Page 15 - Shell_growth_2008

Geo-Mar Lett (2008) 28:309–325 323

18
include variation in δ O seawater values, isotopic disequilib- summer, and samples of shell carbonates from growth
18
rium from so-called vital effects, or time-averaging of δ O increments produced in these periods will yield time-
values arising either during periods of reduced shell growth averaged isotope values which underestimate the highest
or from the sampling strategy. recorded SSTs. In contrast, shell carbonate samples taken
from increments produced in periods of rapid growth
18
Variation in δ O seawater should give better isotopic resolution, as is shown by the
18
high degree of concurrence between δ O SHELL and
18
Mean values of δ O seawater are given in Table 5. Inspection measured SSTs in the autumn and winter.
18
of the monthly values of δ O seawater , in particular how far There will be no isotope record at times when shells fail
they deviate from the mean values for each locality, shows to grow. This will have two principal effects: to widen the
18
no systematic association of these with the degree of offset offset between δ O SHELL and measured SST values, and to
18
between δ O SHELL temperatures and measured SSTs in increase the effect of time-averaging in samples which
those months. For Cala Grande, 4 months have deviations straddle shell increments laid down immediately before and
18
of δ O seawater values exceeding ±0.1‰ of the mean but, in after the episode of growth cessation. The isotope data
three of these months, there are no significant differences presented accord with this model in the hottest months of
18
18
between δ O SHELL temperatures and measured SSTs. For the year (May to September), when δ O SHELL temper-
18
Monte Cofano, in six of the 7 months having deviations of atures produce the largest offsets. The highest δ O SHELL
18
δ O seawater values greater than ±0.1‰ there are significant SST recorded is 25.0°C (Table 11), suggesting that little, if
18
differences between δ O SHELL temperatures and measured any shell growth occurs at temperatures higher than this.
SSTs, but other months with significant temperature differ-
18
ences have only slight deviations of δ O seawater values Effects of sampling strategy
from the mean. Similar disparate variation exists in the
monthly data from Mazzaforno. Sampling, here, includes both the initial selection of the
shells to be used and how the carbonate samples are
Vital effects obtained from these. Only shells with an undamaged
aperture edge should be used. Smaller and faster growing
18
In a discussion of a range of vital effects in the stenotopic shells give lower δ O SHELL temperature offsets than larger
intertidal limpet Patella vulgata, Fenger et al. (2007) found shells and should, therefore, be used wherever possible.
that none could fully account for the offsets found between However, larger shells can yield reliable data—for example,
18
δ O SHELL temperatures and instrumentally measured SSTs. a shell from Mazzaforno with a diameter of 23.3 mm
O. turbinatus is a more mobile species, individuals showing produced a very low non-significant offset (Table 10).
significant movements both up and down and across the Large shells, provided they do not have numerous closely
intertidal zone and being exposed to slightly different water spaced growth checks, can be used with confidence in
18
temperatures and δ O water values. It is therefore less archaeological studies of seasonality, as shown by the
18
likely to be subject to some types of vital effects considered marked shell from Cala Grande in which the δ O values
by Fenger et al. (2007). One possible effect, suggested by from the 24 sequential shell carbonate samples reflected a
Schifano and Censi (1983), is 18 O enrichment during full annual range of SSTs.
precipitation of shell material under conditions of evapora-
tive cooling. They subsequently showed similar enrichment Shell growth checks
in subtidal shells (Schifano and Censi 1986), indicating that
evaporative cooling is not a necessary condition. It is Do shells of O. turbinatus in NW Sicily stop growing
unlikely that this disequilibrium model applies to O. during some period of the year? Field experiments by
turbinatus, or other intertidal gastropod species, because Williamson and Kendall (1981) on the closely related
shell growth occurs under aqueous conditions when the Atlantic species O. lineatus showed that annual growth
mantle edge is in contact with the growing aperture edge cessation lines were produced in the winter. This was
(Crothers 2001), rather than during periods of exposure confirmed by combined growth and oxygen isotope studies
when the animals are retracted into their shells. on shells of this species by Mannino et al. (2003). Such
regular annual growth lines are useful for studies of age-
Effects of variable shell growth rates structuring in populations (Crothers 2001) and archaeolog-
ical assemblages (Mannino and Thomas 2001) of this
Episodes of reduced shell growth will result in more time- species.
18
averaged δ O values in shell carbonate samples (Goodwin We have observed growth lines cutting across the whorls
et al. 2003). Shell growth is less during the spring and of shells of O. turbinatus, possibly indicating periods of