PETKKSON ET AL: MARK-RECAPTl'RE TEST OF HARD CLAMS 



APPENDIX 



Two tests were performed to estimate the efficiency 

 and size selectivity of our hydraulic dredge sampling. 

 First, two sets of quantitative samples were taken at 

 Johnson Creek by placing a circular 0.25 m 2 sampling 

 frame at haphazard locations and then excavating it 

 to a depth of 15 cm. For one set of 24 samples, we 

 used the hydraulic dredge, while we excavated the 

 other set of 10 samples by hand. In each case, the con- 

 tents of the top 15 cm were passed through a 3 mm 

 mesh sieve. By comparing the average hard clam den- 

 sities and size-frequency distributions in these two 

 sets of samples, we have one test of whether the 

 efficiency and size selectivity of samples from the 

 hydraulic dredge differ significantly from analogous 

 hand-collected samples, all collected in the actual 

 field site. 



As a second test of the size selectivity and as a quan- 

 titative estimate of sampling efficiency, 22 marked 

 Mercenaria mercenaria were placed at natural living 

 depths within an otherwise undisturbed bottom in- 

 side our 0.25 m 2 sampling frame. The lengths of these 

 clams ranged from 0.89 to 9.53 cm, with 5 in the 0-2 

 cm range, 3 in the 2-4 cm range, 3 in the 4-6 cm range, 

 7 in the 6-8 cm range, and 4 in the 8-10 cm range. We 

 then used the hydraulic dredge to sample this 0.25 m 2 

 area within the frame in the usual fashion to a 15 cm 

 depth. This trial was repeated five times, moving the 

 frame to a new location each time and recording the 

 numbers and sizes of all clams recovered. 



Average density of hard clams did not differ sig- 

 nificantly (at a— 0.05 in a Student's /-test) between 

 the dredged and hand-collected samples from John- 

 son Creek (Appendix Table 1). Furthermore, the 

 size-frequency distributions (Appendix Table 1) 

 were nearly identical and did not differ significantly 



(at a = 0.05 in a x 2 contingency test). These results 

 imply that the two techniques did not differ in ef- 

 ficiency or size selectivity. Mercenaria mercenaria as 

 small as 0.5 cm long were collected by both tech- 

 niques. In the five trials to estimate the numerical 

 efficiency of the dredging technique, only one clam 

 was missed (5.84 cm long). Thus, the capture ef- 

 ficiency exceeded 99 % and did not vary significantly 

 with clam size within the range of clams used (0.89- 

 9.53 cm). This result implies that we did not collect a 

 biased size (or age) distribution of hard clams in our 

 field sampling. 



APPENDIX TABLE 1. — A comparison of the relative efficiency 

 and size selectivity of hydraulic dredge and hand sampling of 

 hard clams in Johnson Creek, Core Sound, N.C. Also given are 

 results of quantitative estimates of capture efficiency as a func- 

 tion of clam size for the hydraulic dredge technique. 



a) Johnson Creek sampling results 



'Difference not significant at a — 0.05 in /-test. 

 difference not significant at a = 0.05 in x : test. 



779 



