Alarm signals evolution

Chivers DP, Brown GE, Smith RJF (1996) The evolution of chemical alarm signals attracting predators benefits alarm signal senders. Am Nat 148 649-659... 

Latent alarm signals seem to be present in vertebrates, especially in tadpoles of fix)gs, and the existence of such signals may explain some basic evolution questions in e relationships between predators and prey. The presumable inactive compounds in question may be activated by the bacterial flora within the gut of predators, explaining why taxonomically different predators may release diverse responses within one single prey species. Knowing the exact source of the alarm signals will be of utmost importance for future work within predator-prey interactions. 

The aim of this paper is to provide some speculative discussion and propose some hypotheses about the multiple possible functions and interactions, both within a species and between species, that could influence the evolution of chemical alarm signals in fishes. Fish alarm pheromones have been reviewed by Smith (1992) and Chivers Smith (in press) but I will provide basic background information with a few examples. 

The substances that function as alarm pheromones for conspecifics are detectable to heterospecifics and can influence other prey species and alter the behavior of predators. These interspecies effects may have had some influence on the evolution of alarm signal systems. 

Smith, R.J.F. Lemly, A.D. 1986. Survival of fathead minnows after injury by predators and its possible role in the evolution of alarm signals. Env. Biol. Fish., 15, 147—149. 

We dedicate this book to the memory of R. Jan F. Smith who died unexpectedly in the fall of 1998. Jan was a pioneer in the area of chemical communication, especially in the area of alarm signaling by fish. He also made basic contributions to the field of fish migration. An excellent review of Jan s ideas on alarm signals and their evolution appears as a chapter in this book. It is the last major review he wrote on this topic. Jan s enthusiastic and imaginative research resulted in many fundamental advances in our understanding of chemical communication and stimulated many students to pursue careers in our discipline. Jan will be sorely missed by friends, family, colleagues and the broader scientific community. 

Animal alarm signals pose an evolutionary problem (Williams, 1966 Charnov and Krebs, 1975 Maynard-Smith, 1976 Harvey and Greenwood, 1978 Smith, 1977, 1982a Weldon, 1983). What benefit does the sender gain by warning conspecifics It used to be presumed that the benefit to the group or species would account for the evolution of alarm signals but this... 

Costs. The benefits of sending the signal must exceed the costs for the alarm signal function to drive the evolution of the ASCs. In the case of the fish alarm systems the cost will be the production and storage of the alarm pheromone. Injury or death during the release process would presumably occur anyway and are not costs of the signal system. There is no... 

Smith, R, J. F., 1985, Evolution of Alarm Signals Role of Benefits Derived From Retaining Group Members or Territorial Neighbors, (unpublished ms). ... 

Chemical alarm signals operate when interacting individuals are aggregated or in close proximity. It is not unexpected, then, that such signals are well-developed in the social insects and relatively uncommon in pre-social species. Outside the social insects, chemical alarm systems are best developed in two homopteran groups aphids and treehoppers. In this chapter, we discuss alarm pheromones and their social context in these and other pre-social insects and suggest modes of their evolution. 

Figures 6.5 and 6.6 are referred to a slow drift of the output of sensor Sj, i.e., a linearly increasing signal, with a 10-3 K s 1 rate of change, is added to the measured variable for t >t = 9000 s. It can be recognized that the fault is detected a few time instants after the occurrence, while it is isolated about 2000 s after tf. This is due to the slow time evolution of the fault it can be argued that, in the first 2000 s after the occurrence of the fault, its effect is quite negligible and/or almost totally compensated by the observers. In order to reduce the isolation time, the normalization factors could be reduced, at the expense of an increased probability of false alarms. Moreover, Fig. 6.5 shows that the voted measure is the mean value of the measured and estimated data until the isolation is performed, and then it switches to the value of the healthy sensor (5), i).

Johnston (1998) gives several examples of how combinations of neurophysiology and behavioral analysis can sort out the differences between classical pheromones, pheromone-like signals and odor quality signals composed of a mixture of several active compounds. Application of these techniques to fish alarm substances might reveal patterns in the evolution of these chemical signals. 

---