CHAPTER 14

AMPHIBIANS AND REPTILES

W. FRANK BLAIR

Relatively adequate knowledge of communication in the Amphibia and Reptilia is limited to one or two subgroups of each class, and the ensuing discussion will serve in part only to emphasize our lack of knowledge about major subgroups. Aside from this, there is evidence to indicate that evolution has taken very different pathways in producing communication signals in subgroups of the two classes. In general, one particular modality is stressed in each (e.g., vocal signals in anuran amphibians) with, however, evolutionarily interesting departures from this norm and with interesting supplementation in most instances.

The two classes of vertebrates discussed here are structurally, physiologically, and ecologically very different from one another, although they share the conditions of being in general relatively small, ectothermal, terrestrial tetrapods, so it is not surprising that their communication systems are quite different.

Communication in amphibians and reptiles can be categorized rather simply in respect to function:

1. Conveyance of information that an individual of the species occupies a particular segment of space. This information has significance both intraspecifically and interspecifically in respect to dispersion of the population and to use of environmental resources. If the signal indicates conspecificity, the individuals may avoid one another and thus spread the pressure on environmental resources such as food. If the signal provides information of nonconspecificity (or probably more frequently, fails to indicate conspecificity), then individuals of species with different environmental requirements (as different foods) may coexploit the same terrain without detrimental interindividual interaction.

2. Conveyance of information that the individual belongs to a particular species in relation to reproductive activities. Systems of ethological isolating mechanisms permit the coexistence of interfertile populations as separate biological species. Here there is pressure for preservation and for improvement of the specificity of species recognition signals if the hybrids that result from inadequately specific signals are at a disadvantage in competition with the parental species.

3. Conveyance of information as to the sex of the individual. The utility is obvious here in terms of conservation of energy that might be wasted in attempted homosexual matings, possible wastage of gametes, or potential exposure to predators during unprofitable reproductive activity.

4. Conveyance of information about physiological readiness of the individual to mate. The advantages are essentially those mentioned under (3).

The kinds of communication categorized above are largely operative within the species population or in connection with identifying its members. Another sort of communication, and one that I will not discuss at length, involves interspecific communication and may involve either correct information or misinformation that is advantageous to the signaler. The rattling of a rattlesnake is exemplary of the first in that it provides positive identification of an animal that is well protected by its venom apparatus. The second is exemplified by the remarkable leptodactylid frog Pleurodema bibroni, mentioned by Cei (1962), in which the combination of “eyespots” and posture make it look as though the frog were headed in the direction opposite to the one in which it is actually prepared to jump (Fig. 1).

The following discussion follows a systematic arrangement of the two classes of vertebrates, with emphasis on evolutionary trends in each.

AMPHIBIA: APODA

The caecilians (order Apoda) comprise a circumtropical group in excess of a hundred species of fossorial, worm-like amphibians. The life history and ecological relations are more poorly known than for any other group of amphibians, and conclusions about communication channels are mainly inferential. Eyes are vestigial in caecilians, so it may be assumed that visual cues are not used. There is no evidence of vocalization. By this process of elimination, tactile and chemical signals are suggested as likely possibilities. This is supported by recent observations by Gans (1961) on a captive Siphonops annulatus from Colombia. He noted that the head is “extremely sensitive to touch and minor vibrations.” He also noted that his animal would follow objects rubbed with earthworm but was uncertain as to whether the chemoreception was by way of the nostril or by way of the peculiar “tentacle” located between eye and nostril.

AMPHIBIA: CAUDATA

The fewer than three hundred species of salamanders that comprise the order Caudata show ecological diversification mainly in that some remain permanently larval (hence permanently aquatic), whereas the majority metamorphose from an aquatic to a terrestrial stage. Nevertheless, the channels of communication fit a general pattern that contrasts strongly with that of the anurans. Communication involves combinations of chemical, tactile, and visual signals and in all probability makes little if any use of vocal signals. The weak sounds emitted by various kinds of salamanders and propagated in various ways have been discussed by various authors (Maslin, 1950; Bogert, 1960; Blair, 1964a). None of these sounds has been related to behavior unequivocally, and a fair statement of our present state of knowledge is that vocalization is insignificant in the caudate amphibians generally but may have some significance in a few of them.

There is convincing evidence, although much of it is indirect, that chemical communication is of paramount importance in salamanders. Twitty (1959, p. 7), whose work with newts (Taricha) provides the most impressive evidence, states the case succinctly by writing, “In animals as earth-bound as salamanders, living throughout their terrestrial phase in such intimate association with humid soil and all its products, I suppose it is not unlikely that the chemistry of their environment looms larger in their experience than do impressions of sight and sound.” It might be added that the relatively moist skin of all amphibians by comparison with other tetrapods should facilitate close responsiveness to chemical compounds in the ambient environment. Twitty (1961), in a massive field experiment, has demonstrated the ability of displaced and blinded newts to return to the section of the stream in which they breed and the inability of individuals that had undergone surgical section or removal of the olfactory nerves to do so (Twitty et al., 1964).

The significance of chemical signals in population dispersion of newts has not been established. The potency of chemical stimuli in attracting males to reproductively ripe females has been demonstrated impressively by Twitty (1955), who attracted males to cellulose sponges that had been soaked in water in which females had been kept. Davis and Twitty (1964) noted that there were only minor differences in the courtship behavior of the species of Taricha, and they suggested that chemical sex attractants secreted by the females may be involved in species recognition and in the elicitation of a species-specific courtship. They also noted that there “is some indication that reproductive barriers (species-specific sex attractants?) are somewhat less highly developed between allopatric forms” (p. 608), thus suggesting reinforcement of the chemical-isolating mechanism under sympatry.

