Stoner and Davis: Outplanting queen conch, Strombus gigas 



399 



splant portions of the shells followed by long spines 

 on the outer (newer) portions of the last shell whorl 

 support the hypothesis of changing shell shape. Al- 

 though it is normal for spine length to increase 

 proportionally with shell length in queen conch, the 

 posttransplant increase in hatchery-reared stock was 

 extreme and obviously disproportionate in most shells. 



Behavior On all nine dates when burial was exam- 

 ined for free-ranging animals, a higher percentage 

 of wild conch were buried than of hatchery-reared 

 individuals (Fig. 8). Pairwise ANOVA of burial fre- 

 quency on arcsine-transformed data from dates as 

 blocks showed that the difference in burial rates be- 

 tween stock types was significant (F t JS =8.51, 

 P=0.011). Hatchery and wild conch showed nearly 

 parallel patterns of burial frequency over time. How- 

 ever, plots of burial frequency should not be inter- 

 preted as seasonal trends, because juvenile conch 

 appear to demonstrate tidal periodicity in locomo- 

 tory activity (pers. observ). Although the patterns 

 were not quantified, it was frequently noted during 

 field observations that hatchery conch were more 

 active than wild conch. While hatchery-reared indi- 

 viduals were almost always moving, wild conch were 

 frequently found nestled motionless beneath algae 

 or detritus. 



Given the relatively small area of the two outplant 

 sites (1 ha each) tagged conch often dispersed rela- 

 tively far from their original release sites (Fig. 9, A 

 and B). For example, in June 1990 only 15% of the 

 recovered hatchery-reared conch released at site CI 



CL 



o 



Cl 



o 



m 



Apr Moy Jun Jul Aug Sep Oct Nov Dec Jen Feb 



Date 



Figure 8 



Percentage of queen conch, Strombus gigas, buried 

 during each observation for hatchery-reared and wild 

 conch. Values are based upon observations on 30 

 haphazardly chosen conch of each stock type on each 

 date. 



were found in that zone and 35% were found north- 

 west of Cl. Tagged conch tended to move to the north- 

 east and northwest between April and June 1990. 

 By June, hatchery-reared and wild conch initially 

 released at site Cl were widely dispersed and differ- 

 entially distributed (% 2 = 18.01, df=4, P=0.05). Hatch- 

 ery-reared conch released at C2 tended to disperse 

 more widely than wild conch (i.e., from the south- 

 west sandbar to the north zones); the difference was 

 significant <x 2 =255.6, df=6, P=0.05). Wild conch 

 moved toward the center zone between site Cl and 

 C2 (Fig. 9B). Between June and September 1990 a 

 large percentage of conch released at site Cl were 

 found between Cl and C2 (Fig. 9A), while conch from 

 site C2 continued to be found north of the release area 

 (Fig. 9B). In both cases the difference in dispersion be- 

 tween hatchery and wild was significantly different ( C 1 : 

 X 2 =-6.4, df=5, P=0.05; C2: r=17.6, df=5, P=0.05). By 

 November 1990, hatchery and wild conch released at 

 site Cl (x 2 =9.1, df=5, P=0.05) and C2 (x 2 =12.6, df=6, 

 P=0.05) had similar distributions. Conch released at 

 site C2 consistently moved toward the adjacent natu- 

 ral juvenile population centered at Cl (see Fig. 2). 



Enclosure experiments 



Experiment I 



Mortality In the first enclosure experiment ( 7 April 

 to 9 July 1990), mortality data for the midpoint and 

 end of the experiment (Fig. 10) were examined with 

 two separate three-way ANOVAs, by using numbers 

 of dead conch (Table 5). Except when highest mor- 

 tality occured at site C2 at the end of the experiment, 

 mortality did not differ between sites or among the 

 three stocking densities. Differences in mortality be- 

 tween stock types were never significant in the en- 

 closures (Table 5). 



Growth Trends of growth (log-transformed) in the 

 first enclosure experiment (Fig. 11) were similar to 

 those observed in free-ranging conch during the sum- 

 mer (Fig. 5) (i.e. rates were higher at C2 than at Cl, 

 and wild conch grew faster than hatchery conch). In 

 growth period I (7 April-14 May 1990), wild and 

 hatchery conch grew faster at site C2 than at Cl, 

 and wild conch grew faster than hatchery conch at 

 both sites (Table 6). There was also a significant den- 

 sity effect (Table 6); highest growth occurred in conch 

 held at 1.0 individualsm -1 (Fig. 111. 



In growth period II (14 May-9 July 1990), there 

 were numerous two-way interactions in the ANOVA 

 (Table 6), particularly at site Cl. In one-way 

 ANOVAs, there was no density effect for hatchery 

 (F 23 = 5.74, P=0.094) or for wild (F, 3 =7.139,P=0.072) 

 conch at site Cl; however, hatchery and wild conch 



