Hunting California Ground Squirrels: Constraints and Opportunities for Northern Pacific Rattlesnakes

—Our program of research on the predator-prey relationship between rattlesnakes and ground squirrels has emphasized the ground squirrel side of that system. But, we discuss here how some of the antipredator adaptations of ground squirrels have both constrained and generated opportunities for the predatory behavior of rattlesnakes. Constraints: Ground squirrel populations undergoing sustained predation from rattlesnakes have evolved the capacity to neutralize their venom. This has constrained rattlesnake predatory options in several ways. (1) Rattlesnakes can kill squirrel pups but not adults, because the small body volume of pups limits their capacity for venom neutralization. Thus, rattlesnakes can only hunt squirrels during the short season before pups outgrow their vulnerability. (2) Rattlesnakes face the danger of antipredator aggression from protective adults with venom resistance. As a consequence of the assertiveness of adult squirrels, rattlesnakes must at times defend themselves by striking defensively and rattling. (3) This defensiveness creates additional problems for the snakes, as the defensive behavior can disclose a snake’s vulnerability and squirrels make use of these cues by becoming more aggressive when the rattlesnake’s vulnerability is higher. (4) Confrontation by maternal squirrels functions to protect their pups. Finally, (5) these squirrels exhibit spatial and temporal knowledge of how rattlesnakes hunt. Such a “cognitive topography” of risk assessment includes knowledge of how these snakes use microhabitat features to conceal themselves and how the snakes remain hunting for a sustained period in a local area within the squirrel colony. Opportunities: (1) Squirrel populations living in rattlesnake-inhospitable areas lose their venom resistance under relaxed selection in several thousand years, but retain and even intensify their aggressive confrontation of snakes. With climate warming and the return of rattlesnakes to formerly inhospitable areas, the combination of antipredator aggressiveness and low venom resistance can expand the size of prey available to rattlesnakes into adult ranges. Under such conditions, rattlesnakes may hunt larger squirrels for longer seasons and grow to larger sizes. (2) Selection for reliable development of the behavioral antisnake system has led to the precocious expression of this system. As a consequence, pups confront rattlesnakes and so become easy prey. (3) Rattlesnakes appear to use ground squirrel antirattlesnake behavior as a source of cues about the presence and location of vulnerable pups, and at times appear to extract those cues actively through probing. moves in a circle, and returns to the snake from a different direction. She repeatedly reruns such cycles of approaching, tail flagging, evocation of rattling, startled retreating, and then returning until the snake is engaging in sustained defensive rattling. Beginning to vocalize, FC returns to her pups, spends several minutes with them, re-approaches the snake a third time, and goes through more cycles of tail flagging, cautious approaching, elongate investigating, rattling evocation, startled retreating, and re-approaching from a new direction. Another adult male squirrel marked “=4” has been attracted by the disturbance and, like male “1:”, expresses only fleeting interest as he observes from the vicinity of FC’s pups. After FC breaks away from the snake again, she pays another visit to her pups. At 11:05, the snake begins to crawl, revealing a bulge indicating that it has recently fed, probably on a squirrel pup. At this point, the pups spot the snake, perhaps for the first time, and begin to tail flag. The snake’s path brings it obliquely closer to FC’s pups, and FC returns to the snake, approaching to within centimeters of its tail (Fig. 1C). The snake reaches a bush and coils defensively beneath its thick foliage. Squirrel FC harasses the snake there, evoking more defensive rattling and striking as her four tail156 D. H. Owings and R. G. Coss Figure 1. An example of a natural encounter between a Northern Pacific Rattlesnake (Crotalus oreganus), which has just eaten, and FC, the probable mother of the victim, a newly-emerged California Ground Squirrel (Spermophilus beecheyi) pup. Squirrel FC has additional newly-emerged pups nearby. See also the description in the text. A. FC tail flags while closely investigating the snake, who rattles intermittently in response. B. FC evades a defensive strike by the snake. C. As the rattlesnake moves a few meters to a more protected site, FC closely follows. Note swelling at the snake’s midsection, revealing recent meal. D. In view of her pups, FC returns to resume confrontation of the snake, which rattles and strikes in response. Note pup adopting antisnake stance with tail flagging. E. The mother begins to move her pups to another burrow farther from the snake, which has retreated to a nearby burrow. 