Hierarchical Position of Acoustic Signaling In Relation To Other Means of Communication

The hierarchic value of acoustic signaling differs, in relationto other means of communication, for each animal species employing itin various behavioral situations. In certain birds, it is of primeimportance, as Brückner (1933) demonstrated in the hen and Schleidt(1960) inthe turkey. In the latter species, forinstance, the female, surgically deafened, laid and saton her eggs normally, but after the young hatched, she did not seem to be able to differentiate them from predators and killed them as she would predators who approached the nest. The acoustic signal given by the young turkeys had a fundamental hierarchic value which induced recognition by themother and suppressed her aggressiveness. Lacking this signal, the female turkey, in spite of visual information, was dominated by her aggressive behavior pattern. Similarly, the mother hen, unable to hear her chick, but seeing him isolated under a glass bell, abandoned him. In innate behavior in these species, acoustic signalsthus have a privileged position. Lorenz (1950) and his students showed in the duckling and gosling that imprinting depends on acoustic and visual signaling.

In the sexual behavior of theinsect Ephippiger (Busnel and Dumortier, 1954), the female is wholly oriented by the sexual signal, and goes toward a loudspeaker emitting signals of the male even when she is near a silentmale. The sexual behavior of the sow is controlled by association of kinesthetic, optical, chemical, and acoustic signals; in the sequence ofcopulatory behavior, the experimentally isolated acoustic signal of the male, in 80 percent ofthe cases, is alone sufficient to ensure immobilization reflex behavior (Signoret et al., 1960). A porpoise, completely blinded with rubber cups, swims without error through a maze of diverse obstacles utilizing only his echolocation system (Norris, 1961, studying Tursiops), and he is able, under these same conditions, to detect metallic targets as small as 0.2 mm in diameter (Busnel et al., 1965a and b, studying Phocaena). The same is true for the bat (Griffin, 1944; Dijkgraaf, 1946). In some batand dolphin species acoustic signaling by echolocation is hierarchically equal or superior to visual information. These examples may be multiplied. However, generalizations cannot be drawn, because in each species the situation dependsupon sensory factors which predominate and induce one behavior or another at a given moment.

Finally, it must not be forgotten that some acoustic signals may be closely related to gesticulatory or kinesthetic signalssuch as in sea gulls (Tinbergen, 1959). In the case of honey bees, Wenner (1962, 1964) and Esch (1961, 1962, 1965) showed acoustic signals, as much as or more than chemical signals, to be inextricably linked to the signals of the dance. In these cases, communication is composed of a complex of factors, and it becomes increasingly difficult to find a nonrelative value in a given signal form.

Transmission Channel and Background Noise

Each animal species is surrounded by its own Umwelt (air, water, or solid) in whichthe transmission of vibrations follows specificphysical rules. These vibratory phenomena support information during acoustic communication. The transmission channel, even when considered as homogeneous, is not inert, and it plays a role in signal structure or its perception that isindependent of its total effect on the organismunder consideration. In nature, the transmission channel always contains a background noise level which is statistically characteristic of the species' biomes and biotopes and is, of course, related to natural events such as rain, wind, and the breaking of waves. Moreover, when an acoustic signal is propagated in nature (which cannot rightly be considered as a free field), numerous acoustic wavereflections are produced. These, in their turn, constitute a special part of background noise and bring about both a lowering of signal intelligibility and a diminution in the signal carrying power. Returning signal waves are modified by obstacle impedance and volumetric form. This is perceived by certainspecies which employ this information for theirautodirectivity (echolocation). Whatever the importance of the channel background noise, it can be said that in natural surroundings it almost always occurs and that, up to the present time, perception studies have too often neglected this signal-to-background-noise relationship from which the listener's peripheral analyzers have to extract whatever proper information is contained in the signal.

