The notion that speech sounds are complexes of intersecting sets of phonetic features is as old as the earliest attempt to describe speech. Such features are used to describe the languages of the world and also figure importantly in attempts to construct models of speech production and perception. There is, however, still no set of features on which all linguists, phoneticians, speech communication engineers, psychologists, etc., agree, nor is there agreement on whether these features should be defined in physiological-articulatory or acoustic terms or both. This paper will discuss some possible reasons for the continued “confusion,” will examine some phonetic properties which to date have not been proposed as distinctive, and will suggest a mapping procedure by which linguistic “cover features” can be related to physical events. [Work supported by NSF.]