'If adaptation were solely to the inanimate environment, it is easy to believe that evolution will simply track Darwin’s ‘elements of air and water’ in their random walk through time. Selection would be stabilizing until a change in the climate or an accidental geographical displacement introduced a brief interlude of directionality. Each such directional interlude would seem to be as likely to reverse as to continue the previous one. But in fact consistent directionality is introduced because the environment of any one evolving lineage includes other evolving lineages. Above all, it is because adaptations in one lineage call forth counter-adaptations in others, setting in motion the unstable evolutionary progressions we call arms races' (Dawkins and Krebs 1979).
Symmetric conflicts were assumed in the case study I described in Part 100. In reality, both symmetric and asymmetric conflicts between species, or within a species, can occur, and often lead to ‘evolutionary arms races’: Adaptations in one lineage can alter the selection pressure and call forth counter-adaptations in other interacting lineages. If this occurs reciprocally, a runaway escalation of complexity of behaviour may occur, rather like a spiralling arms race between two rival nations.
Arms races take place on evolutionary time scales. It is lineages that evolve, not individuals.
Several factors can make an arms race asymmetric:
1. The specialist vs. the generalist factor. A predator species that is good at hunting several species of prey is unlikely to be strongly effective against any one of them, and therefore any of these prey species stands a greater chance of outrunning the predator in the evolutionary adaptation race. If, on the other hand, the predator has specialized in hunting a particular prey, any adaptively gained advantage by the prey is more likely to be matched by an evolutionary increase in the hunting capability of the predator. Of course, it is also true that a predator able to hunt several types of prey may, on the whole, make a good living; it runs many races.
2. Unequal selection pressure. ‘The rabbit runs faster than the fox, because the rabbit is running for his life while the fox is only running for his dinner’ (Dawkins 1979). This life-dinner principle can explain the asymmetry in the arms race in favour of the prey. The penalty of failure is much higher for the prey than for the predator, resulting in unequal selection pressures.
3. General inequalities in potential rates of evolution. Mammals evolve faster than bivalve molluscs because of the fiercer interspecific competition in the case of mammals. It may also be true that bivalves and, say, frogs have evolved slowly because they are getting on fine as they are.
4. Learning. Learning is similar to natural selection in that both tend to improve the performance of individuals in a population. The difference, of course, is that learning is adaptation over the life time of an individual, whereas in natural selection the improvement is seen only as an average over many generations of many individuals. But in a predator-prey context, learning by a predator can reduce the chances of survival of the prey, and may reduce substantially the number of generations over which the prey would have had an upper hand because of some genetically transmitted set of improved strategies.
Interspecific asymmetric arms races
An asymmetric arms race is like an attack-defence arms race, an example being the contest between parasites and their hosts; in fact, any predator-prey contest. Offensive adaptations on one side are countered by defensive adaptations on the other.
Why are cuckoo hosts so good at detecting cuckoo eggs, but so bad at detecting and rejecting cuckoo nestlings? Dawkins and Krebs (1979) explained it by analogy with the case of the heroin addict who knows that the drug is killing him, and yet cannot stop taking it because the drug is able to manipulate his nervous system, making him crave for the drug. There is evidence which suggests that the orange gape and the loud begging calls of the cuckoo chick have a ‘supernormal’ (irresistible) effect on the brain of the cuckoo. It is conceivable that, in the evolutionary arms race, cuckoos have put their adaptive emphasis on two different things in the life cycle of the chick: On mimetic deception at the egg stage (the cuckoo eggs look similar to the eggs of the host); and on manipulation of the nervous system of the host at the late nestling stage. As the host species became better and better at distinguishing cuckoo eggs from its own, cuckoos responded by evolving eggs with increased similarity to the eggs of the host species. And as the hosts evolved to become less and less susceptible to the cuckoo nestlings’ begging calls for food, the nestlings’ calls evolved to become more and more plaintive and irresistible. The life-dinner principle is applicable here. The cuckoo had to do better in the evolutionary arms race for shear survival of the species. The host species, on the other hand, had no threat to its survival by of the presence of cuckoo eggs and nestlings side by side with its own.
Predator-prey dynamics was modelled as early as the 1920s by Lotka (1925) and Volterra (1926). Population densities N1 and N2 of two competing prey and predator species are modelled on an evolutionary time scale by the famous Lotka-Volterra equations:
∂N1/∂t = N1(r1 – b1N2),
∂N2/∂t = N2(-r2 – b2N1).
In these equations, r1 is the rate of increase of the prey population when there are no predators present; r2 is the death rate of predators in the absence of prey; b1 denotes the rate at which the prey are eaten up by the predators; and b2 is the ability of the predators to catch the prey.
Intraspecific asymmetric arms races
Such arms races are difficult to analyse because the contestants (parents vs. offspring; males vs. females; queen ants vs. worker ants; etc.) belong to the same gene pool. Any advantage gained by either of the contestants goes to the same gene pool. It is an arms race between two branches of the same conditional strategy, rather than an arms race between two independent lineages with different gene pools. There are queen ants and there are worker ants in an antcolony; a similar situation prevails in a beehive. All female honeybees, including the queen bees, develop from larvae which are identical genetically. Those fed on ‘royal jelly’ become fertile queens, while the rest remain sterile workers. Something similar happens in ant colonies. The contest between queen and worker ants over relative parental investment is an example of arms race of this type.
Suppose there is an ant colony in which there is only one queen, who is the mother of all members of the ant colony. What determines the male-female sex ratio in this population? One possible approach for answering this question is to use a game-theoretic model in which the contestants are genetically related (Maynard Smith 1978). The term haplodiploidy is used in biology for the way in which the sex of the progeny is determined by external factors like selective feeding or presence or absence of fertilization of eggs. In some social insects, for example, females develop from fertilized eggs, and males from unfertilized eggs. This ensures that sisters are more closely related to each other than to their own offspring, meaning that the best chance they can give their own genes of surviving is to look after each other, instead of laying eggs of their own. This ESS has evolved in Nature several times, and forms the basis of the stability enjoyed, for example, by beehives and termite mounds (Douglas 2005).
Interspecific symmetric arms races
In a symmetric arms race, the two sides get better and better at doing the same thing. This type of arms race is unlikely to be important. The competitors in this category would rather diverge than escalate the completion. There may be, for example, evasion of competition by niche separation (Lawlor and Maynard Smith 1976).
Intraspecific symmetric arms races
Symmetric arms races are more likely to occur within the same species. They are essentially of a competitive nature, and are the stuff Darwinian evolution is made of. Adaptation to male-male competition within a species for females comes under this category of evolutionary arms races (Maynard Smith 1977). Even adaptation to being eaten up by predators is an arms race of this type, because individuals less competent at this task will gradually get eliminated by natural selection.
How do arms races end?
There are several scenarios:
1. One lineage may drive the other to extinction.
2. One lineage may reach an optimum, and thus prevent the other from doing so.
3. Both sides may reach a mutual local optimum (as in the flower-bee coevolution).
4. There may be no stable end, and the arms race may cycle continuously.
NOTE ADDED ON 9 OCTOBER 2015
For a real-life demonstration of the Life-Dinner Principle in action, watch this video: