90 
Fishery Bulletin 106(1 ) 
20 
(/) 
0 
1 10 
o> 
c 
'sz 
o 
CO 
f 0 
'o 
0 
CT -10 
£ 
LL 
-20 
^ ^ ^ <ysS cf V -^P 1 
Month 
Figure 5 
Frequency distribution of back-calculated hatching dates estimated 
from daily otolith increments for age-0 dolphinfish ( Coryphaena 
hippurus ) collected off North Carolina. The hatching dates were 
pooled by month for both 2002 (solid bars) and 2003 (hatched 
bars). 
Gulf Stream and associated waters off the 
North Carolina coast is variable enough to 
cause a decrease in growth rates during the 
winter (Rose, 1966); the same conclusion was 
reached for Florida dolphinfish (Beardsley, 
1967). It is unknown what proportion of the 
dolphinfish population inhabits these regions 
during winter and whether some dolphinfish 
are not exposed to environmental conditions 
that lead to annulus formation. 
It is surprising that the mean marginal 
increment width dropped during late sum- 
mer and fall in our study. The age-1 dolphin- 
fish sampled in the late summer and early 
fall were smaller than those collected in the 
spring and early summer months of the same 
year, and for that reason perhaps a different 
cohort was sampled (Schwenke, 2004). Sam- 
pling from different cohorts is likely given 
the highly migratory nature of dolphinfish. 
Marginal increment width is correlated with 
dolphinfish body size (Schwenke, 2004); there- 
fore, monthly differences in the mean fork 
length can affect the monthly marginal in- 
crement width. Wide variation in marginal 
growth in any given month was also noted by 
Beardsley (1967). 
The assumption of a winter-formed annulus 
was supported by evidence of deposition of a winter- 
formed annulus in scales from a dolphinfish that had 
previously been tagged. The dolphinfish was tagged 18 
October 2003 off Hudson Canyon (Mid-Atlantic Bight) 
at a fork length of 660 mm through the Dolphin Tag- 
ging Program run by Cooperative Science Services, 
Charleston, SC. The dolphinfish was recaptured on 16 
May 2004 off Morehead City, NC, at a fork length of 
864 mm and was estimated to be a one year old with 
a clearly defined annulus. This finding supports the 
assumption that dolphinfish lay down a new annulus 
in the winter. The likelihood that this dolphinfish had 
already laid down an annulus when tagged is not great, 
because 98% of the fish that were age-1 in our study 
had a fork length greater than 660 mm. 
To date, scales seem to be the most appropriate hard 
part to use to determine annual ages of dolphinfish 
because findings are comparable between studies where 
this structure was used. With scales used for age deter- 
mination, the maximum age of dolphinfish from Straits 
of Florida is four years (n = 511; Beardsley, 1967), three 
years off North Carolina (rc=738, Rose and Hassler, 
1968; n = 339, this study), and three years in the Medi- 
terranean Sea (n = 150, Massuti et ah, 1999). Longevity 
and first-year growth for dolphinfish in Florida (Beards- 
ley, 1967) and North Carolina (Rose and Hassler, 1968; 
this study) have greater similarity to longevity and 
first-year growth of western Mediterranean Sea dol- 
phinfish (Massuti et al., 1999) than to longevity and 
first-year growth of dolphinfish in other regions (Fig. 3). 
First-year growth of dolphinfish in the Gulf of Mexico 
(Bentivoglio, 1988) and the Caribbean (Oxenford, 1985; 
Rivera and Appeldoorn, 2000) is faster and maximum 
age is younger than corresponding measures for other 
regions (Fig. 3). Differences in estimated growth be- 
tween regions can be due to different laboratory meth- 
ods, genetics, or environmental conditions (i.e., water 
temperature, food availability, exploitation levels). For 
example, the aging of dolphinfish solely with otoliths 
is found to underestimate the age of older, larger fish 
(Massuti et al., 1999). Alternatively, regional differenc- 
es may represent different genetic stocks (reviewed by 
Oxenford, 1999). One of the first ways to make progress 
in identifying the factors responsible for this inter-re- 
gion variability in growth would be to standardize age 
and growth methods. 
Lengths of age-0 dolphinfish are highly variable and 
age-0 dolphinfish comprised the majority of the sampled 
population in this study. The large age-0 dolphinfish 
that were caught near the time of the theoretical hatch- 
ing date may have represented fall-spawned dolphinfish 
whose annual marks would not be discernible on scales. 
Previous age and growth studies on the U.S. east coast 
have relied solely on annual marks on scales for their 
age estimates, and for these studies all samples were 
obtained through fishery-dependent sources. Because 
dolphinfish do not become fully recruited to the fishery 
until -400 mm FL, past length-at-age-0 curves may 
be biased because smaller dolphinfish were not repre- 
sented. In our study, small dolphinfish obtained through 
fishery-independent sampling allowed for daily aging of 
dolphinfish, and thus reduced the variability associated 
with length-at-age of age-0 dolphinfish and provided an 
estimate of first-year growth rates. 
