WENNER: SPONGE-CORAL HABITAT FISHES 





species was encountered in more night (53%) than 

 day tows (12%), the frequency of occurrence was not 

 significantly different (x 2 = 0.18, df= 108). Based on 

 the percent of fish with stomach contents, Ross 

 (1977) found that peak feeding activity for eight 

 species of searobins (P. carolinus not included) was in 

 the morning. He indicated, however, that these 

 triglids had some food items in the stomach at all 

 times. Bigelow and Schroeder (1953) noted that P. 

 carolinus burrows with only the eyes and head show- 

 ing, and Bardach and Case (1965) confirmed that this 

 species buries itself in the substrate. The available 

 evidence, then, indicates that P. carolinus is a noctur- 

 nally active species that either buries itself in the 

 sandy portions of the sponge-coral habitat during the 

 day or makes nocturnal feeding forays into this 

 habitat from the surrounding sandy areas where it 

 resides during the day. Conclusive evidence can only 

 be obtained from analysis of its food habits as deter- 

 mined from specimens collected in the sponge- 

 coral habitat. 



Haemulon aurolineatum were taken in significantly 

 more night tows because dense day-resting schools 

 dissociate at night when this species disperses to 

 feed. It is well documented that grunts form resting 

 schools on various reefs during the day (Starck and 

 Davis 1966; Collette and Talbot 1972; Hobson 1973; 

 Ebeling and Bray 1976; Ogden and Ehrlich 1977). At 

 dusk, members of the school become active and un- 

 dertake feeding migrations into areas near the reef 

 which may have sandy bottom or algal cover. The dis- 

 persal of H. aurolineatum from dense daytime 

 schools is reflected in the present study by changes in 

 Morisita's index of dispersion (Elliott 1977) for day 

 and night trawl tows. The index equals 1 for a random 

 distribution, > 1 for a contagious distribution, and 

 < 1 for a regular distribution (Elliott 1977). Day trawl 

 tows had an index of dispersion of 5.51 which in- 

 dicates a highly significant contagious distribution of 

 H. aurolineatum. Night collections also had a signifi- 

 cant contagious distribution, but the value of the in- 

 dex was much lower (1.44). By comparing both 

 Morisita's index and the frequency of occurrence of 

 H. aurolineatum in day and night tows, it can be in- 

 ferred that tomtate disperse over the habitat during 

 the night hours. In addition, the occurrence of tom- 

 tate in three of six night tows in the open-shelf study 

 area provided evidence that the species leaves the 

 sponge-coral habitat at night. 



Although Stenotomus aculeatus occurred more 

 frequently in night tows (79%) than in day tows 

 (39%), the results were not significantly different. 



Southern porgy are daytime feeders (Harris 1979) 

 which, as indicated by trawl results, form relatively 



large schools. The index of dispersion showed signifi- 

 cant contagion in both day (I s =4.14) and night tows 

 (I s = 2.43); however, the lower night values indicate 

 that the cohesiveness of the schools decreases at 

 night. As a result of the greater nocturnal dispersal of 

 this species throughout the sponge-coral habitat, the 

 frequency of its occurrence in trawl tows increases. 

 This accounts for the more consistent but lower catch 

 rate at night, in comparison with day catches. 



Stephanolepis hispidus showed the same pattern as 

 Stenotomus aculeatus in its distribution and abun- 

 dance in sponge-coral habitat tows. Planehead 

 filefish were more aggregated during the day (I s = 

 3.99) than at night (I s = 1.96) and had a slightly 

 higher frequency of occurrence in night trawl tows 

 (97%) than in day collections, although the difference 

 was statistically insignificant. 



In addition to the species discussed above, several 

 other less abundant fishes showed increased avail- 

 ability to night trawl tows in the sponge- coral habitat. 

 The gadid Urophycis earlli was found only in night 

 collections. Parker et al. (1979) reported U. earlli to 

 be a secretive species that hid in and under artificial 

 reef structures off the South Carolina coast. Daytime 

 observations in winter revealed hundreds of Carolina 

 hake under reef material (see Parker et al. 1979, fig. 

 10). Data from historical surveys indicate that U. 

 earlli is a nocturnal species since almost all speci- 

 mens collected by traps, handlines, and trawls were 

 taken at night in the sponge- coral habitat or the 

 rugged areas of the shelf-break (Wenner, unpubl. 

 data). During the day, U. earlli probably hides under 

 small ledges where it is unavailable to the trawl 

 gear. 



Carapus bermudensis is a nocturnal species that was 

 unavailable to day otter trawl tows. This species 

 spends its day residing in the respiratory tree of 

 several species of holothurians and emerges at night 

 to feed on near-bottom plankton (Starck and Davis 

 1966; Smith et al. 1981). Most bothids were encoun- 

 tered in night tows, probably due to these species' in- 

 creased vulnerability at night. 



In summary, day otter trawl catches were more vari- 

 able in numerical abundance but frequently weighed 

 more than night catches. This was due to the dense 

 schools formed during daylight by species such as 

 Haemulon aurolineatum and Stenotomus aculeatus. 

 Night tows had more species because of the noctur- 

 nal availability of such species as Apogon pseudo- 

 maculatus, Carapus bermudensis, the ophidioids, 

 etc., as well as the increased vulnerability of diurnal 

 species such as Holacanthus isabilita and Chromis 

 enchrysurus. 



549 



