276 
Fishery Bulletin 112(4) 
sectioned to determine the most suitable aging struc- 
ture for Gulf sturgeon. Additionally, an archive of air- 
dried fin spines from Gulf sturgeon collected on the Su- 
wannee River in 1980-90 was donated for inclusion in 
future age and growth analysis (K. Sulak, U.S. Geologi- 
cal Survey). Fin spines from sturgeon collected in the 
Pascagoula River likewise were processed for future 
studies (M. Peterson, Univ. Southern Mississippi). In 
many cases, the second marginal pectoral-fin ray was 
sampled along with the fin spine: these samples were 
used to compare band counts between these structures. 
Processing of structures 
Fin spines, fin rays, and otoliths The entire left pecto- 
ral, left pelvic, dorsal, and anal fins were removed at 
the point of articulation from each of the 14 archived 
Gulf sturgeon. With a scalpel, the fin spine and ap- 
proximately 5 leading marginal rays were isolated from 
each fin. Excess tissue was removed from between each 
hemitrich of the fin rays with a stiff brush and for- 
ceps. In extreme cases, samples were boiled (2-5 min) 
to ease cleaning. Air-dried fin rays and small, fragile 
spines (<5 mm in diameter) were mounted in Loctite 
0151 Hysol 3 epoxy (Henkel Corp., Rocky Hill, CT) and 
allowed to harden for 24 hours. Large fin spines (>5 
mm in diameter) were not mounted in epoxy because 
of their rigid nature; to simplify mounting and obtain 
the best sections, the articulating process of the fin 
spine was removed by using a Hillquist Thin Section 
Machine (Hillquist Inc., Denver). 
Through the use of an IsoMet low speed saw 
(Buehler, Lake Bluff, IL) with a Norton grinding wheel 
(Saint-Gobain, Courbevoie, France), a minimum of 3 
transverse sections approximately 0.3-0. 6 mm thick 
were made starting at the proximal end of the fin spine 
or ray. Sections were mounted on microscope slides 
with the clear mounting medium Cytoseal (Thomas 
Scientific, Swedesboro, NJ) and examined under a dis- 
secting microscope at magnifications of 10-75x that 
varied depending on the size of the section. Transmit- 
ted or reflected light was used on the basis of which 
one best elucidated band formations. Digital images 
of each section were taken with a MicroPublisher 5.0 
real-time viewing camera (Qlmaging, Surrey, Canada) 
mounted on an RZ Series CMO stereo microscope (Mei- 
ji Techno America, Santa Clara, CA). Growth bands 
were counted by using standard aging methods (Bren- 
nan and Cailliet, 1989; Morrow et al., 1998; Stevenson 
and Secor, 1999), whereby a band pair consisted of one 
opaque and one translucent ring that was continuous 
around the section. 
Both sagittal otoliths were removed from the 14 ar- 
chived Gulf sturgeon, cleaned in fresh water, and al- 
lowed to air dry. One otolith from each sturgeon was 
3 Mention of trade names or commercial companies is for iden- 
tification purposes only and does not imply endorsement by 
the National Marine Fisheries Service, NOAA. 
mounted in 2 Ton Epoxy (ITW Devcon, Danvers, MA) 
and sectioned according to Bruch et al (2009). Sections 
(0.5 mm) were mounted on a microscope slide with Cy- 
toseal mounting medium. Digital images were taken of 
each otolith section according to the method described 
previously for fin rays. The second whole otolith was 
stored dry and archived. 
Each structure was evaluated qualitatively on the 
basis of 3 criteria: 1) invasiveness of the removal; 2) 
practicality of sampling in the field; and 3) consistency 
and clarity of the banding pattern. The structure for 
which removal was the least harmful, while also pro- 
viding consistent banding patterns, was considered the 
“best” aging structure for Gulf sturgeon. 
Precision and bias To measure reproducibility and 
test whether the second marginal pectoral-fin ray was 
a comparable aging structure to the commonly used 
fin spine, average percent error (APE) (Beamish and 
Fournier, 1981) between band counts for each structure 
was calculated when both were available: 
i lx — X I 
APE: — 100% x—Y'f, I — i, (1) 
J R Xj 
where Xy = the ith age determination of the jth fish; 
Xj = the mean age of the yth fish; and 
R = the number of times each fish was aged. 
Age estimates between the fin ray and fin spine were 
measured for precision with the coefficient of variation 
(CV) (Chang, 1982): 
CV: = 100% x 
Xj 
where CVj = the age precision estimate for the yth fish. 
Average percent error was calculated for each fish for 
which samples of both a fin spine and a second mar- 
ginal fin ray existed, and the results were presented as 
an index. Similarly, CV was averaged across all sam- 
ples to produce a mean value. Percent agreement (PA) 
of age estimates from fin rays and fin spines was also 
calculated. Gulf sturgeon without a fully formed band 
pair (band count=0) were not included in analyses be- 
cause their inclusion artificially inflated the APE, and 
CV could not be calculated for those fish. Precision and 
bias of band counts for each structure was examined by 
using a 1:1 bias plot. 
Field proofing After the best aging structure had been 
determined, a protocol was developed for removal of fin 
rays from live sturgeon. Fin rays were collected oppor- 
tunistically from Gulf sturgeon during ongoing collab- 
orative gillnet surveys conducted by researchers from 
Delaware State University, the University of Southern 
Eh 
X: 
( 2 ) 
