Mating calls of frogs from six populations in southern Illinois, N Mississippi, S Mississippi, Louisiana, N Florida and S Florida were studied by multivariate techniques. Canonical-correlation analyses showed that either collectively or as separate populations, call variables were not functions of the perch height of a frog. Low frequency peak and high frequency peak were inversely correlated with snout-vent length and positively correlated with cloacal temperature in all populations combined. These relationships varied in individual populations however, and none of the call variables were correlated with snout-vent length or cloacal temperature in two of the populations. Discriminant-function analysis showed that significant differences existed among populations based on seven call variables. Calls were correctly classified 66.5% of the time and 98.4% of the variation was accounted for. Low frequency peak and high frequency peak provided the most discriminating power and exhibited clinal variation on a N-S gradient. Population differences fit the biogeographical prediction of greater differences with increased geographic separation. Although there were differences in some call variables between populations that were sympatric or allopatric with Hyla gratiosa, the hypothesis of character displacement of the mating call in areas of sympatry was not supported.
No takes yet. Share an insight, caveat, or question.
Asquith et al. (1988) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: