Tim Caro. 2005. University of Chicago Press, Chicago, Illinois. xv + 591 pp. ISBN 0-226-09435-9 (cloth) and ISBN 0-226-09436-7 (paper). Cloth, $95.00; paper, $38.00.— The author, who has focused his career on empirical and comparative study of antipredator defenses in mammals, has expanded his purview to include birds in this comprehensive, gracefully written book. Although Caro writes mainly on two homeotherm groups, he includes material on other vertebrates and insects where they are particularly illustrative of evolutionary principles. In line with the author's intention “to integrate all functional and evolutionary perspectives on antipredator defenses,” the book's components are arranged sequentially to mimic the series of events in a predator-prey encounter. Thus, after an introductory section, chapters are arranged in the following order: morphological and behavioral adaptations to avoid detection, vigilance, conspecific warning signals, signals of unprofitability, antipredator benefits of grouping, morphological and physiological defenses, nest defense, mobbing and group defense, flight and other behavioral patterns of last resort. A terminal chapter considers questions for future work and is largely devoted to conceptual issues linking material covered in various preceding chapters. The work is technically clean and is graced with lovely and topical sketches at the head of each chapter. The treatment is quite nonmathematical and, where equations are presented, they are explained quite clearly. The level of analysis is largely noninstructive with regard to deficiencies of experimental design and analysis, a weakness perhaps inevitable in a work of such scope.

We are told that the book took five years to complete, a claim readily believable given the 87 pages of references. In considering the book in detail, I have emphasized conceptual issues as applied to birds, particularly issues where opportunities for future research are apparent. The introductory chapter's focus on olfactory predator recognition is confined to mammals. The lack of material on birds suggests research opportunities on taxa with known olfactory competence (e.g. vultures, procellariiforms, kiwis, oilbirds).

Chapter 2, “Morphological Traits to Avoid Detection,” begins with discussion of crypsis—put forward as probably the most widespread predator defense in homeotherms— then follows with sections on countershading and contrasting colors. Discussion of bitterns seems misplaced in the mammal section, rather than with the rest of Aves. In considering the adaptiveness of hornbill females plastering themselves and their young into cavity nests, the author misses the advantage of the female conducting the prebasic molt simultaneously with incubation and nestling feeding (with food passed from the male). The chapter contains brief mention of “masquerading,” whereby an animal closely resembles an inanimate object, and notes that such adaptations are relatively common in insects, amphibians, and fish. We are told that no systematic study currently exists on the distribution and ecological correlates of masquerading in birds. Likewise, the whole subject of disruptive coloration in birds needs systematic analysis featuring attention to phylogenetic-independent contrasts.

Chapter 3 concerns behavioral mechanisms for avoiding detection, the material divided into changing the timing and extent of activities, selecting different habitats from predators, and altering patterns of feeding and reproduction. A major portion of this chapter concerns a comprehensive review of work on nest-site placement and nesting behavior: habitat type, distance from habitat edges, habitat patch size, vegetation around the nest, nest height, proximity to other nests, distribution of nests, proximity to predator-deterring social insects, and other categories of refuge. The section concerning edge effects on nest mortality points out that much is to be learned from intensively following nest predators as they “process” habitat edges. Studies employing artificial nests come in for criticism; one matter that stands out as being particularly difficult concerns olfaction and mammalian predators. The scent cues of both bird nests and researchers are not well understood and, therefore, not well controlled. The section on behavior reducing the probability of nest detection by predators omits Steve Lima's work on optimal food-delivery rates. In the mammal section of the chapter is a passage claiming that kangaroo rats remain in more open habitat than pocket mice because, among other adaptations, they have inflated auditory bullae that permit them better to hear sounds made by incoming owls. There does not appear to be any experimental support for this idea. Discussion of Lima's theoretical work on where to forage concludes by saying that future models need to incorporate energy gained from foraging and risk of predation into a single currency of fitness. This will be a daunting task; what common currency is valid short of number of grandchildren? Throughout the book, the word “carnivore” is taken to denote, solely, a member of the mammalian order Carnivora, rather than being used in the larger ecological sense of “flesh eater” (the latter sense applicable to creatures from ciliophorans to cetaceans). In the chapter summary, the interesting general point is made that because of flight, birds have evolved fewer adaptations to avoid detection by predators than mammals have.

