Andrews et al.: Bomb radiocarbon and tag-recapture dating of Carcharhinus plumbeus 
461 
Year of formation 
Figure 4 
Plot of sandbar shark (Carcharhinus plumbeus) A 14 C measured from vertebrae in relation 
to the estimated year of formation (determined from growth band counts), showing that an 
adjustment of formation dates for sandbar shark specimen SB 745 (by 5 years) was neces- 
sary to match the porbeagle ( Lamna nasus) A 14 C record (filled diamonds). Adjusted age 
was increased to a minimum of 20 years. The assumption was made that the missing years 
were those in the late-adult years (as reflected in the outer part of corpus calcareum) and 
that early growth was well quantified. This determination does not preclude an older age 
because there are no limits to the prebomb levels measured in these samples. 
age-related discrepancies. Instead, feeding habits of 
these sharks in deep offshore waters appear to explain 
the observations. 
For the sandbar shark, a scenario similar to that of 
the white shark would be possible if the single sample 
from year-1 of specimen SB 749 was considered uncon- 
taminated. This A 14 C value was clearly postbomb value 
at -7.8 %o (SD = 4.4). Including this sample would pre- 
clude an increase in age for this sandbar shark, as well 
as the other sharks in our study and explain the un- 
expectedly low A 14 C values. Based on the growth band 
age estimates alone, the temporal distribution of the 
A 14 C data for sandbar shark would be similar to that 
of the white shark (cf. Fig. 1 of this study with Fig, 1 of 
Kerr et al. [2006]). However, the well-documented feed- 
ing behavior and depth-related life history of sandbar 
shark do not support this hypothesis (Springer, 1960; 
Stillwell and Kohler, 1993; Conrath and Musick, 2007). 
In addition, although it is certain there was some non- 
surface-derived 14 C included in the sandbar shark diet 
based on the lowest measured A 14 C values (-78.6%e to 
— 110. 9%e cf. -40.2%c to — 66%c for corals), the levels were 
similar to the lowest A 14 C values measured in the por- 
beagle A 14 C reference chronology (-74.6%o to -114. 7%c). 
Because of important life history considerations, it was 
concluded that the innermost sample (year-1) from the 
vertebra from specimen SB 749 was inaccurately ex- 
tracted and included more recently formed vertebral 
material. In support of this conclusion is the series of 
three additional samples that would have formed after 
this sample, all of which were clearly classified as pre- 
bomb material. 
The most plausible explanations for underestimated 
ages of sandbar sharks in this study are either a lack 
of band pair formation at the oldest adult ages or a 
problem with the interpretation of growth bands. The 
validated age of the youngest specimen (SB 43) pro- 
vides some evidence that age can be determined visu- 
ally with growth-band counts in the earliest years of 
growth. A validated age of ten years for this shark 
provides evidence that the missing years for the larger 
shark were most likely those from the latter years of 
life. This argument was well documented for porbeagle 
shark off New Zealand, for which age estimation was 
accurate to approximately 20 years but was underesti- 
mated by several decades for older sharks (Francis et 
al., 2007). Estimation of age from band pair counts was 
not possible for older sharks because as somatic growth 
of the shark slowed or ended, vertebral growth ceased. 
A similar scenario was described for school shark or 
