36.The
“DON’T TREAD ON ME” Phenomenon
Paper
presented at the American Society of Ichthyologists and Herpetologists Meeting,
Boston , Massachusetts ,
June, 1972
By
Michael
S. L oop
Department
of Psychology
The
Florida State University
ABSTRACT: The evolution
of venom and venom delivery systems in snakes has been viewed chiefly as a prey
capture strategy. Consideration of the probable phylogeny of the elapids,
viperids and crotalids suggests however that venom toxicity has been decreased
as venom delivery systems have improved. This apparent paradox is resolved by
considering the viperid and crotalid delivery system maximally effective which
has allowed toxicity to be decreased thereby holding prey capture capability
constant but increasing the defense value of the venom system through sublethal
predator poisoning.
Regardless of the selective pressure precipitating the
viperid and crotalid venom systems the potential for defense through
conditioned avoidance by the predatory population is viewed as having significance
for the suborder Serpentes as a whole. It is suggested that the mechanism of
mimicry, both Batesian and Mullerian, accounts for the successful radiation of
the majority of snakes without departing from a unique and homogeneous form.
The hypothesis accounts for the apparently concurrent appearance of the
venomous families with the Miocene period of major colubrid evolution. The
hypothesis also accounts for both the behavioral and morphological warning
signals seen in many venomous and non-venomous snakes throughout the
world.
Snakes occupy a special place
in the minds and hearts of the human race; a place inhabited by them alone. The
snakes also occupy a special place in the evolutionary story of the reptiles.
They appeared and radiated into all available major ecological niches as the
mammals were overrunning the earth and supplanting the other reptiles as they
came. The success of the snakes is also unique among the vertebrates which have
assumed the elongated legless form. All others have remained restricted to the
life-style which participated the form. In addition to success despite a
traditionally disadvantageous morphology, their morphology has remained
remarkably homogeneous throughout all members despite their vast range of habitats
The suggestion will be that
these "peculiarities" of the snakes are linked together by a common
denominator, the single and simple fact that contained within the snakes are a
percentage of species which are the most noxious creatures on the face of the
earth. This situation, when viewed over the course of its development, appears
to have come about in a progressive series of stages. This paper will attempt
to outline the development and consequences of venom systems for the snakes
that possess them and the ones that don't,
Origin
The snakes represent the
extension of one lizard family which entered into a lifestyle which
participated the characteristics typical of the snakes, i.e. (1) drastic
reduction or complete absence of limbs, (2) absence of external ear openings,
(3) reduction of one lung and elongation of the other, (4) shortened tail, (5)
spectacles over eye as opposed to eyelids, and (6) regression of the pineal
eye. Some discussion remains about what lizard family adopted which lifestyle,
but the most frequently defended position is that some platynotid lizard,
probably similar to today's monitors (Varandea) underwent the required
morphological changes as the result of a burrowing lifestyle ( (Bellairs and Underwood); Walls, 1940; Brock,
1941; Dowling, 1959)
The primary interest in the
evolutionary development of the snakes, regardless of origin, has centered on
the factors contributing to their startling success. There is unanimous
agreement that the ability of snakes to ingest comparatively large prey has
been an important factor (Schmidt, 1950; Gans, 1961). While the ingestion of
large food items has advantages, more food for less effort, it also creates
some difficulties .In particular the digestive system must be geared for a feast
and famine feeding routine, and the entire snake must be geared for contending
with the fact that no animal wants to be eaten and relatively large animals are
relatively more capable of defending themselves. For some of the snakes part of
the solution to the first problem of digestion seems to have fostered the
solution to the second problem of prey capture. Presumably the ingestion of
large prey created a selective pressure favoring elaborate and copious salivary
secretion which has resulted in the wide array of oral glands found in all
snakes. The advantages are clear of delivering these preliminary digestive
enzymes into the prey’s body with elongated teeth modified to carry the
enzymes. It is simple to visualize the course of events from this starting
point. Improvements toward faster digestion were accomplished by an increase in
enzyme strength and by administration of the enzymes into the body of the prey
with specialized teeth. As these specialized teeth crept forward toward the
front of the mouth and the effects of preliminary digestion became more severe
on the prey, the initial bicarbonate status of the system changed into an
offense of unprecedented force. With fangs to administer the venom, the
selective pressure shifted to killing power and presented the world with
poisonous snakes.
