588 



Fishery Bulletin 101 (3) 



regulations^ should provide some measure of protection 

 for spawning ballyhoo in inshore waters. 



Our study design was limited to the presence and ab- 

 sence of spawning females and did not identify concentra- 

 tions of spawning activity associated with specific habitats. 

 Presumably submerged vegetation is an important micro- 

 habitat. Several authors have noted that hemiramphid 

 eggs, including those of ballyhoo, attach by filaments (of 

 the chorion) to vegetation such as Syringoditim fill forme 

 and Sargassum sp. in waters less than approximately 6 m 

 deep (Graham, 1939; Ling, 1958;Talwar, 1962, 1967; Berke- 

 ley and Houde, 1978). However, Berkeley and Houde ( 1978) 

 collected eggs in plankton tows. The specific importance for 

 halfbeak reproductive success of attached versus floating 

 vegetation, or no vegetation, has not been identified. 



The methods of this study define the macroscale spawn- 

 ing habitat of halfbeaks based on the distribution of spawn- 

 ing females. We demonstrate here that GSl values, even for 

 highly iteroparous species, can distinguish females with 

 hydrated oocytes from females in a less advanced stage 

 of oocyte development. The GSI value is simple and inex- 

 pensive to measure, and by including individual halfbeaks 

 for which we had GSI values but no histological data, we 

 more than tripled our sample size with little additional 

 laboratory cost. We could have instead characterized oo- 

 cyte development macroscopically and such a modification 

 is well suited when conditions affect weighing devices. 

 But macroscopic characterization of oocyte development 

 usually follows an ordinal scale that may vary between 

 observers. 



The distribution of females with hydrated eggs may be 

 a better indication of spawning habitat than the distribu- 

 tion of eggs because hydration occurs for only a few hours 

 (DeMartini and Fountain, 1981; Hunter and Macewicz, 

 1985; Brown-Peterson et al., 1988; McBride et al., 2002), 

 whereas egg dispersal may occur over several days. In 

 this study we assumed that spawning females move only 

 limited distances within the few hours of the hydration 

 process, and although limited movement has not been 

 documented for either ballyhoo or balao, we believe that 

 our interpretation of the data supports this assumption. 

 The size of the study area was approximately 200 km by 

 250 km, and it seems reasonable that spawning halfbeaks 

 were not moving extensively within this spatial boundary 

 on an hourly basis. The approach discussed in the present 

 study may meet the needs of other investigators wanting 

 to generate a first approximation of spawning habitats for 

 management purposes, which was the goal of this study. 

 Also, this approach has good potential for use in areas were 

 species identification of halfbeak eggs or larvae is problem- 

 atic (Noell et al., 2001). Analyses requiring a smaller area 

 or finer spatial resolution will depend on verification of a 

 hydration period that is short in relation to expected fish 

 movements. 



The specific example presented in our study was limited 

 because we collected the fish using commercial fishing 

 vessels on routine fishing operations. This was cost-effec- 

 tive, but wo were not able to identify spawning habitat 

 preference or to define the complete geographic extent of 

 the spawning grounds within south Florida. Gaps in the 



distribution of ripe females, which were particularly evi- 

 dent in the middle Florida Keys, were typically related to 

 gaps in sampling coverage. In addition, both species pre- 

 sumably spawn outside the area we sampled. Still, much 

 of the reported geographic range of ballyhoo and balao in 

 the western Atlantic Ocean has been covered in the pres- 

 ent study. The remaining shortcomings of this specific 

 example could be resolved by using this approach within 

 a statistically valid sampling design and estimating size- 

 specific batch fecundity to map reproductive rates within a 

 spatial and temporal context. The data resulting from such 

 a comprehensive sampling design would be well-suited for 

 identifying essential spawning habitat, for siting habitat- 

 specific investigations of spawning dynamics, or for validat- 

 ing dispersal models for early life stages of marine fish. 



Acknowledgments 



We are grateful to many individuals for assistance in this 

 research. First, to the fishermen and processors in the 

 halfbeak fishery, all of whom participated in this survey. T. 

 Brown, J. Hunt, and R. Moretti provided logistical support 

 in Marathon. R. Beaver, K. Krumm, E. Robillard, D. Snod- 

 grass, and J. Whittington assisted in fish collection and 

 processing. G. Gerdeman, P. Nagle, F Stengard, C. Stevens, 

 and P. Thurman assisted with tissue processing and repro- 

 ductive staging. C. Anderson assisted in preparing Figure 

 1 and GIS habitat analyses. B. Mahmoudi, R. Taylor, M. 

 Zimmermann, and two anonymous reviewers provided 

 constructive comments. Editorial assistance was provided 

 by J. Leiby and J. Quinn. This research was funded in part 

 by a grant from the National Oceanic and Atmospheric 

 Administration (NOAA) to the Florida Fish and Wildlife 

 Conservation Commission (Saltonstall-Kennedy Program, 

 NOAA award no. NA77FD0069). 



Literature cited 



Berkeley, S. A. and E. D. Houde. 



1978. Biology of two exploited species of halfbeaks, Hemir- 

 amphus brasiliensis and H. balao from southeast Florida. 

 Bull. Mar. Sci. 28:624-644. 

 Berkeley, S. A., E. D. Houde. and F. Williams. 



1975. Fishery and biology of ballyhoo on the southeast Flor- 

 ida coast. In Sea Grant Special Report 4, 1-15 p. Univ. 

 Miami Sea Grant Program, Coral Gables, FL. 

 Brown-Peterson, N., P. Thomas, and C. R. Arnold. 



1988. Reproductive biology of the spotted seatrout.CvMosf 10?! 

 nebiihsiis, in south Texas. Fish. Bull. 86:.373-88. 

 Collette, B. B. 



1965. Hemiramphidae (Pisces, Synentognathi) from tropical 

 west Afinca. Atlantidc Report 8:217-235. 

 DeMartini, E. E., and R. K. Fountain. 



1981. Ovarian cycling frequency and batch fecundity in the 

 queenfish, Seriphus politiis: attributes representative of 

 serial spawning fish. Fish. Bull. 79:547-60. 

 Graham, D. H. 



1939. Breeding habits of the fishes of Otago Harbour and 

 adjacent seas. Trans. Proc. Royal Soc. New Zealand 69:361 - 

 372. 



