Chapter 9 - Cowards and Bluffers

9. Cowards and Bluffers

…snakes are first cowards, then bluffers, and last of all warriors.

Clifford Pope, 1958

               

                A farmyard encounter between a domesticated cat and a hognose snake illustrates Pope’s assertion. At first approach, the cat pawed and sniffed the snake. The snake flattened its head and anterior body to form a cobra-like hood and struck at the cat with its mouth closed. Then the snake’s tail coiled, it opened its anal glands, and smeared a foul smelling musk over its body by rubbing its tail on the body. Continuing its investigation, the cat pawed at the snake’s head and seemed unconcerned by the snake’s behavior. More closed mouth strikes and hissing did not deter the cat. When the snake opened its mouth to reveal a black lining, it also contorted its head. The cat continued to harass as the snake rolled on its back, exposing a uniform cream colored belly. The snake was apparently feigning death. The cat continued to play with the snake, but eventually lost interest and returned to my sister’s house.

                Hognose snakes have defensive displays composed of many different behavioral elements that involve many organ systems. While all of this may confuse a predator, exactly how the series of display behaviors is triggered by the snake’s physiology is poorly understood. Hognose snakes have venom, and humans, and other animal react to it. Despite potent venom, many snakes are killed and eaten by many predators.

Predation on Snakes

                Large spiders, predaceous diving beetles, and centipedes have their own venom and will prey on small snakes. Perhaps the most important invertebrate predators on snakes are crabs. Crabs are abundant in coastal habitats and wet forests, and will feed on almost any organic matter they find, but they will also capture, kill, and consume live prey, including snakes. Encounters between live crabs and live Asian Bockadams (Cerberus rynchops) were documented by Harold Voris and William Jeffries with infrared photography. The crabs killed and ate the snakes. Crabs also live in rainforests where they will wander overland far from water in search of food, and snakes are not safe from these heavily armored, eat-anything crustaceans.  Even highly venomous snakes are not safe from crab predation. Amy Cunkelman and Aaron Bauer reported crabs feeding on St. Girons’ Sea Krait (Laticauda saintgironsi), and crabs have been known to kill and eat hatchling Hamadryads (Ophiophagus hannah). Jonas Dehling reported a land crab that killed and ate a Terciopelo (Bothrops asper). David Maitland found a terrestrial crab species on Tobago that preys upon three different species of snakes. Arijana Barun and colleagues collected 219 Virgin Island Racers (Alsophis portoricensis) on Guana Island, and more than 70% of the snakes had tail damage or scarring. The major suspect for damaging snakes was the terrestrial hermit crab Coenobita clypeatus, a diurnal species that may seize snakes but are unable to kill or consume them. Of interest, small snakes did not show scarring, but scarring and tail damaged increased dramatically once the snakes reached 450 mm in body length.

                Fish and amphibians are also snake predators. A two-meter Black Mamba (Dendroaspis polylepis) was found in the stomach of a Brindle Bass (Epinephelus lanceolatus) that weighed only 3 kg, and the snake was undoubtedly captured as it swam across a body of water. A Puff Adder (Bitis arietans) was found in the stomach of a walking catfish (Clarias) and it may have been captured while swimming, or while the fish was walking overland. Snakes that live in water are at risk of being swallowed by fish. Despite the venom found in many coastal and marine snakes they are eaten by sharks and other large fish. Frogs and toads are often eaten by snakes, but turn about does occur. Large species such as the North American Bullfrog (Lithobates catesbeianus), the African Pixie Frog (Pyxicephalus adspersus), the South American Marine Toad (Rhinella marina), and horned frogs (Ceratophrys sp.) are ambush predators with wide mouths capable of swallowing small snakes. Crocodiles (Crocodylus sp.), the alligator (Alligator mississippiensis) and caiman (Caiman sp.) will also eat snakes, as will large, carnivorous lizards such as the monitor lizards (Varanidae).

                There are also snakes that specialize in feeding on other snakes. This is cannibalism only if the snake eats members of its own species, and some do. Many elapid snakes (cobras, kraits, and coral snakes) specialize in feeding on other snakes; among them, the Hamadryad has a reputation for its snake-eating habits. The Asian Sunbeam Snake (Xenopeltis) has a varied diet, but, it too, is well known as a snake predator [Figure 9–1]. In North America kingsnakes (Lampropeltis), coachwhips (Masticophis), and racers (Coluber) are also important snake predators.

INSERT FIGURE 9-1.

Figure 9–1. A Sunbeam Snake (Xenopeltis unicolor) eating a Rainbow Mud Snake (Enhydris enhydris). Khon Kaen, Thailand.

                Hawks (Falconiformes), owls (Strigiformes), secretary birds (Sagittariidae), herons (Ardeidae), hornbills (Bucerotidae), roadrunners (Geococcyx), and a variety of other birds will readily eat venomous and nonvenomous snakes. White-bellied Sea Eagles (Haliaeetus leucogaster) grab venomous sea snakes in their talons when the snakes surface to breathe, and herons wade along shorelines foraging for any small animals including venomous species like the Cottonmouth.

                Carnivorous mammals will eat snakes if they find them. Shrews (Soricidae), while small, are often venomous and will kill snakes much larger than themselves. Mongooses (Herpestidae), badgers (Melinae), foxes (Vulpes), cats (Felidae), and some primates such as tarsiers (Tarsius) are known to eat snakes though no mammals specialize in feeding on snakes. Pigs (Suidae) will kill and eat snakes. And White-tailed Deer (Odocoileus virginianus) and other hoofed mammals will stomp on snakes and kill them, but do not eat them. Snakes are preyed upon by many predators and harassed by many other species but they have evolved a variety of adaptation that allows them to escape ― at least sometimes.

Escape by Flight

                Clifford Pope’s assertion that snakes are cowards is supported by the fact that flight is usually the first response to the perception of a predator. Removing themselves from the presence of a predator by crawling into a burrow or crevice, dropping off a branch into the water, or dropping off a branch or tree trunk and gliding away from the predator are all effective methods of escape.

                The Northern Water Snakes (Nerodia sipedon) and other members of the genus Nerodia often sit on branches hanging over the water. At the least disturbance, they drop off the branch and into the water to escape. The method is effective because it is unlikely that a predator hunting on land or from the air will follow the snake into an aquatic habitat [Figure 9–2A]. William Copper and co-workers examined the flight behavior of N. sipedon in the wetlands of Michigan and Ohio. They found that 44% of the snakes held their position and did not flee, 37% dove into the water, and 19% swam away with their head above water. Females took flight at a greater distance than juveniles or males. The authors suggest that submerged snakes assessed the situation as one of greater risk than those that swam off with their head above water.

INSERT FIGURE 9-2.

Figure 9–2. A. The Northern Water Snake (Nerodia sipedon) basking on branches overhanging the water. Will County. B. The Sand Viper (Cerastes vipera) partially buried in the sand.

                Dropping into the water to escape predation is quick, but some snakes can very quickly bury themselves if they live on loose sand. Some sand dune dwelling vipers of the Middle East and Africa are able to rock their bodies side-to-side and push the sand out from under them. In effect, this allows the snake to sink into the sand and potentially out of the predator’s sight [Figure 9–2B]. This behavior is discussed in greater detail in Chapter 12.

                Rapid movement away from a predator may be the simplest method to avoid being eaten. Some species are faster than others; coachwhips (Masticophis), racers (Coluber and Mastigodryas), mambas (Dendroaspis), and taipans (Oxyuranus) will usually move away from a potential predator. The fastest species are usually slender in build and do not use constriction to kill prey. The side-to-side movements of the body, combined with color and pattern, often work to confuse the predator so the snake can escape. Even relatively slow-moving species, like the vipers, will try to escape by fleeing. Failing escape, the snake may change direction and move toward the potential predator. This behavior has given rise to the belief that snakes will chase humans. Stories about cottonmouths, rattlesnakes, saw-scaled vipers, mambas, and cobras chasing humans are legendary. However, snakes don’t really chase humans, the snakes are simply trying to escape. When a human stands its ground and does not move away from a charging venomous snake, it keeps crawling in a straight path directly passed the person. However, there are incidents that may be difficult to interpret and snakes may, in fact, actually be aggressive if they are fearful or feel trapped.

