Anderson et al : Current velocity and catch efficiency in sampling settlement stage larvae of coral-reef fisties 



407 



Family 



Figure 2 



Abundance of 25 families of larvae that constituted at least 1% of the catch 

 by light traps (mean no. larvae per sampling date) or channel nets (mean no. 

 larvae per 1000 m'') at (Al both study sites combined, (B) S.E. LSI (southeast 

 Lee Stocking Island), and (C) So. Bock (South Bock Cay). Error bars are 1 SE. 



of 2111 larvae were collected from light traps (n=849; 78 

 samples) and channel nets (n=1262; 26 surface and 26 sub- 

 surface samples), representing 20 and 33 famihes, respec- 

 tively (Table 1), combining blenniids and labrisomids as 

 blennioids, and not including clupeid and atherinid fishes. 

 Synodontids were excluded from further analysis because 

 of the presence of large postsettlement individuals in light 

 traps. For families that constituted at least 1% of the total 

 catch by at least one of these sampling methods, carap- 

 ids, chaetodontids, gobiesocids, and holocentrids were 

 not caught in channel nets. Conversely, carangids, chlo- 

 psids, congrids, muraenids, ophichthids, ophidiids, and 



tetraodontids were not represented in light-trap samples. 

 Channel nets had higher family richness and diversity at 

 S.E. LSI, So. Bock, and at both sites combined (Table 2) 

 than did light traps. 



There was no concordance in the rank order of abun- 

 dance of families collected by light traps and channel 

 nets at either S.E. LSI (r^=-0.358, n=25, P>0.05), So. 

 Bock (Spearman's rank correlation coefficient: r, =-0.120, 

 n=25, P>0.05), or at both sites combined (r^=-0.162, 

 n=25, P>0.05), including all families that represented at 

 least 1% of the catch of either sampling device (Fig. 2). 

 Similarly, there was no correlation in rank abundance of 



