FISHERY BULLETIN; VOL. 85, NO. 4 



Table 2. — Tabulation of the number of queen conchs caught In the /th sample last caught in or 

 before the hth sample (after Seber 1982). / and h designate sampling periods, z, = number of 

 different individuals caught before the /th sample which were not caught int he /th sample, but were 

 caught subsequently, and is equal to the sum across the row excluding the diagonal element. 



1 



8 



Total 



h 

 1 

 2 

 3 

 4 

 5 

 6 

 7 

 8 



42 



Table 3. — Population estimates for queen conch derived from the Jolly-Seber multiple mark-recapture analysis. M/N = proportion of 

 marked conchs in the population just before sample i; N = population abundance just before sample / with standard deviation d[N/N] and 

 coefficient of variation 6-/ A/; S = proportion surviving from the /th to (/ + 1 ) th sample with standard deviation (j[S] and coefficient of variation 

 a/S;B = numberof new individuals joining the population between the /th and (/ + 1) th sample with standard deviation d[B] and coefficient 

 of variation d/6. 



recruitment (B) between times 2 and 3. These 

 figures are obviously unrealistic (e.g., survival 

 cannot be greater than one) and are not consid- 

 ered in any subsequent analysis or discussion. 



Survival rates of queen conch varied from 0.839 

 to 0.237 with coefficients of variation varying 

 from 13.3% to 22.5%. Intensive fishing was 

 known to have occurred in the study area be- 

 tween samples 3 and 4, and 6 and 7. Fishing also 

 occurred between samples 7 and 8, but to a much 

 lesser degree. In the former two periods estimated 

 survival rates were the lowest found: 0.237 and 

 0.293, respectively. Survival estimates from all 

 other periods represent the effects of only natural 

 mortality and emigration. The average of these 3 

 values is 0.6817. This would correspond to an in- 

 stantaneous rate of natural mortality per 3-mo 

 period of 0.3832 [= -ln(l - S)] if no emigration 

 occurred. Assuming the effects of emigration are 

 constant between sampling periods, estimates of 

 instantaneous fishing mortality can be obtained 

 by subtracting the average value of instanta- 

 neous natural mortality (which includes any emi- 

 gration effect) from instantaneous total mortality 

 for periods of known fishing activity. For sam- 



pling periods 3-4, 6-7, and 7-8, the instantaneous 

 rates of fishing mortality per 3-mo period were, 

 respectively, 1.056, 0.845, and 0.0935. Dividing 

 the sum of these by the total time period for which 

 mortality estimates are available (1.75 year) re- 

 sults in an estimate of the annual instantaneous 

 fishing mortality rate (F) for the population in 

 the study area of 1.14 between August 1983 and 

 May 1985. 



On an annual basis, the apparent instanta- 

 neous rate of natural mortality for queen conch is 

 1.533. However, this overestimates true natural 

 mortality because permanent emigration was 

 known to occur, and it was not possible to pre- 

 cisely quantify the effects of emigration on the 

 basis of these data (see Discussion). The estimate 

 does provide an upper bound on the value of nat- 

 ural mortality. 



DISCUSSION 



Model Assumptions 



Principal assumptions of the Jolly-Seber 

 method are that sampling is random, with catch- 



800 