Although chemical signals appear to be the most important communication media in the caudate amphibians, other avenues are used as well. Smith (1954, p. 38), writing of European newts (Triturus), recognizes three probable factors in sex recognition:

1. External appearance (there is strong sexual dimorphism which is accentuated during the reproductive season). The author also attributes specific recognition qualities to the male patterns by saying, “The marked difference in appearance between the males of the palmate and smooth newts would be recognized at once by the females, just as they are by us.”

2. The smell of the hedonic (sexually stimulating) scent glands that are distributed over the cheeks, body, cloaca, and tail in both sexes.

3. The response of the female (by remaining perfectly still and not moving off, she informs him that he is to continue the courtship).

Tinbergen (1953) sees the courtship of various newts as a series of signals: (1) visual (involving posturing by the male), (2) tactile (involving a sudden leap by the male that sends a strong water current to the female, which often pushes her aside), and (3) chemical (involving tail-waving by the male which probably carries a chemical stimulant toward the female).

Noble (1931) has attempted to trace the evolution of reproductive behavior through the extant families of caudate amphibians. In the primitive Hynobiidae (Asiatic land salamanders) and their specialized derivatives the Cryptobranchidae (hellbender of United States and Asiatic giant salamander), fertilization is external and sight of the extruded eggs is believed to provide the stimulus for the emission of sperm. In most other salamander groups fertilization is known to be external, with the male depositing a spermatophore which is then picked up by the female, and complex behavior patterns have evolved. In the Ambystomatidae, reproductive behavior is initiated by a Liebesspiel involving both tactile and chemical stimuli, and the sequence also involves posturing leading ultimately to deposition of a spermatophore and subsequent picking up of it by the female. There are speciesspecific differences in the Liebesspiel. Some species (e.g., Ambystoma tigrinum, A. opacum, A. maculatum) are said to show no sex discrimination at the beginning of courtship (Kumpf, 1934).

Noble (1931, p. 389) argued that the courtship of the other families of salamanders is built from the ambystomatid pattern through elaboration of one or more phases of it. The Plethodontidae “seem to have specialized in hedonic glands as the source of stimulation.”

In sum, all but the most primitive of living salamanders communicate species identity, sex identity, and physiological readiness to mate mainly through chemical, tactile, and visual signals and probably in this descending order of importance. Relative to most anuran amphibians, the series of signals involved in species and mate recognition by salamanders appears to be more complex and varied and presumably would provide a more effective bar to interspecific mistakes.

AMPHIBIA: ANURA

Communication among anuran amphibians, particularly with reference to vocal signals and their significance, has been reviewed exhaustively by Bogert (1960). My discussion will summarize main aspects of our present knowledge and mention in detail only a few significant contributions since Bogert’s summary. In general, there is much more firm information about communication in frogs than in salamanders, but even here there are large gaps in our knowledge. The greatest gap exists in respect to knowledge of communication mechanisms associated with dispersion of individuals in the population. This refers not to aggregation in respect to the reproductive phase of the population life history, which will be discussed below, but to the nonreproductive phase or to dispersion of kinds that do not aggregate to breed. Enforced spacing of individuals (territoriality) has been reported for only a few species and seems most likely in groups which spend the reproductive and nonreproductive segments of their lives in essentially the same terrain, e.g., largely aquatic frogs (Rana of various species), terrestrial breeders (family Leptodactylidae), or possibly tree frogs (Hylidae) of the tropical rain forest that breed in small catchments of water in bromeliads, tree holes, and the like.

The work of Martof (1953) with the green frog (Rana clamitans) is usually cited as illustrating organization among males of a relatively aquatic frog. Since the green frog vocalizes extensively and over a long period (mid-May to mid-August in the area studied), it may be that the call is a vocal signal “proclaiming ownership,” but it is possible that, at close range, visual cues are also involved as the frogs call, floating on the surface with body inflated (Bogert, 1960). Observations of the bullfrog (Rana catesbeiana) by A. P. Blair (1963, p. 151) are suggestive of organization of the males of this species. Combat between two males was noted: “They butted, wrestled, shouldered each other around, splashed a great deal, but there was no suggestion of any attempt at amplexus. They called occasionally during the encounter which lasted some 10-15 minutes.” This author also reported that “A very common procedure is for male to call, hop a few feet along the edge of the pond, then stop and call once or several times, move again, etc. However, when a moving male approached another, the stationary male usually moved away” (ibid.). It seems likely that both vocal and visual cues are involved. It is pertinent that, like the green frog, the bullfrog tends to vocalize throughout the entire warm season of the year.

Capranica (1965), working with evoked vocal responses in a laboratory colony of the bullfrog, recognized both “mating” and territorial calls. One call, emitted only by males, evoked response calling by males when broadcast either from taped natural calls or from synthesized calls. Attraction of females to the call was not demonstrated, and the behavioral reasons for the answering by other males are obscure. He recognized three territorial calls, which seemed clearly related to organization of the colony. He found that many of the animals tended to maintain exclusive locations and that “When another bullfrog of that colony chanced to enter this location, a distinct call was often uttered by the proprietary frog (commonly followed by an attack upon the intruder” (p. 91). The three kinds of territorial calls were analyzed spectrographically. One, emitted only by males, is given with the vocal sacs partially inflated. Another is emitted only by the females, and the third by both sexes. The latter is “often given by the proprietary frog immediately after the successful banishment of an intruder” (p. 93). In the absence of evidence for its attractiveness to females and in view of the behavioral responses of other males to it (i.e., answering), the possibility strongly exists that the “mating” call described by Capranica is really a territorial call. In fact, it has not been shown that Rana of any species has a mating call in the sense that the call attracts a reproductively ready female.

Bogert (1960) has suggested a territorial function for the call of the aquatic pig frog (Rana gnjlio). The presumed territorial call differs from the presumed mating call in consisting of a longer series of grunts with, however, the same basic structure.

