The range of sounds that are exploited phonetically by the world's languages represents only a portion of what humans are capable of producing vocally. Moreover, among attested phonetic segments there is enormous variation in the frequency of occurrence across languages: Most segments are relatively rare, while a few occur almost universally. A major task of phonetic theory is to explain these patterns of selection. Traditionally, many phoneticians have believed that two principles—articulatory economy and perceptual distinctiveness—play a role in shaping sound patterns and segment inventories. However, these principles have not often been formulated with sufficient precision to have genuine explanatory content. The focus of this presentation is on the role of auditory factors in structuring vowel systems. First, attempts to predict vowel inventories on the basis of a principle of auditory dispersion (i.e., sufficient auditory contrast) are reviewed. Second, a corollary of the dispersion principle, the auditory enhancement hypothesis, that provides a general account of certain widespread patterns of phonetic covariation in the production of vowels is explored. Finally, how the notion of sufficient contrast may explain some puzzling acoustic differences between male and female tokens is considered.