Stevenson and Secor; Growth of Aapenser oxynnchus 



155 



(« = 114). Sagittae were cleaned in 107c 

 bleach, rinsed several times with deion- 

 ized water, and air-dried. One sagittal oto- 

 lith from each pair was embedded in Spurr 

 epoxy and sectioned as described in Secor 

 et al. (1991). Owing to their fragility, oto- 

 lith sections were polished by hand with a 

 variety of fine sandpapers and a 0.03-pm 

 alumina slurry. 



Annuli in thin sections of fin spines 

 and otoliths were viewed under reflected 

 light at 15x magnification by two experi- 

 enced readers. An annulus was defined as 

 a bipartite zone comprising an opaque and 

 a translucent zone (Fig. 1 ). The first trans- 

 lucent zone was counted as the beginning 

 of the first year of life. In some instances, a 

 secondary fin spine was embedded within 

 the primary fin spine (Feindeis, 1997), in 

 which case care was taken to enumerate 

 annuli only in the primary spine. False 

 annuli were consistently observed in fin- 

 spine sections of older individuals and 

 were excluded from annulus counts. These 

 structures were not continuous around the 

 entire circumference of the section and 

 were thus distinguishable from annuli to 

 be counted (Prince et al., 1985). 



Precision and bias 



Readers counted annuli without knowing 

 collection date, fish size, or previous age 

 determination. They were trained with 

 the aid of an imaging system that permit- 

 ted simultaneous observation of annular 

 growth zones. Paired difference tests were 

 used to statistically evaluate bias and pre- 

 cision among readers in a single blind test. 

 The coefficient of variation (CV) was used 

 to measure precision; bias was assessed 

 visually using age-bias plots (Fig. 2). An 

 age-bias plot showed paired estimates of 

 age for the same fish (Campana et al., 

 1995), with the estimates of reader 2 rep- 

 resented as mean age, and 95';'f confidence 

 intervals corresponding to each of the age 

 classes estimated by reader 1. For exam- 

 ple, if reader 1 estimated five fish to be 15 

 years old, the age-bias plot indicated the 

 mean age of those five fish as estimated 

 by reader 2. Divergence from the equiva- 

 lence line, where A^e^,,,,/,, , = Age„„,^„. ^_, 

 indicates a systematic difference between 

 readers. Paired age estimates for either fin- 



translucent 

 zones 



Figure 1 



Backscatter electron micrographs show (A) a sagittal otolith section; 

 IB) a pectoral-fin spine (lower lobe only) from a Hudson River Atlantic 

 sturgeon; and (C) a fin-spine section from a hatchery-reared Atlantic 

 sturgeon. White arrows (B) indicate annuli. Organic deposits in the fin- 

 spine section are indicated with dark arrows (B). Pits resulting from 

 chemical microanalysis are visible in the otolith section. Dashed circle 

 in (A) indicates annuli that are difficult to interpret. 



