Stevenson and Secor Growth of Aapenser oxynnchus 



157 



persive (WDS) electron microprobe at the Center for 

 Microanalysis, College Park, MD. Phosphorus was 

 also measured because it makes up a large fraction 

 of the fin spine's hydroxyapatite structure ( Steven- 

 son, 1997). Accelerating voltage was 25 kV and cup 

 current was 20 nA. Each measurement represented 

 rastering or scanning by the probe over a 5 x 5 jam 

 area of the hardpart section. Molar weights of cal- 

 cium and phosphorus were measured in peripheral 

 regions of fin spines for 20 fish aged 3-18 years (fin- 

 spine estimate) collected in March, June, September, 

 and November. The mean of three peripheral points 

 was calculated for each individual, and means were 

 contrasted by month of collection with ANOVA. 



Microprobe measurements were used to construct 

 elemental chronologies of 15 fin spine sections taken 

 from sturgeon aged 12-36 years (fin-spine estimate) 

 and 164-236 cm TL. For each chronology, a series of 

 point measurements was taken along an axis that 

 traversed several annuli, with five points per annu- 

 lus; points were assumed to sample seasons in linear 

 proportion. 



A solution of OTC (25 milligrams per kilogram of 

 body weight) was injected into the dorsal muscula- 

 ture of five juvenile laboratory-reared fish of known 

 age (55—65 cm TL) from Hudson River broodstock. 

 Juveniles had been reared in circular fiberglass 

 tanks in a recirculating system and fed a mixture 

 of commercial pellet feeds. They were subjected to a 

 twelve-hour photoperiod; water temperature ranged 

 from 15° to 20°C. Three months after injection, the 

 first fin spine section was removed from three fish 

 with a jeweller's saw, and silver nitrate was applied 

 to encourage clotting of the wound. Fifteen months 

 after injection, a second section was removed from 

 the opposite pectoral-fin spine of one fish only. Fin- 

 spine segments were dried, sectioned, mounted on 

 glass slides, and polished. Thin sections were viewed 

 with epifluorescent microscopy to identify OTC marks. 

 The position of OTC marks with respect to opaque 

 and translucent zones was recorded and a micro- 

 graph was taken. In all cases, the reader was aware 

 that the fish had received an injection of OTC but 

 had no knowledge of the date of section collection. 



Growth 



Reported measurements of dressed carcasses were 

 converted to total length based upon conversion met- 

 rics derived for Hudson River Atlantic sturgeon (Ste- 

 venson, 1997). Sturgeon included in the conversion 

 metric sample (n=235l were collected during spawn- 

 ing season in the Hudson River; all but five pos- 

 sessed mature gonads. Ages derived from fin-spine 

 and otolith sections were used to fit von Bertalanffy 



growth models by using a Marquardt iterative esti- 

 mation procedure for the three model parameters. 

 Length-at-age relationships were first examined to 

 determine variance structure and progression of 

 modal length with age. Growth parameters were 

 estimated iteratively for males and females with a 

 least-squares method. Because of high variance at 

 the point of growth inflection, it was unlikely that 

 a single growth model would fit all portions of the 

 growth curve. Therefore, the juvenile portion (42-152 

 cm TL) was modeled separately with a power func- 

 tion based upon the best fit of residuals. 



PC-SAS (SAS Institute, Inc., 1994) and Statgraph- 

 ics Plus (STSC, Inc., 1992) were used for all statisti- 

 cal tests. Data that did not satisfy the assumption 

 of heteroscedasticity (Bartlett's test, a=0.05) were 

 transformed to satisfy this assumption. Transformed 

 data that did not satisfy this assumption were ana- 

 lyzed with a Kruskal-Wallis nonparametric test to 

 examine differences among groups. 



Results 



Comparison of hard parts 



Otoliths were irregularly shaped and their annuli 

 were difficult to interpret. In contrasting several sec- 

 tioning planes, we observed that annuli on transverse 

 sections yielded the most consistent interpretations. 

 The first three to nine growth zones showed a clear 

 alternation of opaque and translucent zones. There- 

 after, translucent zones were irregularly spaced and 

 often appeared to overlap (Fig. lA). Low optical con- 

 trast between opaque and translucent zones reduced 

 the readers' confidence in assigning annuli, espe- 

 cially in sections with more than twenty annuli. 



In contrast, fin-spine sections exhibited concentric 

 narrow translucent zones and wide opaque zones 

 when viewed with transmitted light. Fin spines con- 

 tained a vascularized core and deposits of organic 

 material in lobe regions (Fig. IB). Interspersed were 

 fibrils that we interpreted as collagen or some other 

 structural protein. Annuli became narrower toward 

 the outer edge in larger (and presumably older) fish. 

 Secondary fin spines (84% of fin-spine sections) and 

 false annuli were observed but were simple to iden- 

 tify and disregard. Belts of two to five narrow annuli 

 were apparent in most female fin-spine sections (96% 

 of a subsample of 48). These belts were not apparent 

 in the juvenile sturgeon examined. 



Imprecision (CV) in age estimates was 4.8% between 

 two readers of the same fin-spine section (Fig. 2A). 

 Mean imprecision was 1.2 years (^=-1.97, P>0.05); 

 estimates by the two readers were not significantly 