Another background noise whose importance has been underestimated is a specifically biological one which is superimposed on the natural background noise and from which message information must be extracted if the intelligibility coefficient is to remain acceptable or reach threshold perception. This biological noise, produced especially in a crowd or dense social grouping which is engaged in a more or less determined social activity (as in a reproduction area, a hive, a colony, a pack, or a fraternity), consists of a certain aggregate noise:the sum of all the emitted individual signals and the byproducts of motor activities (wing or flight noise, noise of legs against the substratum, etc.). This latter carries no specific information. Such biological noises are emitted, for instance, in groupings of sea birds, bee colonies, porpoise schools, or seal harems. It covers exactly the same frequency band as that of the species' signal emission and would thus, theoretically, lead to an auditory saturation, making it impossible to transmit far-field any information other than specific information (suchas the general noise characterizing the species). An animal must therefore have the capacity of perceiving a signal and extracting it from a random background noise. This signal individualizes communication between members of a species. The physical characteristics which allow thisdetection are varied, each animal using its ownspecific sensory organs to this purpose. In some cases these must even be able to detect an information-carrying signal whose intensity is below that of the background noise. Cherry (1957)has described this phenomenon in man as the "cocktail party effect." So far in animal communication it has received little attention from bioacousticians. Its origin can be related to a preferential motivation which triggers an acute selectivity of the nervous centers, the biophysics of which is still little known. However, it can be observed in quite a few higher animal groups, notably gregarious vertebrates, that members of a permanent or temporary couple are capable of personal recognition through individualized acoustic emissions, even though the number of such simultaneous signals sent produces a high background noiselevel [individual recognition experiments with the emperor penguin (Prevost, 1961) and with bats (Möhres and Kulzer, 1956; Griffin, 1958].

These problems of masking background noise in the transmitting canal should be more thoroughly studied. They would surely interest psychophysiologists as well as neurophysiologists, for whom this question should prove particularly interesting since it is not strictly auditory in the usual sense. That is to say, it does not concern just the eighth pair ofnerves. The decoding of special information, selectively chosen from the background noise by means of a guiding motivation, is probably analogous to the mechanism used for specific vision. There it is the eye which, directed by a particular motivation, selects from the mass of data constituting the background noise such small informative details as a word on a page or a letter in a word which alone focus the attention. Possibly this is also the case in olfaction.

Communication Sound Source

Sound sources either have a mechanical origin by specialized or nonspecialized emission organs or result from vibrations imparted to the substrata by a part of the animal body acting as an amplifier. Percussion on the substratum asa sound source is illustrated by ground-tapping(usually with the hind feet as done by prairie hens, rabbits, mice, Neotema), by woodpecking (series of shocks produced by the beak of woodpeckers), etc.

The range of sound emission organs found in the animal kingdom is quite varied; they are usually bilateral in invertebratesand very often unpaired in vertebrates. They may be restricted to one sex, or they may present a considerable sexual dimorphism. They are found on all different parts of the body. For example, the following may be found functioning as sound emission organs in invertebrates: chitinous toothed files which, by friction, stimulate a vibrating body— wing, elytra, antenna, thorax, leg, abdomen (Orthoptera, Crustaceans); friction or vibration of nonspecialized organs such as the wings (mosquitoes and some moths); semirigid plates on a resonant cavity stimulated by neuromuscular contractions (Tymbal method—Cicada); reedlike organs which function by aspiration and expiration of air (death's-head hawk moth, Sphinx atropos). In Myriapoda two species have been found, Scutigera and Rhysida, which can automate legs. These species have no special stridulatory organ; however, when the legs are separated from the body, they emit sounds. When they are intact, they aresilent. The hypothesis is that the noise emitted by the leg attracts predators, leaving the animal free to flee (Annandale et al., 1913; Cloudsley-Thompson, 1961).

In lower vertebrates, there is much polymorphism and a great variety of sound sources: nonspecialized organs may produce friction, as do vomerine teeth in certain fish; osseous, rattle-type apparatuses may be found, made up of moving, oscillating parts which knock each other when agitated, as in rattlesnakes;whistling or vibrating apparatuses which function by air expulsion through a more or less differentiated tube (larynx) ending in an aperture (glottis) with more or less functional lips. The expelled air is supplied by the lungs themselves or by being in contact with an air pocket reserve with or without diverticula (vocal sac of some amphibians); and finally, membranes may be stretched over resonating pockets (as is theswim bladder of fish). These apparatuses are activated either by external percussion (fin beating) or by contraction of muscles disposed in different ways around the cavity.

