In spite of the fundamental role that distinctive features play in phonology and phonetics, current research continues to raise basic questions concerning how features can be defined in terms of measurable language-independent physical properties in the articulatory and acoustic/auditory domains. It has been proposed that for a feature to be recovered from a speech event, not only must its articulatory condition be met but the acoustic definition must be satisfied, in the absence of which further enhancing attributes must be present [K. N. Stevens, “Acoustic and perceptual evidence for universal phonological features,” Procedings of the 15th International Conference of Phonetic Sciences, Barcelona, 2003 (unpublished), pp. 33–38]. This paper addresses this proposal through the study of the phonetic feature (spread glottis), drawing upon acoustic, fiberscopic, and photoelectroglottographic data from Berber and Kabiyé. A definition for this feature is proposed which associates a specific articulatory state with an equally specific acoustic definition, holding for the full set of sounds and contexts in which this feature can be implemented.