Chapter 8 - Fragile Predators
8. Fragile Predators
Snakes have solved a potentially serious problem―how to nourish a heavy body with a small mouth…..
Harry W. Greene 1997
A night hike along a mountain trail in southern Thailand revealed a White-spotted Cat Snake (Boiga drapiezii) that had seized a Kumpol’s Gecko (Cnemaspis kumploi). Predator and prey were perched on top of a huge boulder to the side of the trail. The snake was holding the lizard in its mouth and the belly of the gecko was already discolored from the action of the snake’s venom. The White-spotted Cat Snake is slender, rear-fanged species, and the one eating the gecko was a small specimen. While this individual had used its venom, some rear-fanged, gecko-eating snakes will simply seize and swallow without envenomating the prey, a fact that was emphasized by one Ornate Gliding Snake (Chrysopelea ornata). On capture, the snake was palped to see what it had eaten and a house gecko was regurgitated. The mucus-covered gecko momentarily laid in the hand of the person who captured the snake before it ran off, apparently unharmed from the trip into the snake’s digestive system. Venom immobilizes the prey, reducing the chance of escape as the snake attempts to maneuver the prey into position for swallowing. Once a snake actually has a gecko in its mouth, the lizard has little chance to escape, unless a meddling herpetologist happens to come along.
But snakes do not always feed on such benign prey. Rodents, small carnivorous mammals, fish, venomous invertebrates, and, of course other venomous snakes, all possess sharp teeth, claws, spines, and toxins that could be used against a snake.
A single look at a snake skeleton makes its fragility immediately apparent; there are no massive bones, sharp incisors, grinding molars, or powerful limbs with claws. Instead, there are thin, fragile-looking ribs, many small vertebrae, and a delicate skull and jaw. Even though the teeth are sharp, they are often thin and easily broken. Snakes are not brute force predators like large crocodilians, ancient meat-eating dinosaurs, or carnivorous mammals. They are subtle, sophisticated predators that fill ecological niches that cannot be filled by other existing vertebrates. As Harry Greene noted, snakes have solved the problem of getting large prey into a body with a small head, and they have done this by evolving mobile skull elements, venom, constriction, and highly elastic organs, including the skin.
Prey Recognition
How are snakes able to recognize prey? And, why is it that some snakes specialize in feeding on very specific prey while others are generalists? Undoubtedly, snakes use all of their senses when hunting, but some are more important than others in finding prey. David Chiszar and colleagues experimentally determined that garter snakes are alerted to the presence of prey, first by chemical cues and then by visual cues. Rattlesnakes, on the other hand, are first alerted to the presence of prey by visual cues. Using six Prairie Rattlesnakes (Crotalus viridis) and six Baja Rattlesnakes (Crotalus enyo), Chiszar and colleagues examined the relationship between tongue flicking and hunger. The snakes were observed once a week for 8 weeks in two situations: a clean cage and a cage containing mouse odors. The snakes had been fed once a week until the start of the study, at which point fasting began. To conduct the weekly observations, snakes were placed into each cage for five minutes and all tongue flicks were counted. During weeks 1–4, the mean rate of tongue flicks in the clean cage was statistically indistinguishable from that in the cage with the mouse odors. During weeks 5–8, the tongue- flicking rate increased significantly when the snakes were exposed to mouse odors. Chiszar and colleagues suggested an increase in hunger increased chemosensory response to prey odors and that this, perhaps, is the basis for triggering foraging behavior in rattlesnakes.
Northeastern Illinois streams and quarries support populations of the crayfish-eating Queen Snakes (Regina septemvittata). However, Queen Snakes won’t eat just any crayfish. Instead, they target crayfish that have recently molted their exoskeletons. Newly molted crayfish have about the same consistency as peanut butter, making them easy to capture, swallow, and digest. Crayfish shed their chitinous exoskeleton so they can grow; before they shed the layer that is too small for them, however, they absorb the calcium from the hard chitinous layer, making it a soft chitinous layer. The calcium is temporally stored in small disk-shaped, stone-like gastroliths within the crustacean’s body cavity. The exoskeleton is then shed and replaced with a new layer. A hardening period follows where the calcium is re-established in the new external skeleton, making it hard and snake-resistant. It’s only for a few days that the crayfish are vulnerable to predation, so how do the snakes find them?
Queen Snakes patrol the shorelines of streams and quarries, investigating crevices in search of newly molted crayfish [Figure 8–1]. When discovered, the crayfish is extracted from hiding and quickly swallowed underwater before the snake continues it search. Queen Snakes are forest and forest-edge inhabitants. They do not use the open prairie ponds and streams and often bask in shrubs at the water’s edge. However, the related Graham’s Crayfish Snake (Regina grahamii) is found in aquatic prairie habitats. It uses cattail marshes and wet prairie soils to search for its crustacean prey.
INSERT FIGURE 8-1.
Figure 8–1. A Queen Snake (Regina septemvittata) foraging along a shoreline. Will County, Illinois.
Light, sound, and chemical cues are collected by snakes hunting for food, but one of these sources of sensory information seems to dominate snake hunting behavior, the collection chemical information. Vision and detection of vibration (sound) are important to successful predation. But, olfaction (smell) and the vomeronasal system (VNS) via the Jacobson’s organ are the sensory systems involved in collecting chemical information and are probably the senses most heavily relied upon by the majority of snakes. Understanding the roles of the VNS and olfaction in prey detection are vital to understanding snakes’ hunting behaviors. Researchers had suggested that the VNS was used only for communication between members of a species by collecting the chemical signals known as pheromones. There is evidence to suggest otherwise, however. The habit of tongue flicking is well documented in snakes and the tongue could be collecting information for both systems, although anatomy suggests it is involved with the VNS only [Figure 8–2]. Once the tongue is withdrawn into the mouth the molecules on the two tips or tines are transferred to the blind pouches in the roof of the mouth that lead to the VNS receptors.
INSERT FIGURE 8-2.
Figure 8–2. A Sidewinder (Crotalus cerastes) collecting molecules with its tongue. San Diego County, California.
In the 1960’s, Gordon Burghardt experimentally tested newborn water snakes with odors from fish, earthworms, amphibians, slugs, crickets, mice, and crayfish. Litters of newborn snakes from captive female Northern Water Snakes (Nerodia sipedon), Queen Snakes, and Graham’s Crayfish Snakes were tested with odors. Using distilled water as a control, Burghardt presented each of the naïve, newborn snakes with the odor from each potential food on a cotton swab. He counted the number of times the snakes flicked their tongues to investigate the chemicals on the swabs during a 30 second time period. The results left no doubt. The Northern Water Snake responded most strongly to the chemical signatures of the frog and fish, its natural prey; the response of the crayfish-eating snakes was weak to these chemicals, though strong to the chemical signature of a crayfish. Furthermore, these snakes responded most strongly to the chemicals of a newly molted crayfish. The neonates had never been exposed to the natural environment,; they had never sensed natural prey in any setting, but they responded to what adults of their respective species were known to eat. And, the neonate Queen Snakes could distinguish between a crayfish with a hard exoskeleton and the more edible, newly molted specimen.
In a recent follow-up, R. Mark Waters and Burghardt tested naïve newborn Queen Snakes and Graham’s Crayfish Snakes again and found crayfish chemical signatures elicited tongue flicking and strikes. But they found neonate Graham’s Crayfish Snakes to be less interested in food, suggesting hunger may be involved in the snake’s responses. Investigation of stored lipids revealed that new orn Queen Snakes had 17% of their body mass stored as fats while 24% of the Graham’s Crayfish Snakes’ bodies were composed of stored fat. Female Graham’s Crayfish Snakes were providing their offspring with more energy reserves than the female Queen Snakes. Young Queen Snakes that had not been fed for 21 days responded to the odors of newly molted crayfish and hard-shelled crayfish equally and this may have resulted from hunger. Crayfish snakes are clearly born with the ability to recognize specific prey, but this may not be true for all snakes.
