432 



Fishery Bulletin 91(3). 1993 



ties of blubber, muscle, viscera, bone, and fins, where 

 weightings were the predicted fractions of total wet 

 body mass for each tissue type. Energy density of vis- 

 cera was assumed to be the same as energy density of 

 muscle, based on the relatively muscular nature of 

 these organs (compared with bone or blubber) and the 

 absence of any visible fat deposits upon or within or- 

 gans. Errors were assumed negligible because the vis- 

 cera comprised a relatively small fraction of total body 

 weight and it appeared unlikely that the viscera rep- 

 resented a major lipid depot in these tropical ceta- 

 ceans. Energy density of fins was assumed to be the 

 same as energy density of bone, based on the assump- 

 tion that caloric content of cartilage (the actual com- 

 position of fin material) is closer to bone than to either 

 muscle or blubber. Again, errors due to this assump- 

 tion should be relatively small as the fins contribute 



very little to overall body weight. Estimated total en- 

 ergy density is presented as a function of the regres- 

 sion-predicted values of each parameter, where param- 

 eter variances are ignored. 



Data analysis 



Relationships between morphological measurements 

 and total wet weight were determined by using linear 

 regression analysis of log transformed variables, where 

 morphological measurements and wet weights were 

 converted to logarithms (base 10) prior to analysis. 



Results 



Significant allometric trends (P < 0.05) occurred in 23 

 of the 25 parameters (Table 3; Figs. 2-11). No signifi- 