Sounds produced by nonspecialized organs are also found in higher vertebrates. These include breast-beating in the gorilla, organ-clapping such as wing-beating of the woodpigeon, drum-rolling in the hazel grouse and gold-collared manakins, and trembling of remiges (primaries) and rectrices (tail feathers) in the woodcock and snipe. Owls and storks use their beaks, and some bats (Möhres and Kulzer, 1956) and some insectivores, such as tenrecs (Gould, 1965), use their tongues. In many higher vertebrates specialized organs are found, usually working by propelled or aspirated air in a moreor less differentiated tube equipped with modulating membranes or slit system. These organsarevocal cords, muscular glottal lips, the larynx of odontocetes, and the bird larynx and syrinx. These apparatuses often have additional organs which form air reservoirs or resonators (clavicular and cervical air sacs), as found in the bustard, ostrich, crane, and morse. In some monkeys these features are found in the thyroid cavity, as is the gibbon's vocal sac or the hyoid bone resonating chamber of the New World howler monkey. Curious peripheral sound organs are also found such as the fifteen-spined sound apparatus in tenrecs, the Madagascar Hermicentetes, Centetes (Gould, 1965), and the tail bellof the Bornean rattle porcupine, Hystrix crassispinis. Anatomists and morphologists have described the remarkable richness of the characterof these structures in hundreds of animalspecies. However, much is still unknown. For example, it is not yet known what elements generate echolocation clicks in odontocete Cetaceans and in many bats. Phonation mechanics is a large fieldwhich still needs much exploring, and the microphysics of most phonatory apparatuses remains to be studied in many species.

Whether an animal that is provided with adequate sound emission organs will or will not activate them, depends upon physiological and psychophysiological factors particular to the species and its ecological conditions. Such factors may be abiotic—temperature [applies mostly to poecilothermal animals, but also to some vertebrates as, for instance, to the foot in the flycatcher bird (Curio, 1959)], humidity, or light—or they may be endogenous (hormonal state, age, psychological motivation, etc.). When a sound emission occurs, it follows a code in which different factors come into play: rhythm, pulse repetition rate, amplitude variation, intensity, frequency, etc. The periodicity of biological activity (circadian and seasonal rhythms under the influence of temperature and light) may also influence sound emission. The animal's age determines even the nature of the signals. Insect larva and nymphs do not generally have the sound apparatus which the imago has later. The signals of young mammals and birds do not have the same physical structures as do adult signals (sub-song).

Other interesting aspects of studying sound emission apparatuses of animals may be looked at from a mechanical point of view. It can be considered that sound, when emitted by an organ having a more or less complex mechanism, is the product of this mechanism transmitted via the central nervous system. This system thus governs the physical structure, since it is a result of the structural dynamics of theapparatus under consideration. In this case, acoustic signals are the result of the activationof anatomical elements which are controlled by neuromuscular structures, in turn directed by neuromotor centers.

It is even more interesting to find that there is, in some species, specificprogramming of innate acoustic activity associated with other aspects of behavior and under control of the higher nervous structures. This iseasier to demonstrate in invertebrates, where the acoustic activity is genetically fixed (Leroy, 1962) and where the nervous structures are more circumscribed and less diffuse than in higher vertebrates. It is not surprising to find cellular group localization in the brain which selectively commands the descending neuromotor channels, activating muscular fibers of the sound-producing organs. This coincides with data known about the role of more or less specialized centers of different levels of the central nervous system which command, by a given channel, a corresponding muscular reaction in voluntary orinvoluntary acts. The first example of this localization was experimentally obtained by Hüber (1952) on the field cricket. Interestingly enough, it was shown that the signal emission was regulated, not by one or more stimulating centers, but by the suppression of action from inhibiting centers. That is to say, as soon as these centers were destroyed, the insect transmitted its signal almost continuously. Depending upon the centers and the neuromotor coordination which these governed, corresponding activated muscles put the elytra into different stereotyped positions. These movements were in all respects comparable with those found in the different signaling pattern behavior of the species, such as sexual call, courtship, and rivalry signals.

The higher centers in the cricket controlling soundemission are localized inthe pedunculate and central bodies. Fighting behavior was set off by the caulicles and the corpuscallosum. The center of the pedunculate body was found to govern the courtship song. Thecentral body directed sound apparatus vibrations, coordinated with the placing of the elytra in a "flat position." It is almost certain that the sounds emitted after these stimulations arenot simple noises but a series of pulses organized into a signal and having all the features of natural and normal signals.