Tiger Snakes (Notechis scutatus) live on mainland Australia as well as some offshore islands, and are exposed to different kinds of prey in the different locations. Fabien Aubret and colleagues examined the feeding preferences for two populations of Tiger Snakes using captive- born snakes naïve to prey, lab-reared animals with a controlled diet, and wild-caught adults that had experienced natural prey. The mainland population was from a swamp, and they fed primarily on frogs while the island population fed mostly on skinks and mice as young snakes and shifted to Silver Gull (Chroicocephalus novaehollandiae) chicks as adults. When presented with prey odors, all snakes showed interest in those taken from frogs, mice, and chicks. Interestingly, the island neonates showed significantly more interest in the odor of chicks compared to their mainland counterparts. The authors’ results suggested Tiger Snakes do show a genetic pre-disposition to prey types and that, despite this, prey choice is plastic. The island has no frogs but the offspring responded to their odor, suggesting a genetic basis for the recognition of prey from its mainland ancestors. The island snakes were also of interest because neonates and adults show distinctly larger heads than mainland snakes suggesting they have been selected for feeding on larger prey – like the gull chicks.
The VNS is important in prey recognition as well as for locating mates and finding places, like dens (hibernacula), to aggregate. Experiments with snakes that had the vomeronasal nerve cut did not respond to prey, but did respond to non-biological odors. Snakes with severed olfactory nerves could still find prey using the VNS. The VNS stimulates different areas of the snake’s brain than the olfactory system does, and snakes are highly dependent on the VNS.
Vipers and some elapids strike, bite, and then release prey to avoid the nasty bite a rodent or carnivorous mammal can deliver. They then trail the rodent using the VNS until the venom has disabled it, so the snake can swallow it in relative safety.
Kurt Schwenk reviewed the history of the VNS in snakes, and found experiments done in the 1920s and 1930s established the chemosensory function of snakes’ tongues. Schwenk wrote,
The proposed mechanism of forked tongue function does not suggest that lizards and snakes use it to sense the direction taken by the prey or conspecific. This would require detection of some sort of polarized trail. Trail direction probably is determined by tongue flicking of objects in the environment against which the target has pushed during locomotion.
M. Rockwell Parker and colleagues used Northern Pacific Rattlesnakes (Crotalus oreganus) to test the idea that snakes could follow an edge. Animals had the VNS nerve cut on one side, and were tested in a Y-maze to see if they were indeed detecting the edge of a trail. Because a strike is the stimulus for the searching behavior, the researchers allowed the snake to strike the mouse whose odor would be follow through the maze. Parker and colleagues then made a trail with the envenomated mouse. If edge detection was being used by the snakes, then the damaged nerve should have disrupted the snake’s ability to follow a trail. This, however, was not the case and the snakes with the cut nerve performed as well as unaltered snakes. The authors point out that the rattlesnakes may have been able to compensate for the handicap, but they suggest it is more likely that the tines are forked to increase the area sampled for molecules.
Prey capture and recognition techniques may vary between the sexes of a given species. The highly aquatic Arafura File Snakes (Acrochordus arafurae) show sexual dimorphism in prey detection, with males having smaller heads and bodies than females, and the tendency to to use shallow water, while females use deeper water. Shawn Vincent experimentally demonstrated that female Arafura File Snakes ambush large fish in deep water using movement to detect the presence of prey, while males use odor and chemical signatures to identify prey in their shallow water hunting grounds.
The literature contains stories about pythons swallowing fruits and, if the reports are accurate, it may be due to the snake confusing the odor of the fruit with that of an animal odor, or the results of an animal odor being present on the fruit. There are reports in the early literature on the Tentacled Snake (Erpeton tentaculatus) that report it has been known to eat small fruits. The tentacled snake is a highly aquatic snake that feeds exclusively on fish. It is an ambush predator that assumes a J-shaped posture underwater in the tangle of roots and stems of aquatic plants. When small fish venture into the strike area, an explosive strike is released and the fish may be captured. It is not difficult to imagine the snake striking at a fish and accidentally swallowing a small fruit or other piece of debris in the water, and these events provide the most likely explanation for the accounts in the early literature. Stories about snakes swallowing inappropriate objects such as light bulbs or glass eggs (used in poultry houses to induce egg-laying) in chicken coups are legendary, and they occur because snakes usually depend on the VNS to recognize food.
Snakes are exceptionally tactile; large portions of their bodies are always in contact with environmental surfaces and they undoubtedly have many touch receptors in their skin. Snakes hunting in water have been reported to use tactile cues to capture food. Richard Shine and colleagues presented free-ranging Macleay’s Mud Snake (Enhydris polylepis) with dead fish and plastic fishing lures without hooks in five different ways: (1) lying still underwater; (2) moving like a struggling fish, (3) moving like a struggling fish behind a plastic barrier, (4) moving up to and coming in contact with the snake and (5) wrapped in cheesecloth to eliminate visual cues but with chemical cues detectable. In presentations that involved movement, the authors added a splashing effect produced by a squirt gun. Unexpectedly, they found that tactile cues such as body contact and splashing stimulated the snakes to strike at the potential prey more frequently than the chemical cues. There is evidence that snakes hunting bats may also rely on tactile contact as bats fly in and out of their roosting sites. Hanging on cave walls or in trees or shrubs at the mouth of the cave, snakes strike at bats as they brush against them.
Scavenging
Road hunting for snakes in Trinidad one evening, I came across a Rainbow Boa (Epicrates cenchris) in the middle of the lane. I exited the car and the snake remained on the pavement with its nose down and tongue flicking. Under the snake’s nose was a very flat, very dead, road-killed mammal that had become one with the pavement.
Although generally considered hunters, snakes do sometimes play the role of scavenger, feeding on animals that have died as the result of something other than the snake’s own effort. Some snakes may be simply opportunistic scavengers and feed on carrion when it is found; others may have evolved, or simply learned, to specialize in scavenging food. But, at least some scavenging is the result of the snake being fooled by information they obtain through their VNS.
The Cottonmouth (Agkistrodon piscivorus) appears to be an exceptionally opportunistic species. A list of the species eaten by this snake would take up most of the page, even in tiny font. The Cottonmouth is known to feed on all major groups of vertebrates as well as such unlikely invertebrates as snails, insects, and crayfish. But what makes this snake’s feeding habits most unusual is the frequency with which it will feed on carrion.
The Florida Cottonmouth (Agkistrodon p. conanti) population on Sea Horse Key in Cedar Keys National Wildlife Refuge, Florida was studied by Charles Wharton. The island’s mangrove forests support large nesting colonies of herons (Ardeidae), ibis (Threskiornithidae), and cormorants (Phalacrocoracidae). During the nesting season (April-August), many Sea Horse Key Cottonmouths congregate under the nesting birds. When adult birds bring fish to nestlings, the regurgitated fish sometimes falls out of the nests onto the forest floor. Wharton estimated that the rookery supplied food for about 75% of the island’s Cottonmouths. The snakes fed at night and were alert to the sound of falling fish. Wharton found snakes could be attracted to disturbances in the leaf litter. Tapping the leaves with a snake hook caused a Cottonmouth to come toward it with its head raised. The sound could even elict a strike. Small fish were eaten by large snakes, and small snakes would attempt to eat fish far too large to swallow. Given their varied feeding habits, it is interesting that snakes would not attempt to feed on chicks that had fallen from their nests. In one instance, Wharton placed a live chick in the vicinity of a snake and it was ignored, but he did find a cormorant leg in one specimen, and a cormorant wing in another. Apparently, Cottonmouths will eat portions of dead birds. In another manipulation, Wharton placed 10 pieces of dead fish near a nest where snakes were waiting. One snake was observed eating a piece, and a second snake was found nearby and had eaten two pieces. Surprisingly, they did not gorge themselves on the pieces of fish. The oddity here is that some Cottonmouths will eat substantial amounts of food if the opportunity arises, suggesting that individuals of a species may have distinctly different table manners.
