634 



Fishery Bulletin 101(3) 



8 10 12 14 16 



6 8 10 12 14 



Age (years) 



16 



Figure 6 



Prionace glauca growth curves and size-at-age data based on vertebral 

 band counts. LOWESS (locally weighted regression smoothing) curves 

 have been fitted to the data by sex: (A) males and (B) females. 



308 juvenile blue sharks collected in the Azores were used 

 to model early growth in this species. Silva et al. (1996) 

 calculated an annual growth rate of 30 cm/yr for the first 

 five years of life and aged the samples to seven years. 

 More recently, Henderson et al. (2001) used 159 vertebrae 

 sampled from blue sharks taken from oceanic waters off 

 Ireland. Like the previous two studies, the size range of 

 samples was limited to juvenile fish less than 191 cm FL 

 and the estimated ages ranged from 1 to 6 years. 



Stevens (1975), Silva et al. (1996), and Henderson et al. 

 (2001) modeled blue shark growth with the VBGF. These 

 curves are similar to each other (Silva et al., 1996, Hender- 

 son et al., 200 1 ), yet show slower growth than the current 

 study (Fig. 8) despite the fact that we used criteria similar 

 to those of Stevens ( 1975) for vertebral interpretation. This 

 result is not surprising in light of the fact that these three 

 studies share common methods and sample biases. All 

 three of the previous studies were performed on juvenile 

 sharks from the eastern North Atlantic, the vast majority 



of which were between 100 and 184 cm FL. Because of the 

 lack of samples from very small fish, one study (Silva et al., 

 1996) included vertebral readings from full-term embryos 

 in the growth curve. It is well documented that embryonic 

 growth is not comparable to postnatal gi'owth (Casey et 

 al., 1985; Pratt and Casey, 1990) and, therefore, embryos 

 should not be included in a postnatal growth curve. The 

 lack of large and small specimens in the calculations of 

 these growth curves is particularly problematic because 

 validation of the first growth increment is essential as it 

 forms the basis of further counts. Moreover, the smallest 

 and largest of the specimens are the most influential in the 

 estimation of growth (Campana, 2001). 



All three of the previous studies used similar whole 

 centrum vertebral processing techniques and band count 

 criteria, which would lead to corroborating counts, yet not 

 necessarily to accurate counts (Campana, 2001). Wliole 

 vertebrae simply do not allow for high band resolution in 

 older slower growing fish. Therefore, counts from whole 



