FIGURE 7. Concordance and discordance among call centered and note centered approaches to name sound units in anuran vocalizations. The upper two schematic spectrograms show examples where both approaches lead to the same categorization of sounds. (A) A single tonal sound is repeated after regular silent intervals of longer duration than the sounds. There is no defined duration of the series of sounds; if undisturbed, calling could go on for minutes or hours. In both approaches, one sound unit would be a call, and the note-centered approach would define each call consisting of a single note. (B) Series of rapidly repeated sounds, each composed of a series of bursts of sound energy. Because these bursts are <10 ms in duration they are defined as pulses. The call-centered approach does not define each major subunit as call because the silent intervals between them are much shorter than the units themselves; thus, both approaches agree in defining the units as notes. (C) This species emits clearly defined and stereotyped series of sounds, each series being separated by variable intervals from the next series. The note-centered approach defines one coherent entity of sound emission as a call; hence, each sound series unit is a call, and the subunits are notes. In contrast, the call-centered approach defines each sound unit as a call (and each series as a call series) because it is separated from other such units by a long silent interval. (D) This species emits two distinct kinds of pulsatile sound units, of which one is much longer than the other. Because the combination of sounds is emitted as coherent entity, in the note centered approach the entire sound emission is a call and the sound units are notes of two types, of which one is arranged in a series. In the call-centered approach, each sound unit is a call because they are separated by long silent intervals from the next unit. Two call types can be distinguished and one of these is arranged in a call series.