A laboratory study of Cottonmouth feeding behavior done by Barbara Savitzky found Cottonmouths fed most often at the water-land interface; they took dead fish more often than live fish, and they rarely submerged when foraging. However, captured fish were occasionally lost and the snake had capture success of 75%. When prey handling skills of the Cottonmouth were compared to those of the Diamondback Water Snake (Nerodia rhombifera), the water snake was more skilled at capturing and manipulating prey. The Cottonmouths had longer prey handling times because they frequently grasped the fish at mid-body so small fish were easily folded and swallowed. Most of the successful fish captures and scavenging was done at the water’s edge, and in near shore environments of Savitzky’s arena, where the snake could drag the potential meal onto land for swallowing. Savitzky concluded that while Cottonmouths are not specialists in feeding on fish, they are competent predators, and she suspected scavenging on carrion may well be a feeding strategy for this species. Most pitvipers are terrestrial–arboreal snakes; the only species known to regularly use water is the Cottonmouth, but its transition to aquatic environments seems incomplete.
Harvey Lillywhite and colleagues followed up on Wharton’s work at Sea Horse Key and discovered that some of the Cottonmouths that fed under nests also foraged along the intertidal zone. They documented snakes feeding not only on dead fish, but on green and brown algae and associated debris. The authors point out that snakes scavenging on fish-scented algae may obtain calories and nutrients in an environment when normal prey may be scarce.
Perhaps the most unusual report of Cottonmouth scavenging comes from George Heinrich and Karl Studenroth. They observed a Florida Cottonmouth scavenging on the carcass of a wild hog. The hog had been butchered and dumped from a bridge crossing a creek. They observed the snake manipulating long strips of fat that were eventually swallowed with difficulty. Additionally, Hamel described a one meter Cottonmouth that peeled a dried, flat, road-killed snake off the pavement before consuming it. Clearly, the Cottonmouth is one snake that will readily eat carrion, but there are others.
Western Diamondback Rattlesnakes (Crotalus atrox) will also eat carrion, and they will go to some effort to gain access to decomposing carcasses. Jim Gillingham and Randy Baker buried decomposing mice and found Western Diamondbacks could readily locate them. However, they could not locate fresh, recently deceased mice that had not started to decompose. Other species of rattlesnakes will also feed on carrion. Funderburg reports an Eastern Diamondback Rattlesnake (Crotalus adamanteus) eating a road-killed cottontail rabbit (Sylvilagus sp.). Boas and pit vipers are not the only snakes to eat carrion; colubrid snakes also feed on dead animals.
Indigo snakes (Drymarchon corais couperi) are the largest North American snake and are dietary generalists. Audrey Smith and Fred Antonio found a 2 m specimen feeding on the head of a decapitated shark. The snake would thrust its head into the musculature of the head and bite and twist its head to tear off a piece of tissue. In yet another example of carrion feeding, James Lee found a partially decomposed squirrel in the digestive system of a Black Pine Snake (Pituophis melanoleucus lodingi).
Snakes are not strictly predators, and snake keepers have long known that captive snakes will readily accept dead food. Travis DeVault and Aaron Krochmal surveyed the literature and found 39 publications that reported 50 observations of carrion feeding by snakes, 43 of which were in the field. And, while 35 species from five families were involved, pitvipers and fish-eating species were represented more frequently than other groups. Snakes may scavenge more often than previous thought. DeVault and Krochmal also proposed that snakes may fill a niche not occupied by any other vertebrate; that is, they can feed on dead organisms that cannot be accessed by other vertebrate scavengers. Yet, most snakes don’t make a living as scavengers. Instead they hunt prey by a variety of strategies.
Ambush
The sun had set over Thailand’s Khao Yai National Park, and we were looking for amphibians and reptiles to photograph. The path through the rainforest was bricked and intended for use by tourists who wanted a rainforest experience without getting their feet muddy. Less than a five minute walk from the parking lot, a Vogel’s Green Pitviper (Viridovipera vogeli) was loosely draped over a vine about 15 or 20 cm off the ground. Its head was pointed down. Immediately behind the snake’s perch was a log and, as I leaned in to photograph the snake, movement caught my eye. A microhylid frog was moving along the log toward the snake. My presence caused the frog to reverse direction and probably denied the snake a meal. Vipers and many other snakes ambush prey, and this Vogel’s Green Pitviper was clearly waiting for its next meal to be directed toward its waiting fangs. The vine was an excellent ambush site because the log acted like a wall, which created a natural path directing small animals to follow its length [Figure 8–3].
INSERT FIGURE 8-3.
Figure 8–3. Vogel’s Pit Viper (Viridovipera vogeli) waiting in ambush in central Thailand.
Snakes have specific areas or locations where they forage and where they ambush prey. Ivan Sazima and Octavio Marques observed an Eastern Green Lora (Philodryas olfersii) on the same tree in a suburban environment three times over a period of 54 days. At each observation, the snake had caught a bird. Even snakes that actively forage show site fidelity.
A field study of the Costa Rican Black-headed Bushmaster (Lachesis melanocephala) by Harry Greene and Manuel Santana reported a 90 cm female that used three sites for 3, 6, and 25 days. During the observation period it moved a total of 50 m. The female spent days under small plants, while at night she sat in an alert posture, exposed on the forest floor. On night 24 she ate a rodent that weighed about 40% of her mass. This was followed by nine days of inactivity before moving to a new site. Ambush sites used by this female, and others, were associated with Welfia palms; the palm seeds are eaten and dispersed by the rodents that bushmasters prey on.
Timber Rattlesnakes (Crotalus horridus) used logs as ambush sites when Howard Reinhardt and co-workers radio-tracked 21 individuals over three seasons. They described the ambush posture as coiled next to a log with its head and anterior body directed upwards, toward the top of the log. The snakes were waiting for their rodent prey, usually white-footed mice in the genus Peromyscus.
Using video surveillance equipment, Rulon Clark collected information on 17 Timber Rattlesnakes over a period of two years. He was able to record 87 encounters between the rattlesnakes and their rodent prey. The snakes were studied in a New York nature preserve and had been surgically implanted with radio transmitters. Ambush sites were used for time periods that ranged from several hours to as long as several days, with an average time of 17 hours. The snakes’ successfully captured 13% of the prey that came within striking distance. Clark’s data suggests that Timber Rattlesnakes feed 12 to 15 times per season, and they consume 1250 to 1550 g of prey per season. Previous authors had estimated that a 500 g Timber Rattlesnake would need about 280 g of rodents per year to sustain itself. If these snakes were getting more than 1200 grams of rodents per season, the snakes had significantly more energy for growth, fat storage, and reproduction. Feeding usually took place at night (81% of the events recorded) and the rest occurred in the morning.
In an effort to determine how prey odors may influenced C. horridus ambush sites, Clark used captive-born snakes and an experimental manipulation of prey odors to determine if feeding experience caused snakes to choose one ambush site over another. He found snakes were more likely to pick an ambush site next to chemical trails from prey they had previously experienced. The size of the prey the snake had eaten seemed to make a difference as well. Clark suggested that snakes learn to recognize and remember odors of prey that are more profitable and choose ambush sites accordingly.
Individuals may gain knowledge of the presence of predators or the location of food or water by watching other animals; this is public information. Clark used a t-maze experiment to examine the behavior of Timber Rattlesnakes to see if they were using chemical signals from other snakes that had recently fed. Snakes that had recently fed or had not recently fed were allowed to crawl around a box, then down a channel, and then to the left or right. In one set of experiments, there was only one trail to follow. In a second set of experiments, there were two trails, one trail from a fed snake and another trail from an unfed snake. In a third set of experiments, a water dish and hide box were added and the snakes were allowed to stay in the compartment they chose for 72 hours. In this third set of experiments, both fed and non-fed snakes were used. The results suggested snakes were indeed using the chemical information in the trail laid down by the previous snakes. In the first experiment, all the snakes followed the trail. In the second experiment, nine snakes followed the trail of the fed snake, six followed the trail of the un-fed snake, and one did not follow the trail. In the third experiment, the Timber Rattlesnakes spent more time in ambush posture on the substrate marked with the odor of a snake that had recently fed as compared to the substrate marked by a snake that had not fed. Timber Rattlesnakes used public information in this experiment and it seems likely they use it in the wild.
