FISHERY BULLETIN: VOL. 70, NO. 3 



10 



[ 



4- 



tr. 



O 



V) 

 UJ 



t 2 



0.5 1.0 



LARVAL LENGTH (cm) 



Figure 11. — Estimates of liters/hour searched by an- 

 chovy larvae 0.3-1.3 cm total length. Lines are based on 

 average estimates and are not data points. Solid line 

 is the volume searched when the photographic estimate 

 of swimming activity was used ; dashed line, the volume 

 searched when visual estimate of swimming activity was 

 used; and dotted line, the relationship, liters/hour = 

 L3, where L is total larval length (cm). Rectangles 

 enclose estimates of search rates for other larval fishes 

 from the literature: 1) Clupea (Rosenthal and Hempel, 

 1970) ; 2) Clupea (Blaxter, 1969) ; 3) Pleuronectes 

 (Blaxter, 1969); and 4) Sardina (Blaxter, 1969). 



yields a slightly higher estimate than the aver- 

 age estimate, but the difference does not exceed 

 5% even in the largest larva under consideration. 

 That the change in volume searched with 

 length was nearly proportional to L^ could be 



expected. The volume is the product of the 

 speed and cross-sectional area of the perceptive 

 field for prey. I assumed a length coefficient of 

 one in the calculation of cross-sectional area 

 (area = 0.45L-) , and the length coefficients for 

 the two swimming speed estimates were close 

 to one as is commonly the case in swimming 

 speed studies (Bainbridge, 1958; Hunter and 

 Zweifel, 1971). Thus the product of the length 

 coefficient for speed and the one for area would 

 be expected to be close to L^ On the other hand, 

 that the product of the various constants used 

 in the calculation was close to 1,000 was simply 

 chance. 



The estimates of volume searched made for 

 other larval fishes, shown as rectangles in the 

 figure, are close to the two for anchovy, especially 

 when the variance in such estimates is consid- 

 ered. How much of the difference betw^een an- 

 chovy and other species can be attributed to spe- 

 cific differences and how much to differences in 

 technique and assumptions is unknown. For 

 example, differences in techniques of estimation 

 of the distance of prey at the time of sighting 

 could account for the differences between my re- 

 sults and others. The initial movement of the 

 head toward the prey was easily detected in the 

 films, but it is possible that a significant period 

 elapsed between recognition and movement of 

 the head. If this is true, I have underestimated 

 the size of the perceptive field for prey. Perhaps 

 some of the herring estimates are higher than 

 the anchovy because the anchovy swims more 

 slowly. Anchovy swim more slowly because un- 

 like most pelagic fishes the anchovy swims inter- 

 mittently; that is they glide between beats of 

 the tail. 



ESTIMATION OF FOOD DENSITY 



To fulfill their metabolic requirement larval 

 anchovy must ingest about 686 rotifers/day /mg 

 dry weight or the caloric equivalent. This esti- 

 mate was derived from the following: 4.5 /^liter 

 of 02/mg dry wt/hr is consumed by anchovy 

 larvae kept on a 14-10 hr light-dark cycle (Las- 

 ker, personal communication); 1 /^liter O2 = 

 0.005 cal; caloric value of the rotifer (Brachi- 

 onus) = 5,335 ± 139 cal/g (Theilacker and Mc- 



834 



