374 



Fishery Bulletin 88(2). 1990 



PACIFIC 

 OCEAN 



HALF MOON BAY Vi; 



Figure 1 



Map of San Francisco Bay showing location of release 

 area (arrow) and major Bay Area divisions and names 

 referred to in the text. 



year 1. This gives some weight to a higher mortality 

 these young fish must suffer relative to the adult 

 sharks. The phenomenon of the young being preyed 

 on by larger sharks has been cited by Springer (1960, 

 1967) and by Holden (1974). 



Stock replacement 



Also calculated for various levels of age at first cap- 

 ture (knife-edge selection) and fishing mortality were 

 the percentages of stock which would be replaced due 

 to reproduction. Assuming an even division among the 

 sexes, this was done by summing over ages, from age 

 at first maturity to maximum age, the products of the 

 number of female survivors per recruit and the number 

 of pups produced by females at each age above the age 

 of maturity and multiplying this sum by 100. For given 

 t, and f" 



/? = X 50exp[-M(^ + l)-F(/-^.)d]P, 



(2) 



(>(„ 



where t,n is female age at maturity, R is stock replace- 

 ment per recruit in percent, P, is estimated number of 

 progeny produced annually by an age t female, and 



1 t>tr' 

 t<t,' 



(3) 



Fecundity as a function of maternal body weight was 

 estimated by fitting a monomolecular curve of the form 

 Pi = P- 6p"'^' ', with P the number of progeny, ir{t ) 

 the weight of the female at age t, and /J, 6, and f con- 

 stants. For computation, age at maturity was taken as 

 the age corresponding to the weight which produced 

 a value of zero for P, in the above equation. 



Ackerman (1971, his table 6) gives the numbers of 

 embryos observed in 66 female leopard sharks rang- 

 ing in size from 9.3 to 16.3 kg. Although his data are 

 in a form which allows only the computation of mean 

 number of embryos per female for five weight classes, 

 he does give the number of females observed in each 

 weight category. Using the numbers of females as 

 weighting factors, the curve was fitted to these means 

 with the Levenberg-Marquardt algorithm as imple- 

 mented in the NCSS nonlinear regression program*. 



•Number Cruncher Statistical System. Version 5.3-Power Pack, 

 1988. NCSS. Kaysville, UT 840.37. Reference to trade names does 

 not imply endorsement by the National Marine Fisheries Service, 

 NOAA. ' 