Three calls have been described for the Pacific tree frog (Hyla regilla) by Snyder and Jameson (1965). One of these is interpreted as a territorial signal by them. It “occurs in natural populations when a calling male recognizes the approach of either a calling or noncalling male, at which time both males give a prolonged series of staccato notes. This dialogue continues until either the invading male has retreated or the resident male accommodates to the invader’s presence” (p. 131). These workers were able to elicit the call in response to taped calls from hidden speakers.

In at least one species of anurans, the communication of ownership of territory seems to be mostly by visual cues. The small South American dendrobatid frog Prostherapis (= Phyllobates) trinitatis has been studied by Test (1954) and Sexton (1960). Small territories are defended by males, females, and subadults, with the females showing the greatest territorial defense. As stated by Sexton (p. 108):

The interaction between resident and intruder follows a definite sequence of events, although some of the steps may be omitted. When the resident first observes the intruder, it orients to face the intruder. The resident’s bright yellow throat is fully visible, and the resident pulsates the throat slowly as a challenge. Frequently this challenge is sufficient to deter the outsider, and the intruder will move away. If this action is not sufficient, the resident may hop in the direction of the intruder and repeat the challenge. If this stimulus is still insufficient to cause the withdrawal of the invader, the resident will hop upon the back of its rival. Usually, this is the end of the encounter, and the intruder will leave the defended area.

Certain other South American frogs usually classified in the Leptodactylidae (Ceratophrys, Chacophrys, and Lepidobatrachus) are remarkable among anurans in defending themselves by biting; they also emit an aggressive call (Barrio, 1963, p. 143). Barrio describes the behavior:

Nos referimos al compartamiento agresivo puesto de manifiesto cuando son excitados, y que consiste en la inflatión máxima de los pulmones seguida de rápidos movimientos al frente o a los costados, con la boca abierta, intentando morder ye emitiendo simultáneamente un sonido que podemos denominar “grito agresivo.”

The aggressive cry is emitted by both males and females. Barrio has illustrated the cry of a Ceratophrys ornata female and of a Lepidobatrachus llanensis male with sound spectrograms for comparison with the mating calls of the males of the respective species. The “unvoiced” quality of the aggressive call is probably related to the emission of this call with the mouth open. The effect of this call on the normal enemies of these frogs or on others of the same species has not been reported.

Another kind of territorial defense is found in the large Brazilian tree frog Hyla faber (Lutz, 1960a, b) in which the males build nest pans in shallow water and defend them against other males. Descriptions of this behavior by Lutz do not mention cues that might identify the sex and thus prevent active combat, and they do emphasize the viciousness of the combat which may even result in the death of one combatant.

Actually, there is little definite information about organization of anuran populations, although there are some grounds for speculation. Jameson (1955) working with the cliff frog (Syrrhophus marnocki), a terrestrial-breeding leptodactylid, noted two types of calls, one of which is heard throughout the year. Bogert (1960, p. 188) has surmised that “the call heard throughout the year serves as an ‘advertisement of ownership’ and qualifies as a territorial call.” Many arboreal frogs (Hylidae, Leptodactylidae) are known to vocalize throughout a much longer period than the presumed breeding season. These socalled “tree calls” or “rain calls” have yet to be shown to be of any significance in spacing of the population, although such significance seems highly possible. These calls may differ from the mating call in some aspects (Fig. 2); however, there is no evidence contrary to the possibility that they reflect incipient sexual arousal of the males as atmospheric moisture increases, since such atmospheric changes do normally precede breeding activities.

The basic pattern of reproductive behavior of anurans involves aggregation of the population in the breeding pool, hence it involves movement of the individual from the familiar terrain in which it “makes its living” to the breeding site. The great majority of anurans, particularly those that move from a distance to the breeding pool, vocalize in connection with their reproductive activities. It seems likely that the development of a breeding chorus of males involves homing in of additional males on the calls of the first males to reach the site. These first males may have arrived fortuitously, or they may have lived nearby and known the terrain. It might be argued that the call of any species would serve this function so long as it lay within the auditory range of the recipient individual, but one would anticipate species specificity in the response since it might be presumed that the initial callers would begin calling from a situation ecologically acceptable as a breeding site for the species in question. Frequent observations of aggregations by species support the idea of species specificity, although they do not rule out possibility of orientation to calls of other species. For example, in southwestern Utah I found one species of toad, Bufo punctatus, calling in the water on one side of an impoundment and another species, B. microscaphus, calling in the water on the opposite side. Out of several hundred individuals of both species, one B. microscaphus had found its way into the B. punctatus chorus.

Although the vocalization of “first-arrived” males may communicate location of the breeding pool to other males, as well as to females, there is little published experimental work on this subject. Bogert (1947) regarded the presence of a breeding chorus as influential in permitting homing of displaced Bufo terrestris. Bogert (1958) demonstrated the orientation of released B. terrestris to an outdoor speaker broadcasting a recorded chorus after these animals had been removed from a breeding congress. Awbrey (1965) has provided the only experimental evidence that the calling of a single male can lead to formation of a vocalizing congress. Conditions were favorable for this experiment because the breeding season was delayed by lack of rain. Broadcasts of the tape-recorded call of a single B. valliceps male on three of five nights, with no broadcast on the two alternate nights, resulted in the appearance and vocalization of other males and the appearance of females in significant numbers on the broadcast nights, but no spawning occurred.

Some species in various taxa of anurans have lost the mating call or have had it sharply reduced. These obviously must use other methods of finding the breeding site. Chemical cues may be important but remain to be demonstrated. Savage (1961) working with the common European frog, Rana temporaria, believes that odors of the algae at the breeding site provide cues for this species. Evidence for celestial navigation by displaced cricket frogs (Acris) has been presented by Ferguson et al. (1965) and for Bufo terrestris by Ferguson and Landreth (1966). Attachment to an area rather than attraction by some such obvious cue as vocal note or algal odor is believed by Heusser (1960a) to be the case for the weakly vocalizing European toad Bufo bufo. He believes that the toads search for a specific geographical spot, and he cites the fact that, “after a spawning ground had been covered by a newly constructed road, toads went on migrâting towards it for two consecutive years and even distinguished the original spawning places within the dried up area” (p. 109).