                Consider the case of the two brothers consecutively bitten by the same Monocellate Cobra (Naja kaouthia) in Bangladesh. Two siblings, one 12 and the other 37, operated a restaurant surrounded by a brick wall.  On July 12, 2000 the older brother acted as cashier while the younger man waited on customers. Four customers were eating lunch when the older brother felt something cross his legs. He moved, felt a bite on his right foot, and then saw the snake. The customers panicked and jumped over tables as the older brother chased the snake out of the building. But the snake reversed course and returned to the restaurant. Despite shouts from the older brother to leave the building, the younger brother ran toward the snake, which then bit him in the left knee. Both brothers were in pain and showing symptoms of envenomation; however, both were treated in the hospital and recovered. It seems likely the snake was foraging for food and felt threatened when the first man moved trying to escape the building. The snake may have returned to the restaurant because it was being chased. When the younger man approached the snake, it, again, felt threatened and defended itself.

                Of course there are situations where a snake is incapable of escaping from a predator, one of those conditions may be the recent ingestion of a large meal. Rita Mehta tested hatchling Trinket Snakes (Coelognathus helena) for locomotion performance and antipredator tactics. Her study provided snakes with meals that were 20–35%, 50–59%, or 70–79% of their body weight and then tested them for locomotion and antipredator behavior. Not surprisingly, the snakes that had not been fed and the snakes with the light meals responded actively to the locomotor tests and predatory attempts; those that had eaten large meals were slow to move and used stationary and cryptic behaviors when threatened with predation.

                Tails may be the last structure a predator sees, and the most likely part of the snake’s anatomy to be seized. A few species break their tails to escape. Tail autotomy is well known in lizards, some of which have evolved fracture planes in their tail vertebrae and have the ability to regenerate a lost tail. But snakes, for the most part, are incapable of re-growing their tails; snakes simply heal the wound. A few genera of New World snakes are known to break-off their tails when grabbed and one group, the shovel-toothed snakes (Scaphiodontophis), are capable of breaking their tails at multiple places along its length. In the Eastern Hemisphere, several species of keelbacks (Xenochrophis sp.) will break their tails, but my observations suggest they don’t do it easily, like geckos. Hunting along the grassy edge of a central Thailand reservoir one of my companions grabbed a Xenochrophis by the tail. Not knowing what kind of snake it was, he held it away from his body and the snake went into a spin. Within a few seconds, the snake was on the ground and my companion was left holding a tail.

Crypsis, Escape by Avoiding Detection

The Malayan Pitviper (Calloselasma rhodostoma) has a pattern that blends into the forest floor ecosystem to camouflage it from its predators as well as its prey [Figure 9–4]. While some may view this as a perfect adaptation, mutant coloration and patterns are constantly being tested and this snake comes in a variety of blotched, spotted, and almost uniform patterns and color morphs. Snakes that most closely match the background they are living on survive while individuals with a pattern that is less cryptic are eaten by predators. Bush and tree-dwelling snakes are often green or brown to match the leaves and twigs of their surroundings. But there is more to being cryptic than color and pattern, there is also behavior.

INSERT FIGURE 9-3.

Figure 9–3. A cryptically colored Malaysian Pit Viper (Calloselasma rhodostoma) hides in a brush pile in central Thailand.

                Remaining still to avoid detection is a large part of crypsis. Even well-camouflaged snakes will be betrayed if they move. Many snakes, possibly all snakes, rely on crypsis to some degree. And it is often the combination of pattern and behavior that allows the snake to escape its predators. The vine snakes of Southeast Asia are masters of camouflage; even though they come in a variety of colors (greens, browns, yellows), their long, thin bodies can remain motionless for long periods of time allowing them to remain undetected by potential predators as well as their potential prey. Some North American ratsnakes (Pantherophis) and the tree snakes of South America (Chironius) also have a defense behavior that has been poorly studied but long known. They will kink their body in such a way that it disrupts the body’s outline, making the snake look like a fallen branch or twig.

                However, not all cryptic coloration is dull or in muted colors with camouflage-like patterns. Bright colors and sharply contrasting patterns can also be cryptic in the right circumstances. In other situations, bright colors provide crypsis while simultaneously warning predators of the snake’s toxins or bluffing predators into staying away from non-existing toxins. The potential predator can receive a memorable experience if an unusual pattern or behavior is combined with a painful bite, a bad taste, a bad odor, or some other noxious experience.

Warnings and Mimics

                Warnings can come in many forms and they don’t have to be bright colors―just something distinctive that can be remembered by the potential predator.

Martti Niskanen and Johanna Mappes used plasticine models painted to look like the Spanish Snub-nosed Viper (Vipera latastei gaditana), a snake with a very distinctive mid-dorsal zigzag stripe, and models of the same size that were missing the stripe. They placed the models within the viper’s habitat in Doñana National Park in Spain and positioned them on a variety of natural and white backgrounds. Snake predators were primarily birds of prey, hawks and owls, including the short-toed eagle (Circaetus gallicus), a snake specialist. The models were frequently attacked by the birds. Small snake models were attacked more often than larger models, and solid-colored, large snakes were attacked more often than large, striped individuals. Combined, 39% of their models were attacked and the large striped models, the ones most resembling the snub-nosed vipers, sustained the fewest attacks. Their results suggest the zigzag stripe is aposematic, a warning. Previous workers suggested the pattern was disruptive camouflage, and it may, in fact, be both.

                Looking into a crevice under a large boulder in a Costa Rican rain forest, I could see the coils of a coral snake. With the help of a stick, the snake was soon out of the darkness where it became apparent that the one snake was actually two. A Central American Coral Snake (Micrurus nigrocinctus) had partially swallowed a Clouded Slug-eating Snake (Sibon nebulata). Within a few moments, the coral snake had regurgitated its prey, hid its head beneath a body coil, and was frantically waving its curled tail in the air. As we attempted to get it into a bag, it thrashed from side to side in a jerky motion.

                About 75 species of coral snakes (Micrurus) inhabit the Americas, and many of these have a pattern of red, black and yellow (or white) that forms rings around their bodies. However, some have only red and black, and there are a few that are mostly red or mostly black in color. Over the last century, biologists have debated the value of the bright coloration because it would seem to make the snakes highly conspicuous to predators. The question becomes more complex with the realization that many species of non-venomous, or mildly venomous, rear-fanged snakes also look like or mimic coral snakes. Is there survival value in having red, black and yellow bands?

                One of the earliest experiments that involved snake mimicry was done in 1911 when Hans Gadow painted lengths of flexible tubing to look like coral snakes and placed them in vegetation. He noted that the tubing was eye catching when standing close but the tubes seemed to vanish once an observer was more than 15 feet (4.57 m) away. When he painted tubing with just one color and repeated the experiment, the tubes attracted attention from a greater distance. This experiment suggested that coral snake patterns were cryptic and hid the snakes from predators that were at a distance. Uniform-colored snakes were more visible at a distance.

                This, of course, would explain why so many snakes, harmless and venomous alike, have bright-colored bands. Supporters of the view that bright coloration was only cryptic argued that mammals and birds were color blind and coral snakes were only active at night when the colors could not be seen and that the bands simply made the snakes more difficult to see in the leaf litter. Others disagreed, however.

                An alternative hypothesis for the bright coloration of coral snakes was that it served to warn potential predators of their venom, and that the harmless snakes that looked like coral snakes were mimicking them. A predator that learned to avoid a coral snake would also avoid a non-venomous coral snake mimic. Evidence to support the mimicry hypothesis grew. Coral snakes were found to be more active during the day, and mammals and birds were found to have better color vision than previously thought. Furthermore, not only did the harmless mimics look like coral snakes, they behaved like coral snakes with tail displays, head hiding behavior, and jerky motions of the body [Figure 9–4].

INSERT FIGURE 9-4.

Figure 9–4. The Central American Coral Snake (Micrurus nigrocinctus) from Costa Rica. On leaf litter coral snakes can be cryptic from a distance but aposematic close-up.

                Birds and mammals are coral snakes’ most common predators. Furthermore, both these groups are very dependent on vision and a variety of species of both clades have color vision. It seemed likely that as day active birds and mammals foraged for food by turning over the leaf litter and searching logs, they will encounter coral snakes. Susan Smith tested two species of birds (motmots and kiskadees) known to feed on reptiles in tropical forests. Raised in captivity, they had never seen a coral snake yet, they avoided a red and black banded model while they readily attacked models that had blue and green rings and striped red and yellow models. Additional experiments using free-living birds and plastic models demonstrated that birds would attack plain brown snake models more frequently than plastic models painted to look like coral snakes.

