Schwenke and Buckel: Age, growth, and reproduction of Coryphaena hippurus 
83 
Materials and methods 
Collections 
Dolphinfish from recreational fishery sources were 
obtained every month between May 2002 and May 2004 
(except December 2002 and 2003, January 2003, and 
February 2004) from fishing ports in North Carolina. 
Recreational anglers typically fished for dolphinfish 
in waters associated with the western wall of the Gulf 
Stream. In the summers of 2002-03, samples of large 
fish were provided through various sportfishing tourna- 
ments held in these same areas. To supplement length- 
at-maturity data once peak spawning was identified, 
maturity staging was conducted on male and female 
dolphinfish from April 2005 through July 2005. 
Sampling of commercial dolphinfish catches was done 
in addition to recreational dolphinfish sampling and 
was primarily conducted in the winter months to in- 
crease the sample sizes available for this time period. 
Small dolphinfish were not readily available through 
recreational and commercial sampling; therefore sample 
sizes were augmented by two different methods. First, 
a total of four fishery-independent trips were made in 
August 2003 and July 2004. During these trips, the 
distance traveled offshore averaged 20 km, and small 
lures were trolled, as opposed to large dead bait or 
large lures as is done in the recreational and com- 
mercial fishery. Second, small and intact dolphinfish 
were obtained from stomachs of larger dolphinfish and 
yellowfin tuna (Thunnus albacares) caught by anglers 
from recreational charter boats. 
Dolphinfish were measured to the nearest mm for 
fork length (FL) and total length (TL), sex was deter- 
mined (through macroscopic examination of the gonads), 
and the fish were weighed (to the nearest 0.1 kg) and 
tagged. Date and location of port sampled were recorded 
for each dolphinfish. Scale samples were collected be- 
fore the fish were filleted according to methods estab- 
lished by Beardsley (1967). In some instances, filleted 
dolphinfish carcasses were only available; therefore 
scale samples were not obtained on all sampled fish. 
All tagged carcasses were brought to the laboratory for 
extraction of otoliths and gonads. 
Age and growth 
To determine if daily rings were present on sagittal 
otoliths of age-0 dolphinfish, the otoliths were removed, 
cleaned, and stored dry until mounted in epoxy resin. 
To avoid interotolith variability, only the left otolith 
was used for reading. Otoliths were prepared for read- 
ing following methods described for transverse sections 
in Secor et al. (1992). Reading was done with a light 
microscope equipped with a digital camera. The image 
from the camera was transmitted to a computer and 
examined by using Image-Pro Plus software (Image- 
Pro, vers. 4.5, Media Cybernetics, Silver Spring, MD). 
Growth increments were counted from the core, begin- 
ning at the first clearly defined mark that encircled the 
primordium (Massuti et al., 1999), towards either the 
dorsal or ventral edge, depending on ease of counting. 
To determine the precision of the readings of juvenile 
dolphinfish ages, blind readings of daily growth incre- 
ments were conducted twice by the same investigator. 
Error greater than 10% in reading precision for an indi- 
vidual otolith caused that otolith to be rejected. If error 
in reading precision was less than or equal to 10%, then 
the average between the first and second readings was 
taken as the final age. 
The deposition of increments in dolphinfish otoliths 
begins on the hatching date, and rings are laid down 
daily (Uchiyama et al., 1986; Massuti et al., 1999). 
Thus, no adjustment was required to estimate age from 
incremental counts of sagittae, and it was assumed that 
rings were formed daily. Previous studies on the mi- 
crostructure of sagittal otoliths of dolphinfish from the 
western Mediterranean Sea had found that the daily 
ages from larger dolphinfish (>650 mm FL) appeared to 
be underestimated (Massuti et al., 1999). Furthermore, 
daily ages of dolphinfish have been validated to a size 
of 554 mm FL (Uchiyama et al., 1986). Therefore, our 
analysis was restricted to dolphinfish less than or equal 
to 650 mm FL. To determine individual dolphinfish 
growth rates, the fork length at capture was divided 
by the daily age. 
The annual age of dolphinfish was estimated with 
scales. Eight to ten scales were mounted, sculptured 
side down, on sheets of cellulose acetate 0.5 mm thick, 
and then placed on a scale press to make impressions. 
Scale impressions were examined with a microfiche 
reader at 32 x magnification to permit detection of cir- 
culi, annual marks, and other features of the scale. 
Age groups were classified according to the number of 
annual marks present (see Beardsley, 1967, for a figure 
of an annual mark on a dolphinfish scale). 
To determine the precision of dolphinfish age esti- 
mates, blind readings of annual marks on scales were 
conducted twice by the same investigator. If agreement 
between the first and second reading was not 100%, 
then a scale was reread a third time and was only used 
in the analyses if the third reading agreed with either 
the first or second reading. Additionally, blind read- 
ings of a subsample (n = 50) of dolphinfish scales were 
conducted by an independent reader who was trained to 
identify annual marks on dolphinfish scales. 
To validate annual marks in dolphinfish scales, an in- 
direct validation based on marginal increment analysis 
was used. Marginal increment widths were determined 
by measuring the distance from the outer edge of the 
scale to the closest annual mark. Marginal increment 
width was measured only on dolphinfish with one an- 
nual mark in order to standardize the method, and 
because the majority of dolphinfish aged with annual 
marks were age-1. Measurements (mm) were taken from 
the magnified (32x) scale image on a microfiche reader 
along a straight line from the lateral edge of the scale 
to the outermost annual mark by using a digital caliper. 
Marginal increment widths were analyzed by analysis 
of variance (ANOVA) to test for an effect of month. 