Vipers are capable of distinguishing between envenomated and non-envenomated prey. Using Western Diamondback Rattlesnakes (C. atrox), David Chiszar and colleagues presented C. atrox with envenomated rodents the snakes had bitten and non-envenomated rodents. The snakes selected the envenomated animals more than 70% of the time. In a second experiment, they injected a standard amount of venom into the rodents and presented them to the snakes. Again, the snakes chose the envenomated prey. And, in a third experiment, they injected minimal amounts of venom and the snakes could still discriminate between envenomated and non-envenomated rodents. In a follow-up study, Chiszar and colleagues used C. atrox, the Prairie Rattlesnake (Crotalus viridis), and the Northern Pacific Rattlesnake (Crotalus oreganus). Snakes were presented with rodents injected with their own venom and rodents injected with venom from other members of the same species as well as different species. Again, snakes preferred prey envenomated by members of their own species, and tended to reject prey envenomated by a different species. The results imply that vipers can identify prey that they have bitten. Moreover, rattlesnakes trailing prey ignore the chemical trails of rodents not envenomated, and chemical trails of those that may have been envenomated by a different species.
Hadas Tsairi and Amos Bouskila investigated the ambush sites selected by the desert-dwelling Colored Carpet Viper (Echis coloratus) at an oasis [Figure 8–4]. They found the sites were not random; the snakes used locations that were less than 5 m from water on raised objects. Tsairi and Bouskila used experiments with captives in an effort to determine what cues the snakes were using to select an ambush site. They found the odor of gerbils (Gerbillinae), was probably not involved. Instead, the snakes were choosing locations that had high humidity.
INSERT FIGURE 8-4.
Figure 8–4. The Colorful Carpet Viper (Echis coloratus) is one of about a dozen species in the genus Echis that have heavily keeled scales and create a rasping noise when the sides of the body are rubbed together.
Luring Prey with Tails and Tongues
Snakes waiting in ambush use their tails and tongues to bring prey within striking distance. Cottonmouths do not always scavenge; they also use the hunting techniques employed by other vipers, and those hunting techniques often change with the age of the snake. Evan Eskew and colleagues used Ellington Bay, South Carolina as a study site to examine Cottonmouth hunting techniques. They searched the area on 10 nights, and identified 51 individual Cottonmouths, and collected data on 45 of them. Eight of the snakes were adults and 37 were juveniles. Cottonmouths are born with a yellow-tipped tail that disappears with age, usually when the snakes are 400–600 mm long [Figure 8–5].
INSERT FIGURE 8-5.
Figure 8–5. A neonate Western Cottonmouth (Agkistrodon piscivorus leucostoma). The light (yellow) tail tip is used for luring prey. Union County, Illinois.
Eskew and colleagues found juvenile snakes were almost always tightly coiled at the edge of the wetland habitat while the adult snakes were outstretched or active in the water. About 68% of the juvenile snakes contained newly metamorphosed mole salamanders, while the adults fed on frogs, snakes, and birds.
Brightly colored tails occur frequently in vipers, some elapids, and pythons. Wilfred Neil described caudal luring in juvenile Copperheads (Agkistrodon contortrix) in 1948. A gravid female from Richmond County, Georgia gave birth to 11 young. These were placed in a box with a substrate of dead leaves. Several cricket frogs were placed in the box as food for the young snakes and, about an hour later, Neil peeked into the box. He wrote,
In the shadowy interior of the container I could at first make out only a number of writhing, yellowish objects, for all the world like small worms or maggots. Closer inspection revealed that each little copperhead was coiled up and was holding aloft its bright yellow tail, which was writhing slowly.
In a series of experiments, Randall Reiserer examined stimulus control of caudal luring in four species of pitvipers. When juvenile Copperheads (Agkistrodon contortrix) were presented with caterpillars they did not caudal lure. Instead, they pursued the larva and did not wait for venom to immobilize the insect, swallowing it immediately. On the other hand, when the copperheads were presented with chorus frogs (Pseudacris sp.) or Plains Leopard Frogs (Lithobates blairi), they did lure with the tail but the frogs did not respond. Of 51 trials with frogs, there were five pursuits and 39 feedings but not because the frogs responded to the lure. In one case, a sibling lured another to its tail, though the investigating snake stopped after tongue flicking the lure. While the Copperheads were unsuccessful at luring frogs, they were much more successful with lizards. Western Massasaugua (Sistrurus catenatus tergeminus) were presented with cricket frogs (Acris) and Five-lined Skinks (Plestiodon fasciata) while Desert Massasaugua (S. c. edwardsi) were presented with spadefoot toads (Scaphiopus), Sagebrush Lizards (Sceloporus graciosus), or Western Fence Lizards (Sceloporus occidentalis). The Western Massasaugua from wet environments lured frogs but pursued lizards, while the Desert Massasaugua did not eat frogs, but successfully lured and ate lizards.
Reiserer also tested the Sidewinder (Crotalus cerastes). Caudal luring in this snake could be induced by live prey or models made from dowel rods. While the snakes usually struck at live, slow moving lizards, they ignored the wooden models. The Sidewinders had a low rate of success trailing lizards that had been struck, and previous studies suggested that this rattlesnake was not adept at trailing prey after a strike.
The Horned Adder (Bitis caudalis) from South Africa, a species known to bury itself in sand (like the Sidewinder), was also tested by Reiserer. Adult Horned Adders were presented with lizard, mouse, and dowel rod models. The snakes were partially buried in sand and lizard models elicited caudal luring on two occasions; the presence of the mouse model caused the snakes to strike or purse the prey model, and the dowel model did not stimulate a feeding response of any kind. Caudal luring is stimulated by specific prey in specific predators. Sidewinders and massasaugua appeared to respond to objects that have simple shapes and movement while Horned Adders have better prey recognition skills.
In a follow-up study, Reiserer and Gordon Schuett examined the Sidewinder’s response to lizards they would encounter in their natural habitat as well as lizards not found in their desert habitat. The authors also tested how the snakes would react to a potential predator, a large Colorado River Toad (Ollotis alvaria). The Sidewinders lured more frequently to lizards that used the same habitat, and these species were more attracted to the snake’s tail lures than species from different geographic areas. When a large toad was placed outside the cage, but in view of the neonate sidewinders, the snakes stopped wiggling their tails and went into a defensive posture. A disk was used as a control for a predator, but the snakes never responded to it in a defensive manor. The results suggest that caudal luring is not the result of a generalized stimulus, but that snakes have the ability to recognize visual, olfactory, and perhaps vibration cues from potential prey, at the time of birth or shortly after, without having to learn from experience.
Ali Rabatsky and Jane Waterman used the Dusky Pygmy Rattlesnake (Sistrurus miliarius barbouri) to study shifts in diet and sex differences in caudal luring. Using Brown Anoles (Norops sagrei) for prey they found that adult snakes never caudal lured, and that the juveniles that lured for longer periods were more successful. Juvenile females had shorter tails than males and, in this study, females had to lure for longer times to be as successful as males.
Experiments with the Northern Australian Death Adder (Acanthophis praelongus) by Mattias Hagman and colleagues included videotaping the snakes’ luring. The snakes wiggled their tails more vigorously toward lizards than towards frogs and, while lizards approached the lure, frogs did not. The authors also controlled for the size of the lure by removing tails from dead snakes and mounting them on a mechanisms that would duplicate the snake’s tail movements. Hagman and colleagues found the smaller lures were more effective than large lures.