Staring at a bottle of King Cobras (also known as the Hamadryad, Ophiophagus hannah) on a museum shelf, I was confused. Surely someone must have made a mistake. Here was a dark brown, almost black, body with bright yellow rings. I was sure it must be a rear-fanged Mangrove Snake (Boiga dendrophila). A closer look revealed there was no mistake; the specimen lacked a loreal scale and was, indeed, a juvenile King Cobra. I experienced similar confusion in southern Thailand when I grabbed the tail of a large olive green snake that had its head and most of its body down a burrow. The snake was resisting my pull, and as I looked at the large scales, I realized I was unsure if the snake was an Asian ratsnake in the genus Ptyas or a King Cobra. The snake ultimately escaped into the safety of its burrow.

Mimicry is false advertising; an animal sending a signal that it is something that it is not. Mimicry evolves under a variety of ecological conditions in several situations. Batesian mimicry occurs when a harmless, palatable mimic looks like a non-palatable, toxic or dangerous organism to gain protection from predators. Müllerian mimicry occurs when mimics and models use each other (co-mimics) to teach predators they are not palatable and share the cost of educating the predators. In snakes, Batesian mimicry may be best known between the palatable kingsnakes of the Lampropeltis triangulum complex and coral snakes of the genus Micrurus. Müllerian mimicry is more widespread, and may occur between many mildly venomous rear-fanged South American snakes in the family Dipsididae and the very toxic front-fanged coral snakes; between the mildly venomous Asian wolf snakes of the genus Lycodon and the very toxic kraits of the genus Bungarus; and, between the mildly venomous Cantor’s Mud Snake (Cantoria violacea) the Keeled-bellied Mud Snake (Bitia hydroides) and the very venomous true sea snakes of the genus Hydrophis.

Evidence suggests the Asian green pitvipers are involved in an unusual Müllerian mimicry complex where males have bright red and white markings on their otherwise green bodies, while females tend to be solid green. Kate Sanders and colleagues documented the red, white, and green coloration pattern in male vipers in sympatric and parapatric populations of four genera. They suggested that males are under different selection pressures than females because males move more than their sedentary female counterparts and may be more subject to predation by birds living in the forest understory.

There appears to be a third kind of mimicry, recently proposed by Meredith Rainey and Gregory Grether, that they named “competitive mimicry.” They define competitive mimicry as that which enables access to a defended resource or aids in defense of a resource. Competitive mimicry can occur in three forms. Mimicry of a non-competitor is analogous to Batesian mimicry where resource defenders must decide to attack or not attack an intruder that looks like a competitor of a different species, but is not; this is unknown in snakes. A second type of competitive mimicry is mimicry of a non-competitor, where members of the same species mimic the appearance or odor of another. The transvestite male garter snakes discussed in Chapter 8 are an example of this kind of mimicry. The third form, mimicry of a competitor’s predator to gain access to a resource, is also known among some snakes. Burrowing Owls (Athene cunilaria) compete with ground squirrels for the burrows made by the rodents. Squirrels responded weakly to playback recordings of Burrowing Owl calls in areas where rattlesnakes are rare. In areas where rattlesnakes are common, however, the rodents responded strongly to the owl’s hisses and rattling calls. The owls can keep squirrels out of their burrows by hissing, a sound that mimics a rattlesnake’s rattle.

Bluffs, Threats, and Diversions

                Snakes, both venomous and non-venomous, use a variety of behaviors to avoid being eaten; it may be as simple as inflating their bodies to look larger, opening their mouths to display a colored lining, or making a loud hissing noise by controlling air flow through the respiratory tract.

                Brightly colored parrot snakes (Leptophis) are well known for their defensive displays. These Middle and South American rear-fanged snakes have very large eyes and are day active. When encountered, they attempt to flee, but when harassed they face the predator, open their mouths wide and threaten to bite. The snake’s windpipe or trachea opens in the floor of the mouth and becomes highly visible during the display. Cottonmouths (Agkistrodon piscivorus) are so called because of the white mouth lining that they expose during their defense display [Figure 9–5]. And Black Mambas (Dendroaspis polylepis), which are usually grey in color, get their name from exposing a black mouth lining when threatening a predator.

INSERT FIGURE 9-5.

Figure 9–5. Open mouth displays. Top: A Western Cottonmouth (Agkistrodon piscivorus leucostoma) gaping its mouth in defense. Bottom: A Trinidad Parrot Snake (Leptophis ahaetulla) threatening with an open mouth and a flared tracheal opening.

                In an experiment to document antipredator behavior in the Cottonmouth, Eric D. Roth and Joel A. Johnson tested 46 snakes for seven common behaviors: escape behavior, defensive posture, tail vibration, musk release, mouth gapes, strikes, and bites. Snakes were exposed to a series of three predatory events that escalated from humans opening the cage, to harassing the snake with gentle nudges from a welding glove, to picking the snake up with tongs. When they corrected their data for body size, no differences were found between male and female behaviors. The snakes were scored for their behaviors and analysis revealed that as body size increased the scores decreased. They found aggressive behavior such as a strike or bite rarely occurred with a first attempt to flee, and mouth gaping usually did not occur until after the snake had attempted to flee, exhibited a defensive body posture, and released anal musk.

How is it that a snake can effectively back off predators with an open mouth and colored mouth lining? The answer to this question is probably the memory of the predator. The brightly colored mouth may make the snake more visible and memorable to a large animal that may step on it. Cattle, horses, pigs, deer all have hooves. Hoofed animals don’t have the best close-up eyesight and, as they are moving around, they can easily trample a snake which may bite the mammal in retaliation. Neither animal is interested in the other for food; thus, it benefits both to be aware of the other’s presence. Small and medium-sized carnivorous birds and mammals may have learned to associate the mouth display with a dose of venom and, once bitten, remember the event the next time they encounter the species.

When cornered, many snakes will strike with their mouths closed. Hognosed snakes (Heterodon) and cobras (Naja) are well known for this behavior and it serves a dual purpose; it threatens the predator and reduces the risk of injury to the snake’s fragile mouth. When a snake strikes and then bites an animal it risks having teeth and bones broken. A fake strike gives the potential predator the idea that it should move on, but the message is sent with minimal risk of injury to the snake. Both the cobras and hognose snakes combine these fake strikes with a flattened neck and expanded hood that make the snake appear larger than it actually is and serve to make the experience unforgettable for the potential predator.

                The tiny Three-lined Snake (Atractus trilineatus) of Trinidad has a tail tip that ends in a sharply pointed scale. Similar structures can be found in many burrowing snakes. When handled, the snake will probe with its tail and the handler’s first thought is often that he or she is being bitten. However, this species is so small that it is not capable of delivering a bite to a human. The overall effect of the probing tail is startling, and it seems likely a predator would release the snake when probed with the sharp tail tip.

                Tail displays are common in coral snakes and many other species that live on the ground or in burrows. The tail may be tightly curled, raised off the ground, and waved in the air. At the same time, the snake usually hides its head under a coil of the body or tries to crawl off while keeping the predator’s attention with the tail. This misdirection must be very successful for coral snakes since so many of the species use these tail displays. Similar behavior occurs in many other species that are also brightly colored, including the Asian Pipe Snake (Cylindrophis ruffus) that can dramatically flatten the tail. Asian Pipe Snakes are usually a uniform black above with some narrow white markings on the sides. Should a predator flip it over, however, a belly of bright red, yellow, and black is revealed. When harassed, pipe snakes flatten and raise the tail so the colors are quite visible; the narrow head is pressed to the ground and may be concealed under a body coil. The tail acts like a head with a hood. The distraction of the tail gives the real head time to search the area for a burrow or other refuge and the snake will slide off while the predator’s attention is directed to the flattened tail that is mimicking the head.

Hognosed snakes (Heterodon), Red-necked Keelbacks (Rhabdophis subminiatus), and the cobra known as the Rinkhal (Hemachatus haemachatus) are known to “play dead” or feign death. Interestingly, they also all feed on toads and have displays that involve expanding the neck [Figure 9–6]. In 1955, Hobart Smith and Fred White noted the adrenal glands of the hognosed snakes and the Central American False Terciopelo (Xenodon rabdocephalus) were exceptionally large when compared with other snakes that did not feed on toads, and they hypothesized that large adrenal glands were a physiological adjustment to toad toxins. Toads produce toxins (bufotoxins and bufoagin) in their parotid glands as well as epinephrine and bufotenine, a molecule that stimulates the secretion of epinephrine by the adrenal glands.

