768 
Fishery Bulletin 96(4), 1 998 
(Cummings and McClellan 3 ). Movement patterns 
inferred from mark-and-release experiments carried 
out between 1959 and 1994 (Cummings and 
McClellan 3 ) are consistent with the two-stock hypoth- 
esis. In brief, over 1400 recaptures from approxi- 
mately 14,000 releases revealed cyclical tag-return 
patterns that suggested resident stocks or subpopu- 
lations of greater amberjack along both the eastern 
coast of Florida and the northern Gulf. Exchange 
rates between the two stocks was estimated as ap- 
proximately 1.5% by Cummings and McClellan 3 al- 
though, as noted by these authors, the rate estimates 
were not adjusted for fishing pressure or for poten- 
tial biases due to mortality, tag shedding, lack of re- 
porting, and fishing effort. It also was clear from a 
few tag returns that greater amberjack can migrate 
considerable distances, e.g. from near Charleston, 
South Carolina, to Texas or from northwest Florida 
to Virginia (Cummings and McClellan 3 ). 
Biological information on greater amberjack is lim- 
ited to studies reported in Berry and Burch (1978), 
Shipp (1986), Manooch (1988), and GMFMC (1989). 
Thompson et al. 4 presented data on greater amber- 
jack age, growth, and reproduction, and Cummings- 
Parrack 1 and McClellan and Cummings 2 summa- 
rized most of the available information on landings 
and other fishery statistics. Direct or indirect infor- 
mation on genetic stock structure is even more lim- 
ited. Johnson 5 carried out a pilot study of nuclear- 
gene (allozyme) variation among 225 greater amber- 
jack sampled from the Atlantic (zz =60 ), eastern Gulf 
(n=84), and western Gulf ( /z =8 1 ). Of 72 putative loci 
examined, only one polymorphic (and nonin- 
formative) system was found. On the surface, these 
data do not support the concept of separate stocks. 
However, genetic homogeneity ( sensu stricto) does not 
unequivocally establish the existence of a single 
breeding population (stock), but rather is simply con- 
sistent with the hypothesis that samples are drawn 
from a population with the same parametric allele 
frequencies. In addition, the almost total absence of 
variation effectively precluded rigorous testing of the 
null hypothesis (i.e. the interpretation of genetic 
homogeneity among samples is potentially compro- 
mised by virtue of the absence of significantly vari- 
able nuclear-gene loci). 
3 Cummings, N. J., and D. B. McClellan. 1996. Movement pat- 
terns and stock interchange of greater amberjack Seriola 
dumerili , in the southeastern U.S. Miami Laboratory, SE Fish- 
eries Sci. Center, Natl. Mar. Fish. Serv., Cont. MIA-95/96-14, 
Miami, FL, 24 p. 
4 Thompson, B. A., C. A. Wilson, J. H. Render, H. Beasley, and C. 
Cauthron. 1992. Age, growth, and reproductive biology of 
greater amberjack and cobia from Louisiana waters. Final 
Rep., Marfin Prog., U.S. Dep. Comm., Coop. Agreement 
NA90AA-H-MF722, 77 p. 
Alternatively, the apparent absence of genetic 
variation raises considerable concern about the ef- 
fective size of greater amberjack populations. Com- 
pared with other marine finfish (Smith and Fujio, 
1980; Waples, 1987; Bohlmeyer and Gold, 1991), lev- 
els of nuclear-gene variation in greater amberjack 
(as reported by Johnson 5 ) are low. Richardson and 
Gold (1993) examined mitochondrial (mt)DNA varia- 
tion among 59 greater amberjack sampled primarily 
from the west coast of Florida. Levels of mtDNA 
variation in greater amberjack were low in compari- 
son with red drum and several clupeid species (e.g. 
Atlantic menhaden), but higher than those found in 
black drum, red snapper, and red grouper (Camper 
et al., 1993; Gold et al., 1993; Richardson and Gold, 
1993). Estimates of long-term, effective female popu- 
lation size (computed directly from levels of mtDNA 
variation) paralleled levels of mtDNA variation, sug- 
gesting that effective (female) population sizes of Gulf 
greater amberjack were not atypically low. 
Concerns regarding greater amberjack fisheries in 
the Gulf and Atlantic include the following: 1) pre- 
sumed decreases in average individual size in both 
Gulf and Atlantic fisheries; 2) apparent declines in 
size of the presumed Atlantic stock; 3) a trend of de- 
clining yield in both commercial and headboat fisher- 
ies in the Gulf; and 4) apparent highly erratic recruit- 
ment where success of individual year classes is quite 
variable (Cummings-Parrack 1 ; Cummings and 
McClellan 3 ). These concerns have intensified as the 
economic importance of greater amberjack has grown 
(GMFMC, 1989; Cummings and McClellan 3 ). In this 
study, we employed variation in restriction sites in mi- 
tochondrial (mt)DNA of greater amberjack to determine 
if significant population structure (separate genetic 
stocks) occurs in U.S. waters, i.e. in the northern Gulf 
of Mexico and along the U.S. southeastern Atlantic 
coast. The rationale for this study is the need for 
accurate geographic definition when conducting stock 
assessments (Hilborn, 1985; Sinclair et al., 1985), in 
this case for greater amberjack in U.S. waters. 
Materials and methods 
Appropriate tissues (heart and white muscle) were 
obtained from a total of 444 greater amberjack 
sampled from 11 offshore localities in U.S. waters 
(Table 1; Fig. 1). With exception of a sample of seven 
individuals from near Gulfport, MS, sample sizes 
5 Johnson, A. G. 1990. Progress report: electrophoretic exami- 
nation of greater amberjack ( Seriola dumerili). Panama City 
Laboratory, SE Fish. Sci. Center, Natl. Mar. Fish. Serv., Panama 
City, FL, 34 p. 
