Introduction

Alarm signals offer perhaps one of the greatest challenges to evolutionary explanations of behaviour. Why should animals make alarm signals, at potential risk to themselves, to warn conspecifics of danger? Evolution of such apparently altruistic signals should be difficult if it makes signallers more conspicuous and at greater risk of predation than silent neighbours (see Chapter 11 for predators attracted to alarm pheromones). Benefits to kin probably account for the evolution of alarm signals in most species (Section 8.2) but alarm signals in groups of unrelated individuals are more difficult to explain.

Alarm pheromones may be more widespread than any other type of alarm signal. Their importance is underlined by the observation that in social insects alarm pheromones are the most commonly produced class of chemical signal, after sex pheromones, and have evolved independently within all major taxa (Blum 1985).

Like other pheromones, alarm pheromones are likely to have evolved as secondary or modified uses of existing compounds (Chapter 1). Two common evolutionary routes seem to be from either chemicals used in defence or those released by injury, both events being linked to predation.

Responses to alarm pheromones may be more adaptive than a simple response to the presence of predators as many prey species must live alongside their predators. Only when the predators are hunting (as indicated by release of alarm pheromone from prey) are they actually a danger.

Evolution of alarm signals between related individuals

When already in the predator's jaws, giving an alarm signal might not save the victim, but the victim's signal could evolve by kin selection if it can save the lives of relatives such as offspring, siblings, or clones, by alerting them to danger.