Although the information is limited it is worthwhile to consider the temperature data from A. opercularis in the light of what has been inferred from other evidence about late Holocene climate in the European area. Lamb (1995) considered that the last 800 years BC were marked by relatively cool temperatures, succeeded by relatively warm conditions during the first 400 years AD (essentially the period of the Roman empire in western Europe), then followed by a cool interval until 800-900 AD, a warm interval until 1300-1400 AD (essentially spanning the medieval period in Europe), a cool interval until about 1800 AD (the Little Ice Age), and a period of slight overall warming to the present day. The dates of shells PB1 and PB5 fall within the "Roman" and "Medieval" warm periods, respectively; shells SP1, PB2, PB3, and PB4 derive from allegedly cool intervals. Taking the isotopically-derived temperature ranges from the Holocene shells (Figure 6) to be indicative of the seasonal range experienced (see above), it is noteworthy that PB1 and PB5 record minimum and maximum temperatures almost identical with the median winter and summer extremes of the modern North Sea; the supposed relative warmth of the periods concerned is therefore not borne out by information from these shells, or higher atmospheric temperatures did not cause higher seawater temperatures. Shells SP1, PB2, PB3, and PB4 record somewhat higher maximum temperatures than PB1 and PB5 (outside the range of analytical error) and the minimum temperatures recorded are either very similar (PB2 and PB4) to these two shells or again higher (SP1 and PB3); the supposed relative coolness of the periods concerned is therefore also contradicted by shell-isotope data, or atmospheric coolness was not reflected in seawater temperature. Seawater temperature around northwest Europe is strongly influenced by the North Atlantic Drift, but because the strength of this warm current is related to the strength and persistence of westerly winds, which bring warm air, variation in seawater temperature is likely to mirror atmospheric temperature (Dawson and O'Hare 2000; empirical support in Hickson 1997, figures 6.6.1, 6.6.2). Because the southern North Sea is a shallow, well-mixed water body, surface temperatures (and hence atmospheric temperatures) should be reflected in bottom temperatures, as supplied herein by the oxygen isotopic composition of a benthic mollusc. It is, of course, entirely possible that the years represented by the small number of shells sampled were uncharacteristic, and the "anomalous" temperatures are in any case not beyond what one might expect as the range of variation under the general climatic conditions inferred by Lamb (1995). However, given the consistency with which the isotope data are somewhat at variance with Lamb's interpretations of climate, and in the knowledge that these interpretations are themselves based on limited data and have been called into question by other evidence (e.g., Ogilvie and Farmer 1997), they represent a useful contribution towards accurate definition of late Cenozoic climatic variation. It is worth noting that shells SP1 and PB2, from locations approximately 40 km apart, yield identical carbon dates (1,430 years BP), and the former was preserved with the valves articulated. This may reflect a major episode of sedimentation, and consequent shell burial, at this time.
Because five of the six Holocene shells record lower winter temperatures than those from modern indigenous North Sea shells, it is clear that the winter growth interruptions represented by the prominent growth bands in the latter cannot be due to low temperature. Limited food supply is a possible explanation, but has little supporting evidence (see above). Stress resulting from pollution is another candidate: Pollution is undoubtedly greater now in the North Sea than in pre-industrial times (Middleton and Grant 1990), and stress so caused has been shown to result in reduced "scope for growth" in marine bivalves (Widdows and Johnson 1988; Widdows et al. 1990). Pollution might also be the reason for the relative rarity of A. opercularis at present in the southern North Sea, to judge from the low frequency of shells in surface "grab" samples taken by the BGS as compared to those in cores of Holocene sediments (A. Weller, personal commun., 1997). It is, however, difficult to reconcile the notion that sparse occurrence and reduced winter growth of modern A. opercularis are consequences of pollution with the fact that summer growth is greater. Possibly the effects of pollution may be over-ridden in summer by some favourable condition for growth of the modern North Sea (perhaps higher productivity; see above). However, until such a condition is identified it has to be said that pollution is no more than a possible explanation for the modern occurrence and winter-growth characteristics of A. opercularis in the North Sea.