UJ 



m 



tn 



i 



Q. 



34 

 3 2 

 3.0 

 2.8 

 2.6 

 24 

 2 2 

 2.0 

 1.8 



S 1.6 



4 



1.2- 

 1.0- 



• DECREASING OXYGEN 

 a INCREASING OXYGEN 



YF-C 



YF-B 



BEFORE 

 TEST 



5.5 4.5 3.5 

 DISSOLVED OXYGEN LEVEL (ppm) 



2.5 



FIGURE 2. — Summary of data from the oxygen experiment ob- 

 servations from eight skipjack tuna and three yellowfin tuna. 

 Dots — decreasing oxygen levels, median swim speeds grouped 

 by 1-ppm intervals and "before" test observations. Open trian- 

 gles — increasing oxygen levels, median swim speeds grouped by 

 1-ppm intervals. Solid line — grand median for eight skipjack 

 tuna. Broken lines — medians for each of the three yellowfin 

 tuna, decreasing oxygen levels only. 



containing 1.4 and 1.6 ppm oxygen. They survived 

 a 200-min exposure and a 100-min recovery 

 period. By way of contrast, brook trout, Salvelinus 

 fontinalis, LD 50 'sfor 1.5 ppm and 1.4 ppm were 300 

 and 100 min, respectively (Shepard 1955). The 

 brook trout and the yellowfin tuna were swim- 

 ming at about the same speeds, 1.0-1.5 lengths/s. 

 Although conditions of the two experiments are in 

 no way similar, these data do imply that yellowfin 

 tuna are at least as low oxygen tolerant as brook 

 trout. The higher energy requirements (larger 

 fish, warmer water) of yellowfin tuna allow this 

 conclusion. Perhaps if oxygen levels dropped low 

 enough in my tank (1.4 ppm is about the lowest 

 that could be achieved), an increase in speed simi- 

 lar to that in skipjack tuna would have been ob- 

 served. 



Increased swimming speed should function 

 either to remove the fish from suboptimal areas (if 



coupled with some directive stimuli) or to provide 

 more water to the gills — tunas are ram ven- 

 tilators. Within the skipjack tuna habitat, water 

 deficient in oxygen is found within and below the 

 thermocline (Barkley et al. see footnote 1). Ap- 

 propriate behavior would be to swim up and out of 

 the low-oxygen water. Even without a change in 

 direction, angle of attack of pectoral fins, or body 

 attitude, increased swimming speed alone will 

 cause a tuna to rise due to increased lift ( Magnu- 

 son 1973). 



Faster swimming speeds do not seem to be a 

 response to increase ram ventilation (open mouth 

 swimming). Increased flow over the gills providing 

 more oxygen delivery is offset by increased res- 

 piratory demands imposed by faster swimming. 

 Under conditions of saturated seawater (7.2 mg 

 2 /liter), 15% head loss along the respiratory flow 

 path (Brown and Muir 1970), a conservative oxy- 

 gen utilization factor of 75% (Stevens 1972), and a 

 1 cm 2 mouth gape (Brown and Muir 1970), oxygen 

 is delivered to the gills at the rate represented by 

 the middle broken line (Figure 3). This, of course, 

 also increases as swimming speed increases. Res- 

 piratory demand (solid black line) and oxygen de- 

 livery intersect at two points: the lower is at the 

 minimum swimming speed that can still furnish 

 sufficient oxygen for an animal in an almost basal 

 state and the upper is a point at which exponen- 

 tially increasing respiratory demand again ex- 

 ceeds linearly increasing oxygen delivery. 



The latter would seem to be maximum sus- 

 tained swimming speed; anaerobic metabolism 

 would be necessary at speeds above. However, 

 neither function (anaerobic or aerobic) may be cor- 

 rectly extrapolated to the faster swimming speeds. 

 Respiratory demand might well be less at higher 

 speeds if swimming efficiency increases. 



Yet, if dissolved oxygen concentration drops to 4 

 ppm, increase in swimming speed is an inefficient 

 way to make up the deficit (lower broken line). 

 But, increase gape to 2 cm 2 (I am assuming for 

 argument's sake that this doubles ventilation vol- 

 ume) restores the amount of oxygen delivered 

 (upper broken line). In summary, I suspect that 

 increased swimming speed of skipjack tuna en- 

 countering oxygen-deficient water is not due to 

 ram ventilation needs but rather is a behavioral 

 response to remove an animal from a suboptimal 

 area. Considering the relative expense of faster 

 swimming in terms of oxygen needs, the modest 

 increases in swimming speeds observed are prob- 

 ably very adaptive in that they should cause a 



652 