If only the delivery system is
considered, a clear trend is apparent. The trend has been toward deeper and
faster venom delivery. (Slide 1) The initiation of the fang appears to be
represented by the rear fanged colubrids illustrated by skull A. In this
condition the fangs are grooved, sometimes multiple in number, and set deep in
the mouth so that the prey must be partially ingested before the fangs come
into play. The rear fanged delivery gives rise to the front fang delivery by a
shortening of the maxillary bone. The initial front fang mechanism possessed by
the elapids is represented in skull B. In this condition the fang is rigid and
short but is the most anterior tooth in the mouth. The fang is either deeply grooved
or a hollow tube. The elapid delivery allows envenomation of the prey with
brief chewing motions. The transition from the rear fanged colubrids to the
front fanged elapids represents the transition of the venom system into
primarily a prey capture apparatus. The viperids and crotalids, represented by
skull C , have developed much longer fangs by a still greater shortening of the
maxillary bone which has allowed the fangs to be folded up against the roof of
the mouth when not in use, In these snakes the fang is always a hollow tube.
This condition may well represent the ultimate system for the delivery of a fluid
into the tissue of another animal.
This apparent trend in the
venom delivery system led Bogert (1943) to the conclusion that a colubrid-like
rear fang snake gave rise to the elapids and the elapids to the viperids with
the crotalids the latest model in the viper line. Johnson (1955) argued
convincingly for vertebral characteristics as time and life-style stable within
the recognize ‘families of snakes. Johnson then turned to the
venomous families and concluded that the elapid-viperid-crotalid sequence
suggested by Bogert was correct.
The differences in venom
composition and toxicity of the various venomous snake families has not
received any evolutionary consideration beyond the decision that venom
composition would probably represent questionable characteristic for taxonomic
purposes. However, with phylogeny established by independent and osteological
characters, it is possible to return to the venom and attempt to determine any
trends. Minton (1969) has gathered together toxicity data for a wide variety of
snakes belonging to the front-fanged
families. Toxicity was represented as the amount of venom required to kill 50%
of the population of 20g mice when administered subcutaneously. It is
possible from Minton's data to calculate the average toxicity for the
front-fanged snakes in each family. (slide2) When this is done the elapids
emerge as the most toxic with the viperids a distant second, and the crotalids
lowest of all. The figure illustrates this trend in venom toxicity, expressed
as 1/L D50 micrograms,
across the developmental sequence of the front-fanged snakes (p< 0.01).
It has been the undisputed
contention that the venom system of snakes was evolved primarily as a prey
capture strategy. This contention is undoubtedly true. It is difficult,
however, to explain an abrupt decline in venom toxicity with a concurrent
improvement in delivery if offense, that is killing power, was the only objective.
It could be that the viperids
and crotalids are unable to manufacture venom of the toxicity typically found
in the elapids. This could have occurred through some mutation back toward the
less toxic enzymes that were the initial origin and has been tolerated
selectively by the better delivery. This possibility seems unlikely since
within the crotalids individual species can be found which possess venoms well
above the average toxicity for the elapids.
It could be that the overall killing
power i.e. quality x quantity, has remained constant and the principle of
"what you don't use you lose” has taken hold. Again, relying on Minton's
data, the average venom capacity for the elapids, viperids, and crotalids listed
is 109mg. 138 mg. and 92 mg. respectively. The average quantity of venom
possessed by the families does not conform to the required pattern if killing
power was held constant.
It could be that there are
different patterns of food preferences between the three families and that venom
changes have been directed at manufacturing more appropriate toxins for the
prey. There are in fact differences. The elapids contain a relatively high
percentage of members which feed upon other reptiles, while the viperids and
crotalids tend toward mammalian prey. It should be noted, however, that the
toxicity measures were taken against a mammal so that the trend in toxicity has
been away from the chief prey animals.
It could be that the tissue
destruction which frequently accompanies viperid and crotalid bites contributes
importantly to prey digestion. However, when these snakes are deprived of their
ability to inject venom they feed and grow normally. Furthermore, many
nonvenomous species share identical diets with their venomous neighbors and are
at no apparent digestive disadvantage.
While this apparent decline in
toxicity with the appearance of improved delivery marks the most blatant turn
toward lowered killing power, there is another trend which indicates a move in
the same direction within the lifetime of an individual snake. (slide3) Minton
(1967) studied venom toxicity as a function of age for two species of crotalids
and one elapid. He observed that toxicity increased from birth through the
first 6 to 9 months of life, reaching peak toxicity 3 times that of the
adult snakes for Crotalus and Naja. The relatively constant toxicity for
Agkistrodon was attributed to "the rather poor conditions of the juveniles
between their fourth and ninth months." Minton felt that the toxicity
increase might reflect changed feeding habits. This could well be the case
since most species of snakes are born in early Fall with activity presumably
reduced during the next 6 to
9 months at which time, Spring and
Summer, the food supply materializes. The real question, however, is not why toxicity
increases but why toxicity declines as the snake grows larger and presumably
more capable of delivering his venom.