Identification of the anuran individual as to species is most commonly conveyed by the characteristic mating call. Objective analysis of the characteristics of the mating call dates from the availability of the sound spectrograph in the early 1950s, and descriptions of the call are now available for most North and Middle American species and for others from many parts of the world. The extensive literature has been reviewed by several authors, particularly Bogert (1960), Mecham (1961), and W. F. Blair (1958, 1964a).

The significance of species identification through vocalizations is no longer debatable, although one might think otherwise from the remarkably superficial and biased treatments by Savage (1961) and Wynne-Edwards (1962), both of whom have quoted archaic accounts in an apparent effort to support their beliefs and have ignored the great mass of evidence that has become available in recent years. The greatest bulk of the evidence remains indirect. Sympatric species always differ in one or (usually) more important attributes of the call, although disjunct, allopatric species may show little or no differentiation. Loss of mating call tends to be related to geographical distributions where, because of absence or limitation of related species, the possibility of mismating is absent or reduced (W. F. Blair, 1958). “Cryptic” species have been revealed among anurans since quantification of the physical attributes of the mating call became possible, and it is predictable that additional ones will be found. In these, there is striking differentiation in call, indicating that evolutionary forces have acted on the mating call in the respective populations, although differentiation in the kinds of morphological characters used by taxonomists has been so slight that the separate populations have not received taxonomic recognition. Recent examples have been reported by Johnson (1963) in the Hyla versicolor complex of North American tree frogs and by Barrio (1964a) in the South American leptodactylid genus Pleurodema. Last, there is the evidence from comparison of mating calls within and without the zone of overlap of two closely related species that natural selection under limited hybridization has led to reinforcement of call differences in the zone of sympatry. This evidence has been reviewed by W. F. Blair (1964b), and Littlejohn (1965) has analyzed a significant example in the Hyla exvingi complex of southeastern Australia. Reinforcement of call differences is evident in the zone of sympatry between two species on the mainland, but insular populations of one of the species which have been isolated for about 9000 years and which have been evolutionarily unchallenged by the need for species identification, since they are without closely related congenors on their respective islands, have undergone no significant differentiation in this attribute.

Conclusive evidence that the characters of the call provide information as to species comes from laboratory experimentation in which tape-recorded calls of the correct and of closely related species of males are broadcast to sexually ripe females in a two-choice situation. The abilities of females to discriminate under such conditions have been demonstrated in Pseudacris (Littlejohn and Michand, 1959; Blair and Littlejohn, 1960) and in Hyla (Littlejohn, Fouquette, and Johnson, 1960). Awbrey (1965) used both natural and synthetic calls in testing several anurans, including species of Bufo and Gastrophryne (= Microhyla) in both choice and no-choice situations, and his results indicate that these also used characters of the call as a basis for species identification.

When an anuran approaches a member of its species in a breeding pool, assuming that vocal signals have already communicated species identification, there remain questions of identification that are of importance to the population. What is the sex of the individual approached? What is the reproductive state of the female that is clasped by a male? Wastage of energy and gametes in homosexual (male-tomale) matings or in amplexus of males with spent females is obviously disadvantageous for the population, so it is not unusual that potent signal-response systems have evolved pertinent to these identifications. In many instances, the female is attracted to the calling male, and it is she who initiates the next stage leading to amplexus by making physical contact with the male, who then clasps her.

In large breeding congresses, however, when males are in a high state of sexual excitement, the males of many species may move about attempting to clasp any object of appropriate size. It is here that “release” signals are most important. The considerable literature has been reviewed by Bogert (1960). There are two main classes of cues. One is the release vibration, involving vibrations of the body wall and often elements of the mating call, produced by either males that have been clasped by other males or by nonresponsive females that have been clasped. Even the most highly excited males will usually release the clasped animal in response to this signal. The other class of cue involves girth of the animal. Reproductively ready females are of relatively considerable girth because of the enlarged ovaries. Nonreproductive females, usually recently oviposited ones, of many species will deflate the lungs, drawing in the sides so that the girth is greatly reduced, and this signal is highly effective in securing release by the clasping male. In the laboratory I have seen in Bufo valliceps and in B. granulosus another apparent signal, one involving high inflation of the lungs so that the body wall is essentially nonyielding by comparison with that of a receptive female. This will also secure release by the male that is attempting to clasp. Nonreceptive females that do not effect their release by any of these signals may turn over on their backs or attempt to scrape the male off or to kick him off with their hind legs. This behavior has been well described and illustrated for B. bufo by Heusser (1960b).

In terms of evolutionary pressures, the release call would be expected to be under no functional pressure to differentiate in sympatric species and would differentiate only inasmuch as it had the same mechanisms of production as the mating call, which is under such pressure. Conservation of energy and of gametes is obviously important irrespective of whether the “mistake” involves a homospecific or heterospecific mate. Such an example of essentially no differentiation in release call but marked differentiation in mating call in two species of Odontophrynus (leptodactylid frogs) has been reported by Barrio (1964b).

REPTILIA: TESTUDINATA

Although turtles live on land, in fresh water, and in the oceans, the basic body plan is the same; it is one that emphasizes external armor at the expense of agility. This successful adaptation to the exigencies of life is seemingly influential on the kinds of social organization and intercommunication systems that have evolved in chelonian populations. Much of the background literature pertinent to this discussion has been listed by Evans (1961) and need not be relisted here. Visual cues seemingly outweigh all others in chelonian communication, particularly among the terrestrial turtles, but chemical and vocal signals are also employed.

