6: Chapter Behavioral Adaptations for Survival
6.1 Many people think that an adaptation is a trait that improves the survival chances of an organism. Under what circumstances would such a trait be an adaptation? Under what other circumstances would a survival-enhancing attribute actually be selected against? A trait that improves the survival chances of individuals can be considered an adaptation only when it also improves their ability to produce offspring that survive to the age of independence. A survival-enhancing trait would be selected against if individuals with this attribute consistently had even slightly fewer surviving offspring than relatively short-lived individuals.
6.2 Stephen J. Gould and Richard Lewontin claimed that adaptationists make the elementary mistake of believing that every characteristic of living things is a perfected product of natural selection [463], when in reality many attributes of living things are not adaptations (see Table 6.1 in the textbook). Moreover, in their eagerness to explain everything as an adaptation, adaptationists have, according to Gould and Lewontin, invented fables as absurd as the fictional "just-so" stories of Rudyard Kipling, who made up amusingly silly explanations for the leopard's spots and the camel's hump. How might adaptationists defend themselves against these charges? Do adaptationists have the means to discover whether their tentative explanations for trait X are wrong? In science, there are always going to be incorrect, even silly, hypotheses advanced from time to time, although my reading of the animal behavior journals has turned up no truly absurd ones that have managed to get into print. Why not? Because a scientific hypothesis is intended to be tested, and, without a rigorous test, the hypothesis is very unlikely to be published. Interestingly, Gould and Lewontin picked on a handful of very minor papers, notes really, and they made these examples seem less plausible than they actually were. 6.3 For many evolutionary biologists, the term "adaptation" must be reserved for a characteristic that provides "current utility to the organism and [has] been generated historically through the action of natural selection for its current biological role" [74]. What could "current utility" mean, and what do you think it should mean? Make use of the terms "fitness benefits" and "fitness costs" in your answer. If a trait originated for function X and later took on a different, but still adaptive, biological role Y, does that mean it is not an adaptation? Track down the evolutionary history of the flight feathers on the wings of modern birds (see, for example, [982]). Where did these feathers come from, and what function did their predecessor feathers exhibit? If you go back far enough in time, will the ancestral form of any current trait have the same function that it does now? Current utility could mean a large number of things, ranging from survival-enhancing to body-strengthening but it should mean reproduction-enhancing or gene propagation
improving; in other words, an adaptation is a trait whose fitness benefits must exceed its fitness costs by a greater degree than any alternative form of that trait. Adaptations are measured in terms of their contribution to individual genetic success. Some of the definitions of adaptations insist on consistency of past and present utility, but every trait, if traced far enough back, will be shown to have antecedents with different functions. Feathers are an excellent example because the flight feathers of today's birds almost certainly evolved from feathers that served thermoregulatory or display functions, and these feathers, in turn, evolved from defensive quill-like protuberances from the skin of an extinct reptile. To give a different name to traits whose evolved functions have changed over time would make little sense if all adaptations have had a history of changed function.
6.4 The arctic skua is a close relative of gulls that also nests on the ground and mobs colony intruders, including the great skua, a larger predator that eats many arctic skua eggs and chicks. In one study, hatching success and post-fledging survival were greater for arctic skuas that nested in dense colonies than in low-density groups (Figure 6.4 of the textbook); the number of near neighbors was, however, negatively correlated with the growth rate of their chicks [948]. Rephrase these findings in terms of the fitness costs and benefits of communal mobbing by the arctic skua. If the term adaptation meant a perfect trait, would communal mobbing by arctic skuas be labeled an "adaptation?" It appears that nesting in close association with others has some fitness benefits, probably mainly those associated with protection of the young against predators, but there are fitness costs as well, namely those that affect chick growth rate, such as increased competition for food or increased disturbance of chicks by neighboring arctic skuas. If the term adaptation were reserved for traits that were perfect (i.e., had no fitness costs at all), then there would be no attributes of any sort for which the label adaptation was appropriate.
