Archer et al.: Estimates of the incidental kill of Stenella attenuata attenuata calves in the tuna purse-seine fishery 



235 



In Equation 4, an age distribution was generated for each 

 color phase, and then the number of dolphins in each age 

 class was summed across all color phases. 



To estimate the number of calves in each set, we used 

 this age distribution in conjunction with a weaning model 

 developed from a study of the stomach contents and ages 

 of calves (Archer and Robertson, in press). The model 

 predicts the probability that an animal of a given age (a) 

 will be suckling: 



Pi milk) 



(6) 



1 + e 1 



The estimated number of calves (JV„„; f ) in a set is then 



N. 



calf 



IK 



calf 



Pmilk). 



(7) 



In our estimate of N cal ? we chose to use only the first 

 four age classes (0 to 3) because P(milk) 4 was extremely 

 small (2xl0~ 4 ). These age classes allowed us to decrease 

 computational time without significantly affecting the 

 estimates. 



Number of lactating females 



Observers visually examined the mamillaries of the 649 

 females used in the age distribution above (Eq. 4) for the 

 presence of milk as part of the suite of biological data 

 collected. Using these data in conjunction with the color 

 phase of these females, we calculated the fraction of lactat- 

 ing females in each color phase (Flac v ), 



Flac 



■S/ac 

 Sfem, 



(8) 



where Slac v and Sfem c 



the number of females that were 

 lactating and the total number 

 of females in color phase c of the 

 samples examined. 



Flac c was 0.00, 0.01, 0.04, 0.22, and 0.50 for neonate, two- 

 tone, speckled, mottled, and fused specimens, respectively. 

 The estimated number of lactating females (N lac ) in a set 

 was then 



N lai . = ^(nf; Flac v 



(9) 