Snake tails used as lures often have yellow or white coloration with contrasting darker pigments. Tails usually do not resemble any specific organism, and they may be best regarded as a non-specific mimic. However, as we have seen, specific types of prey usually respond to the tail lure while others do not. There is one spectacular exception to the non-specific nature of tail lures. One snake has altered its tail morphology dramatically. The tip of its tail mimics an arachnid.
Hamid Bostanchi and colleagues described an unusual viper in 2006. A specimen of False Sand Viper (Pseudocerastes) first collected in 1968 had greatly elongated scales on its tail and a swollen tip. It was not until a second specimen was collected, in 2003, that this snake was recognized as a new species, the Spider-tailed Viper, P. urarachnoides. Bostanchi and co-workers wrote that the tail ornamentation resembled, “… an arthropod clinging to the tail tip.” The name urarachnoides means “spider-like.” The authors speculated that the unusual tail is used for caudal luring, and one of the two known specimens had a bird in its stomach. My own examination of the tail tip suggests that it is indeed a remarkable mimic of a spider or solfugid, and a video posted on the Internet shows this snake moving its tail in a fashion that makes it look remarkably like an arachnid [Figure 8–6].
INSERT FIGURE 8-6.
Figure 8–6. The tail ornamentation of the Spider-tailed Viper (Pseudocerastes urarachnoides). The Spider-tailed Viper was described in 2006 from southern Iraq and has the most remarkable tail found in any snake. The elongated scales and swollen tail tip mimic an arachnid as the snake moves the tail over the surface of the sand. The structure and movement apparently attracts birds. Below is a Solifugae of the genus Galeodes, also known as sun spiders, camel spiders, and wind spiders. These arachnids live in the same habitat with P. urarachnoiodes and it seems likely their tail is meant to mimic them.
Tails are not the only structure snakes use to lure prey. The Northern Water Snake (Nerodia sipedon) and the Diamondback Water Snake (N. rhombifera) have been observed “fly casting” with their tongues in order to attract fish. James Czaplick and Richard Porter observed captive water snakes approach a water pan with goldfish; when they were 2 cm from the water, the snake would stop, hold its head above the water, flatten its body, and glide onto the water. While afloat, the snake flicked the water with its tongue several times so the tips of the tines only slightly broke the surface. This behavior attracted fish, and the snake would orient toward the fish and capture it.
A more detailed study of lingual luring was provided by Hartwell Welsh and Amy Lind. They studied the Aquatic Garter Snake (Thamnophis atratus) in northwest California and found adults feed on the Pacific Giant Salamander (Dicamptodon ensatus), while juvenile T. atratus feed on tadpoles, small salmon, and trout. The neonate Aquatic Garter Snakes assumed an ambush position on the shore, extending their tongue onto the water’s surface in a rigid position with the tines quivering on the surface, possibly mimicking insects. Small fish were attracted to the vibrations or the sight of the tines and the snakes struck when the fish were within striking distance. The Aquatic Garter Snakes extended their tongues for an average of 4.7 seconds while luring fish, compared to only 0.43 seconds for a normal tongue flick used to collect information for the VNS. The increased time spent with the tongue extended suppots the idea that the snake is indeed luring prey, not just routinely flicking its tongue.
The Mangrove Water Snake (Nerodia clarkii compressicauda) inhabits southern Florida and Cuba, and is known to feed only on fish. Kerry Hansknecht used a litter of neonate Mangrove Water Snakes to study lingual luring, a behavior not known to occur in adults. Snakes in the water and in the presence of fish, protrude the tongue so that the tines curl back making contact with the rostrum. As more of the tongue was protruded, the tines would curl and form a loop that would extend away from the snake’s head and flex in the direction of the fish. The tongue would also twist during the display. Displays lasted an averaged 10.95 seconds, compared to normal tongue flicking with a mean time of 0.31 seconds.
Actively Foraging
The Brown Treesnake (Boiga irregularis) is a generalist predator that actively forages and has been extensively studied as a result of its invasion of Guam. Since Brown Treesnakes are primarily nocturnal, Steven Campbell and colleagues examined the influence of moonlight and prey on its hunting habits. A large environmental chamber was used to control light, and mice and lizards were used as prey. The snakes were tested to see where the snake would hunt under different lighting conditions and with different prey. Moonlight was the factor that determined where the snake hunted. With little or no moonlight, Brown Treesnakes hunted on the ground. As moonlight increased, the snake moved into the canopy and the presence or absence of prey was not a determining factor. The authors proposed three possible reasons for the snakes’ behavior. Bright moonlight may make snakes more susceptible to predation; terrestrial prey may be less active in moonlight; or Brown Treesnakes may need visual cues for hunting.
Foraging snakes take food that is usually more sedentary than they are, and bird nests provide an opportunity for snakes to locate stationary food. Patrick Weatherhead and Gabriel Blouin-Demers report that six of eight studies using video cameras to monitor bird nest in the Western Hemisphere found snakes to be the most important nest predators. Nests, particularly those in fragmented habitats, seemed to be the most susceptible. Parent birds moved to and from the nests and attracted the attention of snakes. Birds may also have selected nesting sites that are difficult for the snakes to reach.
In a tropical, moist forest W. Douglas Rob inson and co-workers used surveillance cameras to continuously monitor 17 nests of two different species of antbirds on Barro Colorado Island in Panama. Ten predation events were recorded on the bird nests. A Capuchin Monkey (Cebus) was responsible for one event, a Coatimundi (Nasua) was responsible for a second predation event, and the Bird Snake (Pseustes poecilonotus) was responsible for eight of the predation events. In one instance, a pair of Chestnut-backed Antbirds (Myrmeciza exsul) arrived at the nest four minutes after the arrival of a snake. The parents attacked the snake for 45 minutes with vocalizations, wing-spreading displays, and pecked at the snake with their beaks while the snake ate their chicks.
Egg-eating is a specialized diet found in many snake lineages. Alan de Queiroz and Javier Rodríguez-Rob les have shown that egg-eating habits in snakes were likely derived from species already feeding on the corresponding animals. This probably also holds true for species that specialize in eating fish and frog eggs.
Six species of African egg-eating snakes (Dasypeltis) and one poorly known Indian species, (the Indian Egg-eating Snake, Elachistodon westermanni) are believed to feed exclusively on bird eggs. The African egg-eating snakes have modified their quadrate, dentary, and supratemporal bones. They lack functional teeth, and have vertebrae with enlarged, forward- projecting processes of bone that are capped with enamel. The modified vertebrae act as teeth that fit into loose folds of the esophagus to prevent the shelled egg from proceeding into the gut. Once the snake has slit and collapsed the shell, its pieces are regurgitated and the contents swallowed.
Gabriel Gartner and Harry Greene compared the egg-eating performance of the African Forest Egg-eating Snake (Dasypeltis atra) with the Common Kingsnake (Lampropeltis getula). Kingsnakes occasionally eat eggs, but are not specialized for the task. Kingsnakes struggled with eggs, which often slip out of their mouths. Only the largest kingsnakes could actually eat an egg, and they lacked the ability to collapse the egg into a smaller package, making eating multiple large eggs impossible. On the other hand, the specialized African Forest Egg-eating Snake can handle huge eggs, relative to its head size, and eat several because the shell is slit and regurgitated, making room for more eggs. The open savanna and desert habitats of eastern and southern Africa have a large number of small, ground-nesting finches and weaver birds that place their nests in low shrubs or grasses, making it possible for the egg-eating Dasypeltis to specialize on bird eggs. Penn Lloyd reported the Rhombic Egg-eating Snake (Dasypeltis scabra) may account for 13–70% of nest predation in South African rangelands.
