LANGTON: FOOD HABITS OF YELLOWTAIL FLOUNDER 



Although there was a certain degree of hour-to- 

 hour variability, a peak in the weight of stomach 

 contents occurred during the afternoon-early eve- 

 ning period (Fig. 1). In Figure 1 the day has been 

 divided into four periods— dawn (0300-0800 h), 

 day (0900-1400 h), dusk (1500-2000 h), and night 

 (2100-0200 h)— which accounts for seasonally 

 variable day length in the dawn and dusk period. 

 Despite a seasonal change in day length, the com- 

 position of the diet also changed over a 24-h peri- 

 od. Polychaetes were less important prey during 

 the night than during any of the other three time 

 periods. They dropped from values ranging from 

 41-47% to 24% as a percentage of the weight of 

 stomach contents. Conversely, crustaceans, am- 

 phipods in particular, were more important at 

 night (values ranging from 15 to 23% vs. 34%). 

 Unidentifiable animal remains also accounted 

 for their smallest percentage of the diet (13.0%) 

 in the dusk period when the fish stomachs were 

 fullest. The greatest percentage of empty yellow- 

 tail flounder stomachs was found during the 

 night (46%) and the smallest ( 19%) during the day 

 with intermediate levels occurring at dawn 

 (34%) and dusk (26%). 



To evaluate whether the diurnal feeding pat- 

 tern shown in Figure 1 is statistically significant, 

 an analysis of variance, including time of day 

 and seasonal factors, was conducted. The results 

 of this analysis are given in Table 3 for trans- 

 formed data using an inverse hyperbolic sine 

 transformation (Y' = sin h" 1 i\fY)) to account for 

 the extreme skewness of the data (i.e., a large 

 number of empty and almost empty stomachs) 

 (see Bartlett 1947). Both time of day and season 

 are significant factors in determining the weight 

 of stomach contents for yellowtail flounder. This 

 analysis confirms that there are statistically sig- 

 nificant differences in stomach content weight 

 over a 24-h period. These results are, however, 

 influenced by the level of interaction between 

 time of day and season, such that it is not clear 

 which of these two factors is the most important 



Table 3.— Analysis of variance of the weight 

 of stomach contents for yellowtail flounder, ex- 

 pressed as percent body weight, for time of day 

 and season. See text for details. 



"5- 0.30 



cu 



5 



O 



CD 

 <v 

 o 



0.20 



O.IO- 



IO 12 14 

 Time (hours) 



16 18 20 22 24 



Figure 1.— Weight of stomach contents, as a percentage of the 

 body weight, for yellowtail flounder collected during the 

 spring (open circles) and autumn (solid circles) over a composite 

 24-h day from 1973 through 1976. Data points are 6-h weighted 

 averages for the periods identified in the figure as dawn, day, 

 dusk, and night. 



in determining the shape of the curve for the 

 weight of stomach contents over the composite 

 24-h period. Further study specifically evalu- 

 ating the effects of time and season on stomach 

 content weight is warranted. 



Determining what influence size might have 

 on the feeding periodicity of yellowtail flounder 

 is difficult because of the small number of sam- 

 ples when the data are distributed among both 

 size classes and time. To distinguish between 

 mature and immature fish the data were divided 

 into two size classes, 0-15 cm and 21-49 cm fish, 

 which accounts for the bimodal distribution of 

 fish collected for stomach content analysis (see 

 section on Methods). For the 1-15 cm fish the 

 sample sizes were small and unevenly distrib- 

 uted, and no conclusions can be made as to feed- 

 ing periodicity. However, 66% of these smaller 

 fish were caught at night and an additional 24% 

 were caught during the dawn period. In con- 

 trast, the 21-49 cm fish were taken both day and 

 night in much more equal proportions. Only 30% 

 of the catch was taken at night and 20% was 

 caught during the daytime period. The feeding 

 periodicity of these larger fish is adequately 

 represented by the data in Figure 1. There was a 

 gradual increase in the mean weight of stomach 

 contents from dawn to a peak in the dusk period. 

 The major influence of the smaller, 1-15 cm, fish 

 is that they have relatively more in their stom- 

 achs which increases the overall mean in the 

 night period when the numbers of juveniles 

 caught was at a maximum. 



19 