6.5 The ability to hear ultrasound in one species of noctuid moth is considered an antipredator adaptation because it apparently enables individuals to hear and avoid nocturnal, ultrasound-using bats. Imagine that you wished to test this hypothesis via the comparative method. Identify the utility of each of the following lines of evidence on the hearing abilities of other insect species. Specify whether these cases involve convergent evolution, divergent evolution, or neither. 1. Almost all other species of noctuid moths also have ears that respond to ultrasound. 2. Almost all the species in the evolutionary lineage that includes the noctuid moths and many other moths belonging to several other superfamilies also have ears that respond to ultrasound [1346]. 3. Some diurnal noctuid moths have ears, but are largely or totally incapable of hearing ultrasound [422].
4. Almost all butterflies, which belong to the same large evolutionary grouping as the noctuids but are usually active during the day, lack ears and so cannot hear ultrasound [420]. 5. Six species of noctuid moths found only on the Pacific Ocean islands of Tahiti and Moorea have ears and can hear ultrasound, but do not react to this stimulus with anti-bat responses [424]. 6. Members of one small group of nocturnal butterflies have ears on their wings and can hear ultrasound; they respond to ultrasonic stimulation by engaging in unpredictable dives, loops, and spirals [1346]. 7. Lacewings and praying mantises fly at night and have ears that detect ultrasound and lead to anti-bat defensive behavior (see page 111 in the textbook). (1) Neither, since this shared feature could be traced to shared ancestry. (2) Neither, because this lineage owes its existence to a distant ancestor that may have had ears as well. (3) Divergent evolution, because it appears that in the absence of bat predators, a subset of noctuids has evolved a different trait from its nocturnal relatives. (4) Divergent evolution, for the same reason as given in (3). (5) Divergent evolution, assuming that bats are not found on Tahiti and Moorea, which is the case. (6) Convergent evolution, on the grounds that these butterflies almost certainly evolved from diurnal ones that had lost their hearing while these species have regained the trait under selection from bats at night. (7) A clear case of convergence, with unrelated insects having evolved analogous traits as a result of predation by bats at night.
6.6 Some persons would say that the fact that most noctuid moths have ultrasoundsensitive ears is "simply" a reflection of their shared ancestry, a holdover from the past, and therefore that ultrasound sensitivity is not adaptation an in these species [150]. Others disagree, arguing that it makes no sense to limit adaptations to just those traits that have diverged from the ancestral pattern [1007]. Who is right? I personally find Reeve's argument more persuasive. A trait that has been maintained by selection over time after having evolved in a distant ancestor is a trait that advances individual fitness more than any alternative that has appeared in populations of these species. Thus, the mechanism responsible for the persistence of an adaptive trait is no different from the mechanism (natural selection) responsible for the spread of a novel mutant trait that confers greater genetic success on individuals. One could make the distinction between traits maintained by stabilizing selection versus those that are spreading because of directional selection (many evolution textbooks cover this material) but in either case one can argue that both categories of traits owe their existence to selection, and so are adaptations.
6.7 In my front yard, I sometimes find several hundred male native bees clustering in the evening on a few bare plant stems (Figure 6.14 of the textbook). An assassin bug sometimes approaches the cluster and kills some bees as they are settling down
for the night. Devise at least three alternative hypotheses on the possible antiassassin bug value of these sleeping clusters, and list the predictions that follow from each hypothesis. The dilution effect: Members of a large sleeping cluster should be safer simply because of a lower probability of being killed per night, which would be likely if the predator were to kill only one or two bees per evening. The fighting back effect: Members of sleeping clusters combine forces to repel assassin bugs, attacking it when it comes to their sleeping places. The improved vigilance effect: Members of sleeping clusters can react to the presence of an assassin bug by taking advantage of the responses of their neighbors and so flee to safety more reliably. The actual data support the first hypothesis. 6.8 "In the years since 1950, pollution controls have reduced the amount of soot deposited on tree trunks and the melanic form of Biston betularia has correspondingly become increasingly scarce in Europe [135, 237] and North America [468], where the species also occurs." Put this statement in the context of a scientific investigation into whether the typical salt-and-pepper coloration of some members of this species constitutes an adaptation. Begin with a research question and proceed through hypothesis, prediction, test, and conclusion. Question: What causes the typical salt and pepper moth to have its speckled appearance? Hypothesis: That color pattern is an adaptation that reduces predation on resting moths during the daytime. Prediction: In areas in which the usual resting habitat has become available again after a period of soot pollution of woodlands, then the typical form should become more common again, after a period in which the alternative melanic phenotype was abundant. Actual results: The data presented above. Conclusion: Hypothesis supported; the lighter salt-and-pepper color pattern of the moth is an antipredator adaptation that works in unpolluted woodlands.
