If you need to identify a snake, try the Snake Identification Facebook group.
For professional, respectful, and non-lethal snake removal and consultation services in your town, try Wildlife Removal USA.

Wednesday, September 25, 2013

Basics of Snake Fangs

Solenoglyphous fangs of a Gaboon Viper
Snake fangs are specialized, elegantly modified teeth. Some are like hypodermic needles, others are more like water slides. But all serve essentially the same purpose: to inject venom into the snake's prey. Occasionally, the fangs are also used in defense, but studies show that snakes striking in defense are far less likely to inject venom than when they're striking at a prey item, a fact that has assuaged the fears of many an ophidiophobic. I wanted to write a brief review of snake fang types, because their anatomy is very interesting and also because of their important role in classifying snakes and understanding how different groups of snakes are related to each other.

Cross-sections of fangs:
F is an aglyphous tooth.
G is an opisthoglyphous fang.
H is a proteroglyphous fang.
I is a hollow solenoglyphous fang.
From Bauchot (2006)
Many snakes produce venom, which is essentially very strong saliva, in glands in their heads (which is where you produce your saliva, too). We call these glands venom glands if they are well-developed, complete with an interior cavity, a duct connecting to a hollow fang, and compressor muscles that generate high pressures when the jaws are rapidly closed. If they lack these features, we usually call them Duvernoy's glands instead. Because there is a lot of variation among snake species in the structure of these glands and their associated teeth, there is some debate about whether or not venom glands and Duvernoy's glands are really two forms of the same thing. Either way, three groups of snakes (atractaspidines, elapids, and viperids) have independently evolved an advanced apparatus to deliver large quantities of venom during a brief strike, and many other snakes (and a few lizards) have evolved less sophisticated, but still relatively effective, modifications to their teeth in order to deliver venom after they have grabbed their prey and are "chewing" on it. The teeth of modern snakes are classically divided into four types, three of which are typically called fangs. The four tooth types have fancy names, all of which involve the Greek word glyph, one of the meanings of which is "groove". They are as follows:


Folding of solenoglyphous fangs.
Fang is in red, maxilla green,
prefrontal orange, pterygoid yellow,
ectopterygoid purple. Vipers lack
premaxillary and palatine teeth.
From Bauchot (2006)
This most sophisticated fang type evolved once, in the ancestor to all modern vipers, which lived in Asia about 40 million years ago. Fossils suggest that solenoglyphous fangs have changed little since that time, even though vipers have undergone an incredibly successful radiation into 320 extant species found on all continents except for Australia and Antarctica. Solenoglyphous fangs are long and tubular and are attached to the snake's maxillary bone. Most snakes have several tooth-bearing bones, including four (the premaxilla, maxilla, pterygoid, and palatine) in the upper jaw, and one (the dentary) in the lower. In humans, three of these bones (the premaxilla, maxilla, and dentary) also bear teeth - your premaxilla holds your top incisors, while your maxilla holds your upper canines and molars and your dentary all your lower teeth - while the others form part of the roof of the mouth. In vipers, the maxilla bears only a single tooth (the fang) and is hinged so that the fangs can be folded back parallel to the jaws when the mouth is closed, or erected perpendicular to the jaws, the position when striking. The teeth in the pterygoids and dentaries work together to manipulate food once it gets into the mouth. Solenoglyphous fangs are strikingly similar to hypodermic needles. They have a hollow core that receives venom from the venom gland at the entrance orifice near the base and injects it from a slit-like exit orifice on the front of the fang near the tip. If the opening were at the very tip of the fang, its strength would be compromised and it would lack the sharp point needed to penetrate the target. Even under normal use, vipers shed their fangs every two months.