157 Hunting California Ground Squirrels flagging pups look on from a distance of 4.5 m (Fig. 1D). During this fourth encounter with the snake, FC leaves the snake’s vicinity twice to tail flag and cautiously inspect the snake’s previous location beneath the log. The second of these excursions begins when FC loses track of the snake, and while she is away, the snake begins to crawl again, moving obliquely away from FC’s pups and into a squirrel burrow about 2 m north. Squirrel FC spots this snake movement from a distance, begins to call repetitively, and moves to inspect the snake’s new burrow refuge cautiously, once again eliciting defensive rattling. She continues to visit her pups and inspect the snake’s current and original location for another 24 min. At 11:42, 1 h and 17 min after the first snake encounter, FC begins to move her pups, one at a time, to another burrow system about 15 m farther from the snake (Fig. 1E). This encounter by FC with a Northern Pacific Rattlesnake (Crotalus oreganus) is representative of the episodes that we have observed repeatedly during the short period each year when California Ground Squirrel (Spermophilus beecheyi) pups are emerging for the first time from their nursery burrows (Hennessy and Owings, 1988). At that time, multiple rattlesnakes hunt the pups in this area of the coast range east of San Francisco Bay. The intensity of squirrel FC’s engagement with the snake leads us to suspect that the snake had just killed and eaten one of her newlyemerged pups, and that the burrow below the log was a part of FC’s home burrow system. This episode illustrates how rattlesnakes hunting these squirrels face both constraints and opportunities from the squirrels’ antisnake defenses, properties that we summarize as follows. Constraints: (1) Rattlesnakes typically kill and eat pups, but not adult squirrels, which are resistant to the snake’s venom. (2) Adult squirrels confront rattlesnakes, at times assertively enough to elicit defensive rattling and striking. (3) Rattling and striking can leak cues about the snake’s vulnerability. (4) Such confrontation by maternal squirrels functions to protect their pups, as evidenced by both maternal transfer of pups to safer locations and much greater involvement with rattlesnakes by maternal females than nonmaternal females and adult males (Swaisgood et al., 1999a). (5) Adult squirrels exhibit spatial and temporal knowledge of how rattlesnakes hunt, which enhances vigilance and detection of rattlesnakes. Snake-related behaviors induce greater vigilance in nearby observing squirrels. Opportunities: (1) Squirrel populations living in rattlesnake-inhospitable areas lose their venom resistance under relaxed selection, thereby becoming more vulnerable to rattlesnake predation if they reestablish contact. (2) Squirrel FC’s pups began to tail flag once they themselves saw the snake, indicating that they recognize snakes and their predatory implications at the first day of nursery burrow emergence. Such recognition can engender snake confrontation by vulnerable pups, even when the mother is present. (3) Squirrel confrontation of rattlesnakes has the potential to leak cues about the presence and location of defended Figure 2. Variation among ground squirrels in resistance to Northern Pacific Rattlesnake venom. A. A female adult California Ground Squirrel from a highly resistant population (Folsom Lake, California) has received a 20 mg/kg venom injection under anesthesia in the left rear thigh. Note the small lesion (circled) and the squirrel’s alert state. B. A male adult California Ground Squirrel from a population with low resistance (Finley National Wildlife Refuge, Oregon) has received a 15 mg/kg venom injection under anesthesia in the left rear thigh. Note the large lesion and that the squirrel has died. C. Approximate LD50s of laboratory mice injected i.p. with Northern Pacific Rattlesnake venom after injection with ground squirrel sera or saline control. Ground suirrl sera were derived from three different resistant California Ground Squirrel populations and from nonresistant Arctic Ground Squirrels (Spermophilus parryii) that have been free of rattlesnake predation for 3-5 million yr. 158 D. H. Owings and R. G. Coss pups. This chapter will elaborate on these and other ways in which ground squirrel adaptation to rattlesnake predation has generated both constraints and opportunities for rattlesnakes hunting ground squirrels. The relationship between ground squirrels and predatory snakes is ancient and doubtless reflects a predator-prey relationship (Coss, 1999). Fossil deposits indicate temporal and regional contiguity of rattlesnakes, large colubrid snakes, and the Otospermophilus progenitor of modern ground-dwelling squirrels as early as the late Miocene (ca. 10 million yr ago; Black, 1963; Holman, 1979). Rattlesnakes and California Ground Squirrels appear in the same Rancholabrean fossil assemblages of late Pleistocene age (>100,000 yr ago; Miller, 1912; Stock, 1918; Brattstrom, 1953). Whe