**3. Materials and Methods**

GeoB 8502-2 is 14.78 m long with an average sedimentation rate of ≈6 cm/ky and extends back 200 ky to Marine Isotope Stage (MIS) 6. Samples were taken at a resolution of ~1700 years (10 cm sampling interval) by extracting a slice of material, 1 cm in thickness, which corresponds approximately to an average of 170 years of sedimentation. All samples were freeze-dried, weighed, and washed through a 63 µm sieve. They were subsequently dry-sieved into several sieve fractions and non-fragmented *G. bulloides* shells from the 315–355 µm size fraction were picked for mass analysis. The very narrow size interval (40 µm) should be sufficient to overcome the greater proportion of natural size variability without further normalization [26]. Furthermore, this size fraction is widely used in paleoceanographic studies and can allow comparisons. However, any record of shell weight is a composite signal of dissolution superimposed upon initial shell weight variability.

In order to assess whether the influence of dissolution is responsible for the observed shell weights, the preservation state of the specimens underwent multiproxy assessment. Moreover, to better understand the increase in *G. bulloides* shell mass during Termination II (T-II), the selected specimens were analyzed both geochemically and by high resolution X-ray microcomputed tomography (µCT). µCT was used to inspect the interior and the internal structure of the foraminiferal tests, and apart from addressing the test's integrity (manifestation of dissolution), we were also able to assess to what degree the increased recorded masses are the result of interference from shell inclusions or of changes in test thickness. Finally, the XMCT analysis led to total shell volume estimates that allowed the calculation for the first time of volume normalized shell weights or *G. bulloides* shell densities, presenting a more precise method of eliminating the contribution of shell size to shell weight.