There is a dearth of evidence that populations of turtles are organized, with individuals holding territory against others of the same sex, and in fact for a few thoroughly studied species there is strong evidence that such is not the case. This may relate to the dependence on armor; for unless the individual can punish the invader, there is no good mechanism for excluding him. In the apparent absence of territorial organization it is not unexpected, therefore, that no signals communicating ownership of territory can be discerned. Examples of well-studied species that support this generalization are a box turtle, Terrapene Carolina (Stickel, 1950), the desert tortoise (Gopherus agassizi) (Woodbury and Hardy, 1948), and slider turtle (Pseudemys scripta) (Cagle, 1950).

Vocalization has been reported in association with reproductive activities of various turtles, but its significance appears rather obscure. The male wood turtle (Clemmys insculpta) “is known to make a distinct yet subdued note not unlike that of a tea-kettle and audible at thirty or forty feet” (Pope, 1946). The whistle is interpreted as “a by-product of a courtship gesture.” In the aquatic mud turtles, family Kinosternidae, there are patches of horny scales on the inner sides of the hind limbs of males, but not of females (Pope, 1946). Use of these as stridulatory organs in the presence of females has been reported by Evans (1961). Vocalization is widespread as an accompaniment of courtship in the tortoises, which are also the best known turtles in respect to the cues by which sex identification and species identification are achieved. DeSola (1930) gave an account of mating in the giant Galapagos tortoise (Testudo elephantopus) that indicates vocalization to be common by writing, “the males could be heard shouting their deep, bass, resounding roar.” In describing the approach of a male to a female he wrote, “Making his advances he carefully approaches and observes her and if she shows any signs of response, i.e., an approach toward him, he will quicken his pace and commence the deeply resounding guttural tortoise shout” (p. 79).

The most complete analyses of the signals indicative of sex and species in turtles have been made for certain of the tortoises. Eglis (1962) has shown that there is species-specific head movement when various species of tortoises smell any object. These movements fall into two major classes, (1) dirolent (forward-backward movement) and (2) latolent (lateral movement), and there are species-specific variations within these major classes. Auffenberg (1965) has analyzed sex and species discrimination signals in two sympatric South American tortoises (Geochelone carbonaria and G. denticulata). The speciesspecific head movements of the males, important in species and sex discrimination, are thought to be evolved from the primitive motor patterns associated with olfaction. The patterns differ between the two species, with G. denticulata having a smooth latolent movement and G. carbonaria having a jerky latolent movement. “All turtle-like objects are challenged by . . . sexually active, territorial males” (p. 335). Reciprocal head movements are given if the challenged object is another male of the same species, but they are not given if it is a juvenile, a female of either species, or a male of the other sympatric species. If the challenge is not returned, the male moves to the rear of the other animal and smells the cloaca, thus making final identification of an adult female of the same species by means of olfactory cues. Auffenberg noted vocalization in males of both species comparable to that described by Snedigar and Rokosky (1950) for G. denticulata, but unlike them, he heard it only during copulation. As a male followed an unresponsive female about the cage, they observed vocalization resembling, “the efforts of an amateur ventriloquist to imitate the noise of a hen just teaching a brood of chicks to scratch” (p. 47).

Auffenberg (1966, p. 114), in discussing courtship in a gopher tortoise, Gopherus polyphemus, noted that the subdentary glands of the male were everted during the head bobbing and suggested that the bobbing, “may serve as a method of wafting scent through the air as much as being an obvious visual signal.”

REPTILIA: CROCODILIA

The relatively meager information about communication in these remnants of a once mighty reptilian line has been discussed by Evans (1961), and there is virtually nothing I can add to it. Much of the information pertains to the American alligator (Alligator mississipiensis). Vocalization by males, answered by other males in natural populations, is regarded as a territorial device. Females are attracted by the roar of the males during the breeding season, and in addition the secretions of the musk glands located lateroposteriorly to the jaws of the male provide an additional “releaser” during courtship.

REPTILIA: LACERTILIA

More is known about communicatory behavior in lizards than in any other reptiles or amphibians. A classic paper is that of Noble and Bradley (1933), although some of their conclusions must be reexamined in the light of subsequent investigations. Nevertheless, they established the important fact that basic patterns are characteristic of the species belonging to relatively large taxonomie units, i.e., genera. The literature has been reviewed extensively by Evans (1961), who has stressed a behavioral dichotomy among living lizards that parallels evolutionary trends in locomotor methods. Ascalabotids (taxa in which there is no trend toward limb loss or snake-like elongation of the body) depend heavily on visual signals, and among these lizards only in the nocturnal geckos is there “significant” utilization of olfactory stimuli. The autarchoglossids (taxa in which there is evolutionary tendency toward limb loss) depend heavily on chemical signals. This latter group either contains the ancestors of the snakes or has evolved from a common ancestor.

Recent comparative studies of hearing in lizards (Wever, 1965; Wever et al., 1963a, b, 1964, 1965, 1966) by measurement of cochlear potentials and by histological inventory of hair cells are consistent with behavioral evidence as to the role of auditory cues in lizards. The geckos, noted above as being truly nocturnal lizards, are also notably the group in which there is extensive vocalization. Wever and his co-workers found a close correlation between the number of hair cells and maximum electrical output but no close relationship between number of hair cells and sensitivity in examining thirteen species of five families (Wever et al., 1965). They found about 1600 hair cells in the auditory papilla of the Tokay gecko (Gecko gecko), a nocturnal, strongly vocalizing ascalabotid, versus “less than 100 in most iguanids,” which are diurnal ascalabotids that make much use of visual signals. The smallest number of hair cells and poorest sensitivity, but not lowest maximum electrical output, was found in Eremias velox, an autarchoglossid (family Lacertidae). Wever et al. (1966) investigated the leopard lizard (Crotaphytus ivislizenii) because it is remarkable among iguanid lizards in having a distinct vocalization. They found no structural features that “represent a departure from the iguanid type,” but they did find relatively keen sensitivity, with the best sensitivity in the low-frequency region from 300 to 700 Hz and concluded that “this ear more closely resembles the gecko ears than the iguanid ears that have been included in our measurements” (p. 105). Aside from the gekkonids and possibly a few rare examples such as that of the leopard lizard, vocalization seems to be of essentially no significance in communication among lizards. However, whiptail lizards (Cnemidophorus) may squeak when handled. There is no evidence that this sound has any significance, but it is intriguing to note that the one Cnemidophorus investigated by Wever et al. (1965) ranked high in number of hair cells, in sensitivity, and in electrical output.

