26 



Fishery Bulletin 101(1) 



Table 2 



(A) ANCOVA and (B) component least squares regression 

 statistics for log-linear (LnV = Ina + 6LnX) relations of 

 fecundity (F) to carapace length {CD for P. margmatus 

 caught at Necker Bank during three periods: 1978-81, 

 1991, and 1999. Natural logs are used throughout. Under- 

 lines illustrate that the least-square means for each period 

 differ from one another. MSE = mean square error 



LnCL X period (P=0.27 — ns; not included in final model ) 

 Error 105 0.03 



Total 108 



12.426 > 12.284 > 12.119 



1999 > 1991 > 1978-81 



B Regression models: LnY = Lna -i- 6LnA' 



1978-81 LnF= 2.7994 + 2. 1569 LnCL, 



r2=0.708, 



/!=35, 



P<0.001, 



SEa=1.0367, 



SE 6=0.2412 

 1991 LnF = 1.5859 + 2.4778 LnCL, 



r2=0.881, 



n=34, 



P<0.001, 



SE a=0.6934, 



SE 6=0.1607 

 1 999 LoF = 1 .3977 -i- 2.5533 LnCL, 



r-'=0.886, 



/I =40, 



P<0.001, 



SE a=0.6452, 



SE 6=0.1485 



measurements was sufficient for the observed change in 

 egg diameter to have had an 87'* chance of being detected 

 at a = 0.05. 



Individual and population egg production 



Based on the IRP of Kanciruk and lierrnkind (1976), 

 most egg production by the Necker Bank population of 

 P. margincitus in 1999 was by small adults (<60 mm TW) 

 that now dominate the population (Table 4). Large adults 

 (>60 mm TW), although highly fecund, are now too rare 

 to contribute substantially to total population egg produc- 

 tion (Table 4). 



Discussion 



Size-speciFic fecundity and egg size 



Fecundity The initial 16% increase in body size-specific 

 fecundity between 1978-81 and 1991 occurred while 

 commercial CPUE decreased fivefold. Unlike commercial 

 data, research CPUE data were collected at fixed sta- 

 tions (including juvenile habitat), were uninfluenced by 

 increased catchability (the targeting of larger adult lobster 

 in more productive habitats by commercial fishermen ), and 

 continued to show a decline of similar magnitude during 

 the 1991-99 period when size-specific fecundity increased 

 an additional 18''i (Fig. IB). Thus both observed fecundity 

 responses occurred simultaneously with declining lobster 

 densities. The cumulative 36% increase in size-specific 

 fecundity observed for Necker Bank P. marginatus over 

 a >20-yr period of exploitation is not unreasonable given 

 the evidence for concurrent, compensatory declines in 

 body size at sexual maturity in this population (Polovina, 



