Abstract The chronology of the events associated with the late Paleocene thermal maximum (LPTM, Chron C24r) has been established through the construction of a composite reference section that involved chemomagnetobiostratigraphic correlations and assumed minimum diachrony of biostratigraphic events. On this basis, discrepancies between correlations in different sections were explained by inferred unconformities. However, diachrony between distant sections cannot be ruled out. We report here on two geographically close sections drilled onshore New Jersey that yield different records of chemomagnetobiostratigraphic correlations in the interval representing Chron C24r. Because of their proximity ( approximately 40 km apart), diachrony of biostratigraphic events between the two sections can be ruled out. In contrast, the marked lithologic disconformities in the sections explain well the different records of events. We thus conclude that the current relative chronology for Chron C24r is firmly based and that the upper Paleocene-lower Eocene stratigraphic record yields multiple unconformities, with Subzone NP9b rarely sampled. We examine the implications that undeciphered unconformities may have on the identification of proxies for paleoceanographic reconstruction, in particular with regard to the identification of the carbon isotope excursion (CIE) that reflects a dramatic latest Paleocene disturbance of the carbon cycle. We propose biostratigraphic means (short-lived calcareous nannoplankton and planktonic foraminifera taxa) that permit the unequivocal identification of the CIE not only in the oceanic realm but also in neritic settings.