The most thorough investigations of the numerous avenues of communication in a group of inguanid lizards are those of Hunsaker (1960, 1962), who worked in the laboratory with several members of the Sceloporus torquatus species group. Visual signals are paramount in species recognition. He found species-specific patterns of head display (Fig. 3) to be the most important signals identifying the individual as to species. Such head-bobbing is most common in males but is also present in females and in newly born young. In these highly territorial lizards, the species-specific head bob apparently aids in spacing the males, and the “females use this as one factor in choosing a partner and an area in which to live” (p. 174). The courtship bob is restricted to males and did not differ among the seven species studied (Fig. 3). Hunsaker also implicated chemical signals as being more important than had been heretofore thought in these lizards. He found that lizards “tasted” strange objects and strange lizards, and he demonstra ted experimentally that secretions of the femoral glands (best developed in males) contributed to species identification.

Comparative studies of various other iguanids have been made in recent years. Carpenter (1963) found elaborate posturing as a component of species-specific display patterns of two species of fringe-toed lizards (Uma). “A challenging Uma orients laterally towards its opponent, compresses the trunk in an asymmetrical manner (shifts the abdominal area towards the side of the opponent), extends the dewlap, and tilts toward the opponent” (p. 406). Push-up movements, which differ markedly between the two species, are given from this position. Courtship signals are of a type “typical of many iguanids” but have been elaborated by the “simultaneous up and down waving of alternate front limbs” (p. 412). The desert iguana (Dipsosaurus) investigated by Carpenter (1961a) showed complex species identification behavior, one component of which was reciprocal tail-slapping. Carpenter believed that the tail-slapping “is probably significant in establishing dominant-subordinate relationships between combatants” (p. 404). In Urosaurus ornatus, Carpenter and Grubits (1960) noted that the dominant male in a captive colony was darker than others and that he became lighter when removed from the colony. In comparing three closely related iguanid genera, Carpenter (1962) noted that males of Urosauruas, a tree-living lizard, “vividly display intense ventral coloration with lateral compression of the trunk and extension of the dewlap” (p. 151). Males of Uta, a related ground-living iguanid, “are marked dorsally distinct from the females and have much brighter coloration, but ventrally there is little difference that would be enhanced by marked lateral compression of the trunk or extension of the dewlap” (ibid.). He attributes the differences as being possibly a reflection of the ground-dwelling habits of the Uta, “where an elaborate display would broadcast itself but little, in contrast to the colorful display of the posturing Urosaurus performing on a tree, bush or other noticeably raised station” (p. 152). The species-specific challenge of the canyon lizard (Sceloporus merriami) has been described by Carpenter (1961b) from studies of captive lizards. Milstead (1961), working with natural populations of this lizard, found that females as well as males “exhibited strong antagonism toward trespassers, male or female” (p. 47). Along with this, he found the females of this species developing traces of the male warning colors, and in one population “these colors have been developed so strongly by the females that it is not easy to distinguish the sexes on this basis alone” (ibid.).

The highly conspicuous and bizarre ornamentation of African chameleons (Chamaeleo) has been of debated significance. Rand (1961) has noted that no two species have “identical ornamentation where they are sympatric,” and he suggests that these structures are significant in species recognition (p. 413).

REPTILIA: SERPENTES

Information about communication among snakes is scarce. Their serpentine specializations make snakes difficult to observe in their natural populations, and much of the available information comes from observations under the crowded conditions of zoo life. Nevertheless, the general trends can be outlined, and they have been by Lowe and Norris (1950) and by Evans (1961). Derived as they are from the autarchoglossan branch of the lizards, it is not surprising that the utilization of olfactory cues has been stressed in the evolution of snakes. Visual signals also are important, and rearing above the substrate not only enables a snake to see what the local environment holds but possibly also identifies it to others of its kind. Sound propagation is minimal and, aside from the rattling of rattlesnakes and tail vibration of others, consists of glottal hissing (as in Pituophis), cloacal popping (as in Micruroides and Ficimia), or stridulation (as in Echis and Dasypeltis) (Bogert, 1960).

Structuring of natural populations is virtually unknown, and there is scant evidence of territoriality such as that of lizards. The general impression is that territoriality scarcely exists among snakes, and this is consistent with the great difference between snakes and most lizards in their general, daily behavior patterns. Typical lizards, such as the iguanids, are readily observable as they patrol their territories, stereotyped displays providing identity of species, occupation of territory, sex, age, and readiness to mate. Typical snakes are secretive and seemingly thus unlikely to be identified with territory, since intraindividual communication is nullified by the secretive habits. There are indications (Lowe and Norris, 1950) that some elapids, cobras (Naja and Ophiophagus) and mambas (Dendroaspis) defend territories, but it is not known “to what extent the display of cobras and mambas is aggressive behavior directed towards members of the same species, and to what extent, on the other hand, is it employed for the intimidation of other animals” (p. 9). This display includes the familiar inflation of the hood in the cobras and inflation of “the greater portion of the neck and body” in the mambas. Lowe and Norris (1950) have compared distention of the throat in the boomslang (Dispholidus) to the dewlap extension and body compression of many lizards. A comparable display in the bird snake (Theletornis) involves exposure of an “otherwise invisible necklet of white spots” (Rose, 1950, p. 285).

