554 



Fishery Bulletin 94(3). 1996 



Exuma Cays and 13 km across the bank and sound 

 to comprise 456 km 2 . A full-time warden enforces 

 park regulations, which strictly prohibit the collec- 

 tion of any plants or animals. This investigation was 

 conducted near the island of Waderick Wells (Fig. 

 IB) where the park headquarters are located and 

 fishing regulations are best enforced. The study area 

 was well within the park boundaries, 10 km from 

 the southern boundary of the park and 24 km from 

 the northern boundary. 



Adult surveys 



For a comparison of fished and unfished areas, den- 

 sity surveys for adult queen conch were conducted 

 on the bank and the island shelf, from beach to shelf 

 edge, in the Exuma Sound near LSI between March 

 and September 1991 (Stoner and Schwarte, 1994) 

 and near WW in 1994 in this study. Annual surveys 

 for adult conch conducted at selected sites off LSI, 

 where fishing is allowed, showed little interannual 

 variation in population size (Stoner and Sandt, 1992) 

 and provide justification for the comparison of popu- 

 lation densities in different years. Between 1988 and 

 1991, maximum variation from the mean population 

 size was just 19%; the population was 4% above av- 

 erage in 1991. Adult densities at an important re- 

 productive site at 18 m depth off LSI (Stoner and 

 Sandt, 1992) differed only 8% between 1991 and 

 1994. Given the relative stability of adult conch popu- 

 lations at LSI and the long adult life (probably >20 

 years; Coulston et al., 1987), we concluded that sur- 

 veys from WW and LSI could be compared. In addi- 

 tion, intrasite comparisons of conch density and age 

 structure provided some of the most compelling evi- 

 dence for the effects of fishing on the recruitment 

 and population size. For direct comparison, survey 

 methods for this study were similar to those used by 

 Stoner and Schwarte ( 1994). 



Near WW in the MFR, the bank survey area ex- 

 tended 7.5 km along the long axis of the island and 5 

 km southwest from the channels north and south 

 (Fig. IB), comprising 3,245 ha. Throughout this area, 

 a snorkeler was towed behind a small boat for a to- 

 tal of 47 km while counting the number of adult conch 

 within a 6-m-wide field of view and recording the 

 locations in which adults were abundant. Tows were 

 made along a total of 12 transects which ran perpen- 

 dicular to WW. These transects started near the west- 

 ern shores of the primary islands and were less than 

 1 km apart. Latitude and longitude for the begin- 

 ning and end of each transect were recorded with 

 Global Positioning System (GPS). This resulted in 

 direct observation of a surface area of 28.0 ha. Den- 

 sities of adult conch were calculated for the transects, 



and the number of conch was extrapolated for the 

 entire area. 



To test the hypothesis that depth distribution of 

 adults differed between the 2 sites, depth-stratified 

 surveys for adult conch were conducted in Exuma 

 Sound along 6 transects which ran perpendicular to 

 WW. Transect A ran northeast from the north tip of 

 WW, and transect F ran northeast from the north 

 end of Hall's Pond Cay (Fig. IB). Other transects 

 were spaced as uniformly as possible between A and 

 F, 1-2 km apart. Along each transect, estimates of 

 conch density were made at 7 depth intervals: just 

 off the beach (when present), 2.5-5 m, 5-10 m, 10- 

 15 m, 15-20 m, 20-25 m, and 25-30 m. Within each 

 interval, two SCUBA divers swam for 8-30 minutes, 

 depending on depth, holding a line (8 m) between 

 them and counting the number of adult conch that 

 lay beneath the line. One diver carried a calibrated 

 flow meter to calculate the distance traveled. To com- 

 pensate for the potential influence of current on dis- 

 tance measured, the divers swam from the boat an- 

 chor into any discernible current and back, covering 

 two parallel, nonoverlapping paths that normally ran 

 parallel to the isobaths. Thirty-six dives were made 

 during this shelf survey for adult conch density. The 

 mean swim distance was 360 m (SD=6 m). Sampling 

 effort was lowest for the 25-30 m interval because of 

 diving limitations imposed by depth; however, the 

 surface area of this deepest sampling zone was rela- 

 tively small (Table 1). Again, results were directly 

 comparable with Stoner and Schwarte (1994). 



A bathymetric chart of the island shelf east of WW 

 (to 45 m) was created from 250 points for which depth 

 (corrected for tidal state) and position (determined 

 with GPS) were recorded from a small boat. Computer- 

 generated contour lines ( 1 m) were resolved with nor- 

 mal smoothing algorithms. Surface areas for each con- 

 tour interval were calculated, and the mean density of 

 conch found during the dive surveys was extrapolated 

 to determine total number for each depth interval. 



To examine age distribution and to evaluate po- 

 tential migration patterns at WW, shell length and 

 shell lip thickness were measured for at least 50 in- 

 dividuals at each depth interval on the shelf and for 

 213 individuals collected from several locations on 

 the bank. Although young adults laid down lip ma- 

 terial faster than older adults and although shells 

 are subject to erosion, shell thickness at the lip pro- 

 vides a relative indication of conch age (Appeldoorn, 

 1988; Stoner and Sandt, 1992). 



To test for differences in size and age distribution 

 associated with depth, frequency distributions of 

 shell length and shell lip-thickness data were ana- 

 lyzed by depth interval with the nonparametric 

 Kolmogorov-Smirnov test. Several aspects of shell 