INSERT FIGURE 9-6.

Figure 9–6. A: An Eastern Hognose Snake (Heterodon platirhinos) death feigning. The tongue is out, the labial scales are flared, the snake is on its back, and its cloacal opening is gaped. B: An American Toad (Anaxyrus americanus), the toxic prey of the hognose snake.

Dekay’s Snake (Storeria dekayi) is also known to play dead, but they don’t eat toads; instead, they feed on earthworms and slugs. Gary Gerald experimented with the effects of temperature, body size, and locomotor ability in newborn Dekay’s Snakes. Gerald tested 26 snakes from two litters at 10, 20, and 30º C. The snakes were picked up and dropped into water, and the time it took them to swim one meter was then recorded. He found death feigning was consistent when the snakes were in water and were more likely to feign death at warmer temperatures. Body size was negatively correlated to playing dead, and swimming velocity was negatively correlated to the length of time the snakes feigned death. Thus, smaller snakes were slow swimmers and played dead more often than larger snakes.

But is playing dead really a defense behavior? Harry McDonald made observations on 43 Eastern Hognose Snakes (Heterodon platirhinos) and measured their heart rates during displays of death feigning. Three of his specimens were spontaneous death-feigners, and the frequency of the behavior increased in captivity, so much so that all three eventually died exhibiting the behavior. McDonald noted that while death feigning occurs in some species known to eat toads, it also occurs in some snakes that do not (the Worm Snake Rena [Leptotyphlops] dulcis; the Ringneck Snake, Diadophis punctatus, the Ground Snake, Sonora episcopa; the Coachwhip Snake, Masticophis flagellum; and the Eastern Coral Snake, Micrurus fulivius). Other snakes that are well known toad predators (Causus rhombeatus, Xenodon sp.) have not been reported to feign death; however, death-feigning behavior may have more than a single cause in snakes. Hypotheses that death-feigning in hognose snakes and other toad-eating species results from a physiological failure to deal with stress hormones or an interaction between toad toxins and excessive amounts of adrenaline produced by the snake need testing. A few authors have compared death-feigning behavior to seizures, implying the behavior is not adaptive defense behavior, but a pathological behavior created by stress hormones and toad toxins. Some snakes use a different strategy to deal with predators, they use intimidation.

                Hooded cobras are iconic throughout the world, despite the fact that extant species occur only in Africa and Asia. The hood is formed by the skin of the forebody expanded by about 27 modified ribs. Frank Wall described the skeleton of a cobra, presumably Naja naja. He observed that the first three vertebrae (atlas, axis, and the 3rd vertebrae) have no ribs, but the subsequent 27 vertebrae have ribs that were much less bowed than those in the rest of the body, and they showed a greater range of movement. When the ribs are folded toward the vertebral column, the hood is not visible.

                Markings on the dorsal hood of cobras are limited to relatively few species; the Spectacled Cobra (Naja naja), the Monocellate Cobra (Naja kaouthia), and the Chinese Cobra (Naja atra) regularly have hood markings. Other species may have them, but markings appear far less frequently. Hood markings may be interpreted as eye spots intended to startle a potential predator, and the spots that occur on the ventral side of the hoods of some cobras may indeed function as eyespots. However, there are often irregular, double, or single circular markings on the dorsal surface of the hood. They, too, could act to startle a predator if the cobra’s back is turned to the predator. But this would only work as the cobra flees-and, when they do, they rarely have the hood erected. Given that the Spectacled Cobra is often associated with dense human populations (as are the other two species), Harry Greene speculated that the snake’s hood markings may be the result of artificial selection. The markings often bear a remarkable similarity to the Hindu and Buddhist Om symbol, either a single symbol (monocellate) or a double (bicellate) symbol. A scenario where people killed snakes without the markings and protected snakes with markings that resembled their religious symbol might explain the hood patterns seen in these snakes.

Evidence to support religious symbols on cobra hoods comes from Pakistan. The patternless, black Spectacled Cobras are called “caro” meaning fierce, while snakes with hood markings are called “padam” and they are considered peaceful. Folklore suggests the Caro lives for 125 years before it turns into the spectacled version or 'Padam.'  The Padam is peaceful, non aggressive, and does not bite. Only snakes without hood markings are used as source of venom to make antivenom. These snakes are often referred to the subspecies Naja naja karachiensis.

It is hypothesized that the hood’s function is to make the snake look larger than it is and possibly to provide a memorable experience for a potential predator or trampler. Uma Ramakrishnan and colleagues studied snake-related behavior of wild Bonnet Macaques (Macaca radiata) in southern India. They used models of the Green Keelback (Macropisthodon plumbicolor), the Indian Krait (Bungarus caeruleus), the Indian Ratsnake (Ptyas mucosus), the Indian Python (Python molurus), and the Spectacled Cobra (Naja naja). The snake models were made to display normal postures for each species; the cobra’s hood was displayed, the python was stretched out, and the rat snake was in a loose coil. Monofilament fishing line was attached to each so the models could be moved by a researcher from a distance. The models were placed near feeding stations and the reaction of the Macaques was videotaped. The cobra elicited a significantly faster reaction time than the other snake models, and the monkeys often ran from the cobra model.When presented with ratsnake models the monkeys just observed the snake’s behavior.

Startling sounds may deter a predator from eating a snake or warn a large animal of the presence of a dangerous snake so the snake is not injured by the animal, or the reverse. In a series of papers investigating sound production by snakes, Bruce Young and co-workers found sound is produced in a variety of ways and is an important aspect of defense in some snakes.

Perhaps one of the more unusual sounds made by snakes are pops produced by expelling air from the cloacal opening. Both the Sonoran Coral Snake (Micruroides euryxanthus) and the Western Hook-nosed Snake (Gyalopion canum) are known to use cloacal popping. Since these two species have a similar distribution, the behavior may be directed at a specific predator, but, at this time, the specific organism is unknown.

Hissing is often attributed to snakes, but the actual number of snakes known to hiss is not great. In North America, hognose snakes (Heterodon) hiss, bull snakes and gopher snakes (Pituophis) hiss, usually with an open mouth, but there are species in other parts of the world known to make a hissing sound.

 One of these is the highly venomous Russell’s Viper (Daboia russelii) of South and Southeastern Asia that accompanies its closed-mouth hissing with unpredictable jerking and whole-body contractions that result in most of the body being lifted off the ground from a coiled position. Russell’s Viper anatomy does not suggest any specialization in the respiratory passage that would account for the hissing other than exceptionally large external nares. Bruce Young placed short pieces of plastic tubing in the nares of one of these snakes and found a decrease in the amplitude of almost 12 decibels. Flaring the nostrils appears to be responsible for increasing the volume of the defensive sounds.

Two large Southeast Asian snakes have been reported to produce a growling sound; the King Cobra (Ophiophagus hannah) and the Dhaman or Indian Ratsnake (Ptyas mucosus). Young and co-workers noted the ratsnake compresses the anterior portion of the body and expands the throat during defensive displays to produce a long, low-frequency growl. The researchers wrapped the anterior portion of a snake’s body with surgical tape and found the snake incapable of producing the display, but still able to make the sound. The tracheal membrane of this snake is exceptionally wide and expands away from the cartilaginous rings of the trachea into the body. They hypothesized that the membranes collapses and forms isolated pockets that act as resonance chambers during the display. They also suspect that the sound is meant to mimic sound production by the King Cobra, a snake with which the Dhaman shares a similar physical appearance and much of its geographical distribution.

Kenneth Kardong argued the case for the Gopher Snake (Pituophis catenifer) mimicking the Northern Pacific Rattlesnake (Crotalus oreganus) in both color pattern and behavior. Both snakes have similar geographic distributions, and they both have a cryptic pattern of dark blotches separated by light coloration. Furthermore, they both have an auditory display; the gopher snake expels air over a larynx-like structure, producing a sound that can be mistaken for the vibration given off by a rattling tail. The advantage to the harmless gopher snake, of course, is that the rattlesnake is highly toxic and capable of providing a predator with a nasty dose of venom. Samuel Sweet disagreed. He questioned mimicry in this case, stating that while both snakes are cryptic, they use different habitats over much of their range and noted considerable geographic variation in pattern and behavior. He considered gopher snakes poor mimics of rattlesnakes.

