[The anatomic and physiologic foundations of the measuring of time in hearing (author's transl)]

Hornbostel and Wertheimer proved in 1920, that human hearing is able to measure a time difference of 0,03 msec. This "Hornbostel-effect" surpasses by more than a hundredfold the average efficiency of the nervous system. It is only possible by: 1. unusually even diameter of the nervus acusticus fibres 5 micron (Lorente de No 1933), 2. the short way of the nervus acusticus (ca. 4,0 cm) and only two switch-communications (Keidel 1975) from the cochlea to the analysing nucleus accessorius (Galambos and Schwartzkopff 1959), 3. the "miniaturisation" of the cochlea (Keidel 1975), 4. the equal length of the nerve fibres in the cochlea through spiral of the cochlea tapering off to the top (Heermann 1958), 5. ascertainment of statistic average results of numerous simultaneous nerve actions by a neuro-physiologic computer. By altering the speed of speech records and tapes by more than 130% the time pattern, absolutely necessary for understanding language is distroyed, whereas a reproduction of music remains intelligible. By installing an adjustable loop between the high and deep pass sound heads of Matzker's binaural summations test apparatus monaural presentations showed a small deterioration at a retardation of more than 10 msec. At a 90 msec. retardation complete deterioration of speech was observed. At a retardation of 20 msec. the plosives and the short consonants disappear first; next follow the other consonants, whereas vowels remain intelligible at a 100 msec. retardation.