MAGNUSON: ADAPTATIONS OF SCOMBROIDS AND XIPHOIDS 



J» 



-I 

 o 





10 



(a) r»+0.86 



-> 20 



8 



o 

 o 



z 



UJ 



X 



15 - 



</> 



UJ 



z 



y 20 



(b) r-+0.83 





i 



X 



10 

 45 



(c) r-+0.55 



• : . 



3„ 



2to 



= .<" 40- 



Z^ 



OO 



2 



35 



0.5 10 1.5 2.0 



OBSERVED TYPICAL SPEED (Jt/sec) 



2.5 



Figure 9. — Relation between observed swimming speeds 

 and (a) the relative size of the dark muscle and (b) the 

 concentration of blood hemoglobin, (c) maximum thick- 

 ness of body, and (d) the position of maximum thickness 

 along the length of the fish. Data from Table 7. 

 Correlation coefficients, r, are given on graph. 



globin are also highly correlated (r = +0.82, 

 n = 8, P < 0.005). As expected, the dark 

 muscle used for continuous slow speed is 

 larger for species with faster typical swimming 

 speeds. However, concentrations of blood hemo- 

 globin also appear to be keyed to typical 

 speeds and routine metabolism. Differences in 

 need to transport oxygen sufficient for burst 

 speeds apparently influenced hemoglobin levels 

 little. Perhaps this is so because an oxygen 

 debt can be temporarily incurred for burst 

 activity. 



Typical speeds in body lengths per second 

 (Figure 9c, d) were also correlated with maxi- 

 mum thickness of the body (r = +0.55, 

 // = 10, P < 0.05) and with the position of 

 maximum thickness (?• = +0.56, n = 10, 

 P < 0.05). If speed is given in centimeters per 

 second, the correlation +0.21 and +0.49 with 

 maximum thickness and the position of maxi- 

 mum thickness are not statistically significant 

 at P < 0.05. 



Maximum thickness of the body ranged from 

 12.5 to 23.6% of fork length but only Ac. .^olandri 

 had a value less than 18% (Table 7). Only 

 30% of the variability in thickness is associated 

 with typical speeds. Thus, these variations 

 do not appear to be closely related to dif- 

 ferences in typical speeds (Figure 9c) with the 

 exception that the species with the lowest 

 typical speed, Ac. sola)idri, also had a low 

 thickness ratio that would not be expected to 

 minimize drag. Minimum drag should result 

 from thickness ratios near 22% (Alexander, 

 1967). Differences among A//.r/,s', Kaf.^uwoiius, 

 Sarda, and Tliuitnus vary over a narrow range 

 near 22% in a manner not related to typical 

 swimming speeds. 



Scombroids with faster typical speeds tend 

 to have the thickest part of their body more 

 posteriad (Figure 9d). Only 30% of this varia- 

 tion is associated with basal speed, but the 

 trend is sensible in that the more laminar 

 flow profiles should occur in the faster fishes. 



In conclusion, all four characteristics, size 

 of dark muscle, hemoglobin concentration, body 

 thickness, and the position of the thickest part 

 of the body are statistically related in a sensible 

 manner to the variations in slow typical speeds 

 of scombroids. The relations are poorer for 

 the two characteristics involved in reducing 

 drag, apparently because these features are 

 more important at burst speeds than at typical 

 speeds. 



PROBLEMS ASSOCIATED 

 WITH LARGE SIZE 



Among the scombroid and xiphoid fishes, 

 there is a great diversity in maximum body 

 size, areas of pectoral fins, and the presence 

 or size of the gas bladder. Data on these 



349 



