This paper focuses on British English. Such rules as might be devised for specification of segment length would have to take into account, among other factors, the lenis/fortis nature of the segments at syllable margins, degree of stress, the relation between a given syllable and others in the same rhythm group, the same relations within the word, possibly the overall utterance length, and certainly the overall tempo. For the purpose of the piece of research under discussion recordings were taken of the first mode, stress-unstress, with the open front vowel /ae/. It was already known that duration figures for all segments would be different with various syllable nuclei and that they would vary systematically. The intention was to find out what relations existed between syllable nuclei and syllable margins, what difference in terms of duration occurred when there was a falling pitch movement of the final stressed syllable, and what evidence was there for compensation withing the syllable and within the word. If it can be proved that major durational regularities exist in mathematical terms, then a description of broad classes of sounds, such as fortis stop consonants and peripheral/tense front vowels, in terms of standard duration, would have predictive value. (Author/HW)