6.9 Weiss (see page 194 of the textbook) also collected information on the growth rates of caterpillars forced to inhabit shelters that she contaminated with their feces and of others she allowed to mature in clean shelters (see page 194 of the textbook). She found no difference in the weight of the pupae that experienced these two different conditions as larvae; moreover, the days required for the larvae to pupate also did not differ between individuals growing up with and without waste pellets in their shelters. Why did Weiss gather these data? Having demonstrated that the caterpillars were probably safer from wasps as a result of ejecting their feces from their shelters (as described in the text), Weiss also wanted to examine some alternative hypotheses for this waste management behavior. It is possible
that the larvae could be infected by pathogens that grew on their waste, thus providing an advantage to individuals that safely disposed of their fecal frass. If, however, this hypothesis were true, then experimental contamination of the living space of the caterpillars should have resulted in sickly, lighter-weight, slower growing larvae. Because this was not the case, Weiss could rule out the anti-contamination hypothesis for house cleaning by fecal pellet ejection.
6.10 The Bonaire whiptail lizard runs a short distance from potential predators and then raises one foreleg, which it waves about ostentatiously (Figure 6.28 of the textbook) [240]. This arm-waving behavior might be another example of a pursuit deterrence signal. What predictions follow from this hypothesis with respect to when the arm-waving behavior should be performed in response to the approach of a human being (a predator substitute)? That is, should arm waving occur more often when a person approaches slowly or rapidly? In response to a direct or a tangential approach? And which arm should be waved when the lizard is not directly facing the human? Slowly approaching potential predators should be signaled in an attempt to short-circuit a really serious attack; rapidly approaching predators require immediate, rapid escape. Likewise, tangential approaches should elicit a signal more than direct approaches because the tangentially approaching predator has evidently not yet decided to mount a full attack. If the arm-waving really is a signal to the predator, then the forearm closest to the "predator" should be waved.
6.11 Consider Figure 6.33 of the textbook, a game theory diagram based on the concept of a selfish herd (with thanks to Jack Bradbury). In a population of prey animals, most individuals are solitary and stay well apart from others. But some mutant types arise that search out others and use them as living shields against predators. The mutants take fitness from the would-be solitary types by making them more conspicuous to their predators. We will set the fitness payoff for solitary living in a population composed only of solitary individuals at P. But when a solitary individual is found and used by a social type, the solitary animal loses some fitness (B) to the social type. There is a cost (C) to social individuals in terms of the time required to find another individual to hide behind, and a cost arising from the increased conspicuousness to predators of groups composed of two individuals rather than one. When two social types interact, we will say that they each have one chance in two of being the one that happens to hide behind the other when a predator attacks. If B is greater than C, what behavioral type will come to predominate in the population over time? Now compare the average payoff for individuals in populations composed entirely of solitary versus social types. If the average fitness of individuals in a population of social types is less than that of individuals in a population composed of solitary types, can hiding behind others be an adaptation?
If B > C, then the social type will become more common over time. Note that whether a social individual is interacting with another social opponent or a solitary one, it comes out ahead in fitness compared to a rival solitary individual. The average payoff for a social phenotype in an all-social population is less than that of a solitary phenotype in an all-solitary population. But since an all-solitary population is vulnerable to invasion by a mutant social type (under the conditions specified here), then once that type appears and begins to spread, social behavior becomes the adaptation by definition. The point is that an adaptation is defined by the higher genetic success of its carriers, on average, relative to the other type(s) in the population.