FISHERY BULLETIN; VOL. 72, NO. 4 



with the wide-range Reynolds number motions 

 and large amplitude wave forms encountered in 

 this study must be done cautiously and with ap- 

 propriate consideration of hydrodynamical dis- 

 similarities. However, using the most obvious be- 

 havioral differences between the two types of 

 studies, a higher overall efficiency might be sus- 

 pected based on the viewpoint of Lighthill (1971) 

 that the large amplitude tail motions exhibited by 

 some fishes be interpreted as a means of producing 

 reactive thrusts which balance the enhanced vis- 

 cous drag produced upon the commencement of lat- 

 eral movements. Lighthill thus implies that large 

 amplitude movements interspersed with periods 

 of gliding are more efficient than continuous small 

 amplitude oscillations as a mode of propulsion. 

 This appears to be confirmed in the results of this 

 study where the behavior is of this type and the 

 efficiency apparently high. It should be stressed 

 that a range of efficiencies can exist due to the 

 intrinsic variability in O2 consumption values 

 and associated activity measurements and the 

 fact that synchronous determinations of both have 

 not yet been performed. The purpose of the swim- 

 ming efficiency calculation and the associated 

 comparison curves with O2 values (Figure 9) is to 

 demonstrate the relationship the theoretical val- 

 ues determined here have to the available 

 physiological parameters obtained with simple 

 experimental designs. If excursion energies could 

 be obtained by simpler means, one could circum- 

 vent the involved procedures presented in this 

 paper. 



It is interesting to note that the Pacific sardine, 

 Sardinops caerulea, whose ecological niche was 

 primarily taken over by the anchovy, Engraulis 

 mordax, in the California Current (Murphy, 1966) 

 does not exhibit, in the larval stages, the same 

 swimming behavior as the anchovy, i.e., swim- 

 ming bursts followed by glides. Instead it swims 

 by constant, small amplitude oscillating move- 

 ments of the body. In light of the results here and 

 theoretical work by Lighthill it is possible that the 

 propulsive efficiencies in the larval stages of the 

 sardine and anchovy are slightly different, the 

 sardine being less efficient. Thus a small 

 behavioral-propulsive difference between the an- 

 chovy and the sardine might have permitted the 

 anchovy to compete more favorably when there 

 was a decline in sardine population. 



The evaluation of propulsive energetics as 

 outlined in this study is directed at only one 

 size of the anchovy larva because the method 



10" 



r o 



o 

 o 



CURVE OBTAINED FROM Oj 

 CONSUMPTION MEASUREMENTS 

 {SEE TEXT). 



THEORETICAL MODEL VALUE 

 COMPUTED WITH WAVE PARA- 

 METERS FITTED TO 1.4 cm 

 LARVAE (SEE TEXT). 



HUNTER AMPLITUDE INTERCEPT 

 MODIFICATION OF THEORETICAL 

 MODEL FOR LENGTHS OTHER 

 THAN 1.4cm (SEE TEXT). 



Figure 9. — Energy consumption of swimming based on theoret- 

 ical model (open circles and open square) and total energy con- 

 sumption based on O2 utilization (closed circles) as a function of 

 length. Vertical lines on both curves span one standard error of 

 the data. 



requires detailed knowledge of the various 

 wave-form parameters as functions of time for 

 each length of the organism studied. Valid re- 

 sults cannot be obtained for other sizes by a 

 mere alteration of the length of the organism in 

 the wave-parameter functions. By the method 

 outlined here, the only way to properly evaluate 

 propulsive energetic costs for different lengths 

 would be to repeat the course of wave- 

 parameter determination completely. However, 

 with such limitations in mind it is interesting 

 to compare results obtained when modification 

 of the existing wave-parameter functions is 

 made using extensions of known length- 

 dependent wave-parameter quantities which 

 have been measured for larval anchovies. The 



896 



