WEIHS; RESPIRATION AND DEPTH CONTROL IN ENGRAULIS MORDAX 



where ^ is a diffusion coefficient obtained from 

 experimental correlations of the diffusional flux 

 with the Reynolds (Re) and Schmidt numbers (Sc). 

 (The Schmidt number is the ratio of the kinematic 

 viscosity to the diffusivity and nondimensionally 

 indicates the relative importance of these two ef- 

 fects in a given flow situation.) For the present 

 circumstances 



D 



k = — {2 + 0.6 Re '/2 Sc'/3) 



2a 



(14) 



for average swimming speeds of approximately 

 5 cm/s ( Hunter 1972) and a Schmidt number of 600 

 we have 



J,„„ = 1.27 X 10"^ g/min. 



(15) 



Hence, oxygen transport due to convective diffu- 

 sion is over 20 times higher than for the motion- 

 less larva (Equation ( 12)). The calculation leading 

 to Equation (15) is approximate, as the larva's 

 shape will influence the coefficient 0.6 in k (Equa- 

 tion (14)) and also change the form of Equation 

 ( 13). It is, however, accurate to at least an order of 

 magnitude (Levich 1962). Thus, once the larva 

 starts swimming, the mass transfer of oxygen to 

 its surface increases by at least an order of mag- 

 nitude. Recently, an additional mechanism for 

 oxygen transport to stationary eggs was identified 

 by O'Brien et al. (1978) who showed that under 

 certain riverbed conditions natural convection, 

 due to the oxygen and metabolite gradients, may 

 contribute to the oxygen transfer. This effect may 

 play a supplementary role in the present (pelagic) 

 case as the natural convection effects are much 

 smaller than the forced convection. 



Tests with Larvae 



Egg batches were obtained once a week from 

 groups of adult northern anchovy maintained in 

 the laboratory and induced to spawn. Measure- 

 ments were made each week during a 6-wk period 

 to minimize bias due to a single cohort group. 

 Water temperature ranged from 19° to 21° C, and 

 overhead fluorescent lighting was used. The 50% 

 hatching point was determined and defined as 

 "day 0" for each batch. Experiments were carried 

 out on age day larvae every week (six times). 



A set of five 2,000 ml graduated cylinders filled 

 with filtered seawater was used for the environ- 

 mental tests. Oxygen concentrations of 100, 80, 



60, 40, and 20% of saturation at the measured 

 temperature were produced by bubbling nitrogen 

 through each of the cylinders. After the larvae 

 were added (about 25 individuals/cylinder), the 

 cylinders were sealed off with rubber stoppers. 

 Oxygen concentrations were measured periodi- 

 cally during the experiments with a Beckman In- 

 strument Model 160 Physiological Gas Analyzer^ 

 to check on initial values and possible drift;. 



Individual fish were monitored for a 5-min 

 period, and duration and number of swimming 

 bursts were recorded on a Esterline- Angus Opera- 

 tion Recorder Model AW. Records were also made 

 of approximate swimming direction (measured 

 from horizontal) as well as the change in orienta- 

 tion of motionless larvae while they were sinking 

 during the resting periods. Five active larvae were 

 monitored in each container every week, for both 

 day and day 1 tests. 



After the day experiments were finished each 

 week, the equipment was reset and the day 1 tests 

 conducted 24 h later with additional larvae from 

 the same batch. The latter larvae were kept in 

 oxygen-saturated water from hatching to 

 minimize stress due to oxygen starvation. 



RESULTS 



No appreciable change in the proportion of time 

 spent in burst swimming was observed when the 

 measurement at 100% of saturation concentration 

 of oxygen (which is the oxygen level in the natural 

 state in the sea because of turbulent interchange 

 with the atmosphere) was compared with the time 

 spent in motion at the 80 and 60% oxygen levels 

 (Figures). 



When oxygen levels were <60% of saturation, 

 large increases in the time spent swimming were 

 observed. The rate of increase of swimming time in 

 both ages (day and day 1) were similar. Various 

 attempts at describing all five data points for each 

 age-group by means of a single empirical exponen- 

 tial function were not successful (low coefficients 

 of determination). Thus, it seems that a different 

 behavioral mechanism is triggered when oxygen 

 levels fall below 60%^^ of saturation at the given 

 temperatures, i.e., much lower than expected oxy- 

 gen concentrations in the upper layers of the sea, 

 where the anchovy larvae are usually found. 



^Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



113 



