The uncal apex is an anatomical landmark frequently used for segmenting the hippocampus into its anterior and posterior segments, a necessary step for computing many measurements of its long axis. It functions well, as it is both local to the hippocampus and easy to identify. However, in spite of widespread use and definition in the EADC‐ADNI Harmonized Hippocampal Protocol (HarP), how the uncal apex is influenced by gross hippocampal changes during normal aging has not been established, nor has the possible impact on measures of anterior hippocampus (aHPC) and posterior hippocampus (pHPC) volume. Here I drew upon three large data sets to describe and confirm these relationships, investigating them in one large data set and replicating my findings in the two others, evaluating a total of 4,434 hippocampi. I found the uncal apex fell in an increasingly more anterior position with increasing age. This age‐related retraction of the uncus began after age 36, with the sharpest effects arising after age 60. This phenomenon exaggerates age‐related aHPC volume decreases while simultaneously underestimating age‐related pHPC volume decreases, a pattern I confirmed by juxtaposing uncal apex and MNI space‐based landmarking. A hippocampally based reference frame was also rendered unstable by age‐related shifts in the posterior extent of the hippocampus. Both the uncal apex and hippocampal reference frame should therefore be used with caution in aging research, or in research involving other demographic or disease factors known to evoke gross changes in the hippocampus. Instead, MNI coordinate‐based heuristics may be appropriate for segmenting the hippocampus in study designs involving such factors. Apex‐based segmentation is still attractive, however, in study designs where advanced age and atrophy are not used as regressors, including investigations into long‐axis effects in healthy young adults. Progress toward localizing functional divisions within the hippocampus is needed to identify best practices for the field.