Different Functions of "Alarm" Calling for Different Time Scales: a Preliminary Report On Ground Squirrels

The predator-evoked calling of California ground squirrels (Spermophilus beecheyi) was studied in the field during the reproductive season. Three different sources of data indicated that adults call more after, than before their young have reached the age of first emergence from natal burrows. During exposure to a tethered rattlesnake (Crotalus viridis oreganus) and to a freely-moving dog (Canis familiaris), and in natural encounters with a coyote (Canis latrans) or bobcat (Lynx rufus), calling was more frequent after than before young first emerged. We concluded that California ground squirrels call in order to warn their offspring about predators, like other ground squirrel species do. In order to see the increase in mammalian predator evoked calling after pup emergence, we had to separate calling on the basis of its temporal organization. Nonrepetitive calling involved spacing a few vocalizations irregularly in time. Calls patterned in this way were more common early in an encounter, became more frequent after pup emergence, and more consistently elicited immediate reactions. Such calling was probably used to warn pups. Repetitive calling comprised rhythmic emission of a series of vocalizations. Calls organized repetitively were more common later in an encounter, were not emitted more frequently after pup emergence, and less consistently evoked immediate reactions. These and other differences between the two temporal patterns of vocalizing led us to propose that repetitive calling represented a "tonic" communicatory effort (as in SCHLEIDT, 1973). Repetitive inputs to other squirrels may act "cumulatively" in a longer time scale than nonrepetitive calling, so as to cultivate or maintain vigilance in other squirrels. The repetitive caller could benefit by using the enhanced reactions of these more vigilant squirrels as a source of information about the predator. We propose that predator-prey episodes may be understandable from an "epigenetic" perspective. That is, the first alarm calls during an encounter should shift squirrels from an unwarned to a warned status; subsequent calling must then function in some other way than as a warning.