Stoner and Davis: Outplanting queen conch. Strombus gigas 



397 



bers of untagged, wild conch at the CI area. Hatch- 

 ery stocks were recovered in about equal proportions 

 at the two sites. 



Growth Free-ranging wild conch had higher growth 

 rates (log-transformed) than hatchery-reared conch 

 during all three seasons examined (Table 4). During 

 summer, the difference was approximately two times 

 (Fig. 5), but the rates began to converge in the fall. 

 Growth rates were highest during summer and fall, 

 and lowest during winter, following patterns of wa- 

 ter temperature (Fig. 3). Conch grew significantly 

 faster at site C2 than at CI during both summer and 

 winter; site differences were not significant in the 

 fall (Table 4, Fig. 5). 



Morphology At the beginning of the free-ranging 

 experiment, shells of hatchery-reared conch were 

 significantly lighter than those of wild conch from 

 the Children's Bay Cay nursery site (slopes were 

 homogeneous, P=0.833, P=0.365; ANCOVA: P=92.62, 

 P< 0.001) (Fig. 6A). Lower shell weight in hatchery- 

 reared conch is a function of either thinner shells or 

 differences in shell form compared to wild conch. 

 Regressions of shell width with shell length (Fig. 6B) 

 showed that the spines were, in fact, longer in wild 

 conch than in hatchery-reared stock (slopes were ho- 

 mogeneous, F=l. 76, P= 0.190; ANCOVA: F=73.99,P< 

 0.001). Regressions of tissue wet weight with shell 

 length show no significant difference in tissue weight 



Table 4 



Results of two-way AN OVAs for growth rates in free- 

 ranging queen conch (Strombus gigas). "Site" refers 

 to the two experimental sites CI and C2. "Stock type" 

 refers to hatchery-reared versus wild conch. 



Source 



di 



MS 



Period I (April to September 1990—64 days) 



Site x stock type 1 <0.001 1.205 0.273 



Site 1 0.003 22.157 0.001 



Stock type 1 0.045 322.407 <0.001 



Error 396 <0.001 



Period II (September to November 1990—100 days) 



Site x stock type 1 <0.001 1.332 0.249 



Site 1 <0.001 0.418 0.518 



Stock type 1 0.004 16.967 <0.001 



Error 396 <0.001 



Period III (November 1990 to February 

 1991—66 days) 



Site x stock type 1 

 Site 1 



Stock type 1 



Error 115 



between wild and hatchery-reared conch (Fig. 6C), 

 (slopes were homogeneous, F=1.76, P=0.190; 

 ANCOVA: F=3.24, P=0.077). 



Measurements made on shells of hatchery-reared 

 and wild conch at the end of the experiment in Feb- 

 ruary 1991 show that lines for shell weight and width 

 had converged (Fig. 7, A and B). Shell weights of 

 hatchery-reared conch were still lighter than those 

 of wild conch (slopes were homogeneous, P=0.189, 

 P=0.665; ANCOVA: P=7.44, P=0.008) (Fig. 7A), but 

 the lines were closer than in April 1990 (Fig. 6A). 

 Stock type did not affect the relationship between 

 shell length and shell width in February (Fig. 7B) 

 (slopes were homogeneous, F=2.01, P=0.160; 

 ANCOVA: P=0.957, P=0.33 1 ). Hatchery-reared conch 



hatchery wild 



Site C1 



hatchery wild 



Site C2 



03 



E 



E 0.10 



Penod II 

 Sept- Nov 1990 



E 



E OlOf | 



a> too ^^_^^ ioo 



I "I Ml 



hatchery wild 



Sited 



hatchery wild 



Site C2 



Penod III 

 Nov 1990- Feb 1991 



hatchery wild 



Site C1 



hatchery wild 

 Site C2 



Figure 5 



Comparison of growth rates of free-ranging hatch- 

 ery-reared and wild queen conch, Strombus gigas, 

 at sites CI and C2. Growth periods I— III represent 

 summer, fall, and winter, respectively. Values are 

 mean ± SD, with the number of conch measured 

 shown inside the vertical bars. 