The bird known as the Greater Roadrunner (Geococcyx californianus) is a generalist predator and well known for feeding on lizards and snakes. Wade Sherebrooke and Michael Westphal tested three roadrunners who were naïve to snakes. The birds had been raised in captivity and fed live prey, but not snakes. They asked: do coral snake color patterns cause an innate alarm response in roadrunners? How do roadrunners respond to clay models and live snakes? Do roadrunners exhibit innate behaviors when confronted with rattlesnakes, and, if so, are the same behaviors used with gopher snakes? The study revealed no innate behaviors toward coral snakes (models or live snakes); the birds directed pecking at the snake’s head; however, the birds were cautious towards both the Western Diamondback Rattlesnake and the Gopher Snake. They responded to the presence of these snakes by leaping and flapping their wings. The snakes neither rattled nor hissed at the birds. The authors concluded that the birds’ behavior supported the view that gopher snakes are a rattlesnake mimic because the birds respond to both the venomous rattlesnake and the non-venomous gopher snake with the same cautionary behavior.

Some snakes also make sounds by rubbing their scales together. The African egg-eating snakes (Dasypeltis, also discussed in Chapter 7) have a defense display that involves open-mouth strikes, inflating the body with a sac attached to the trachea, coiling its body into a series of C-shaped-folds, and moving the coils so they rub against each other. The scales of the lower rows have strong central keels and as they rub against each other, they produce a hissing sound. Most of the African egg-eaters are cryptic in pattern with dark blotches or bars on a slightly lighter colored background, but some are uniform in color. It has been suggested that they mimic small vipers in the genera Bitis, Echis, and Causus. Remembering that African egg-eaters are virtually toothless, it would seem likely these small snakes could benefit from looking and sounding like a more dangerous snake. Carl Gans and Neil Richmond examined the defensive display of African Egg-eaters and suggested that the displays may not only serve to threaten predators, but may also be used for intimidating parent birds guarding the nest.

Sea snakes are subject to predation from a wide range of species, from crabs to birds to large fish. While actively searching for food, they have their heads in crevices or burrows and leave their tails and posterior body exposed for any predator to see. Some species have compensated for this in an unusual way. Arne Rasmussen and Johan Elmberg observed the Yellow-lipped Sea Krait (Laticauda colubrina) performing an unusual trick with its tail. The snakes twist the tail tip around the long axis of the body, to make the tail end look like a dorsal view of the head. This behavior was discovered when Rasmussen was diving off the coast of the Bunaken Island in Indonesia. He was following a large krait for about 30 minutes; the snake was swimming between corals and crevices hunting for food. Rasmussen was momentarily distracted by a second snake,; but when he looked back, he saw a “head” facing him while the tail probed the coral. Rasmussen’s surprise grew as he saw a second head emerge from the coral instead of the expected tail. It was only when the snake swam away that the first head was clearly seen to be a paddle tail. Examination of 98 sea kraits’ (Laticauda) tails, as well as the tails of other species of sea snakes, showed heads and tails have similar coloration. They found similar coloration in the Thick-tailed Sea Snake (Hydrophis pachycercos) and Curtus’ Sea Snake (Lapemis curtus), suggesting that this defense adaptation has evolved at least twice in sea snakes.

Rattles

Lawrence Klauber’s classic work on rattlesnakes devoted considerable space to the role of the rattle and the historical viewpoints regarding its function, including those that extend as far back as the 16th century. Klauber viewed the rattle as a way to warn predators and large mammals that could injure the snake to stay away, a view he supported with a considerable amount of observation.  In a study of rattling sounds made by 27 rattlesnake species, Bruce Young and Ilonna Brown found the size of the proximal rattle segment was a good predictor of the sound the snake would make and that the sound of the rattle would change as the snake grows.

Considerable variation can be found in rattle size, form, and sound. They range from small, barely audible rattles like that of the Eastern Massasaugua (Sistrurus catenatus) to the large, loud rattles of the Eastern Diamondback Rattlesnake (Crotalus adamanteus). The rattle forms from the style, a series of fused tail vertebrae to which the muscles that shake the rattle are attached. The pronged shape of the style probably encourages the retention of segments of the rattle as the snake grows and sheds it skin, adding a segment at each shedding. Jesse Meik and André Pires-daSilva examined the morphology of the rattlesnake style in 34 species and found an inverse relationship between tail vertebrae and style size. The transition to larger, more globular styles occurred twice: once in the Eastern Massasaugua (S. catenatus) and once in other rattlesnakes (Crotalus). The basal Pygmy Rattlesnake (Sistrurus miliarius) has a small style and relatively small rattle compared to the massasaugua. The authors suggest that the small style found in the Sinaloan Long-tailed Rattlesnake (Crotalus ericsmithi) represents the path that encouraged the retention of rattle segments. The Catalina Island Rattlesnake (C. catalinensis) had the most-derived style and it is the only rattlesnake that has lost the ability to retain segments―a rattleless rattlesnake.

Rattling rattlesnakes are performing one of the fastest sustainable movements known in vertebrates, which make tail rattling energy expensive. Brad Moon and colleagues studied Western Diamondback Rattlesnakes (C. atrox) and found rattling frequencies ranged from 23 to 100 Hz at temperatures between 10º and 35º C. They found the rattle twists as it swings from side to side. This same torsion has been observed in other rattlesnakes, and the authors have proposed that it is an ancestral trait and will lead to understanding how the rattle evolved. Because tail vibrating has been reported in many species of snakes that lack rattles, it seems likely that tail vibration is, indeed, an ancestral trait. However, tail displays in rattlesnakes may not always involve the rattle.

Ali Rabatsky and Jane Waterman described the Dusky Pygmy Rattlesnake (Sistrurus miliarius barbouri) using a tail undulation behavior that did not involve sound production. The entire tail is undulated in a wide arc while the distal half of the tail is wiggled in a sinusoidal fashion. How this behavior complements other tail displays by this rattlesnake is unclear. However, the Dusky Pygmy Rattlesnake has a small rattle that is difficult to hear and this visual display may compensate for the low volume. Or, it may have some other purpose. Rattles may effectively ward off mammals or birds, but they apparently do nothing for the defense against snake predators.

Body-Bridging

                The behavior of a rattlesnake toward the snake-eating kingsnake was first noted by Klauber. He describes how a rattlesnake raises its body in a vertical loop, with its head and tail pressed to the floor of the cage. He hypothesized that this made it more difficult for the kingsnake to secure a hold on the rattlesnake’s head.

                Raymond B. Cowles noted a similar interaction between a Northern Pacific Rattlesnake (C. oreganus) and a California Kingsnake (Lampropeltis getula). With its head and tail pressed to the ground, the rattlesnake would raise its body into a loop, bring the loop down on the predatory kingsnake, and strike the kingsnake with a heavy blow. Cowles also observed similar behavior used toward the odor of the spotted skunk (Spilogale putorius).

                In a series of experiments, Charles Carpenter and Jim Gillingham presented predatory Speckled Kingsnake (Lampropeltis getula) and the Prairie Kingsnake (L. calligaster) to Copperheads (Agkistrodon contortrix), Cottonmouths (A. piscivorus), the Pigmy Rattlesnakes (Sistrurus millarius), and the Hopi Rattlesnakes (C. viridis). All of the pitvipers responded by elevating, inflating, fliping, and jerking their bodies. Carpenter and Gillingham termed this behavior “body bridging.” The pitvipers coiled tightly and hid their heads beneath a coil in an attempt to keep the kingsnakes from seizing the head. While the behaviors were similar, each species had its own variation of body bridging. The authors believed that while they used whole kingsnakes in these experiments, it was the odor of the kingsnake that was eliciting the behavior.

                To follow up on the previous experiments, Paul Weldon and Gordon Burghardt used 21 additional species of New World pitvipers as well as several other snakes known to be snake predators, such as the Indigo Snake (Drymarchon corais), the Black Racer (Coluber constrictor), the whipsnake (Masticophis) and two other species of kingsnakes (Lampropeltis). Instead of presenting a live snake to each pitviper, the snakes were presented with a cotton ball that had been moistened with alcohol and rubbed on a snake-eating species’ back. The pitvipers responded to these scented cotton balls with the anticipated body bridging behavior.

Body bridging has also been reported in some Australian elapids such as the Bandy-bandy (Vermicella annulata) and the Mud Adder, (Denisonia devisi). Lifting a body coil allows the snake to use its mid body like a club. The lack of limbs has limited the options of snake defensive behavior, and snakes have invented some new tools to protect themselves from predators.