Some snakes specialize in feeding on squamate eggs. For many years, it was thought that the two species of North American Leaf-nose Snakes (Phyllorhynchus) were dietary generalists, but the Spotted Leaf-nose Snake (P. decurtatus) [Figure 8–7] was later suspected of feeding on gecko tails. Stephanie Gardner and Joseph Mendelson examined 410 preserved specimens of the two species and found 36.5% of the snakes that contained food, contained remains of eggs. Only 1.8% contained gecko tails. In Australia, the small, burrowing shovel-nose elapids of the genus Brachyurophis have modified, blade-like teeth in the back of the mouth, and many of them feed on reptile eggs. In Asia, at least some of the kukri snakes of the genus Oligodon specialize in feeding on reptiles eggs.
INSERT FIGURE 8-7.
Figure 8–7. The Spotted Leaf Nosed Snake, (Phyllorhynchus decurtatus) is an egg-eating specialist from western North America. Pima County, Arizona.
Two species of sea snakes have reduced fangs and venom glands and feed only on fish eggs. Harold Voris discovered that the Turtle-headed Sea Snake (Emydocephalus annulatus) was a fish egg predator in the mid-1960’s. Twenty years later, Michael Guinea of the Northern Territory University observed the Turtle-headed Sea Snake feeding on Ashmore Reef north of Australia and saw the snakes rubbing their labial scales on coral, scraping the eggs off the coral reef and eating them. Richard Shine and colleagues studied this snake’s feeding behavior in New Caledonia and found it swims slowly, investigating crevices and burrows in search of fish nests. What distinguishes this snake from others is that it ingesting tiny eggs, eggs hundreds of thousands of times smaller than the snake.
Small snake species often feed on invertebrates and some specialize. Otavio Marques and colleagues examined 146 specimens of the Agassiz’s Rear-fanged Snake (Pseudablabes agassizii). They found spiders and scorpions made up 72% of the diet, insects made up another 24%. Agassiz’s Rear-fanged Snake is found in the shrub grasslands of southern South America known locally as the campo sujo. It forages during the day, probing burrows in search of spiders. While some of the spiders it feeds on are small and easily overpowered, it also feeds on mygalomorphs― tarantulas. These are potentially dangerous prey that could kill or severely injure the snake. When handling these large arachnids, the snake envenomates them and usually swallows them abdomen-first to avoid the tarantula’s venom as well as the spiders’ bites. Pseudablabes agassizii is a dwarf species, rarely exceeding 0.48 m, though it is part of a clade of snakes that often attain a meter in length. Its larger relatives feed on lizards and birds, and it seems likely Agassiz’s Rear-fanged Snake evolved a small body size and an invertebrate diet to fill a niche not used by other clade members.
Snails and slugs are slow-moving prey and must be actively sought by the snakes that specialize on them. Despite the fact that gastropod-eating snakes are in several different families, they show some convergent morphology and behavior. All tend to have large eyes, short, rounded heads, and they all wipe the sticky secretions from the mollusks off their mouths after eating [Figure 8–8]. In North America, the brown snakes of the genus Storeria (family Natricidae) feed on slugs and earthworms, and in the tropics several genera of the family Dipsididae (Dipsas, Sibon, Tropidodipsas and Sibynomorphus) tend to specialize in gastropods. In Asia the members of the small family Pareatidae usually feed on gastropods. And, in Africa the snakes of the genus Duberria (family Pseudoxyrhophiidae) are the snail and slug-eating species.
INSERT FIGURE 8-8.
Figure 8–8. Two gastropod-eating snakes in different clades. A: The Trinidad Snail-eating Snake (Dipsas trinitatis) endemic to the island of Trinidad in the West Indies is in the family Dipsididae. B: The Keeled Slug-eating Snake (Pareas carinatus) from Thailand is in the family Pareatidae.
Snakes that eat snails face the problem of extracting them from their shells, a feat that requires the snake to get its teeth into the soft tissues and pull the snail from its shell. Snails can coil their shells to the right (dextral) or to the left (sinstral), but shells are most often coiled to the right. Snails that coiled their shells to the left were thought to suffer disadvantages because so many individuals coil to the right. Masaki Hoso and colleagues found Iwasaki’s Snail-eating Snake (Pareas iwasakii), and other members of the Pareatidae that specialize in feeding on snails, have increased the number of teeth on the right side of the jaw. In tests using Iwasaki’s Snail-eater with dextral shelled snails, the authors found the snakes could extract the snail with fewer mandible retractions than if the snail’s shell was sinstral. Snakes presented with sinstral snails dropped them more often and took longer to extract the snails from the shells.
The European Ladder Snake (Rhinechis scalaris) is a large colubrid snake that may reach 1.6 meters in length, and it forages for endotherms. About 87% of its diet is composed of mammals and 11% is bird; of these, 46% of the prey items were nestlings. The Ladder Snake actively searches for nests and burrows. Juan Pleguezuelos and colleagues examined 368 Ladder Snakes and correlated their morphology and diet to their foraging mode. Many snake species have adults that feed on endothermic prey, while hatchlings or neonates feed on ectotherms (frogs, fish, lizards or invertebrates). Pleguezuelos and colleagues found this was not the case with the Ladder Snake. Instead, hatchling snakes ate nestling mice and birds and they continued to feed on small prey until they were 900 mm in body length.
Specialized diets have been considered disadvantageous because they limit the snake’s options, but there can be advantages. E. J. Britt and colleagues compared populations of the Western Terrestrial Garter Snake (Thamnophis elegans) that ate fish to populations that ate slugs for efficiency of food assimilation. The coastal population of T. e. terrestris fed on the nutritionally poor slug, Ariolimax columbianus. On the other hand, the inland population of T. e. elegans fed on the nutritionally rich fish, the Speckled Dace (Rhinichthys osculus). In the lab, both groups of snakes were fed both prey. The authors found the slug-eating coastal snakes could assimilate 62% more energy from the slugs than could the fish-eating inland snakes. The coastal population was well adapted for feeding on slugs in terms of behavior and physiology while they retained the ability to digest fish.
Shifting Diets
Snakes adapt as evolution tinkers, and island populations are good places to examine these changes. The Eastern Garter Snake (Thamnophis sirtalis) is a generalist predator feeding on a large variety of species ranging from leeches and earthworms to fish and frogs, though birds and mammals are uncommon in its diet. Martin Greenwell and colleagues collected gravid female Eastern Garter Snakes from High Island in Lake Michigan and from nearby Pellston, Michigan. After the young were born, they tested them with odors from earthworms, frogs, birds, fish, mice, and distilled water as a control. They found the responses from the two populations to be about the same.While the two populations had no statistical difference in dietary preference, and both showed the strongest response to the odor of earthworms, the island population showed a strong response to frogs and the mainland population had a strong response to birds. Despite that, the island garter snakes fed heavily on nestlings of the Common Tern (Sterna hirundo), and the authors concluded the snakes had only recently learned to exploit the bird resource.
While a species’ diet may differ from one geographic area to another, diets may also be altered by events in the environment. Paul Hampton and Neil Ford studied snake diets in flood plain forests and compared prey taken in years with flooding and years without. They used three natricids, the Plain-bellied Water Snake (Nerodia erythrogaster), the Banded Water Snake (Nerodia fasciata), and the Ribbon Snake (Thamnophis proximus). In years with flooding, dietary overlap between the three species increased; in years without flooding, diet overlap decreased. Additionally, two species, the Plain-bellied Water Snake and the Ribbon Snake took less prey in years without floods. Rising water transported fish into side pools that were usually fishless, and the diversity of fish available to snakes increased. In years without flooding, the Plain-bellied Water Snake ate fewer frogs and more salamanders.
