ABSTRACT: The ATI (assemblage turnover index), which is inversely related to the concept of complementarity, detects points of enhanced assemblage change.Whereas complementarity calculates the permeability of boundaries to a communityâ€™s continuance, the ATI quantifies boundary impermeability. Sample-wise ATI values for planktonic foraminifera (ATIsp) are here reported for the Late Quaternary at the ODP(Ocean Drilling Program) Hole 994C, Blake Ridge, western subtropical Atlantic Ocean, and the pattern compared with that for a piston core from a comparable latitude south of the Azores archipelago. There were no statistically significant correlations in Hole 994C between ATIsp and either interpolated d18O data or benthic foraminiferal ATIs. Thus, turnover in the planktonic and benthic foraminiferal assemblagesmust have been driven by different factors. Assemblage turnover peaks inATIsp occurred at someGlacial Terminations (specifically, T-II and T-IV) and within some interglacial MISs (marine isotope stages). This reflects rapid restructuring of the planktonic foraminiferal community at those times. Few comparable peaks inATIsp were recorded during the commencement of glacials, the protracted onset of which apparently induces only low levels of assemblage turnover. There were no ATIsp peaks in the interglacial MIS 3,which was a muted event. The conditioned-on-boundary index (CoBI), which is derived from theATI, showed that different tropical to subtropical species contributed to each of these peaks in turnover, such that there was no distinct cycling between glacial and interglacial planktonic foraminiferal assemblages. The migrations of zonal foraminiferal provinces is unlikely to be the dominant process, as thatwould have induced repetitive changes in assemblage composition between provincial types thatwere not observed. The piston core taken south of the Azores archipelago contained peaks in assemblage turnover that occurred shortly after Terminations. This might reflect that that authors for that core reported only the eight most abundant species. Alternatively, it might be inferred that the situation of the Azores core, taken north of the tropical-subtropical provincial boundary, was such that it does not reflect provincial migration.