Fishery Bulletin 102(1) 



and fertilization rates are 100% in the absence of fishing, 

 but the population must have multiple males for high fer- 

 tilization rates. For all the results presented in our study 

 we assumed a fixed sex-change pattern, mating among 

 males and females at each site proportional to gamete 

 production, and larval export among mating sites. We also 

 assumed, unless otherwise noted, a sharp size-selective 

 fishing pattern (r=l) and that the probability of sex change 

 and removal of sex-changing fish by the fishery are cen- 

 tered at the same mean size or Lj=L c . Clearly, the results 

 presented in our study may not apply to cases where these 

 assumptions are not met. 



General patterns predicted by the model 



First, we examined the general effect of fishing mortality 

 on the sex-changing stock for the case when one mating 

 site exists. When L f = L c , eggs produced per recruit decrease 

 only slightly with fishing mortality (e.g. a 3% drop as fish- 



ing mortality increased from to 3, Fig. 3A). However, the 

 mean number of eggs fertilized (both total and per recruit) 

 decreases sharply as fishing mortality increases (e.g. a 30% 

 drop as fishing mortality increased from to 3, Fig. 3A). 

 The number of recruits per year decreases as well. As fish- 

 ing mortality increases, male spawning stock biomass per 

 recruit decreases dramatically, whereas changes in female 

 spawning stock biomass would be practically undetectable 

 (90% drop for male SSBR, compared with a 3% drop for 

 female SSBR as F increases from to 3, Fig. 3B). Because of 

 the drop in male SSBR, total spawning stock biomass (males 

 and females) per recruit also decreases as fishing mortal- 

 ity increases. Sperm production per recruit is predicted to 

 decrease with increasing fishing mortality (Fig. 3C). 



Sensitivity of stock dynamics to fishing pattern 



In general, mean population size decreases as fishing pres- 

 sure increases (Fig. 4A). The adult sex ratio (measured as 



