626 



Fishery Bulletin 91(4). 1993 



complete discrimination. Snappers are an important 

 component of both the recreational and commercial 

 fisheries in the Southeastern United States, Gulf of 

 Mexico, and Caribbean. Analysis of the distribution of 

 eggs and larvae for life history and recruitment stud- 

 ies of the fishery has been hampered by the inability 

 to identify eggs and larvae of most snapper species. 

 Our results significantly increase the number of eggs 

 and larvae that can be unambiguously identified. In- 

 complete separation by our methods for all thirteen 

 species is due to sharing of identical haplotypes be- 

 tween some species, which could be caused by possible 

 occurrence of natural hybrids between snapper species 

 (Domeier and Clarke, 1992). The lack of between spe- 

 cies resolution may also be attributed to the fact that 

 the amplified fragments are too short for RFLP analy- 

 sis, since the number of restriction sites is subjected to 

 the fragment length. 



Average sequence divergence observed between the 

 snapper species is also low compared with those be- 

 tween congeneric species of other fish taxa where the 

 entire mtDNA molecule was analyzed (see Billington 

 and Hebert, 1991). Bartlett and Davidson (1991) using 

 direct nucleotide sequence of 307 bp cytochrome b gene 

 fragment found several distinct nucleotide substitu- 

 tions between four tuna species of the genus Thunnus. 

 Since many of the nucleotide substitutions observed 

 between these tuna species are found to create or de- 

 stroy palindromic sequences, PCR-RFLP analysis on 

 the 355 bp cytochrome b gene fragment alone could 

 identify these tuna species using only a few endonu- 

 cleases (Chow and Inoue, 1993). They also found that 

 RFLP analysis of cytochrome b and 12S rRNA gene 

 fragments provided insufficient variation to separate 

 all eight species of the genus Thunnus, but all tuna 

 species could be separated using a 940 bp fragment of 

 flanking region between ATPase and cytochrome oxi- 

 dase subunit III genes. Similar analysis on the cyto- 

 chrome b gene fragment could separate six billfish Pa- 

 cific species and five Atlantic species, but separation of 

 all species was incomplete (Chow 1 ). 



Although it appears that RFLP analysis may fail to 

 pick up distinct nucleotide substitutions between spe- 

 cies, analysis on longer fragments may overcome this 

 disadvantage. Thus, differentiation between closely re- 

 lated species appears to vary among regions within 

 the mtDNA and in a given gene may vary among taxa. 

 Recently, Silberman and Walsh (1992) using PCR-RFLP 

 analysis on the 28S ribosomal RNA gene simply and 

 unambiguously distinguished three spiny lobster spe- 

 cies of the genus Panulirus. Nuclear rDNA may be 



'Chow, S. 1992. Identification of billfish species using mitochondrial 

 cytochrome b gene fragment amplified bv polymerase chain reac- 

 tion. ICCAT Working Doc. SCRS/92/66. 



better suited for species descrimination, where little 

 intraspecific polymorphism is expected (Dover, 1982). 



We have observed relatively high intraspecific poly- 

 morphism in some snapper species, in which the 

 nucleon diversity (h) of L. campechanus, L. cyanopterus, 

 L. griseus, and L. jocu are 70, 67, 42, and 83%, respec- 

 tively. Gyllensten and Wilson (1987), using restriction 

 analysis on the entire mitochondrial DNA molecule, 

 reported nucleon diversity of brown trout (Salmo trutta) 

 to be 72%. Carr and Marshall ( 1991 ), using direct nucle- 

 otide sequence of a 298 bp cytochrome 6 gene, reported 

 that of the western Atlantic and Norwegian popula- 

 tions of Atlantic cod (Gadus morhua) to be 36 and 

 88%, respectively, comparable with those of some snap- 

 per species. It is highly probable that some snapper 

 species contain a large number of maternal lineages 

 within the population, which can be detected by PCR- 

 RFLP analyses as demonstrated in this study. 



The present study indicates that PCR-RFLP analy- 

 sis is much simpler and less expensive than conven- 

 tional mtDNA and nucleotide sequence analyses and 

 better suited for large-scale genetic surveys requiring 

 large numbers of specimens. With this method, not 

 only are quantitative analyses of species composition 

 at early life history stages possible, but also intensive 

 genetic analysis. Since relatively little is known about 

 the spawning habits, dispersal, and migration of these 

 snapper species, it would be very interesting to com- 

 pare, for example, genetic structures between adults 

 and their early life stages. Such genetic analyses would 

 aid in evaluating the utility of management schemes, 

 such as reserves and protected areas. 



Acknowledgments 



We thank D. Benetti and M. Domeier, University of 

 Miami, for their kind help in collecting snapper 

 samples. Thanks are also due to C. R. Robins for giv- 

 ing us the opportunity to use valuable specimens from 

 the Ichthyological Museum of the University of Mi- 

 ami, and to J. D. Silberman for his superb technical 

 advice and for commenting on an earlier draft of this 

 manuscript. Research was supported by NOAA/CIMAS 

 contract No. NA90-RAH-00075. 



Literature cited 



Allen, G. R. 



1985. FAO species catalogue. Snappers of the world. 

 An annotated and illustrated catalogue of lutjanid 

 species known to date. FAO Fish. Synop. No. 125, 

 vol. 6, 208 p. 



