This paper considers some basic problems inherent in the instrumental analysis of intonation. The problems are illustrated by a detailed analysis of one intonation contour in American English. The material studied consisted of two sets of data. The first set involved sentences recorded by one speaker with determined stress and pitch patterns. These sentences consisted of 1263 CNC words produced in an identical frame, with primary stress and the peak of the intonation contour occurring on the CNC word in the frame. The second set of frame sentences involved 70 minimally different words, uttered by five different speakers of the same general dialect. The fundamental frequency values for the various levels of the intonation contours of the sentences were measured from narrow‐band sound spectrograms, and the measurements were correlated with the segmental phonetic structure of the utterances in which the intonation contours were produced. The intrinsic fundamental frequencies of the various syllable nuclei of the words from both sets of data, and the influence of preceding and following consonants on these frequencies, are described. The relationships among successive intonation levels within the sentences also are discussed.