In the early 1990’s, the Round Goby (Neogobius melanostomus), a small fish from the Caspian Sea, was introduced into the North American Great Lakes from ballast water discharged from a cargo ship. The alien goby feeds on two other invasive species from the Caspian region, the zebra mussel and the quagga mussel. But the goby also feeds on native invertebrates as well as the eggs and larvae of native fishes. The island-dwelling Lake Erie Water Snake (Nerodia sipedon insularis) has recently adapted to this new prey. Richard King and colleagues examined Lake Erie N. s. insularis specimens collected between 1948 and 2004. Prior to the goby infestation, the Lake Erie Water Snake fed on native fishes and the salamander known as the Mudpuppy (Necturus maculosa). Lake Erie Water Snake specimens collected prior to 1992 showed no gobies in their diet, but gobies made up 25% of the diet in specimens collected between 1996 and 1998. By 2002, the gobies composed more than 90% of the snake’s diet. Gobies have a higher fat content than the native species. As a result, the water snakes’ growth rate increased, and the population now attains larger body sizes, matures earlier, and produces larger litters. Snakes adapt to new prey and, as they do, the new diet brings about changes in morphology and physiology.
A similar situation has been documented in Australia’s Red-bellied Black Snake (Pseudechis porphyriacus) and the Common Bronzeback (Dendrelaphis punctulatus). Ben Phillips and Richard Shine predicted that snake populations exposed to the highly toxic, introduced Marine Toad (Rhinella marina), would adjust their gape and body size. Snakes are gape-limited predators, meaning the size of the prey they can consume is limited by the degree to which the snakes can open their mouths. Snakes with larger heads have a larger gape. The presence of toxic toads should favor snakes with a smaller gape because a small head would limit the snake’s ability to swallow a toad large enough to kill it and give the snake a better opportunity to handle the toad toxins. Comparing populations of snakes that had been exposed to toads at various points in time since 1935 (the date of the toad’s introduction), the authors found the average body size had increased, while the average head size decreased. The same calculations for the non-toad-eating Marsh Snake (Hemiaspis signata) and the toad-eating Australian Keelback (Tropidonophis mairii) showed no changes in head or body size over the same time period. In a follow up study on the costs associated with eating toxic toads, John Llewelyn and colleagues examined Marine Toads as prey for the Australian Keelback. They measured the time involved in swallowing a toad, the snakes’ ability to move after consuming toads, and the possibility that toads are poor in nutritional value. Indeed, toad toxins did affect the snake’s ability to swim; the larger the dose of toad toxin, the slower the snakes swam. Snakes took longer to consume toads than tree frogs. Additionally, snakes that fed exclusively on toads lost body weight while those fed on tree frogs gained weight. Unlike the Lake Erie Water Snakes that thrived on the introduced gobies, the Australian snakes did not fare well on a diet of toxic Marine Toads.
The arms race analogy is important to understanding why snakes have such powerful toxins, toxins that are often described as being able to kill hundreds or thousands of prey items with micro-amounts of venom. Alternatively, the concept explains how snakes have evolved to cope with their prey and the defenses those prey have developed as a means to avoid being eaten by snakes.
The Sierra Garter Snake (Thamnophis couchii) and the Eastern Garter Snake (Thamnophis sirtalis) have populations that feed on the California Newt (Taricha torosa) and the Rough-skinned Newt (Taricha granulosa) respectively. Unlike the Australian snakes that were feeding on introduced toxic toads, the garter snakes have co-evolved with the toxic salamanders. Edward Brodie and colleagues discovered the newts have increased their production of tetrodotoxin (TTX), a molecule that blocks sodium channels in nerve and muscle cells and prevents nerve impulses from being generated. The toxicity of TTX is high and it works on a wide range of species. All populations of garter snakes seem to have some resistance to TTX but populations that live with Taricha newts have elevated their resistance to TTX, by changing the affinity of their sodium channels to bind to TTX.
This tale of coevolution took an unusual turn with the discovery that larval newts are capable of detecting TTX at very low concentrations. Apparently the larvae use this ability to avoid adult newts that will eat them. Richard Zimmer and colleagues argue that the arms race proposed by Brodie may be one sided, and the TTX acts as general predator deterrent to a variety of potential invertebrate and vertebrate predators. Supporting their hypothesis is the fact that garter snakes are dietary generalists, and do not forage specifically for newts. Zimmer and colleagues experimentally demonstrated that larval newts immediately sought cover when TTX was detected. The response was strongest when the larvae were between 3–5 weeks old. By seven weeks, however, the larvae were too large to be eaten by adults and the response no longer occurred.
Snakes also shift their diets as they grow. Juvenile snakes need smaller prey, but as they grow taking larger prey is more efficient. Juvenile snakes often use a different habitat than adult snakes and changes in habitat lead to changes in prey availability. Henry Mushinsky and co-workers found two water snakes, the Plain-bellied Water Snake (Nerodia erythrogaster) and the Banded Water Snake (N. fasciata), feed on fish until they are about 50 cm in length, at which time they switch to frogs. Similar reports have been made concerning the Yamakagashi (Rhabdophis tigrinus) and the Viperine Water Snake (Natrix maura). In these instances, the young snakes changed from small fish-eating snakes to large frog-eating snakes. But the reverse also occurs. The Cape Garter Snake (Thamnophis validus) feeds upon frogs until it is about 50–70 cm long and then begins feeding on fish.
Hyperpredation
The high altitude lakes of California were formed by Pleistocene glaciers scouring mountain slopes. Fish have not been able to naturally colonize these lakes because of the steep terrain but for the last two centuries humans have stocked the lakes with trout and salmon. The introduced fishes have had long-term, unintended consequences for the ecosystem. The Cascades Frog (Lithobates cascadae) is endemic to the higher elevations of the Cascade Ranges of Washington, Oregon and northern California, and its populations have been in decline for several decades from air pollution. The Aquatic Garter Snake (Thamnophis atratus) and the Eastern Garter Snake (Thamnophis sirtalis) also occur in the Cascade ranges to elevations well above 1900 meters, and both are natural predators of the Cascades Frog.
Karen Pope and co-workers presented data that suggested the Cascades Frog is the subject of hyperpredation by the Aquatic Garter Snake. Hyperpredation is the abnormally high predation of native prey species by a predator population that has increased because of highly abundant exotic prey. Stomach analysis of the Eastern Garter Snake showed that it fed exclusively on amphibians while the Aquatic Garter Snake feeds on both fish and amphibians. The two snake species were seldom found using the same landscapes, and Pope and colleagues suggest the overlap in diets prevents the two species from coexisting. However, the Aquatic Garter Snake’s presence was closely correlated with the presence of trout that prey on the Cascades Frog’s tadpoles. The introduced trout provided more food for the snakes and, the more abundant garter snakes took more tadpoles and frogs. Since trout and amphibians feed on insects they might be considered competitors; and since they are also directly involved as predator and prey, it makes for a complex set of relationships.
Another series of complex food web relationships exist on Guam where the introduced Brown Treesnake (Boiga irregularis) has eaten its way through much of the native fauna. This snake will be discussed in more detail in Chapter 10. But it is interesting to note that concern for this snake spreading to other islands centers around introduced rodents. The rodents may serve as a food source that allow B. irregularis to reach high population densities while it simultaneously decimates the native endemic fauna. Andrew Wiewel and colleagues wrote,
By the time Brown Tree Snake predation pressure began to reduce introduced prey densities (forcing a decline in Brown Tree Snake density), many native species were already extinct.
Hyperpredation is a consequence that was unforeseen. As humans move species into new environments around the globe, unexpected consequences will become more frequent.
The Co-Evolution of Prey
The arms race between predator and prey has already been discussed in several different contexts. Primates evolved better vision and fear of snakes to avoid being bitten, and numerous prey species evolved resistance to venom, forcing snakes to evolve more toxic venom. In turn, snakes have evolved cryptic morphology and behaviors, the capacity to mimic other snakes, and behaviors that help them evade predators. How do the prey species push-back and defend themselves against snakes?