In amphibians, reflex croak of the "warning" type has been obtained in the leopard frog Rana pipiens by mechanicallystimulating the anterior part of the spinal cord (Aronson and Noble, 1945). Working along the same lines, Schmidt (1965) has done excellent experimental research on central respiration mechanisms and their relation to acoustic signal transmission in different Anura, Rana and Hyla. In these species all vocalizationcontrol is localized in the trigeminoisthmic-tegmentum, which activates the phonatory motor mechanisms. In fact, the vocal cords and glottal system which regulate air flow are excited by the intermediary of the vagus; abdominal muscles making up the air pump system are controlled by the spinal nerve and the hyoid depressorsby the hypoglossal nerve.

Mating calls can be obtained by nerveimpulse intensity or electrical stimulation when taking into account the animal's hormonal state, which may be modified with androgens. Schmidt also concludes that mechanisms concerned with phonoresponses, just like the sexual calls, appear to be localized in the preopticzone. Recent research on birds indicates that stimulation of different nervous centers sets off signals which seem in all respects comparableto natural signals. In the male red-winged blackbird (Agelaius pheniceus) vocalizationswere elicited from an area beneath the optic tectum including the torus semicircularis and theunderlying gray matter. The vocalizations evoked by electric stimulation resembled the calls given at the approach of a potential predator orcalls given by birds being caught in cages and held in the hand (distress call). Whistle-type calls were evoked from the hypothalamus (Brown,1965). Different types of calls have been evoked by electrical stimulation of the brain in thecock or hen, but responsive areas were not delineated anatomically except in the hypothalamus, where androgen implants were used (von Hoist and St. Paul, 1963). In ducks, quacking is governed by the archistriatum and the tegmental midbrain (Phillips, 1964). The mesencephalon directsthe squawks of the southern lapwing (Belonopterus chilensis)(Silva et al., 1961) andthe alarm call and "ruckcuh" threat coo of the feral pigeon (Columba livia) (Heinroth and Heinroth, 1949; Akermann et al., 1960; Rougeul and Assen-macher, 1961). Anatomically, there are similarities between the central gray matter of mammals and the torus of other vertebrate classes. In all cases, the sounds thus produced are innate signals; furthermore, they are associated with all the corresponding attitudes or postures attending these signals as for example, hair-bristling or feather-ruffling.

In mammals, knowledge acquired by electrical stimulation of the deep layers of the diencephalon in unanesthesized animalsshowed diffuse localization of the command of vocalization. These motor zones have never induced more than instinctive signals of the distress type, whether on cat, monkey, or porpoise. Among the first of many studies done in this field were those of Karplus and Kreidl (1909) and Hess (1928). Those of Hunsperger (1956) on the cat and of Lilly (1958a, 1958b and 1962) on porpoises can be more closely examined.

In the diencephalon and mesencephalon of the cat, Hunsperger described zones which set off defense or escape behavior accompanied or not by roaring, cries, and growling, constituting, however, a phonation rather than amessage. Control of varying behavior connected with this defense is found in very different parts of the brain. The distress behavior scheme is localized in the mesencephalic central gray matter and in the perifornical region of the hypothalamus. This zone extends into the corticalregion, especiallyin the periform lobe and theanteroventral partsof the temporal lobe. Thus, this structural entity extends from the mesencephalon to the anterior brain. The primary cat mewing which occurs in uneasy situations is governed by the mesencephalic level under thesubstratum and is a complementary of the defense reaction. The secondary mewing which appears after stimulation is set off from Papez' circuit, the hippocampus, and also from canals which connect these structures to the preoptic region. It is probable that these vocalizations are at least partially related to distress signals. As yet, the sounds thus elicited have never been really studied as communication.

According to Lilly, attainingthe motor area in the monkey resulted in vocalization; however, as yet no true signal has beenrecorded, although distress cries were probablyemitted during stimulation. Behind the parietalcortex in the dolphin, and above what corresponds to the orbital cortex in man, is the thalamic region at the caudal nucleus level which controls the distress signal and other sounds. These latter may be extremely varied owing to the different vocalization systems used by the animal (larynx, lips of the blowhole, etc.) (Lilly and Miller, 1962).

The higher we go in the animal kingdom, the more diffuse and heterogeneous become the motor zones, introducing a notion of degrees of freedom. The production of complex signals depends upon numerous centers which interfere with each other, and thus no longer permit the "all or none" responses of invertebrates or lower vertebrates. In mammals, zones corresponding to a specific signal are not found. Instead, generalized phonation zones can be described which are diversely activated by other centers concerned with different emotional behavior patterns.

