LAURENCE: moENERUETIC MODEL FOR WINTKK FLOUNDER LARVAE 



«.oo 



J. 75 



3.50- 



3.15 



3.00 



t. 75 



t. SO 



t.li 



s.oo- 



1.75- 

 1.50- 

 1.25- 



1.00- 

 0.75- 

 0.50 

 0. ZS 

 0.00 



0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 BOO. 300.0 1000.0 1100.0 

 DRY UEIGHT (US) 



FIGURE 13. — Index of body weight consumed per day by winter 

 flounder larvae at 8°C over the range of dry weights from hatch- 

 ing to metamorphosis and at different planktonic prey concen- 

 trations. Numbers for each simulation indicate prey concentra- 

 tion in calories per liter; 6.7-21.7 simulations are in ascending 

 order from top to bottom. 



0.0 100.0 200.0 300.0 HOO.O 500.0 600.0 700.0 300.0 300.0 1000.0 1100.0 

 DRY UEIGHT (UG> 



FIGURE 14. — Gross growth efficiencies of winter flounder larvae 

 at 8°C over the dry body weights from hatching to metamor- 

 phosis and at different plankton concentrations. Numbers for 

 each simulation indicate prey concentration in calories per liter. 



at lower prey densities, while larger, older larvae 

 approached maximum feeding ration at increas- 

 ingly higher densities. The higher coefficient of 

 proportionality (a, Equation (7)) values for the 

 smaller larvae suggests that they have an easier 

 time capturing their maximum ration. In fact, 

 they reach their maximum ration at lower prey 

 densities because their stomach capacity is very 

 small and limited, while large larvae with greater 

 stomach volumes can take advantage of higher 

 plankton densities. From the standpoint of suc- 

 cessful captures to obtain the maximum ration, 



smaller, younger larvae are actually much less 

 efficient than larger. 



This size effect on feeding ration over a range 

 of prey densities has not been specifically exam- 

 ined for fish larvae before. Powers (1974) theoret- 

 ically evaluated tha Ivlev relationship with 

 laboratory feeding data for an amphipod, Aniso- 

 gammarus confervicolus. He examined changing 

 coefficients of proportionality (a) at constant 

 maximum ration. The results showed that the 

 asymptote is approached more quickly at higher 

 a's, similar to the results noted in this research. 

 Powers did not analyze maximum feeding ration 

 as a function of animal size at changing a's. He 

 did, however, state that animal size would prob- 

 ably have an effect since larger animals are better 

 predators than smaller ones. 



The initial sharp reduction in feeding times pre- 

 dicted by the model following hatching until a dry 

 weight of 75 /ug (Figure 6) was undoubtedly due 

 to the increased ability of growing winter flounder 

 larvae to capture prey. This is supported by Schu- 

 mann (1965), who reported that larvae of Pacific 

 sardine, Sardinops sagax, which were initially 

 successful at feeding increased their searching 

 ability and the probability of capturing a sub- 

 sequent prey. The increase in predicted feeding 

 times from 75- to 500-^g size was due to the 

 exponential increase in metabolic rate for pre- 

 metamorphosed larvae (Laurence 1975). The re- 

 duction in predicted feeding time from the initia- 

 tion of metamorphosis until its completion 

 (500-1,000 fMg) was related to the decrease in 

 absolute metabolism due to behavioral changes 

 of metamorphosing winter flounder (Laurence 

 1975) and their greatly increased efficiency at 

 capturing prey, which required less energy expen- 

 diture. The decrease in predicted feeding time 

 with increase in prey concentration was due to 

 the increased chance of prey encounter and cap- 

 ture. Zaika and Ostrovskaya (1972) also con- 

 firmed this for Baltic smelt and Pacific herring, 

 Clupea harengus pallasi, larvae when they theo- 

 retically showed that the time spent searching for 

 food decreased exponentially with an increase in 

 food concentration. 



Most larval fish have been reported as visual 

 feeders (Houde 1973) and require daytime light 

 intensities for optimum feeding (Blaxter 1969). 

 In view of this, it is surprising that little research 

 has been done on the relationship of feeding pa- 

 rameters and available time for feeding. Ivlev 

 (1961b) combining field and laboratory data for 



539 