Modified solenoglyphous fang of
African Burrowing Asp (Atractaspis engaddensis)
A similar fang type evolved a second time about 29 million years ago in a group of African snakes, currently placed in the family Lamprophiidae, subfamily Atractaspidinae. Two genera, Atractaspis (mole vipers, burrowing asps, or stiletto snakes) and Homoroselaps (African dwarf garter or harlequin snakes), possess elongate anterior fangs, although only those of the stiletto snakes are movable. Stiletto snake fangs pivot on a socket-like joint that is more flexible than those of vipers, allowing these snakes to strike beside and behind them with their mouth closed. This is an adaptation to living underground and envenomating small mammals and other reptiles in narrow subterranean burrows. The fang morphology of atractaspidines and viperids is remarkably similar, considering that these two snake lineages last shared a common ancestor over 40 million years ago.


Proteroglyphous fangs of an Eastern Green Mamba
(Dendroaspis angusticeps). Don't try this.
From Bauchot (2006)
This fang type also evolved only once, in the ancestor to all modern elapids, which lived 25-40 mya in Asia or Africa. Proteroglyphous fangs are in the front of the mouth and are about three times shorter than solenoglyphous fangs. This is because they are not hinged. Snakes with proteroglyphous fangs typically strike their prey and hang on until the venom has taken effect, as opposed to releasing they prey and then tracking it down. Some elapids constrict their prey at the same time as envenomating it. Over 350 species of elapids exist today, including well-known groups such as cobras, mambas, death adders, taipans, coralsnakes, and sea snakes, and less-well-known species, mainly found in Australia, of which a good number are small, secretive, and not considered dangerous to humans.

Maxilla of a proteroglyphous snake showing the almost
completely closed groove along the anterior edge connecting
the two orifices, as well as the aglyphous tooth at the
rear of the maxilla. This line may be obscured in longer fangs.
From Shea et al. 1993
Unlike solenoglyphs, some proteroglyphs have other teeth on the maxilla behind the fang. However, the fang is always separated from the other teeth by a gap, called a diastema. Some elapids have more than one functional fang on each side. In both vipers and elapids, there are usually at least two fangs on each maxilla at any one time, one that is in use and one that is a reserve fang. Both fangs are draped in a layer of connective tissue and skin called the fang sheath. Some proteroglyphs have partially movable fangs, including many of the most dangerous species such as mambas, taipans, and death adders. A few, such as spitting cobras, have modified exit orifices to their fangs that are smaller and rounder than in other cobras, a modification that increases the velocity with which venom is ejected. Modifications to the muscles and the fang sheath also facilitate spitting in these cobras. A few elapids, such as sea snakes that eat only fish eggs, have lost their fangs and their venom glands, which suggests that the primary role of venom, at least among elapids, is in feeding rather than in defense.


Opisthoglyphous fang of Eastern Hog-nosed Snake
These are commonly known as "rear-fanged" snakes. Opisthoglyphous fangs are grooved rather than hollow and are found near the back of the maxilla, behind the normal teeth. Typically, snakes with rear fangs must chew on their prey to bring their fangs into a biting position. There is considerable variation in the size, shape, and number of opisthoglyphous fangs from species to species, and sometimes even within a species. Most opisthoglyphous fangs are connected to Duvernoy's glands, which differ from true venom glands in several important ways, most notably in that they lack associated muscles to generate the pressure needed to evacuate venom, as in solenoglyphous and proteroglyphous snakes. The pressure on the venom glands of biting solenoglyphs and proteroglyphs can exceed 30 psi, the pressure of a car tire, whereas the pressure inside the Duvernoy's glands of opisthoglyphs is generally less than 5 psi. Because Duvernoy's glands also lack a chamber for storing venom, the idea is emerging that opisthoglyphous snakes probably secrete their venom only during chewing, which explains why prolonged bites by opisthoglyphs generally have more severe effects.

