FISHERY BULLETIN: VOL. 85, NO. 2 



son and Chess 1976; Helfman 1986), probably due 

 to visual limitations (Durbin 1979). The absence of 

 nocturnal prey and the strongly diurnal periodicity 

 of stomach fullness (Fig. 6) indicate that D. punc- 

 tatus feeds during the day. Round scad rarely con- 

 sumed mysids, tanaids, and cumaceans, which are 

 abundant in the water column at night but dwell in 

 the benthos during the day (Kaestner 1970; Hobson 

 and Chess 1976). The diel periodicity in empty 

 stomachs (Fig. 6) is not biased by sizes of fish differ- 

 ing among collection times (see Results). Individuals 

 of different sizes have similar gut evacuation rates 

 (Perrson 1981) but different gut capacities; there- 

 fore, small individuals empty their guts more quickly 

 than large individuals. The lack of a size difference 

 in this analysis substantiates the daily feeding period 

 by round scad. 



Scales were a common item in round scad 

 stomachs, but were deleted from analyses because 

 several observations indicated that round scad were 

 feeding on debris (including scales) generated by 

 trawling: 1) oral chambers of specimens often con- 

 tained scales; 2) scale size and type varied; and 3) 

 the extent of presumed lepidophagy was not corre- 

 lated with fish size. In addition, scales were seldom 

 found in latter portions of the gut. Most other lepi- 

 dophages generally prey on a small number of 

 species (Sazima 1983), or for only a portion of their 

 life history (Carr and Adams 1972). Thus, round scad 

 probably consume scales on occasion, but not to the 

 extent that the data would indicate. It seems likely 

 that round scad were feeding on scales abraded from 

 fishes during their avoidance of or capture by the 

 trawl. Yamaguchi (1953) also attributed the occur- 

 rence of scales in the stomachs of D. pinnulatus to 

 gear bias. 



Ontogenetic changes in the diet occurred with 

 growth. Larger individuals had more diverse diets 

 which included larger prey. Small fish fed primar- 

 ily on small, abundant copepods and copepodites. 

 Stomach contents of several (10) small juveniles 

 (13-26 mm FL) collected during an ichthyoplankton 

 survey in 1973 also contained mostly copepods (S. 

 Hales pers. obs.). Data from these specimens were 

 not included in previous analyses due to the possi- 

 bility of differential prey digestion. 



Seasonal changes in the diets of round scad prob- 

 ably reflect fish size and the relative abundance of 

 zooplankton components. Copepods and mollusks 

 were the most important prey in all seasons except 

 during spring when barnacle cyprids were the most 

 numerous and chaetognaths contributed the great- 

 est volume of prey. Zooplankton volumes and diver- 

 sity reach their peak in the spring (Deevey 1960; 



Reeve 1964), and cyprids are more abundant in the 

 spring than in any other season (Lang and Acken- 

 hausen-Johns 1981). The preponderance of copepods 

 in the diet of round scad in summer partially re- 

 flects the abundance of juvenile round scad, which 

 appear to feed primarily on small abundant cope- 

 pods. 



Molluscan veligers may be an exception to the 

 general pattern of prey selection. They have not 

 been reported to be abundant in the zooplankton of 

 the South Atlantic Bight (Paffenhofer 1980, 1981) 

 or elsewhere (Deevey 1960; Reeve 1964), yet oc- 

 curred frequently in the diets of round scad. Shells 

 of both gastropods and bivalves and the opercula of 

 gastropods were all that remained in the stomachs 

 of round scad on occasion; such parts are apparent- 

 ly digested slowly and retained in the stomach. How- 

 ever, such an explanation alone does not account for 

 the frequency and abundance of mollusk veligers. 

 Brewer and Kleppel (1986) have reported the para- 

 dox of low bivalve density yet frequent occurrence 

 in the guts of some larval fishes, and suggested that 

 bivalve veligers may occur in microscale patches just 

 above the bottom. Planktonic stages of gastropods 

 are important prey for horse mackerel {Trachurus 

 trachurus), which school near the bottom (Macer 

 1977). Thus, the frequency and abundance of mol- 

 lusk veligers in the stomachs of round scad may be 

 attributed to both the abundance of mollusk veli- 

 gers being greater than generally recognized and 

 their low digestibility and long retention in stom- 

 achs. 



Age and Growth 



Round scad grow rapidly to sexual maturity at 11 

 cm FL in 4-5 months, and apparently achieve a 

 major proportion of their total size in their first year. 

 Because of the problems encountered in age deter- 

 mination in this study, little can be said about the 

 age of most adults. The asymptote predicted by the 

 von Bertalanffy model (161 mm FL) is much shorter 

 than the maximum size observed in the South Atlan- 

 tic Bight. Thus, the growth rates observed in this 

 study should not be extrapolated to older and larger 

 fish. Houde et al. (1983) reported the mean size of 

 round scad in the eastern Gulf of Mexico to be 136 

 mm FL at age 1, 160 mm FL at age 2, and 177 mm 

 FL at age 3. In addition, they reported considerable 

 variation (10-20 mm) in the mean lengths at age. Dif- 

 ferences in the growth rates observed in the two 

 studies are believed to be due mainly to method- 

 ology, but may also be due to slight differences in 

 the growth of round scad between the two areas. 



264 



