‘Calibrating’ amino acid ratios with uranium-series dates requires an accurate knowledge of current mean annual temperatures (CMATs) over the region studied. To measure these, test-tube sized ‘diffusion sensors’ were emplaced for 1 year (in 1984, 1985 and 1986), both outside and inside Minchin Hole sea-cave in South Wales and Belle Hougue sea-cave in Jersey, both of which have yielded Oxygen Isotope Substage 5e uranium-series ages on speleothems. Our outside temperatures agreed with meteorological ones. Our inside temperatures were over 1°C lower. To allow for this, a mean of ‘empirical’, ‘linear’ and ‘parabolic’ epimerisation calculations suggests that ratios from molluscs inside the caves should be multiplied by over 1.1 for comparison with outside ratios. This raises Bowen et al.'s ‘Pennard’ stage ratios from inside Minchin (and Bacon) Hole up towards the ‘Unnamed’ stage ratios outside, and suggests that the Unnamed sites are also from Oxygen Isotope Substage 5e, as proposed by Proctor and Smart. The same conclusion is reached more strongly by comparisons with the ratios and temperatures inside Belle Hougue to the south, and at Eemian (assumed 5e) sites in The Netherlands, Germany and Denmark to the east. The Pennard ratios from outside sites may provide further evidence for global sea stands close to the present level later in Oxygen Isotope Stage 5.