in the same direction to give these new signals meaning, resulting in two
functional and different mate-recognition systems (Andersson 1994).
The problem in sexual selection has been a bit more complicated and
controversial (e.g., Kirkpatrick and Ryan 1991; Andersson 1994; Ryan
1997). In some mating systems, the female’s choice of a mate does not influence her immediate reproductive success. In these mating systems, the male
signals evolve under selection generated by female mating preferences. But
how can preferences, the neural and cognitive processes biasing receivers
toward certain signal variants, evolve if all receivers are producing the same
number of offspring?
Here, we review some of the hypotheses for maintaining congruence
between the signal and the receiver during evolution. We use the term
coevolution in the more general sense—when the evolution of one suite of
phenotypic characters influences the evolution of another suite of characters. Evolution of the two sets of characters need not proceed simultaneously; there could be substantial lag. Furthermore, it must be
remembered that not every change in a signal requires a change in the
receiver. Receivers seem not to have highly tuned accept–reject filters but
are more than capable of ample generalization (Enquist and Arak 1998).
The three major factors we will consider that might contribute to the functional integration of signal–receiver systems are pleiotropy, genetic correlation, and sensory exploitation.
6.1. Pleiotropy
Pleiotropy is multiple phenotypic effects resulting from the same gene. This
offers the simplest explanation for how signals and receivers remain functionally integrated during evolution. If signals and receivers are controlled
by the same gene or tightly linked sets of genes, then genetic changes will
simultaneously and similarly affect signal and receiver.
6.1.1. Scaling
Because mate-recognition functions are so important to both sender and
receiver, we assume that selection plays an important role in bringing about
the match between peripheral tuning and long-distance signal. But there
might be other, more parsimonious processes at work.
There are two variables to which signal variation often scales in a predictable manner: body size and, in ectotherms, temperature. Dominant or
carrier frequencies tend to decrease with larger size because the characteristic vibration pattern of the morphological substrate is negatively correlated with mass. This relationship has been especially well-documented
in frogs (Martin 1972; this chapter, Section 2.2), birds (Bowman 1983), and
mammals (Morton 1977). In many ectotherms, ambient temperature
influences rates of behavior, and sound production is no exception. The
5. Selection on Signals
247
functional and different mate-recognition systems (Andersson 1994).
The problem in sexual selection has been a bit more complicated and
controversial (e.g., Kirkpatrick and Ryan 1991; Andersson 1994; Ryan
1997). In some mating systems, the female’s choice of a mate does not influence her immediate reproductive success. In these mating systems, the male
signals evolve under selection generated by female mating preferences. But
how can preferences, the neural and cognitive processes biasing receivers
toward certain signal variants, evolve if all receivers are producing the same
number of offspring?
Here, we review some of the hypotheses for maintaining congruence
between the signal and the receiver during evolution. We use the term
coevolution in the more general sense—when the evolution of one suite of
phenotypic characters influences the evolution of another suite of characters. Evolution of the two sets of characters need not proceed simultaneously; there could be substantial lag. Furthermore, it must be
remembered that not every change in a signal requires a change in the
receiver. Receivers seem not to have highly tuned accept–reject filters but
are more than capable of ample generalization (Enquist and Arak 1998).
The three major factors we will consider that might contribute to the functional integration of signal–receiver systems are pleiotropy, genetic correlation, and sensory exploitation.
6.1. Pleiotropy
Pleiotropy is multiple phenotypic effects resulting from the same gene. This
offers the simplest explanation for how signals and receivers remain functionally integrated during evolution. If signals and receivers are controlled
by the same gene or tightly linked sets of genes, then genetic changes will
simultaneously and similarly affect signal and receiver.
6.1.1. Scaling
Because mate-recognition functions are so important to both sender and
receiver, we assume that selection plays an important role in bringing about
the match between peripheral tuning and long-distance signal. But there
might be other, more parsimonious processes at work.
There are two variables to which signal variation often scales in a predictable manner: body size and, in ectotherms, temperature. Dominant or
carrier frequencies tend to decrease with larger size because the characteristic vibration pattern of the morphological substrate is negatively correlated with mass. This relationship has been especially well-documented
in frogs (Martin 1972; this chapter, Section 2.2), birds (Bowman 1983), and
mammals (Morton 1977). In many ectotherms, ambient temperature
influences rates of behavior, and sound production is no exception. The
5. Selection on Signals
247