450 



Fishery Bulletin 94(3), 1996 



females in our sample was slightly but significantly 

 larger than that of males. This is also consistent with 

 a small but significant difference in sex-specific 

 growth rates; however, this finding is confounded by 

 the significantly older median age of females, which 

 would also contribute to the observed difference in 

 the sex-specific size-frequency distributions. There 

 is no readily apparent biological explanation for the 

 slightly older age-frequency distribution of females 

 in relation to males. 



The only other estimates of von Bertalanffy growth 

 parameters for bonefish are by Morales-Nin (1994) 

 for 31 fish caught off Mexico's Pacific coast. She used 

 a length-frequency analysis to estimate Zf=0.275 and 

 L m = 473 mm; her estimate of K is close to our Florida 

 estimates, but her estimate of L^ is much less. Her 

 small sample size could have seriously biased her 

 parameter estimates. Furthermore, Pfeiler (1996) 

 presented allozyme data that suggested that an 

 undescribed species of Albula occurs off Mexico in 

 the Gulf of California. It is unclear whether A. vulpes 

 actually occurs off Mexico's Pacific coast or whether 

 Morales-Nin's growth parameter estimates apply to 

 A. vulpes or to Pfeiler's undescribed species. 



The estimated instantaneous total mortality rate 

 for bonefish was low. Our estimated range for natu- 

 ral mortality (0.2-0.3) included our total mortality 

 estimates, suggesting that there is little fishing mor- 

 tality. This was expected because the commercial 

 harvest of bonefish from Florida waters is prohib- 



ited and most bonefish caught by recreational an- 

 glers are released. The low estimate for total mor- 

 tality suggests that the mortality rate of bonefish 

 caught with hook-and-line gear and then released 

 was low, and that there was little mortality of bone- 

 fish as bycatch in commercial net fisheries operat- 

 ing in Florida Keys waters. 



Otolith weight is not a precise estimator of bone- 

 fish age, but it could be used to approximate age dis- 

 tributions in situations where annulus counts are not 

 available. Otolith weight accounted for more of the 

 variability in age (84.9%) than fish length (77.0^ for 

 males and 79.2% for females). The Florida Keys are 

 on the northern edge of the range of bonefish in the 

 western North Atlantic. In specimens from the more 

 tropical and less seasonal portions of the species' 

 range, annuli may be more difficult to count and thus 

 be less useful in age determination. This has been 

 observed in other elopomorph species. Crabtree et 

 al. (in press) reported that estimating age by sec- 

 tioned otoliths was more difficult with tarpon from 

 Costa Rican waters than it was with those from 

 Florida waters; they rejected as unreadable more 

 than twice as many otoliths from Costa Rican tar- 

 pon as those from Florida tarpon. 



We cannot explain the presence of the two excep- 

 tionally small, old, and sexually mature female bone- 

 fish in our sample. Variable growth rates and stunt- 

 ing of some early maturing females is a possible ex- 