Outright aggression toward a snake predator can be effective. Wen-San Huang experimentally manipulated predation on free-living Long-tailed Skink (Mabuya longicaudata) nests by the Taiwan Egg-eating Kukri Snake (Oligodon formosanus). While most populations of the Long-tailed Skink do not show parental care, at least part of the population on Orchid Island off Taiwan’s coast does. Huang placed several species of intruding reptiles into the nests of the skink and most were ignored. But the lizard-eating Keeled Ratsnake (Elaphe carinata) caused the lizards to permanently abandon their burrows. However, female skinks (but not males) would defend their nests against the Taiwan Egg-eating Kukri Snake. In 70 encounters, female skinks with eggs attacked the snake 53 times (11 times they escaped and 6 times they ignored the snake). The major source of egg mortality in unguarded nests was snakes, and the percentage of eggs that hatched was greater for nests that were guarded. Some females guarded nests violently and pushed snakes away, biting the head and forebody. When the snake backed off, the female immediately returned to the nest. But, some species have evolved more sophisticated means of dealing with snake predators.
Ground squirrels of the genus Spermophilus and rattlesnakes have been involved in an arms race for hundreds of thousands, if not millions, of years and have adapted to each other’s adaptations. The ability of the ground squirrels to resist rattlesnake venom was previously discussed in Chapter 5, but ground squirrels have also evolved ways to test the snake for dangerousness and adapted some other tricks to avoid being eaten.
Using audio recording of rattling rattlesnakes of different sizes and at different temperatures, Ronald Swaisgood and colleagues replayed the rattling tapes to free-living squirrels in an attempt to determine if the rodents could distinguish between snakes that were dangerous (large or warm) and snakes that were not dangerous (small or cold). Ground squirrels frequently harass Northern Pacific Rattlesnakes (Crotalus oreganus), placing themselves at some risk of being eaten. Swaisgood and co-workers found squirrels associated rattling with snakes and could determine if the snake was dangerous based upon the sound cues, approaching the speaker closer if the rattling sound was from a slow, high frequency rattle (a cold or small snake) and staying farther away from fast, low frequency rattles (a warm or large snake). Squirrels waved their tails, stood on their hind legs, and were hesitant to approach the area where the rattle was heard for up to 10 minutes after the recording had stopped. But that’s not all.
Aaron Rundus and colleagues found ground squirrels have evolved a specific communication signal for dealing with rattlesnakes. When a squirrel was confronted by a snake, the squirrel raised its tail, a behavior known as tail flagging. Tail flagging occurs whether the snake is a rattlesnake or the non-venomous Gopher Snake (Pituophis catenifer). However, squirrels have added an infrared component when tail flagging for rattlesnakes, a signal that shifts the rattlesnake’s behavior from predatory to defensive. The infrared component is not used when the squirrel is confronting a gopher snake which lacks infrared receptors. Normally, the tail flagging occurs when the squirrels are harassing the snakes. Tail flagging tells the snake, “I know where you are,” decreasing the chance of an ambush. At the same, time the tail flagging behavior tells other squirrels to be alert for snakes. The infrared mechanisms by which the squirrel alters the snakes behavior has yet to be described in detail. However, it is likely that the snake is responding to increased blood flow in the squirrel’s tail; this would increase the mammal’s apparent size if held aloft, and attract the snake’s attention.
Rodents in several different clades have been observed chewing shed snake skins and applying their saliva and the substances obtained from the snake skins to their fur. Barbara Clucas and colleagues tested different hypotheses to explain this behavior in the California Ground Squirrel (Spermophilus beecheyi) and the Rock Squirrel (Spermophilus variegatus). The behavior could be to offer protection from rattlesnakes, or it may reduce ectoparasites like fleas. It’s also possible that the behavior helps to deter other ground squirrels during aggressive encounters. Adult female and juvenile squirrels applied the snake scent more to the flanks and tail than did adult males. Clucus and colleagues found no evidence that it reduced the parasite load, or that it had an impact on aggressive encounters between rival squirrels. The authors examined the behavior of rattlesnakes toward scents of ground squirrels and the scent of rattlesnake mixed with squirrel. They found that the snakes were more attracted to the scent of ground squirrel alone, suggesting that squirrels are using the snake scent from shed skins to repel or confuse the snakes.
Other mammals have also devised various behaviors in response to snakes. The Banner-tailed Kangaroo Rat (Dipodomys spectabilis) responds to snakes with footdrumming; the rodent drums its hind feet against the substrate when in the presence of a snake. Jan Randall and Marjorie Matocq recorded the sound and played it back to the rodents and found it had little or no impact on the Kangaroo Rat’s behavior. But when females were in the presence of snakes, females with pups footdrummed more than those without pups. Randall and Matocq hypothesize that the vibrations are being used like the tail flagging behavior in ground squirrels to tell the snake, “I know where you are.”
Aliza le Roux tested the alarm reactions of the colonial, burrowing Brants’ Whistling Rat (Parotomys brantsii) that inhabits the semi-arid, sparsely vegetated habitats of southern Africa. In the presence of a predator, the rodent produces a high pitched whistle-like call. Interestingly, despite having a similar bandwidth and frequency pattern, the vocalizations elicited in response to a snake lasted longer than those made in response to a bird or human. The whistle alerts other members of the colony to the snake’s presence. Once a snake is detected, the rodents are usually able to escape predation by remaining above ground.
Lacertid lizards are numerous on Mediterranean Islands and, like many lizards, they have caudal autotomy. That is, they can shed their tails. Panayiotis Pafilis and colleagues examined the tails of 15 species of lizards in five genera from the islands and the Mediterranean mainland. Study sites were in three categories: mainland, Pleistocene islands, and pre-Pleistocene islands. The authors compare rates of tail breakage between the field and laboratory settings. In addition to snakes, each field site played host to a variety of predators such as fox, jackals, and birds (including raptors, crows and shrikes). They found predator diversity greatest on the mainland and lowest on the pre-Pleistocene island. Increased tail damage was correlated to the presence of predators, and they found a statistical correlation between increased tail autotomy and the presence of vipers. The authors concluded that lizard tail autotomy has evolved as a defense against venom and, that by dropping their tails, bitten lizards may prevent venom reaching their bodies. This is an intriguing hypothesis, and it will be interesting to see if it can be verified. However, viper hunting strategies may pose a problem. Vipers are mostly ambush hunters, and juvenile vipers and some adults caudal lure their prey. Consequently, it is most likely the head or body of the lizard that is struck, rather than the smaller target of the tail. If Mediterranean vipers pursed their prey, the situation would increase the chance of a lizard’s tail being envenomated. Under these circumstances it would seem probable that tail autotomy evolved in response to venomous snake predators. However, pursuit by vipers is not a common strategy. Birds on the other hand are likely to pursue lizards and grab them by the tail, causing the lizard to drop its tail in an effort to escape.
Early Experiences
Two neonate Elephant Trunk Snakes (Acrochordus javanicus) live in an aquarium in my office. When I first obtained these snakes, they could not have been more than a few months old. The aquarium they live in is well stocked with fish, and I noticed that they would often kill fish and lose track of them. I often found dead fish floating at the surface or sitting on the bottom of the tank during the first week or so. However, the snakes soon learned to hold onto the fish during handling so the prey was swallowed and not lost. The two snakes were soon consuming 30 to 40 fish per week.
Rita Metha used hatchling Trinket Snakes (Coelognathus helena) to see how performance with prey of different body weights changed with experience and if feeding performance improved with practice. She found that during the first four feedings the hatchlings were learning the best direction to ingest prey and that snakes that used constriction reduced their prey handling time. As snakes gained experience with prey, they incorporated more behavioral techniques into their feeding repertoire. Snakes that encountered prey of the same size narrowed their behavior, although it did not affect the overall capacity of snakes to deal with larger prey items. Snakes that had early experience with large prey were able to deploy more complex behaviors when they encountered prey of variable sizes.
Snakes learn from experience and, while some of their behaviors are innate, others are unique to the individual. As actors in ecosystems, snakes have had long relationships with their prey and many novel adaptations have evolved in both the snakes and their prey. Venom, constriction, modified tail tips, altered physiology, and subtle changes in behavior have contributed to make snakes highly successful predators. But of course, snakes are themselves vulnerable to predation and environmental change.
Snake
(Nerodia sipedon) in photo 7.
Comments
Post a Comment