Chapters 4 and 5 review the recent deluge of empirical and theoretical work on vigilance as an antipredator adaptation. An important conundrum concerns the extent to which living in groups is an adaptation for sharing vigilance or for diluting the probability of being taken by a predator at any given level of group vigilance. Furthermore, despite much effort, we still do not understand why cheating on extent of vigilance by animals in a group setting does not seem to exist, though the need to continually monitor the flock-departure behavior of flock-mates may provide a clue to the absence of cheaters. In considering vigilance in mixed-species foraging groups, the author fails to emphasize adequately that cost:benefit ratios for various component species may be influenced by their relative social dominance status within the mixture.

Further along in chapter 5, we are told that “male vigilance may be determined by both natural and sexual selection.” Presumably, notwithstanding this shorthand prose, the author realizes that sexual selection is but one variety of natural selection. This ornithologist is curious to know how the ears of mule deer (Odocoileus hemionus) “have interesting parallels with heterophyid rodent ear structure.” The chapter concludes by acknowledging that there seem to be so many factors affecting vigilance that, for now, we cannot get past the checklist stage to theory except to notice that vigilance before and after a predator has been detected may have different causes and functions.

Chapters 6 and 7, respectively, are devoted to warning signals directed to conspecifics and heterospecifics. Birds, ground squirrels, ungulates, and primates provide the bulk of examples, with the behavior of ground squirrels having undergone the deepest analysis. (Except for one mention, behavior of another group of social mammals, the cetaceans, is absent from the book.) The author notes the rather puzzling lack of high-frequency alarm calls in Australian passerines. How the structure and frequency of normal contact calls might be responses to predation risk receives no attention, but the author notes that the energetic cost of alarm-calling has not been studied. Ultrasonic alarm calls apparently occur in ground squirrels; do they occur in birds? Richard Dawkins's suggestion that one function of alarm-calling may be to cause other members of a group to behave cryptically, reducing the likelihood of a predator spotting the group, has apparently never been tested. Also needing work is the notion that alarm-calling is mutually beneficial to all members of a group because any predator thwarted by an alarm call is less likely to hunt again in the same area. (This is another example of potential payoffs to study of predator behavior.) The author claims that such reciprocal altruism is more likely to apply to mammals than to birds, but he forgets that many permanent-resident birds live in flocks with stable long-term membership. False-alarm warning calls are known to occur in several avian species, particularly subordinate species in permanent-resident mixed-species flocks. We await application of signal-detection theory to this phenomenon. Could it be that deceptive alarm calls are most common in visually occult environments where independent detection of predators by each individual in a group is less certain? We are told that “the excitement of research on alarm calls” stems from the possibility of evaluating broad questions at both comparative and experimental levels.

Chapter 7, “Signals of Unprofitability,” considers antipredator adaptations occurring once an animal has been detected by a predator. Topics covered include behavior letting the predator know that it has been detected, informing it that the prey is in good condition, displaying powerful physical defenses, or letting the predator know the prey is noxious or poisonous. The question of “cheating” in signals of prey quality to predators is a fertile area for research. The author's mention that aposematism evolves only with kin selection is certainly countered by the example of the striped skunk (Mephitis mephitis). Using Great Tits (Parus major) as model predators, two experiments designed to discover the evolutionary trajectory of aposematic coloration have produced disparate results. In one case, the conclusion was that unpalatibility selected for gregariousness, which, in turn, selected for warning coloration; but a repetition of the experiment in another lab produced contradictory results. This situation sets up the delightful prospect for a third, disinterested, party to evaluate simultaneously the conclusions now in place. Considerable attention is paid to the iterative process by which predators come to associate prey characteristics with unprofitability. An experiment had students learning about “prey” unprofitability by experiencing the consequences of choosing displayed pixels. An interesting hypothesis is that heightened conspicuousness of prey is selected because predators conclude that because a prey item is so conspicuous it must have been detected and rejected by other predators in the past, and so must be unprofitable. Although study of aposematism in birds has been hindered by the prevailing notion that bird flesh is neither noxious nor poisonous, comparative study has found negative relationships between visibility and edibility in 38 species of Palearctic birds and in 200 avian species from Africa. The author indicates that a point for future research concerns whether the distastefulness of some bird species affects the hunting tactics of predators. Apparently, the only example of Müllerian mimicry in birds concerns five species of pitohuis in New Guinea that are both conspicuous and toxic.