Chemical Warfare

                Snakes have cloacal glands on each side of their vent, or cloaca, that produce foul- smelling molecules. When disturbed, snakes can release the bad-smelling stuff, sometimes along with the contents of their intestine. The substance may be smeared onto the predator holding the snake or the snake may role in the musk, spreading it over its body, perhaps in an attempt to make itself less tasty.

                The scent gland molecules from a kingsnake were used in a series of experiments by Andrew Price and Joseph LaPointe. An experiment of particular interest involved food laced with snake scent gland material and food that was not mixed with the material. The food was presented to a variety of mammals (Spotted Skunk, Spilogale putoris; Badger, Taxidae taxus; Bobcat, Lynx rufus; Coyote, Canis latrans; and a Gray Fox, Urocyon cinereoargenteus), all of which were potential snake predators. The mammals refused the food with snake musk, warily approached the food, or ate the food only after first eating the untreated food.

                Brent Graves and David Duvall tested for a sex-linked difference in the odor of snake cloacal gland musk using a double-blind technique and found that female garter snake (Thamnophis) and ratsnake (Pantherophis) musk was more malodorous than that made by male snakes of the same species. But, cloacal musk may also have another function; it may contain an alarm molecule, a chemical message to members of the same species that says “danger is near.”   Prairie Rattlesnakes (Crotalus viridis) produce such a molecule and experiments by William Wood and colleagues showed a rattlesnake exposed to the alarm molecule, increased the heart beat dramatically. The musk functions as a warning message to other snakes. Studies of scent gland chemistry found that snakes in different families shared some common molecules.

Scent gland secretions of snakes are also thought to repel predators, but few predator species have been tested for responses to these exudates. Domestic Cats (Felis catus) were tested by Jeannie Wright and Paul Weldon for responses to scent gland secretions of the Gray Ratsnake (Pantherophis spiloides), or to chloroform extracts of snake gland molecules applied to filter paper or food. More cats salivated or rubbed on filter papers treated with scent gland secretions than on control papers. Scent gland exudates elicited rubbing and pawing in cats more frequently than did chemicals from a shed snake skin. Cats offered food treated either with water or with scent gland secretions ate fewer of the latter; this result is consistent with the hypothesis that scent gland secretions deter feeding, and protect snakes from being eaten by predators.

Scent glands play yet another role in snake life history. Female Brown Treesnakes (Boiga irregularis) have larger glands than males, and the males’ secretions have been shown to be used by females to reject unwanted male suitors. Weldon and colleagues have shown that only 6% of the cloacal gland secretions can be extracted with organic solvents, suggesting that the remaining portion of the secretions are macromolecules. Given the presence of proteins and peptides in cloacal glands, and the fact that mammals avoid them when mixed with food, implies cloacal gland molecules may be more than noxious, they may be toxic.

                A night walk in a tropical Thai forest is always interesting; there are always new animals and new animal behavior to see. One June evening, Red-necked Keelbacks (Rhabdophis subminiatus) were particularly active and numerous. While this snake may exceed a meter, most adults are smaller. The R. subminiatus were sitting on vegetation 40–60 cm off the ground, and a few were found crawling across the forest floor. Individuals sitting in the vegetation would remain in position even when we stuck a camera in their faces and took flash photos. However, the ones on the ground would usually attempt to flee and, when restrained with a gloved hand, they would flatten their forebody, arch their head, and expose a bright-red patch of skin [Figure 9–7]. The Red-necked Keelback (Rhabdophis subminiatus), the Yamakagashi (Rhabdophis tigrinus), and several of their relatives have enlarged nuchal glands (or a single gland) in the skin behind the head.  When disturbed, they draw attention to the area with a display that involves color, flattening the forebody, and bending their head downward to expose the forebody to a predator. Damage to the head and forebody can lead to serious injury.  Understanding the function of this behavior requires some knowledge of anatomy and diet.

INSERT FIGURE 9-7.

Figure 9–7. The Red-necked Keelback, (Rhabdophis subminiatus) from southern Thailand. The snake is displaying typical defense behavior, expanding its forebody and displaying the red colored skin over the nuchal glands.

                Rhabdophis often eat toxic toads with no ill effects. The snake absorbs the toxic molecules obtained from the toads, transports them through their blood stream to the nuchal glands, and stores them. When a predator damages the skin, the glands rupture, spilling the toxin onto the skin where the predator’s mouth will likely contact the poison. The toad toxins are steroids that will affect the heart rate of a potential predator. Experiments done by Deborah Hutchinson and colleagues provided captive, pregnant female Yamakagashi a diet of fish, non-toxic frogs, or toxic toads. Females on the fish and non-toxic frog diet gave birth to young that did not have toxins in their nuchal glands. However, the pregnant females fed a diet of toxic toads were able to provision their offspring with the toad toxin in utero and the neonates were born possessing nuchal gland toxins.

                Akira Mori and Gordon Burghardt examined the antipredator behavior of 27 species of natricid snakes with and without the nuchal glands. The Yamakagashi is known to use the head- arch, neck-butt, and a dorsal-facing posture and the authors found three of the four species with nuchal glands also use these behaviors. The glands lack a duct, but the skin covering them is relatively thin and would be easily broken by a predator that grabbed the snake. The neck-butting behavior may be intended to encourage the predator to break the skin and receive the toxins. Mori and Burghardt found only species with nuchal glands exhibited these behaviors but could not elicit any of the behaviors in the Green Keelback (Rhabdophis nigrocinctus), a species also reported to have nuchal glands. Rhabdophis are relatively passive when it comes to delivering their nuchal gland toxins; this is not the case with the venom sprayed from spitting cobras.

                Spitting cobras actively spray venom in the face of a potential predator or trampler. The following story may be the earliest written account of a spitting cobra defending itself. A fort was being repaired at Mouree on Africa’s Gold Coast when the workmen observed a large snake behind a pile of stones. William Bosman reported the following incident in 1705.

…a mason…laid hold of his tail, designing to pull him out from betwixt the stones; but finding that impracticable, cut off as much of his body as was in reach with his knife; and believing he had disabled him from doing any farther mischief, without the least shadow of fear removed the remainder of the stones; but as soon as the snake was at liberty to turn himself, he clung about the mason (who thought to have caught him in his hand) and spit his venom all over his face: Which proved so forcible that the fellow at that very instant became stark blind; in which condition he remained some days, but was at last restored to his sight.

                Ejecting venom from the fang opening under enough pressure that the venom can travel as far as three meters has become known as spitting, an ability found in many species of Asian and African cobras. Travelers’ tales talked about snakes that “spouts poison into your eye,” and Andrew Smith (the same Andrew Smith who believed boomslangs to be harmless in Chapter 4) considered stories of spitting cobras in South Africa to be folklore. However, William Bosman accurately described venom spitting in the 18th century, and it was another 100 years before Heinrich Boie named a Javanese cobra Naja sputatrix in 1827. Teo hundred years before ,Goring Jones provided an early description of “spitting venom” from an Asia snake (near Mandalay, Burma).

Yesterday evening as Lieut. Gibson of our regiment was going to his quarters to dress for Mess he saw a snake at which he proceeded to throw a stone. He then called for a light and a stick, and as he was bending down with the light to look for the snake, it made a dart at him but fortunately missed its aim. Some of the poison of saliva, however, was ejected into Lieut. Gibson’s eye causing instant and great pain, and the eye lids and parts round swelled up quickly to the size of a large hen’s egg. The snake was killed, and was found to be a small black cobra about 3 feet 4 inches in length. Lieut. Gibson went to the hospital and after a painful night recovered his eyesight.

                Despite the 1900 report, it was not until 2000 that the species of cobra that sprayed Lieutenant Gibson was described to the scientific community. It is a species restricted to an arid savanna in central Burma and is now known as the Burmese Spitting Cobra, Naja mandalayensis. However, other species of cobras also eject venom from their fangs and in Asia there are more species of spitting cobras than there are non-spitters.

                David Warrell and L. David Ormerod reported on nine cases of African Black Spitting Cobra (Naja nigricollis) venom getting into the eyes of humans. In five instances, the only effect was simple conjunctivitis, but four had corneal ulceration and one developed inflammation in the middle layer of the eye, suggesting venom had been absorbed into the eye’s anterior chamber. Two of the patients were permanently blinded.

