significantly. Red muscle temperature was moni- 

 tored to test if changes in metabolic rate reflected 

 changes in it. All fish showed red muscle tempera- 

 tures as stable as those shown in Figure 1. Fish 

 were in the respirometry box approximately 15 to 

 30 min before data recording began and red mus- 

 cle temperatures generally approached thermal 

 steady state during this period. 



The SMR of 33 fish (0.317-4.737 kg) was success- 

 fully determined. A regression line of SMR versus 

 body weight was fitted by Gauss-Newton iteration 

 technique (Biomedical Computer Programs, pro- 

 gram number BMDP 3R), rather than by a linear 

 regression technique based on log-log transforma- 

 tion of the data (Figure 2). The advantages of the 

 former method and disadvantages of the latter 

 have been discussed by Zar (1968) and Glass 

 (1969). 



The best-fitting allometric equation was found 

 to be: 



SMR 



412.0 ( ±27.1) W 



0.563( *0,07) 



(1)2 



where SMR = standard metabolic rate in milli- 

 grams O2 per hour and 

 W = body weight in kilograms; 



values in parentheses are the standard deviations 

 of the parameters. The coefficient of determina- 

 tion (r2) is 0.72. 



The exponent in the allometric equation de- 

 scribing the effect of body size on the SMR of other 

 teleosts ranges from approximately 0.65 to >1 

 (Winberg 1956; Fry 1957; Beamish and Mookher- 

 jii 1964; Beamish 1964; Glass 1969; Brett 1972). 

 The lower value for the exponent in Equation (1) 

 indicates that the weight specific SMR (i.e., mil- 

 ligrams O2 per gram per hour) of skipjack tuna 

 decreases more steeply as body size increases than 

 does the weight specific SMR of other species. 



The strong influence of body size on the SMR 

 may be a unique characteristic of thermoconserv- 

 ing species such as skipjack tuna. However, the 

 value of the weight exponent could also be 

 influenced by the technique used to measure SMR. 

 For comparative purposes, it would be useful to 

 determine the SMR's (and corresponding allomet- 

 ric equation) of species (e.g., salmonids) where 



^If the allometric equation to describe the effect of body size on 

 whole body SMR is: SMR =aW'' then the corresponding equation 

 to describe weight-specific SMR versus body weight is: SMR; W = 

 aW-WorSMR' =oH"' 'whereSMR' = weight-specific SMR. W 

 = body weight, and a and h are fitted parameters. 



3 04 5 10 2 3 4 50 10 



BODY WEIGHT (kg) 



Figure 2. — A double logarithmic plot of the standard metabolic 

 rate (SMR) versus body weight (W). The line represents the 

 allometric equation SMR = 412.0 w" 56.) jy^^ coefficient of de- 

 termination IS 0.72. 



these parameters are already known, but employ- 

 ing the methodology outlined in this study. 



The SMR has been postulated to be a function of: 

 the diffusing capacity of the respiratory system, 

 whole blood sugar concentration, and the rate at 

 which the circulatory system can deliver sub- 

 strates and oxygen to the cells (Schmidt-Nielsen 

 1970; Ultsch 1973; Coulson et al. 1977; Wilkie 

 1977; Hughes 1977; Umminger 1977). Specifically 

 which of these factors most influence the SMR of 

 skipjack tuna is unknown at this time. Selection 

 pressures apparently favor significant reductions 

 in the weight-specific SMR of skipjack tuna as body 

 size increases (hence the lower exponent m Equa- 

 tion (D). How the factors that determine SMR 

 could be affected by such selection pressures is also 

 unknown. 



The SMR's for skipjack tuna are approximately 

 5 to 10 times greater than those reported for other 

 teleosts of equal body size (Pritchard et al. 1958; 

 Beamish 1964; Brett 1965, 1972). However, the 

 great difference in the effect of weight on SMR, 

 and the unique methodolog>' employed in this 

 study, makes direct comparisons tenuous. 



Application of the Results 



Careful application of my results in energetics 

 models is advised for several reasons. First, skip- 

 jack tuna attain maximum body size of approxi- 

 mately 22 kg iKitchell et al. 1978); although the 

 weight range of fish I used in this study covers 

 more than an order of magnitude, there is still a 

 large untested size range. Because skipjack tuna 



496 



