Although fullness peaked 

 during the day, notably during late afternoon through 

 dusk, there was no significant difference in fullness 

 between daytime collection time intervals, including 

 those containing dawn, noon, or dusk. Diana (1979) 

 remarked that when fish with diel feeding cycles are 

 collected at different intervals of time, some stom- 

 achs should be empty. We found that 20% of the stom- 

 achs we examined were empty. 



Field observations in this study did not include 

 reference to regurgitation of prey (i.e. everted stom- 

 achs or partially digested food in the mouth). Bow- 

 man ( 1986) stated that undetected regurgitation rela- 

 tive to fish food studies could produce severe under- 

 estimation of prey consumption, and he 2 reported 

 that regurgitation is frequently observed in fishes 

 trawled from depths exceeding 100 m. In our study 

 only two cobia were collected at depths >40 m (i.e. 

 56 m and 66 m), and the stomach of each contained 

 food. Incidentally, the juvenile from the 66-m site 

 contained the second highest volume of food mea- 

 sured from all juvenile stomachs. Most cobia in our 

 study (94%) were trawled from waters <30 m in 

 depth. Even though somewhat limited by the small 

 sample size of stomachs with prey, we were able to 

 observe that increasing size (length) in juveniles 

 appeared to influence an apparent ontogenetic shift 

 from crustacean to piscine prey. Diet composition of 

 juvenile cobia showed no marked changes in prey 

 taxa between seasons or geographic locales. 



Cobia are strong swimmers, quick enough to catch 

 fish, squid, and decapod crustaceans. Personal ob- 

 servations, made by the senior author of cobia in 

 culture systems and in natural habitats, reveal that 

 R. canadum is a very aggressive feeder and quite 

 capable of catching motile prey. Interestingly, prey 

 species in this study demonstrate considerable di- 

 versity in the foraging behavior of juvenile cobia. An- 

 chovies and squid are important components of the 

 open-water nekton. Portunid crabs are both nektonic 

 and benthic organisms. The shrimps and the 

 Brachyuran crab, though capable of swimming, arc 



2 Bowman, R. E. 1981. Examination of known and potential 

 causes of variation in fish feeding studies. Northeast Fisher- 

 ies Science Center, Natl. Mar. Fish. Serv., NOAA, 166 Water St., 

 Woods Hole. MA 02543. Woods Hole Lab. Ref. Doc 