Opisthoglyphous fangs of Boomslang (Dispholidus typus)
Don't do this either.
Most of these snakes are not harmful to humans, with a few notable exceptions. Boomslangs and Twigsnakes are arboreal, diurnal African colubrines that prey on lizards and birds. They have short heads, rear fangs situated comparatively close to the front of the mouth, and partially muscled Duvernoy's glands. They also have potent venoms and their bites have killed several people, including two prominent snake biologists, Karl Schmidt and Robert Mertens. Bites from other rear-fanged snakes are known to cause relatively mild, transient, and local symptoms, but clinical documentation of these bites and their effects is scattered, incomplete, and frequently anecdotal. Many are written by the victim themselves! The above notwithstanding, bites from opisthoglyphs are generally less medically important than those from proteroglyphs and solenoglyphs. As a result, snake venom research has not focused on them, so there is still much that we do not know about this group of snakes, some of which are becoming increasingly common in the pet trade. Based on what little we do know, the composition of opisthoglyph venom/Duvernoy's secretion is fairly similar to that of viperids, elapids and atractaspidines, which makes sense given that each of these groups is more closely related to certain opisthoglyphs than they are to one another.

A: python, B: viper, C: rear-fanged colubroid, D: cobra
The f  marks the portion of the maxilla where the fang develops.
E shows the elongation of the posterior part of the
maxilla pushing forward the developing fang of a
night adder (d.a.o. = days after oviposition)
From Vonk et al. 2008
Unlike the first two groups, opisthoglyphous fangs appear to have evolved more than once, in snakes as diverse as Quill-snouted Snakes, Neck-banded Snakes, and Boomslangs. At least, that's what we used to think. Actually, it is likely that both solenoglyphous and proteroglyphous fangs evolved from opisthoglyphous fangs, as revealed by an ingenious study that used evidence from embryology and genetics to reveal the evolutionary origins of the three types of snake fangs. In a snake embryo, tubular fangs are formed by the infolding of ridges on the front and back sides of the fang, such as those that form the groove of opisthoglyphous fangs. Furthermore, front fangs develop from the rear part of the upper jaw, and are strikingly similar in their formation to rear fangs. They are pushed into the front of the mouth by disproportionate growth of the initially small part of the maxilla that is behind them. Finally, in the anterior part of the maxilla of front-fanged snakes, expression of a gene called sonic hedgehog, which is responsible among other things for the formation of teeth, is suppressed.

Relative size of the venom gland (VG) in
A: rear-fanged colubrid (Helicops leopardinus),
B: boomslang, C: homalopsid,
D: cornsnake, E: African egg-eater
SG = supralabial salivary gland
From Fry et al. 2008
Although developmental similarity is not conclusive proof of structural homology (similarity due to inheritance rather than due to other factors), these findings are consistent with the hypothesis that solenoglyphous, proteroglyphous, and at least some opisthoglyphous fangs are homologous structures. The hypothetical evolutionary trajectory was thus: some snakes evolved grooved fangs in the rear of their mouth. In a few cases (viperids, elapids, and atractaspidines), they subsequently lost the preceding teeth as what was formerly a rear fang became a tubular front fang. Other snakes retained their anterior teeth (at least some non-front-fanged colubroids), and still others developed fangs but then lost them (aglyphs such as ratsnakes). Evidence for this surprising final part comes from the formation of the maxilla and its teeth, which takes place in a single piece in pythons, but from two pieces in all fanged snakes as well as in ratsnakes, a pattern which supports a single evolutionary origin and subsequent loss of fangs. Additionally, vestigial Duvernoy's glands have been found in ratsnakes, egg-eaters, pareatid slug-eaters, and other nonvenomous aglyphs, a discovery that has led to the misleading generalization that all snakes are venomous and much subsequent misunderstanding among the non-scientific community. Toxic saliva does not a venomous animal make, as evidenced by the fact that even human saliva injected subcutaneously will produce pain and swelling.


Both boas and pythons have only
aglyphous teeth, which is about
the only thing this film got right.
This word is used to describe unmodified teeth, essentially non-fangs. All snakes, even those that possess fangs of the first three types, have aglyphous teeth which they use for gripping their prey as they manipulate it during swallowing. As I just mentioned, many advanced snakes that today have only aglyphous teeth probably evolved from fanged ancestors. Several of these snakes, such as North American kingsnakes, ratsnakes, and bullsnakes, have atrophied Duvernoy's glands that lack toxin-producing serous cells. These snakes employ other sophisticated techniques, such as constriction, which is also used by more primitive snakes like boas and pythons (which did not evolve from fanged ancestors).