A venomous snake has one goal
with respect to its prey. That goal is immediate incapacitation which is
achieved through injection of a sufficient quantity of venom. If a snake
possesses a modest delivery system, as with the elapids, or is small in size,
as with young individuals, toxicity must remain high to insure efficient prey
capture. As delivery is improved either through morphological adaptations or
increased size of the individual, the toxicity of the venom need not remain as
high to achieve the desired result.
A venomous snake has one goal
with respect to its predators. That goal is to be left alone. There are in
principal two ways to achieve this end. One is to adjust the reproductive
probabilities of the predator; the other is to adjust the behavioral
probabilities of the predators. While these two paths have the same end, the
means are very different. It seems that the venomous snakes have been presented
with a clear choice point. The simplest means to adjust the reproductive
probabilities of the predatory population is to kill those individuals given to
attacking snakes thereby creating a selective pressure favoring those individuals
which do not. This route would have as its outcome an innate avoidance. The
other possibility is to adjust the behavioral tendencies of the individual
predators by administering a noxious or punishing stimulus, thereby reducing
the probability of future attack on an individual basis. While the innate
avoidance would be beneficial in the long run, natural selection gazes only at
the moment. "In order to make it clear how, as I believe, natural selection
acted, I must beg permission to give one imaginary illustration “(Darwin 1859). Consider
two venomous snakes, A and B. Snake A possesses a highly toxic venom and snake B a less toxic venom. Both
have equal deliveries. In dealing with their prey A and B both kill within a criterion time period. When dealing
with the respective predators, however, A kills them all while B kills 15%,
leaving 85% sick but surviving. Snake A has succeeded in opening a number of
slots for the survival of new predators or the migration in of neighboring
ones. Snake B, on the other hand, has left the predator population by and large
intact, but now possessing individuals with rather bitter memories. Snake A
will continue to face roughly as many aggressors as before, but snake B now has
a percentage of the predator population avoiding him. Assuming only the
territorial tendencies of the organisms involved and the capability of a
predator to form an association between a previously neutral stimulus, the
snake, and an inherently noxious event, the venom, it is suggested that snake B
is at an immediate advantage over snake A. The proposal I wish to make is that
the decline in venom toxicity seen both across the families of venomous snakes
and within the lifetime of the individual is a move toward lowering the
probability of killing a predator during defensive biting since in principle it
would appear more effective to address a predator’s CNS than his DNA.
Regardless of the selective
pressures which have participated the toxicity declines mentioned, the outcome
will still be a lowered incidence of predators killed or rather an increasing
incidence of poisoned survivors. The only criterion for the venomous snakes to
capitalize upon this situation is the acquisition of some warning stimuli
either morphologically or behaviorally. This criterion appears to have been met
chiefly through behavioral signals, i.e. the animal does something during
defense interactions which renders him more detectable. Many examples are
available. (Slides 4, 5, 6, 7) These are by no means exhaustive but represent some
of the more notorious characters.
The mechanism of Batesian
mimicry whereby one harmless organism gains a selective advantage by
approximating the appearance of another noxious animal has been documented in
many instances, primarily among insects. In Batesian mimicry the noxious animal
is termed the model, the harmless animal the mimic, and the similarity in
appearance between them the signal. It has been demonstrated that unpalatable
stings, chemical sprays and a host of other stimuli will function as
sufficiently effective deterrents to a predator to allow both model and mimic
to enjoy reduced predation.
Consideration of both
laboratory and field data indicates that the most important variable in a
Batesian mimicry complex is the intensity of the noxious stimulus. Duncan and
Sheppard (1965) explicitly demonstrated that if a mild and strong noxious event
is associated with a particular signal, animals receiving the strong noxious
stimulus will generalize their avoidance to a much wider range of similar stimuli.
They translated this finding into the next SL IDE.
These data, when translated into Batesian mimicry, indicate that if a model is
mildly noxious then natural selection will insist upon a good signal match by the
mimic. On the other hand, if a model is very noxious, then a mimic may achieve
protection with only a modest approximation of the model signal.
Johnson (1956) suggested that
the vertebral characters indicated that the elapids probably represent a more
primitive family than today's nonvenomous Colubridae family, with the elapids
springing from some pre-Colubrid. While the fossil record is scarce for snakes
in general, Johnson noted that elapid fossils appeared slightly before the
fossils of today's family Colubridae in the early Miocene. Tihen, at these
meetings last year, suggested that the fossil record indicated that for all
intents and purposes the appearance of most of our modern families was
simultaneous in the Miocene. Johnson posed an interesting problem at the
conclusion of his proposed phylogeny of the venomous families. He stated,
"It is possible that venom and the venom apparatus were developed when
competition for food was intense. As this condition was alleviated, the
nonvenomous Colubrids had the opportunity to undergo their major radiation
without the advantage of venom. Unless some such postulate of lessening
competition between venomous and nonvenomous snakes is made, it seems difficult
to understand why the venomous snakes have not completely dominated our present
herpetofauna."
