RESPIRATION AND DEPTH CONTROL AS POSSIBLE REASONS FOR 

 SWIMMING OF NORTHERN ANCHOVY, ENGRAULIS MORDAX, 



YOLK-SAC LARVAE 



Daniel Weihs' 



ABSTRACT 



Larval northern anchovy in the yolk-sac (nonfeeding) stage exhibit regular bursts of continuous 

 swimming during the first 3 days after hatching. It has been suggested that this behavior may have a 

 respiratory function. A different possibility is depth control, countering the tendency of the larvae to 

 sink when motionless. This paper includes a theoretical and experimental investigation of the possible 

 functions of these swimming bouts. 



The theoretical approach was to define a model and calculate the oxygen available to the larva 

 when resting and while moving, and experiments were jjerformed as a check of the theoretical 

 results. The experiments were conducted on yolk-sac larvae in sealed tanks with varying dissolved 

 oxygen concentrations to determine the effects of reducing the available oxygen on the frequency and 

 duration of the swimming bursts. Results of the experiments confirmed the theoretical model. They 

 indicate that the swimming bouts both help the larva stay at a constant depth and have a respiratory 

 function when the oxygen concentration in seawater is less than 60% of saturation. 



Newly hatched northern anchovy, Engraulis 

 mordax, larvae exhibit a pattern of regular short 

 bouts of continuous swimming interspersed with 

 periods of resting. These larvae are still in the 

 yolk-sac stage and are not feeding so that the 

 locomotory behavior must have some other pur- 

 pose, as these motions are energy consuming and 

 also endanger the animal by attracting predators 

 (Lillelund and Lasker 1971). Hunter (1972) 

 suggested that these swimming bouts might have 

 a respiratory function. Respiration has to be by 

 cutaneous diffusion through the 2-3 /xm thick skin 

 (Lillelund and Lasker 1971) of the larvae as the 

 gills develop only at a later stage. The purpose of 

 this paper is to test this hypothesis and another 

 possibility, depth control, to counter sinking due 

 to the negative buoyancy, using theoretical and 

 experimental methods. 



First, I develop a theoretical model for oxygen 

 transport to motionless and swimming yolk-sac 

 larvae and estimate the possible oxygen uptake. 

 Next, I describe the experiments to test the predic- 

 tion of the theory for both proposed mechanisms 

 and compare their results. 



'Southwest Fisheries Center La Jolla Laboratory, National 

 Manne Fisheries Service, NOAA, La Jolla, Calif.; present ad- 

 dress: Department of Aeronautical Engineering, Technion, 

 Hidfa, Israel. 



METHODS 



Analytical Model 



A mathematical model is now introduced to con- 

 sider the possible respiratory function of the bouts 

 of continuous swimming of yolk-sac anchovy lar- 

 vae. First, we calculate the oxygen transport to a 

 motionless larva. This transport is then compared 

 with the metabolic requirements. If the metabolic 

 requirements are not met, larval motion (and the 

 resulting convective diffusion) is required. 



The size of yolk-sac larvae (2.7-4.0 mm total 

 length) and their swimming speeds (Hunter 1972) 

 lead to tj^ical Reynolds numbers, based on larval 

 length (Weihs 1980) of <20. (The Reynolds 

 number is a nondimensional factor indicating the 

 relative importance of pressure and viscous effects 

 on a body moving in a fluid under given circum- 

 stances — the higher the Reynolds number, the 

 smaller the influence of the viscosity.) The larvae, 

 as a direct result of their small size, are in a highly 

 viscous laminar flow situation in which turbulent 

 effects can be neglected. Thus, the larvae and 

 their immediately surrounding water would be 

 transported together in oceanic turbulent eddies, 

 which are of the order of tens of centimeters in 

 diameter. As a result, a nonswimming larva would 

 stay for a relatively long period in the same mass 



Manuscript accepted; July 1979. 



FISHERf BULLETIN: VOL. 78, NO. 1, 1980. 



109 