There are very few dangerous species of aglyphs, but one, Rhabdophis tigrinus, is becoming well-known as one of the only snakes capable of sequestering toxins from its prey for use in its own defense. This species has enlarged posterior maxillary teeth that lack grooves, so they are by definition aglyphous. However, it has relatively potent venom and has caused the deaths of several people. Among colubroids, the distinction between opisthoglyphs and aglyphs has never been entirely clear, but I'm distinguishing between them here because they are two of the four traditionally recognized types of snake teeth. Although the four types of snake teeth in this article are commonly discussed, a more accurate classification for snake teeth might be to divide them into tubular (the fangs of viperids, elapids, and atractaspidines), grooved (the rear fangs of non-front-fanged colubroids), and ungrooved (all other snake teeth).

Aglyphous (ungrooved) teeth and rear fangs of
Rhabdophis tigrinus
From Mittleman & Goris 1974
Happily for snake biologists like myself, the evolution of fangs opened the door for a massive evolutionary radiation of advanced snakes (>2800 species, or >80% of all living snake species). Although sophisticated venom delivery systems, of which fangs are just one of many integral parts, were clearly evolutionary advantageous, they have obviously also been costly at times, leading to their loss in ratsnakes, egg-eaters, and other lineages of advanced snakes. Also worth noting is that many lineages of basal snakes have got along just fine without venom, so there is not an inherent superiority of being venomous as the word "advanced" seems to imply. Rather, some have suggested that during periods of transition from forest to grassland, such as that which took place simultaneous to the dramatic colubroid radiation during the Miocene, snake taxa that were characterized by slow locomotion and constriction (boas & pythons) were supplanted by those characterized by rapid locomotion (many aglyphous colubrids) or passive immobilization (tubular- and grooved-fanged vipers, elapids, and atractaspidines that could use venom to catch their prey). Of course, both slow locomotion and constriction have subsequently been re-evolved among the colubroids, but there has been a lot of time since the Miocene.


Thanks to Daniel Rosenberg (boomslang fang) and Nick Kiriazis (hognose fang) for use of their photographs.


Bauchot R, editor. 2006. Snakes: A Natural History. New York, New York: Sterling Publishers. <link>

Cundall, D., (2002) Envenomation strategies, head form, and feeding ecology in vipers. In: Biology of the Vipers: 149-162. G. W. Schuett, M. Höggren, M. E. Douglas & H. W. Greene (Eds.). Eagle Mountain Publishers, Eagle Mountain, UT <link>

Greene, H. W. (1997) Snakes: The Evolution of Mystery in Nature. Berkeley: University of California Press <link>

Fry BG, Scheib H, van der Weerd L, Young B, McNaughtan J, Ramjan SR, Vidal N, Poelmann RE, Norman JA, 2008. Evolution of an arsenal: structural and functional diversification of the venom system in the advanced snakes (Caenophidia). Mol Cell Proteomics 7:215-246 <link>

Hayes, W. K., S. S. Herbert, G. C. Rehling & J. F. Gennaro, (2002) Factors that influence venom expenditure in viperids and other snake species during predator and defensive contexts. In: Biology of the Vipers: 207-234. G. W. Schuett, M. Höggren, M. E. Douglas & H. W. Greene (Eds.). Eagle Mountain Publishers, Eagle Mountain, UT <link>

Jackson K, 2002. How tubular venom‐conducting fangs are formed. J Morphol 252:291-297 <link>

Kardong, K. V. & T. L. Smith, (2002) Proximate factors involved in rattlesnake predatory behavior: a review. In: Biology of the Vipers: 253-266. G. W. Schuett, M. Höggren, M. E. Douglas & H. W. Greene (Eds.). Eagle Mountain Publishers, Eagle Mountain, UT <link>

Kardong KV, 1996. Snake toxins and venoms: an evolutionary perspective. Herpetologica 52:36-46 <link>

Kuch, U., J. Müller, C. Mödden & D. Mebs (2006). Snake fangs from the Lower Miocene of Germany: evolutionary stability of perfect weapons. Naturwissenschaften 93, 84-87