The proposition I wish to set
forth is that the sudden expansion of the nonvenomous colubrids was the result
of the appearance of venomous snakes. With the advent of these extremely
noxious animals, apparently bent upon keeping the predatory population alive,
the elongated legless form of the snake became a signal paired with an
exceedingly noxious stimulus.
While the possibility exists
that the snake form per se could mark the outer limits of a generalized
avoidance, such a suggestion is insulting to the intelligence, i.e.
discrimination ability, of the signal receivers. In all regions of the world
save Australia
the nonvenomous species out number the venomous which implies that by chance
alone a significant proportion of predators will encounter nonvenomous snakes
first. Furthermore, the nonvenomous snakes represent comparatively defenseless
prey for their size. For the most part they are slow in escaping and incapable
of inflicting a bite of any consequence. A one pound rat represents a much more
favorable prey item than a one pound nonvenomous snake. These considerations
strongly suggest that predatory signal receivers should have both opportunity
and motivation to discriminate nonvenomous snakes from their venomous
neighbors. Therefore the expectation would be that nonvenomous species should
approximate, to some degree, the signal(s) of sympatric venomous species.
(a) Southeastern
United States -The Southeastern United States represents a
convenient "test area" since three genera from two families of
venomous snakes are found there. The three genera will be addressed
individually in an attempt to assess the signal(s) available to a potential
predator. The remaining genera of snakes will then be reviewed to determine the
extent to which these signals are approximated by nonvenomous species.
(1) The Moccasins (Agkistrodon,
Family Crotalidae) - The moccasins are represented in two species, the
cottonmouth (A. piscivorus) and the copperhead (A. contortrix). The cottonmouth
inhabits the Gulf Coastal States and the copperhead has its stronghold in the
interior regions. The two snakes have very similar patterns at birth. The
copperhead retains its juvenile pattern while the cottonmouth becomes dark,
almost black, by its second year. When alarmed both species rapidly vibrate
their tails and the cottonmouth exposes the white interior of its mouth (hence
the name).The classic distinguishing characteristics are the facial pit , thick
body, triangular shaped head, vertical pupils, and single row of scales under tail
(Conant, 1958). Of these characteristics only four seem useful signals, i.e. tail
vibration, exposure of mouth interior, thick body, and triangular shaped head.
The other characteristics would necessitate unduly close inspection.
(2) The rattlesnakes (Crotalus,
Family Crotalidae)- Three species of rattlesnakes are found in the Southeastern United States . They differ markedly in
pattern and size. The uniting character and unquestionable signal is the rattle
borne on the tail. The rattlesnakes share all the morphological and behavioral
characteristics of the moccasins but greatly accentuate the audibility of the
tail shaking with the rattle. Frequently they hiss when aroused.
(3) The coral snake (Mircrurus,
Family Elapidae)- One species is found in the S.E. United States and possesses
the triad ringed pattern(black-yellow-red). The coloration of the coral snake
has been taken to be warning in its function (see Mertensian mimicry).
If attention is turned to the
twelve open air (nonburrowing) genera of snakes found in the S.E. United
States, the following defensive behaviors have been noted by Conant (1958):
Flattening of the body 33% (Natrix, Stoseria, Thamnophis, Heterodon); hissing
50% (Heterodon, Drymarchon, Elaphe, Pituophis, L ampropeltis,
Stilosoma); tail rattling 58% (Coluber, Masticophis, Drymarchon, Elaphe,
Pituophis, L ampropeltis, Stilosoma).
The only remaining genera is Cemophora and this snake possesses triad
red-yellow-white banding, like the coral snake
Extensive personal field
observation has also shown that a number of genera (Natrix,Thamnophis ,
Heterodon, Coluber, Masticophis, Elaphe) extend horizontally the posterior ends
of the mandibles when assuming a defensive posture. This behavior is not seen
when the snake is on the offense against a prey item. This behavior has the
result of giving the head a decidedly triangular shape
(b) Worldwide - Similar
situations appear to be the case in other parts of the world.
The South
Seas : "The sea snakes are naturally fish eaters with a strong
predilection for eels, curiously enough for the banded eels of tropical seas,
which many of the sea snakes greatly resemble.” (Schmidt 1950) In all
probability it is the eels which resemble the sea snakes for reasons now
obvious.
