ENEast Baltic *vakaras and Slavic *večer ‘evening’ can be projected in the late Indo-European protoforms *uokviro- and *uekvero- respectively. If Armenian gišer ‘night’ is derivable from *uekero-, the reconstruction should have been *k. In the article the old etymology of Bezzenberger (1878) is discussed. His solution is based on relation of the (Armenian-)Balto-Slavic isogloss ‘evening’ with Baltic *väka- and Slavic *veko ‘eyelid, cover’, i.c. ‘evening’ = ‘eyelid / cover of day’. The irregular vocalic correspondences can be explained by apophonic processes operating independently in both branches: vrddhi-lengthening *yoko- > *vaka—> *väka- in Baltic and *yekko- > *veko- —> *veko- in Slavic.