LaDuc, T. J., (2002) Does a quick offense equal a quick defense? Kinematic comparisons of predatory and defensive strikes in the Western Diamond-backed Rattlesnake (Crotalus atrox). In: Biology of the Vipers: 267-278. G. W. Schuett, M. Höggren, M. E. Douglas & H. W. Greene (Eds.). Eagle Mountain Publishers, Eagle Mountain, UT <link>

Mittleman M, Goris R, 1974. Envenomation from the bite of the Japanese colubrid snake Rhabdophis tigrinus (Boie). Herpetologica 30:113-119 <link>

Pyron, R. A., F. T. Burbrink, G. R. Colli, A. N. M. de Oca, L. J. Vitt, C. A. Kuczynski & J. J. Wiens (2011). The phylogeny of advanced snakes (Colubroidea), with discovery of a new subfamily and comparison of support methods for likelihood trees. Mol. Phylogenet. Evol. 58, 329-342 <link>

Savitzky AH, 1980. The role of venom delivery strategies in snake evolution. Evolution 34:1194-1204 <link>

Shea G, Shine R, Covacevich JC, 1993. Elapidae. In: Glasby C, Ross G, Beesley P, editors. Fauna of Australia. Canberra: AGPS <link>

Vonk FJ, Admiraal JF, Jackson K, Reshef R, de Bakker MA, Vanderschoot K, van den Berge I, van Atten M, Burgerhout E, Beck A, 2008. Evolutionary origin and development of snake fangs. Nature 454:630-633 <link>

Weinstein SA, Warrell DA, White J, Keyler DE, 2011. "Venomous" Bites from Non-Venomous Snakes: A Critical Analysis of Risk and Management of "Colubrid" Snake Bites. Amsterdam: Elsevier <link>

Weinstein SA, White J, Keyler DE, Warrell DA, 2013. Non-front-fanged colubroid snakes: A current evidence-based analysis of medical significance. Toxicon. 69, 103-113 <link>

Weinstein S, White J, Westerström A, Warrell DA, 2013. Anecdote vs. substantiated fact: the problem of unverified reports in the toxinological and herpetological literature describing non-front-fanged colubroid (“colubrid”) snakebites. Herpetological Review 44:23-29.

Wüster, W., L. Peppin, C. Pook & D. Walker (2008). A nesting of vipers: Phylogeny and historical biogeography of the Viperidae (Squamata: Serpentes). Mol. Phylogenet. Evol. 49, 445-459 <link>

Young BA, Dunlap K, Koenig K, Singer M, 2004. The buccal buckle: the functional morphology of venom spitting in cobras. J Exp Biol 207:3483-3494 <link>

Creative Commons License

Life is Short, but Snakes are Long by Andrew M. Durso is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.

Tuesday, September 10, 2013


Head of Xenodermus javanicus
One of the weirdest-looking snakes in the world is Xenodermus javanicus, also called the Javan Tubercle Snake, Javan Mudsnake, Rough-backed Litter Snake, or, best of all, the Dragonsnake. Although they don't breathe fire, their anatomy is strange enough to evoke images of mythical creatures. The first Dragonsnake was described in 1836 by Danish zoologist Johannes T. Reinhardt. He named it Xenodermus, Greek for "foreign skin", because of its peculiar scales, of which Reinhardt described three types. A triple row of large keeled scales runs down the center of the back, flanked by two rows of huge keeled tubercles that resemble crests more than scales. The gaps between these knobby rows are covered in small irregular smooth diamond- and pentagon-shaped scales, whereas the space below the tubercles is coated in more traditional parallel rows of keeled dorsal scales, which nevertheless resemble a highly organized bed of oysters more than typical colubrid scalation.