SL IDE
12
3. One signal receiver's
opinion: man
Section I revealed that in
general man was appreciative of the signal match between models and mimics in a
variety of situations. Human opinions, however, were of no use in evaluating
proposed mimicry systems since these were neutral with respect to Homo sapiens.
Neither model nor mimic played any role in human affairs and vice versa.
Poisonous snakes, on the other
hand, do interact with man on occasion. In fact snakes kill more people every
year than all other non-human vertebrates combined, about 40,000, This figure
represents approximately a 15% death rate which moves the total venomous bites
to at least 300,000. These estimates are probably too low with the total number
of venomous bites running possibly up to one million annually (Milton ,1969). Therefore it
would seem reasonable to allow Homo sapiens' opinion to count as evidence for
or against the proposition under consideration.
People don't like snakes.
Minton (1969) puts it, "The primates that were to become men were
indulging in a unique sort of cerebral activity, They were taking images that
came into their brains by way of perfectly good mammalian sense organs,
coloring them with emotions, and projecting them somewhat distorted upon a
screen of inner consciousness,...The snake was one of these images,., the snake
remains today the one animal that man universally respects and fears, covertly
loves, and intensely hates." Minton however also suggests, "There is
no good biological reason why this should be so. "
Many propositions have been
suggested to account for man's attitude toward snakes. The kindest thing which
can be said about most of these proposals is that they are imaginative. It
rather seems, however, that the phenomena of snake avoidance may be accounted
for without mention of inner consciousness screens, or phylaic symbols, or
colored images. People avoid snakes for the same reason that toads avoid yellow
and black buzzing insects, and for the same reason that birds avoid butterflies
of a particular color pattern, i.e. an indeterminante percentage of all these creatures
are best left alone. Man is simply a signal receiver of the snakes' mimicry efforts.
It seems likely that if Homo sapiens is generally reluctant to bet his hand or
his life on proper identification, the rest of earth's creatures may be equally
reticent. And from the snakes' point of view nothing could be more useful than
to be left alone by these creatures too large to be eaten, at that moment.
SL IDES
1.
Skulls Opisthoglyphs, Proteroglyphs, Solenoglyphs
2. Venom toxicity as a
function of family
3. Venom toxicity as a
function of age (Milton ,
1967)
4. Cobra
5. Eastern diamondback
rattlesnake
6. Cottonmouth, open
mouth display
7. Coral snake
8. Selective advantage
as a function of similarity for two noxious models
(Duncan & Sheppard,
1965 Fig. 5)
9. Hognose snake
display
10. Cottonmouth
11. Banded water snake
12. Diamondback water snake head
References
Bellairs, A. A. and Underwood, G. "The Orgin of
Snakes." Biological Review (1951): 193-237.
Bogert, C. M. "Detitional phenomena in cobras and other
elapids with notes on adaptive modification of fags." Bulletin of the American Museum of Natural History (1943):
8K3, 285-60.
Brocks, G. T. "The skull of Acontias meleagrie with a
study of the affinities between lizards and snaks." Zoological Journal
of the L innean Society (1941):
71-88.
Conant, R. A field guide to reptiles and amphibians. Boston : Houghton Mifflin
Co., 1958.
Dowling, H. G. "Classification of teh serpents: A
critical review." Copeia (1959): 1: 38-52.
Dowling, H. G., S. A. Minton and F. E. Russell. Poisonous
Snakes of the World. Washington D.C. : United
States Government Printing Office, 1965.
Duncan, C. J. and P. M. Sheppard. "Sensory
discrimination and its role in the evolution of Batesian mimicry." Behavior
(1965): 24: 269-282.
Gans, C. "The feeding mechanisms of snakes and its
possible evolution." American Zoologist (1961): 1: 217-227.
Johnson, R. G. "The adaptive and phylogenetic
significance of vertebral form in snakes." Evolution (1955): IX:
367-388.
Johnson, R. G. "The orgin and evolution of the venomous
snakes." Evolution (1956): 10: 56-65.
Minton, S. A. Venoms from juvenile snakes. In. F. E. Russel
& P.R. Saunders (Eds.) Animal Toxins, Oxford : Pergamon Press, 1967
Minton, S. A. and M. R. Minton. Venomous Reptiles. New York : Charles
Scribner's Sons, 1969.
Schmidt, K. P. "Modes of evolution discernible in the
taxonomy of snakes." Evolution (1950): 4: 79-86.
Walls, G. L .
"Opthalmological implications for the early history of snakes." Copeia
(1940): 1-8.