Chapter 8 details antipredator benefits of grouping, other than shared vigilance. Covered are the dilution effect, transmission of information about danger from predators, erratic escape that may confuse a predator, and several other miscellaneous factors. The dilution effect is a function of the risk per capita of detection times the risk per capita of capture once detected. If this product is lower for animals in groups than for solitaries, grouping should be selected, but presently, data on encounter rates between predators and grouped or solitary prey are quite rare. The Trafalgar Effect is defined as transmission among group members about the presence of a predator, and is named after the intership communication about the presence of an enemy in Nelson's fleet. (The British seem very fond of such metaphorical remnants of the Empire; e.g. red queen hypothesis, Beau Geste hypothesis.) The chapter concludes with the generalization that some combination of food type, predator pressure, and body size determines the grouping pattern of homeotherms, and that the grouping pattern coupled with habitat type selects for antipredator alternatives, such as crypticity in solitary species or corporate vigilance in group-dwelling species.

Chapter 9 is devoted to morphological and physiological antipredator defenses. The morphological attributes focus on body size and escape speed, whereas the physiological characteristics concern examples of unpleasant odor or taste. After noting that some prey are physically too large or powerful for a given predator to subdue, the discussion of body size as a defense focuses on energy-input and handling-time spent by the predator on prey of various sizes. The discussion of gape-size limits on prey size considers birds only as prey (e.g. of snakes); but as predators, birds can be gape-size-limited. How large a fish can a heron or cormorant swallow whole? It is claimed that individuals of larger prey species could be subject to fewer predation attempts than individuals of smaller prey species because larger predator species live at lower densities than smaller prey species. But the generalization is also true that larger prey species live at lower densities than smaller prey species, so that capture attempts per prey individual is not necessarily lower for large prey individuals. Picked from the plethora of hypotheses about reversed size-dimorphism in raptors is the notion that the female is larger so she will be better able to defend the nest. Caro notes the different escape tactics found within Aves and points out the opportunity to relate escape tactic to wing morphology. Although birds are far less endowed with morphological means of defense than mammals, many taxa do have spurs. In most spurred species, such structures seem the product of sexual selection, because only males are so endowed; but in 16 species, females also are spurred. We lack evidence as to how spurs might be used against predators and whether individuals experimentally despurred might have higher predation rates. In the matter of avian chemical defenses against predation, fully 80 genera of 17 orders have been reported as being malodorous to humans. Eider ducks (Somateria mollissima) defecate on their eggs before withdrawing from a predator, and such deposits have been shown to deter food-seeking ferrets and Norway rats (Rattus norvegicus). Brought into the discussion is the complex of nesting adaptations in response to predation risk (e.g. clutch size, nesting attempts per year, nestling developmental rates).

The 44 pages of Chapter 10 are devoted almost entirely to nest defense in birds. A survey of waders found that nest defense was more likely to occur in larger species, in colonial nesters, and in species with biparental care of young. During distraction displays before nest predators, birds appear to use interactive techniques similar to those recently developed for playback experiments; the birds monitor a predator's response, adjusting strength and orientation of distraction display appropriately. Surprisingly, time and energy costs of nest defense have apparently never been measured. Also discussed are the methodological limitations of quantifying nest defense using approaching humans as surrogate predators or employing artificial, or artificially placed, nests. A major theme of the book, well exemplified in this chapter, is that little is known about the long-term effects of such anti-predator behavior on the responses of potential predators. Aside from being deterred in the instant, how does a predator respond over succeeding, hours, days, and beyond? We are told that in both altricial and precocial species, parents give a specific warning call that has the effect of silencing the offspring, an advantage Caro considers to apply only after the young hatch. Is this true? Might eggs stop “peeping” in response to warning calls? A related question asks: because of presumably lower predation risk, do eggs and young of hole-nesting birds vocalize more that those of open-cup nesters? Regarding nest defense and renesting potential, it is not clear why the author concludes (on page 354) that nest defense of later clutches in a season should be greater because such clutches tend to be smaller than earlier clutches. Four pages later, we are told the contrary, that Tawny Owls (Strix aluco) may defend late clutches less because they are smaller. Opportunities exist for experimental evaluation of the several explanations concerning differences between the sexes in nest defense. On page 364, the author discusses “predatory strigifomes.” Are there any nonpredatory strigiformes? Stemming from the theoretical notion that parental investment (as exemplified by nest defense) should be influenced by reproductive value of the current brood, some studies have shown that broods of heavier young are more vigorously defended that broods of lighter young. These results, however, do not appear to control for parental condition. One might expect the parents of better-nourished (i.e. heavier) young to be better nourished themselves and, therefore, better equipped to spend time and energy on nest defense. The author concludes that study of nest defense in birds is “the most theoretically grounded aspect of antipredator defense in homeotherms, because it takes its predictions directly from parental investment theory using parental defense as a testing ground.”