Reproductive behavior of snakes “is based chiefly upon their mode of predation, which is geared to excellent olfactory and tactual perception” (Evans, 1961, p. 165). Visual cues also seem important in many snakes. Courtship in colubrids is well represented by the description given by Fitch (1963, p. 408) for the racer (Coluber constrictor):

Courtship and mating is divisible into the following well-defined stages: 1) the finding of a receptive female by the male; 2) the persistent following of the female by the male, who courts her by lying extended along her body and performing writhing movements, with periodic interruptions during which he momentarily leaves the female and courses rapidly through the grass around her; 3) the acceptance of the male by the female, signalled by the raising of her tail and the almost instantaneous intromission.

Viperids are characterized by a seemingly ritualized performance that has been called a “combat dance” and that, as reported in the literature, may involve male-male or male-female pairs. Evans (1961, p. 165) describes this in pit vipers, “the two individuals concerned raise the anterior third of their bodies off the ground as their necks entwine or rub and push against each other as their heads infrequently dart about or wave rhythmically.”

REFERENCES

Auffenberg, W., 1965. Sex and species discrimination in two sympatric South American tortoises. Copeia MCMLXV, 335-342.

——, 1966. On the courtship of Gopherus polyphemus. Herpetologica XXII, 113-117.

Awbrey, F. T., 1965. An experimental investigation of the effectiveness of anuran mating calls as isolating mechanisms. Diss., Univ. Texas, Austin.

Barrio, A., 1963. Consideraciones sobre Comportamiento y ‘Grito Agresivo’ propio de algunas Especies de Ceratophrynidae (Anura). Physis XXIV, 143-148.

——, 1964a. Especies Cripticas del Género Pleurodema que Conviven en una Misma Area, Identificadas por el Canto Nupcial (Anura, Leptodactylidae). Physis XXIV, 471-489.

——, 1964b. Caracteres Eto-Ecológicos Diferenciales entre Odontophrynus americanus (Duméril et Bibron) y O. occidentalis (Berg) (Anura, LeptodacÎ tylidae). Physis XXIV, 385-390.

Blair, A. P., 1963. Notes on anuran behavior, especially Rana catesbeiana. Herpetologica XIX, 151.

Blair, W. F., 1958. Mating call in the speciation of anuran amphibians. American Naturalist XCII, 27-51.

——, 1964a. Acoustic behaviour of amphibia. In Acoustic Behaviour of Animals, ed. R.-G. Busnel, Amsterdam, Elsevier Publ. Co.

——, 1964b. Isolating mechanisms and interspecies interactions in anuran amphibians. Quarterly Review of Biology XXXIX, 334-344.

——, and M. J. Littlejohn, 1960. Stage of speciation of two allopatric populations of chorus frogs (Pseudacris). Evolution XIV, 82-87.

Bogert, C. M., 1947. A field study of homing in the Carolina toad. American Museum Novitates No. 1355, 1-24.

——, 1958. The biological significance of voice in frogs. Booklet accompanying Folkways Records Album No. FX 6166, 1-18.

——, 1960. The influence of sound on the behavior of amphibians and reptiles. In Animal Sounds and Communication, ed. W. E. Lanyon and W. N. Tavolga, Washington, D.C., American Institute of Biological Sciences Publ. No. 7.

Cagle, F. R., 1950. The life history of the slider turtle, Pseudemys scripta troosti (Holbrook). Ecological Monographs XX, 31-54.

Capranica, R. R., 1965. The evoked vocal response of the bullfrog: A study of communication by sound. Massachusetts Institute of Technology Research Monograph No. 33.

Carpenter, C. C., 1961a. Patterns of social behavior in the desert iguana, Dipsosaurus dorsalis. Copeia MCMLXI, 396-405.

——, 1961b. Patterns of social behavior of Merriam’s canyon lizard (Sceloporus m. merriami—Iguanidae). Southwestern Naturalist VI, 138-148.

——, 1962. A comparison of the patterns of display of Urosaurus, Uta, and Streptosaurus. Herpetologica XVIII, 145-152.

——, 1963. Patterns of behavior in three forms of the fringe-toed lizards (Uma—Iguanidae). Copeia MCMLXIII, 406-412.

——, and G. Grubits, III, 1960. Dominance shifts in the tree lizard (Urosaurus ornatus—Iguanidae). Southwestern Naturalist III, 123-128.

Cei, J. M., 1962. Batracios de Chile. Ediciones de la Universidad de Chile, Santiago.

Davis, W. C., and V. C. Twitty, 1964. Courtship behavior and reproductive isolation in the species of Taricha (Amphibia, Caudata). Copeia MCMLXIV, 601-610.

DeSola, R., 1930. The Liebespiel of Testudo vandenburghi, a new name for the Mid-Albemarle Island Galapagos tortoise. Copeia MCMXXX, 79-80.

Eglis, A., 1962. Tortoise behavior: A taxonomie adjunct. Herpetologica XVIII, 1-8.

Evans, L. T., 1961. Structure as related to behavior in the organization of populations in reptiles. In Vertebrate Speciation, ed. W. Frank Blair, Austin, Texas, Univ. Texas Press.

Ferguson, D. E., H. F. Landreth, and M. R. Turnipseed, 1965. Astronomical orientation of the southern cricket frog, Acris gryllus. Copeia MCMLXV, 58-66.

Ferguson, D. E., and H. M. Landreth, 1966. Celestial orientation of Fowlers toad Bufo fowleri. Behaviour XXVI, 105-23.

Fitch, H. S., 1963. Natural history of the racer Coluber constrictor. Univ. Kansas Pubis. Mus. Nat. Hist. XV, 353-468.

Gans, C., 1961. The legless tetrapod. Natural History LXXI, 26-27.

Heusser, H., 1960a. Uber die Beziehungen der Erdkrote (Bufo bufo L.) zu ihrem Laichplatz II. Behaviour XVI, 93-109.

——, 1960b. Instinkterscheinungen an Kroten under besonderer Berucksichtigung des Fortpflanzungsinstinktes der Erdkrote (Bufo bufo L.). Zeitschrift fur Tierpsychologie XVII, 67-81.