                Clarification of cobra systematics and nomenclature has only recently begun. For much of the 20th century, Asian cobras were believed to compose one wide-ranging species, the Indian or Spectacled Cobra, Naja naja. Today about 10 species of Asian cobras are recognized and eight of them are known to spit venom. In the early 1940’s, Charles Bogert first recognized the fang modifications necessary for spitting venom. Spitting species have a smaller opening in the fang, and it is located farther away from the fang tip than in non-spitting species. Bogert also suggested spitting cobras had shorter fangs than non-spitting species. However, Wolfgang Wüster and Roger Thorpe have shown this to be incorrect.

                Calling what cobras do spitting is, of course, misleading in that it suggests the snake sprays venom by blowing it out of its mouth with air coming from the lungs― much in the same way a human would spit. Instead, snakes are spraying venom out of the fang’s discharge opening. As the venom travels through the fang’s canal, it is directed slightly upward when it reaches the slightly restricted discharge orifice the fluid pressure is increased. The snake orients its head to aim the venom.

                Bruce Young and co-workers examined spitting cobra morphology and behavior and found that spitting venom requires a two-part mechanism. First, the protractor pterygoidus muscle displaces and deforms the maxillary and fang sheath so that soft tissue is not impeding venom flow. Secondly, as the protractor muscle is acting, the adductor mandibulae externus muscle is placing pressure on the venom gland. Young and colleagues found that by having the muscles act simultaneously, venom pressure at the fang tip is greater than the pressure that could be produced by each muscle individually; the pressure was 200 times greater at the fang tip than in the venom duct. Also of interest, Young and colleagues suggested that the evolution of spitting has been modified from specializations of prey ingestion as opposed to prey capture.

                Guido Westhoff and colleagues compared the spitting behavior of the Black Spitting Cobra (Naja nigricollis) and the Red Mozambique Spitting Cobra (Naja pallida) and found not all spitting cobras spit the same way. The Red Mozambique Spitting Cobra discharges the venom stream in two distinct jets while the Black Spitting Cobra’s venom forms a fine spray, producing a venom cloud. A face with or without eyes, or with just one eye, stimulated the snakes to spit, while just isolated eyes without a face did not elicit spitting. This suggests spitting cobras do not aim directly for the eyes, but for the face. While other literature and observations suggest that venom may be sprayed up to 3 m, these authors did not succeed in getting snakes to spray venom more than a meter. During spitting, the snakes do move their heads, and the Black Spitting Cobra made a complete circular movement with its upper jaw within 16 milliseconds. In a second paper, Ruben Berthé and colleagues used the same two species of cobra to see if the snakes adjusted their venom spray for target distance. They found a positive correlation between target distance and spitting pattern dimension, and the snakes did compensate for target distance, but not perfectly. The cobras adjust their behavior within 40–60 milliseconds to account for changes in target distance.

                Spitting cobras meter their venom expenditure. R. C. H. Sweeney reported a Black Spitting Cobra sprayed venom 57 times in 20 minutes. If it spit all of its venom, it was using 1.75% of its total venom in each spray; therefore, remarkably small quantities of venom are involved in each spraying event. Jennifer Cascardi and co-workers collected venom after it had been sprayed from a Red Mozambique Spitting Cobra. They found the average volume spit was less than 1.7% of the volume in the venom gland, a number almost exactly the same as predicted by Sweeney’s observation. Interestingly, the chemical composition of the venom spit differed; the first five spits contained proteins that were 9.0 and 75 kiloDaltons in weight, but they were absence from later spits.

                The geography of the spitting cobras is also of interest. Spitters occur in Africa and in Southeast Asia, but are absent from the Middle East, India, and Sri Lanka. The Monocellate Cobra (Naja kaouthia) is a relatively widespread species and shows considerable variation in the size of the fang’s discharge orifice, suggesting that some populations should be able to spray venom. The Chinese cobra (Naja atra) also has a fang orifice that is as small as that found in spitting cobras. Despite this, neither species is known to spray venom. Wolfgang Wüster and Roger Thorpe found the Philippine Cobra (Naja philippinensis) shows sexual dimorphism in the size of the fang’s orifice, with the males having a smaller discharge orifice in the fang than females. They did not provide a hypothesis that accounts for this difference, but it seems likely male cobras may move more than females, and, therefore, would encounters predators more frequently than females.

Spitting venom is a low-risk defense behavior because it allows the snake to remain at a distance from its potential predator. If the snake has to actually bite the predator defensively, the risk of injury to the snake is much greater. And, while spitting is low risk, it may be somewhat expensive in terms of the cost of venom production.

Speculations On Rattling and Spitting

                Harvard University herpetologist Thomas Barbour wrote a 1922 paper that linked rattlesnake rattles and spitting behavior in cobras as ways to repel the “…hordes of grazing ruminants…” In other words, Barbour viewed both rattles and spraying venom as ways to avoid being trampled by large mammals. Before and since his paper, other hypotheses suggested to explain these extreme adaptations. Co-evolutionary scenarios involving extinct fauna are at best highly speculative, but are useful for stimulating experiments and research; they are also of interest just for the fun factor.

Rattlesnakes are endemic to the Western Hemisphere and, in 1940, Howard Gloyd suggested rattlesnakes evolved on the Mexican Plateau. Lawrence Klauber and Harry Greene also viewed the ancestral home of the rattlesnakes as somewhere in Mexico. This hypothesis was maintained based on the fact that the species considered morphologically primitive occur in Mexico, and about 27 of 38 rattlesnake species live in Mexico today. Aaron Place and Charles Abramson used the cladistic method of ancestral area analysis to predict the geographical origin of rattlesnakes. They concluded that it was Mexico’s Madrean Occidental, and the ancestral habitat was pine-oak forest.

Harry Greene had previously proposed the Madrean-talus-predator hypothesis, suggesting the rattle had evolved to repel predators such as mustelids (weasels and badgers), procyonids (coatis, ringtails, and raccoons), and bears that will search and probe crevices looking for prey. Biting and envenomating a nose or front foot of a mammal probing the crevice where a snake is hiding, combined with the sound of a rattle, would be an unforgettable experience for a potential predator and one they would most likely avoid in the future. Place and Abramson applied their analysis to hypotheses of rattle evolution scenarios and they rejected Barbour’s trampling-ungulate hypotheses in favor Greene’s Madrean-talus-predator hypothesis.

A phylogenetic analysis of the pitvipers based on four mitochondrial genes by Todd Castoe and Christopher Parkinson found rattlesnakes to be monophyletic and Sistrurus (the Massasaugua and Pygmy Rattlesnakes) to be the sister to Crotalus (all other rattlesnakes). What this implies is that rattlesnakes may have not evolved in Mexico, but farther north. Wolfgang Wüster and colleagues published a phylogeny and historic biogeography of vipers based upon four mitochondrial genes and provided dates for various divergence events. Using Bayesian molecular dating in conjunction with dispersal-vicariance analysis, Wüster and colleagues estimate a pitviper colonized the Western Hemisphere 22.1 MYA (26.9–17.9 MYA) during the late Oligocene or early Miocene.

Pit vipers are not known from Europe. By 22.1 MYA, the Atlantic separated Eurasia and North America, making it unlikely pitvipers moved from Europe into North America. At this time, however, northeastern Asia and Alaska (Beringia) were connected and covered with mixed hardwood and deciduous forest. These forest habitats are used by some Asian and American pitvipers even today. Wüster and colleagues also retrieved a date of 9 MYA or earlier for the most recent common ancestor of Sistrurus and Crotalus based upon a fossil vertebrate from Sistrurus; their Figure 4 suggests the ancestral rattlesnake was present about 12 MYA. Pitvipers may have had a window of about 3 MY to evolve rattles (between 12–9 MYA).

The idea that rattlesnakes used the rattle as a lure goes back to at least Nathaniel Shaler’s 1872 paper, but was probably established in folklore well before this. Shaler hypothesized that the rattle’s sound mimicked cicadas and functioned as a lure to attract birds to the rattle. Of course, not all rattlesnakes eat birds and there is really no evidence for rattles serving this function in adult snakes today. There are several species of modern rattlesnakes that engage in caudal-luring, though this is known in juvenile snakes and was discussed in Chapter 8. The caudal luring hypothesis proposed by Gordon Schuett and co-workers suggested the rattle originally evolved to increase the visual attractiveness of the lure. Using this scenario the rattle, or perhaps just the button, was first used to bring prey within striking distance.