Chapter 11 concerns mobbing and other forms of group defense. Virtually all empirical data concerning mobbing stem from study of birds. Work on swallows has lead to the distinction between “active mobbers,” which approach closely and may even strike a predator, and “passive mobbers,” an oxymoronic label for animals that approach less closely and are silent. While mobbing apparently only rarely results in the mobber being killed by the mobbee, there are certain to be time and energy costs of mobbing, neither of which has been quantified for any species. A central theme of the book—the lack of information on predator behavior—is repeated in this context; we know very little about how a predator's subsequent behavior is affected by an episode of mobbing. Is an Eastern Screech-Owl (Otus asio) likely to reduce or desist its hunting in an area where it has been mobbed? Experimental playbacks to radiotagged owls would likely produce an answer. Conversely, does an individual of a prey species learn by cultural transmission, through mobbing, not only the identity but the location of a predator? The author recounts the notion that mobbing may be adaptive in demonstrating fitness to conspecifics, an idea calling to mind Amotz Zahavi's thoughts on “prestige.” An arresting fact is brought out in the discussion of mixed-species assemblages as group defense against predators. In Japan, Azure-winged Magpies (Cyanopica cyana) breed most successfully near nests of Lesser Sparrowhawks (Accipiter gularis), apparently because the hawks drive away the Large-billed Crows (Corvus macrorhynchos) that otherwise would depredate the Azure-winged Magpies' nests. Remarkably, Azure-winged Magpies breeding within 100 m of a Lesser Sparrowhawk pair tended to synchronize nesting with the Lesser Sparrowhawks, but Azure-winged Magpies more than 1 km from a breeding pair of Lesser Sparrowhawks were much more variable in breeding phenology. The disparity suggests that Azure-winged Magpies near Lesser Sparrowhawks time their laying after observing something that indicates the Lesser Sparrowhawks' breeding state—perhaps the intensity of the latter's nest defense.

Chapter 12 is devoted to flight and other “behaviors of last resort.” Covered are behavioral freezing in place, startle responses, fighting back, and fleeing. Apparently, freezing of young in response to parental warning is accompanied by bradycardia “in which the heart rate plummets.” Why the bradycardia? Does it occur in birds? Although these questions are not addressed, we are told that the whole subject of immobility and the factors controlling it have been understudied. “Regurgitating” stomach oil seems a vast understatement of the projectile vomiting that procellariiforms use as a “last resort” against predators. The section on flight contains nothing of the considerable modeling of avian body-mass trade-offs between starvation risk (smaller body mass) and predation risk (heavier body mass). On days when a female is yolking up an egg, her body mass increases steadily until the egg is laid. Might acceleration and speed of flight from predators have been a constraint on egg size? In the section concerning autotomy of body parts, a last-resort response notable in herptiles, the “fright molt” of birds is mentioned in passing. Development of a standard methodology could provide comparative data on what appears to be a legitimate form of autotomy in some avian species.

Chapter 13, the last chapter, describes relationships between the forms of antipredator response treated separately in earlier chapters. Following a section on the synergism between morphological and behavioral responses, material is structured as answers to three questions: why does a given species respond differently to different predators, why do different species use different responses to the same predator, and why does a given prey species show a number of different antipredator responses to the same predators. The section on coevolution contains an interesting example of the complexity of cetacean behavior. Off the northwest coast of North America, pods of killer whales (Orcinus orca) in permanent residence hunt salmon and use pulsed sound to communicate. Pods that are transient hunt marine mammals and do so without using pulsed sound except just after making a kill. The difference in behavior between the two whale types is attributed to the fact that marine mammals can detect killer whale calls, responding with antipredator behavior, but fish cannot detect such sounds. It also seems possible, however, that if resident pods socially dominate transitory pods, the latter may remain silent so as not to reveal their location. The complexity of sorting out coevolutionary relationships among predator and prey is well illustrated by the striped skunk, which uses its aposematic coloration to warn mammalian predators of its noxious odor, but thereby becomes more vulnerable to Great Horned Owls (Bubo virginianus) that are undeterred by the odor.

This is an important book. It treats a limited topic encyclopedically, including older references of the sort often overlooked by reviewers in this day of limited electronic databases. It is heuristic throughout in pointing to opportunities for further work, and concludes the last chapter by explicitly detailing “Ten pressing questions.” Excellent material for graduate-level seminars, the paperback is not unreasonable in price, and the cloth edition will be a worthwhile addition to institutional libraries.—