Dorsal scales of Xenodermus javanicus
Given the dorsal side only, it's difficult to tell what the taxonomic relationships of Dragonsnakes are. However, you can tell this snake is a colubroid if you examine the ventral side, where wide, well-developed ventral scales are present, unlike the smaller ventral scales of more primitive snakes. The tail, which constitutes up to a third of the total length, has a single row of scales on the underside, a characteristic reminiscent of vipers but also found in some colubroids. Overall, the head is probably the weirdest attribute. The top, sides, and bottom of the head are covered by small granular scales, similar to those of pythons and other henophidians. But a few specialized scales grace the nose and lips of Dragonsnakes. These include about 20 labial scales, a small rostral scale at the tip of the nose (impossible to see from above), two large nasal scales, directed forward, enclosing the nostrils, and several small shields in the vicinity of these nasal scales, separated by bare skin. What are all these weird scales for? Why didn't Dragonsnakes evolve more specialized scales, like the other descendants of their common ancestor with colubrids? These are open questions, but the Dragonsnake's environment probably has something to do with it. Another rare Bornean reptile, the Earless Monitor Lizard (Lanthanotus borneensis), has a similar mixture of high- and low-entropy scalation.

Ventral side of Xenodermus javanicus
Dragonsnakes' closest relatives are 16 species of obscure snakes in 4 genera that together make up the family Xenodermatidae. Xenodermatids occupy a position similar to that of pareatids, near the base of the colubroid family tree. Traditionally, both xenodermatids and pareatids were considered subfamilies of the "junk family" Colubridae, but recent phylogenetic analyses all agree that they are only distantly related to other Colubridae, and must therefore be recognized as separate families. Xenodermatids are the most distantly related colubroids, or the "sister group" to all other colubroids, having diverged nearly 50 million years ago at the beginning of the Cenozoic, four times closer to the extinction of the dinosaurs than to today. Intriguingly, one recent article found strong support for a sister-group relationship between xenodermatids and the awesomely bizarre acrochordids, or filesnakes. This small family of three freshwater and marine species is found from northern Australia and the Solomon Islands west to India and Sri Lanka.

Xenodermus javanicus
Dragonsnakes are found in southeast Asia, including southern Burma and Thailand and peninsular Malaysia, as well as on the islands of Borneo, Sumatra, and Java. Although this area was once connected, the isolation of the islands and mainland has probably resulted in geographic isolation among populations of Dragonsnakes. Although no study to date has examined either morphological or genetic differences among Dragonsnakes from different islands, I wouldn't be surprised to see them split up into multiple species once this information eventually becomes available. We know a little about the ecology of Dragonsnakes in the wild. They are active at night, when they hunt for frogs along rocky streams in tropical rain forests, as well as in rice fields. They lay several clutches of 2–4 eggs each year during the rainy season, from October to February. The young hatch after an incubation period of about two months. Apparently when you grab one, it freezes, holding its body stiff, a behavior which you can see well in this video, taken by field herpers in Indonesia.

Dragonsnake plate from Dumeril's Erpetologie Generale,
showing in detail the scale anatomy.
This plate hangs on the wall in my dining room.
Apparently Dragonsnakes are increasing in popularity in the pet trade, so we may learn more about certain aspects of their biology through the keeping of captive individuals. Hopefully collection from the wild will be kept to a minimum, and somebody one day will conduct a detailed study of Dragonsnake ecology.


Thanks to Tom Charlton, Mister Pupkin, and LOB for use of their photographs.


de Rooij, N. (1917) The Reptiles of the Indo-Australian Archipelago. Il. Ophidia. Leiden: E. J. Brill <link>

Kopstein, F. 1938. Ein beitrag zur morphologie, biologie, und ökologie von Xenodermus javanicus Reinhardt. Bulletin of the Raffles Museum 14:168-174. <link>

Pyron RA, Burbrink F, Wiens JJ, 2013. A phylogeny and revised classification of Squamata, including 4161 species of lizards and snakes. BMC Biology 13. <link>

Reinhardt, J. T. 1836. Afhandling om Xenodermus javanicus. Oversigt over det Kongelige Danske videnskabernes selskabs forhandlinger. Kjobenhavn 3: 6-7 <link>

Creative Commons License

Life is Short, but Snakes are Long by Andrew M. Durso is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.