Ali Rabatsky surveyed snakes that caudal lure and found caudal luring has been reported in 21 species of vipers, and that 67 species of vipers have contrasting tail colors. The behavior was found in four other families of snakes and they, too, had contrasting colored tails. Rabatsky also found reports of the pygopodid lizard, and Burton’s Snake-gecko (Lialis burtonis) luring prey with its tail. Caudal luring has most likely evolved multiple times in various clades of squamates.

Rabatsky tested the idea proposed by Schuett and co-workers that segmentation of the rattle enhanced the visual attractiveness. Rabatsky made the prediction that snakes with two or more segments would attract more prey. In experiments with the Pygmy Rattlesnake (Sistrurus miliarius barbouri), 36% of her snakes lured lizards to within striking distance, but snakes with two or more segments on the rattles did not catch any more food than those snakes with fewer than two segments. 

Norman Sisk and James Jackson tested two hypotheses regarding the evolution of the rattle. The authors used the Fence Lizard (Sceloporus undulatus) and a mechanical, vibrating model to test visual attractiveness of the lure, one bilobed like that found in rattlesnakes and one that was conical like a normal tail tip. They found no difference in the response of the lizards. They also tested the hypothesis that the bilobed lure would produce more sound intensity than a conical lure when it was pressed against the substrate; they found it did not. Sisk and Jackson rejected the hypotheses though, in all fairness, we do not know what kind of prey the proto-rattlesnake was luring, and we really don’t know much about the proto-rattlesnake. It does, however, seem probable that the first rattlesnake, or its ancestor, used the tail tip to lure lizards, given what we now know about caudal luring in Sistrurus. Here it is important to revisit the facts that Sistrurus, apparently the most basal of the living rattlesnakes and have relatively small rattles. Additionally, the juveniles caudal lure. The Sistrurus rattlesnakes, also use grasslands, savannas, longleaf-pine, and scrub oak woodlands.  All of these are open, relatively dry habitats.

North American grasslands and open woodlands supported numerous medium and large browsers and grazers between 12 and 9 MYA. During this time the ancestral rattlesnake was evolving and after 9 MYA, there may have been considerable selection for larger, louder rattles.

Matthew Kohn and Theodore Fremd reported carnivore and ungulate diversity in western North America abruptly increased 17.5–17.0 MYA, the diversity decrease about 11 MYA, which was then followed by a period of stability in ungulate diversity. Kohn and Fremd hypothesized that this resulted from climate change induced by tectonic movements in the Basin and Range, Mojave, Columbia Plateau, and Rio Grand Rift. The high-angle extensional faulting increased topography and ecosystem diversity. Between 15 and 8 MYA, the flora became dry adapted in the Great Plains. This coincides well with the DNA clock dates for evolution of the dry adapted rattlesnakes. Atmospheric carbon dioxide increases had started as long as 25 MYA. The warming climate stimulated the expansion of the C4 grasslands that spread world wide by 8–7 MYA. These grasslands were subject to seasonal precipitation and intense seasonal fires. C4 plants have a mechanism that allows them to collect and store more carbon dioxide than C3 plants and as a result the C4 plants can grow faster than those plants using the C3 photosynthesis pathway. The overall effect is to increase photosynthesis and the food available to animals.

The North American Miocene was rich in browsing and grazing mammals with hoofs; there were at least 15 families and 143 genera of ungulates. Camels and horses were the most numerous of these genera, but pig-like mammals and antelope were also very numerous.  There were another eight families of carnivores (Carnivora) with at least 83 genera; many may have been omnivorous but still capable of killing a snake for food, and they included dogs, weasels, and raccoon relatives.

Christine Janis and colleagues described the mid Miocene (12–8 MYA) Great Plains environment as woodland savanna with 500–1000 mm of annual precipitation, and a temperature 10ºC warmer than today.  The mammalian fauna included plant-eaters with low crowned teeth that browsed (brachydont ungulates). As many as 19 species coexisted at a single location, and about 9 of these were mixed browsing-grazing species. By 4.5 MYA the browsers had disappeared and grazers dominated the grasslands. Janis and colleagues suggest woodland flora with mixed browser – grazer communities were likely global in distribution. At this same time productivity was exceptionally high, that is food was abundant.

W. David Lambert compiled a list of 28 species of large Miocene ungulates from North America’s Great Plains and Gulf Coast that had body weights between 36 and 1,792 kg. A meter-long ground-dwelling snake stepped on by a 36 kg Nannippus (an extinct horse) or a 1792 kg Aphelops (an extinct rhinoceros) produce similar results. The point here is that the evolution of rattlesnakes did not necessarily have to co-evolve just with large grazing herds that were around 4.5 MYA; there were plenty of large animals around to step on them prior to that date.

We know very little about the ancestral rattlesnake or its prey. While studies have shown that the rattle does nothing for caudal luring in its current evolutionary state, what if that first rattle segment originally worked as a visual enhancement for the snake’s prey? Several authors have noted the similarity of the rattle’s button to a caterpillar head or larval insect. The hard, swollen button of keratin at the tip of a neonate rattlesnake has been termed the sclerophyma by Sisk and Jackson. A segment is added to the rattle each time the snake sheds, and the rattle grows from the base. Therefore, the ancestral rattlesnake may have enhanced its ability to attract prey with just the swollen tail tip, the sclerophyma. Imagine the shape, coloration, and movement of the tail were involved in luring prey, sound was not involved. Then, a mutation altered the tails of some snakes in the population to elongate the rattle with segments. When the rattlesnake vibrates its tail, it now makes a distinctive sound. The sound could then be associated with painful envenomation. Adult snakes could no longer use the rattle to lure prey because it produced sound. Large herbivorous mammals and smaller carnivorous mammals would soon learn to avoid rattlesnakes by associating the sound with envenomation. Thus, a scenario that involves selection for an enhanced caudal lure to improve feeding success; followed by selection for a mutation to elongate the rattle as an aposematic auditory display; may have produced a lineage of cryptic, ground-dwelling pitvipers with rattles on their tails.

However, Thomas Barbour also suggested spitting behavior in cobras evolved in response to large mammals. Today, African spitting cobras use open habitats such as grasslands and deserts. Wolfgang Wüster and colleagues suggested the African spitting cobra clade evolved in the early Miocene, about 15 MYA. Their confidence limits are quite wide for this clade’s origin, it ranges from the late Eocene to the late Miocene a time interval of almost 21 MY. This timeline does not match what was known about the evolution of the high-crowned teeth ungulates that evolved to feed on the C4 grasslands implied by Barbour’s statement. Again, this does not mean large mammals were absent and snakes were not trampled.

René Bobe examined the evidence for the evolution of arid ecosystems in east Africa and found the Fort Ternan fossil site in western Kenya yielded evidence that both woodland and grasslands were present 14 MYA. At least 18 genera of large herbivorous mammals with body masses that reached 350 kg existed in Africa’s early to middle Miocene. Similarly, the paper by Christine Janis and co-workers previously referred to, reports 18 and 23 browsers, grazers, and mixed-feeding ungulates from East African localities 14 and 10 MYA, respectively. Loïc Ségalen and Julia Lee-Thorp wrote about the early Miocene of southwestern Africa’s Namib. They state that between 19 and 17.5 MYA the extinct fauna evolved in more open environments compared to those suggested by previous paleontological reconstructions. Open habitats used by large mammals (as well as medium and small carnivores) were present before, during, and after 15 MYA when the DNA clock suggests spitting cobras were evolving in Africa.

The Asian cobras also appear to be monophyletic and were derived from an African ancestor. The Asian cobras may have shared an ancestor about 17 MYA, and two of the spitting species shared an ancestor about 7 MYA. Today, the Asian spitters tend to use more wooded environments and would have also encountered browsing and grazing mammals of a variety of sizes as well as small and medium-sized carnivores. Being able to spit venom into the eyes of a potential trampler or predator was a life-saving, low risk behavior for cobras. Frank Wall reported the observations of a Mr. Reid. Wall wrote,

A herd of buffaloes that were standing, feeding out of a row of “nands,” suddenly became very excited and broke loose, stamping and snorting, and to all appearances were terrified. On investigation, a cobra was found close by, which was killed, one old cow when she saw it rushed upon its body and trampled it. This, by the way, is the method by which deer and pigs are reported to attack and destroy snakes.

                Barbour’s hypothesis that rattlesnake’s rattles and spitting in cobras evolved to prevent snakes from being trampled by herbivores or harassed and eaten by carnivores remains viable, but unfortunately not verifiable without time travel.