586 



Fishery Bulletin 105(4) 



On many occasions over several years while scuba 

 diving on hard-bottom habitat in the study area in 

 depths from 25 to 30 m, I have observed red porgy of 

 the size that dominated the distribution of those col- 

 lected in the present study, but virtually none of the 

 size in the left-hand tail (smaller fish) of the distribu- 

 tion from the NMFS video survey (Fig. 1). Similarly, 

 individuals smaller than that typically seen in hook- 

 and-line samples have rarely, if ever, been observed by 

 personnel from the Panama City NMFS Laboratory, 

 either from ROVs or stationary camera arrays at many 

 sites in the N.E. Gulf of Mexico. 



Lastly, juvenile red porgy have been found primarily 

 on sandy bottom and typically in shallower waters than 

 those inhabited by adults (Manooch and Hassler, 1978; 

 Labropoulou et al., 1999). These findings, together with 

 the scuba and video observations, support the validity 

 of the assumption that the range of sizes of red porgy 

 in a given collection in this study were the same or at 

 least very similar to those actually present at the site 

 at the time the sample was taken. 



If the size structure of the hook-and-line samples in 

 this study accurately reflect the true size distribution 

 at the site at the time of the collection, then the sex 

 ratios of those samples should also accurately reflect 

 true values, and hence the finding of no evidence of sex 

 selectivity is also valid. The absence of behavior-related 

 size or sex selectivity in hook-and-line samples means 

 that any evidence of truncation in size structure or 

 skewing of sex ratios in exploited red porgy populations 

 should not be attributed to greater aggression or "hook 

 attraction" in males. Instead, such size truncation or 

 skewing of sex ratios can be easily explained as the 

 result of simple size-selective harvesting caused by the 

 targeting behavior of fishermen, not by the behavior of 

 the fish, and is a common phenomenon not unique to 

 protogynous fishes. 



Although there is historic evidence of such size-se- 

 lective harvesting in red porgy fisheries off the south- 

 eastern United States, there is none indicating that 

 behavior-related selectivity contributed to the crash 

 of those fisheries in the 1980s. From the late 1970s 

 through the mid 1990s, annual mean weights of red 

 porgy in the southeastern United States were rough- 

 ly 20-30% less in the recreational headboat fishery 

 than in the commercial hook-and-line fishery (Fig. 2 in 

 Vaughan and Prager, 2002). Such size-selective harvest- 

 ing (taking larger individuals more likely to be male) 

 may have skewed sex ratios temporarily, but there was 

 no evidence of a long-term effect, i.e., an increasingly 

 female-biased sex ratio. Males composed 15.8% 13.3%, 

 and 20.4%' of red porgy taken in a fishery-indepen- 

 dent survey in U.S. Atlantic waters during 1979-81, 

 1988-90, and 1991-94, respectively (Harris and McGov- 

 ern, 1997). The increasingly female-biased sex ratios ob- 

 served in gag catches in the 1980s and 1990s have been 

 attributed to behavior-related selectivity (Coleman et 

 al., 1996; McGovern et al., 1998). An increasing trend 

 in the proportion of males in smaller size classes off 

 the southeastern United States indicates that red porgy 



compensated for the effects of fishery-driven selectivity 

 for size by changing sex at smaller sizes (Harris and 

 McGovern, 1997). 



The targeting behavior of fishermen has produced 

 selectivity in red porgy fisheries in the GOM, as well. 

 Mean lengths of recreationally and commercially 

 caught red porgy from the Florida Middle Grounds and 

 off west central Florida have differed significantly (321 

 vs. 350 mm TL) (Hood and Johnson, 2000), in a man- 

 ner similar to the pattern in weight differences in the 

 Atlantic fisheries noted by Vaughan and Prager (2002). 

 It is unlikely these length differences are related to 

 fish behavior. More likely they reflect gear selectivity 

 as well as spatial differences in areas fished. DeVries 

 (2006) found persistent, significant, small-scale (<10 

 km) spatial differences in size structure and many 

 other demographics and life history traits of red porgy 

 in the N.E. GOM — differences that indicate that the 

 species has a complex subpopulation or metapopulation 

 structure. Further analysis of data from Hood and 

 Johnson's (2000) study indicated that sex ratios also 

 differed between the two fisheries: males composed 

 40.5% of the recreational and 49% of the commercial 

 samples, although the sex of 78 of the 274 fish col- 

 lected from the latter samples was not determined and 

 these 78 fish were not included in the calculations. 

 These differences in sex ratios likely reflect, and are 

 consistent with, the differences in size structure. Ad- 

 ditionally, as in the Atlantic, there was no evidence 

 that sex ratios have become increasingly female bi- 

 ased. In 1978 and 1979 males composed 30%- and 38%, 

 respectively, of red porgy sampled from charter boat 

 and headboat catches in northwest Florida (Salomon 

 and Fable, 1981). Two decades later Hood and Johnson 

 (2000) found that males composed 40.5% of recre- 

 ational and 49% of the commercial samples, and in 

 the present study the percentage of males averaged 

 39% from five of the sites and 66% at the remaining 

 two sites. 



Size or sex selectivity resulting from fish behavior 

 does not appear to characterize the hook-and-line fish- 

 eries of all exploited protogynous species. Besides the 

 apparent absence, in the case of red porgy, of selection 

 for males with the use of hook-and-line gear, there is 

 also no indication of such selection in the hook-and-line 

 fishery for the red hind (Epinephelus guttatus), a ser- 

 ranid (Shapiro et al., 1993; Coleman et al., 1996). It 

 may be that rapid skewing of sex ratios in protogynous 

 fishes caused by behavior-related size or sex selectivity 

 is the exception rather than the rule. The evidence that 

 aggressive behavior in large, primarily male scamp and 

 gag can lead to sex selection in their fisheries and rap- 

 idly skew sex ratios (Gilmore and Jones, 1992; Coleman 

 et al., 1996) is primarily circumstantial or observational 

 and there have been no controls or statistical tests. Al- 

 though the case for behaviorally caused sex selectivity 

 in gag is fairly convincing, more statistically rigorous 

 and direct testing of the hypothesis would strengthen 

 these conclusions. Until more studies, like the present 

 one, are conducted on other species, the prevalence of 