Hunsaker, D., 1960. Ethological isolating mechanisms in the Sceloporus torquatus group of lizards. Diss., Univ. Texas, Austin.

——, 1962. Ethological isolating mechanisms in the Sceloporus torquatus group of lizards. Evolution XVI, 62-74.

Jameson, D. L., 1955. The population dynamics of the cliff frog, Syrrhophus marnocki. American Midland Naturalist LIV, 342-381.

Johnson, C., 1963. Additional evidence of sterility between call-types in the Hyla versicolor complex. Copeia MCMLXIII, 139-143.

Kumpf, K. F., 1934. The courtship of Amhystoma tigrinum. Copeia MCMXXXIV, 7-10.

Littlejohn, M. J., 1965. Premating isolation in the Hyla ewingi complex (Anura: Hylidae). Evolution XIX, 234-243.

——, and T. C. Michaud, 1959. Mating call discrimination by females of Strecker’s chorus frog (Pseudacris strecken). Texas Journal of Science XI, 86-92.

——, M. J. Fouquette, Jr., and C. Johnson, 1960. Call discrimination by female frogs of the Hyla versicolor complex. Copeia MCMLX, 47-49.

Lowe, C. H., Jr., and K. S. Norris, 1950. “Aggressive behavior in male sidewinders, Crotalus cerastes, with a discussion of aggressive behavior and territoriality in snakes. Natural History Miscellanea, Chicago Academy of Sciences No. 66, 1-13.

Lutz, B., 1960a. Noçao de Território em Anfibios Anuros Hyla faber Wied. Anais da Academia Brasileira de Ciencias XXXVIII, 143-145.

——, 1960b. Fighting and an incipient notion of territory in male tree frogs. Copeia MCMLX, 61-63.

Martof, B. S., 1953. Territoriality in the green frog, Rana clamitans. Ecology XXIV, 165-174.

Maslin, T. P., 1950. The production of sound in Caudate Amphibia. Univ. Colorado Studies Series in Biology I, 29-45.

Mecham, J. S., 1961. Isolating mechanisms in anuran amphibians. In Vertebrate Speciation, ed. W. Frank Blair, Austin, Texas, Univ. Texas Press.

Milstead, W. W., 1961. Competetive relations in lizard populations. In Vertebrate Speciation, ed. W. Frank Blair, Austin, Texas, Univ. Texas Press.

Noble, G. K., 1931. The Biology of the Amphibia. New York: McGraw-Hill.

Noble, G. K., and H. T. Bradley, 1933. The mating behavior of lizards; its bearing on the theory of sexual selection. Annals New York Academy of Sciences XXXV, 25-100.

Pope, C. H., 1946. Turtles of the United States and Canada. New York: Alfred A. Knopf.

Rand, A. S., 1961. A suggested function of the ornamentation of East African forest chameleons. Copeia MCMLXI, 411-414.

Rose, W., 1950. The Reptiles and Amphibians of Southern Africa. Cape Town: Kew Miller, Ltd.

Savage, R. M., 1961. The Ecology and Life History of the Common Frog (Rana temporaria temporaria). New York: Hafner.

Sexton, O. J., 1960. Some aspects of the behavior and of the territory of a dendrobatid frog, Prostherapis trinitatis. Ecology, XLI, 107-115.

Smith, M., 1954. The British amphibians and reptiles. Revised ed. London: Collins.

Snedigar, R., and E. J. Rokosky, 1950. Courtship and egg laying of captive Testudo denticulata. Copeia MCML, 46-48.

Snyder, W. F., and D. L. Jameson, 1965. Multivariate geographic variation of mating call in populations of the Pacific tree frog (Hyla regilla). Copeia MCMLXV, 129-142.

Stickel, L. F., 1950. Population and home range relationships of the box turtle, Terrapence c. Carolina (Linnaeus). Ecological Monographs XX, 352-378.

Test, F. H., 1954. Social aggressiveness in an amphibian. Science CXX, 140-141.

Tinbergen, N., 1953. Social Behaviour in Animals. London: Methuen, 1953.

Twitty, V. C., 1955. Field experiments on the biology and genetic relationships of the Californian species of Triturus. Journal of Experimental Zoology CXXIX, 129-147.

——, 1959. Migration and speciation in newts. Science CXXX, 1735-1743.

——, 1961. Experiments on homing behavior and speciation in Taricha. In Vertebrate Speciation, ed. W. Frank Blair, Austin, Texas, Univ. Texas Press.

——, D. Grant, and O. Anderson, 1964. Long distance homing in the newt Taricha rivularis. Proceedings National Academy of Sciences LI, 51-58.

Wever, E. G., 1965. Structure and function of the lizard ear. Journal of Auditory Research V, 331-371.

——, and E. A. Peterson, 1963a. Auditory sensitivity in three iguanid lizards. Journal of Auditory Research III, 205-212.

——, D. E. Crowley, and E. A. Peterson, 1963b. “Auditory sensitivity in four species of lizards.” Journal of Auditory Research HI, 151-157.

——, E. A. Peterson, D. E. Crowley, and J. A. Vernon, 1964. Further studies of hearing in the gekkonid lizards. Proceedings National Academy of Seiences LI, 561-567.

——, J. A. Vernon, D. E. Crowley, and E. A. Peterson, 1965. Electrical output of lizard ear: Relation to hair-cell population. Science CL, 1172-1174.

——, M.-C. Hepp-Reymond, and J. A. Vernon, 1966. Vocalization and hearing in the leopard lizard. Proceedings National Academy of Sciences LV, 98-106.

Woodbury, A. M., and R. Hardy, 1948. Studies of the desert tortoise, Gopherus agassizii. Ecological Monographs XVIII, 145-200.

Wynne-Edwards, V. C., 1962. Animal dispersion in relation to social behaviour. New York: Hafner.