The genetic transmission of anatomical structures of a species does not giverise to any particular problem as far as the phonating organs are concerned. It is logical to consider the nervous structures associated withthese organs and, by the same token, the morphology of innate signals to be genetically controlled. This has been demonstrated to be true fora great number of species, whether invertebrates (Fulton, 1933-1952; Perdeck, 1958; Leroy, 1963-1966), lower vertebrates (Blair, 1956-1963) or certain birds (Thorpe, 1961; Thielcke, 1961-1962; Messmer, 1956; Sauer, 1954). However, if the activation of all phonation systems followsMendelian laws, as does their morphology, and if innate sound emission follows the same rules, it is noted that besides the mechanical (and therefore cerebral) rigidity of the basic framework of physical signal production, a certain measure of choice eventually appears. This degree of freedom may be explained by a preestablishedhiatus in the command center containing learnedinformation memorized at certain stages of the individual's life.

An autocontrol, ensured by the ear through a feedback mechanism, must be added to the description of control of the physical characteristics of emitted sounds given above. Certain birds, for instance, deafened at birth, are no longer capable of developing the complete structure of the species' specific songs (Huchtker and Schwartzkopff, 1958; Konishi, 1964). Bred solitarily in soundproof rooms, they develop that portion of the complete species repertoire which is inbred, but none of the learned part.

Control of phonation and sound emission by the ear is also associated with learning mechanisms which occur in certain social conditions during the animal's development (imprinting in young birds, for instance) (Lorenz, 1950). This learning can eventually lead (especially in many birds) to great variations in the physical form of the signal, bringing aboutlocal dialects on the one hand and psittacism or mimetism on the other. In all instances studied, dialects had the value of a true signal no matter what their physical characteristics. Imitations, on the other hand, can probably be considered as a meaningful signal for most speciesexcept possibly for parasite birds, where they simply reflect learning and an empty phonatory activity.

In this learning process, genetic learning affects the structure of vocalization as well as the memorization processes. Here may lie one of the most fundamental differences between man and animals, as far as vocalization is concerned. In man the only aspects of vocalization under genetic control are those qualities of voice determined by the structure of the larynx. But language as such is usually learned through audition-phonation feedback.

The Message

The sound message, more than any other communication method, lends itself to comprehensionand correct analysis by the experimenter, assuming of course, the electroacoustic equipment tobe precise. Sonagram frequency spectra and oscillogram transcriptions have become classical inthese studies, and they give us a rather precise picture of the signal's diverse physical parameters. There are hundreds of publications concerning this aspect of the acoustic communication problem.

Certain analyses may be done by series ofelementary, orthogonal, translatedfunctions instead of by harmonicanalysis. These elementary functions are damped sinusoidals, synchronous with the pitch, and may be analyzedby autocorrelation. Taking human voice analysisas basis, it is possible to determinethe number of bits per second (or bauds) and obtain a partial quantification based on different metrologies. Putting sonagram-type signals on a digital machine is made possible by point analysis. A transcription into machine code may then be made, thus rendering the whole in a geometrical form easier to manipulate than analysis made directly from the sonagram. The original processes of form recognition are now rather well known, and wemust think about applying them soon to animal acoustic signals. Probably, this field will be first approached through studies on echolocation signals of porpoisesand bats. These signals are recorded by oscillographic traces which may be directly interpreted by computer.

Acoustic frequency bands varywidely according to species and zoological groups. Low-frequency signals are emitted by fish and certain amphibians and in flight noises. Very high frequency signals are emitted by many insects and some porpoises. The highest frequencyrecorded is 350 kHz for the echolocation clicksof Steno bredanensis (Norris and Evans, 1966).

For many of the lower species, the signal's morphology is a close physical expression of the mechanical structure of the emission apparatus. These signals have thus a sort of obligatory physical form rigidly determined by the elementary movement of the organs. Other signals, on the contrary, have a flexible physical structure due to the possibility of varied uses of the same organ (such as the bird syrinx, higher vertebrate vocal cords, and the delphinid larynx) and to a directing brain capableof making choices.

More than any other communication method, suchasolfaction or vision, the acoustic signal's physical nature lendsitself to reemission in rather satisfactory conditions. Theexperimenter can also intervene in the signal'sstructure by means of recording tapes. He may cut, fractionate, inverse, filter, and isolate certain factors. Thus, the signal's physicalnature makes possible, and relatively easy, synthetic signal creation by means of electronictechnology. Tectonics of the acoustic message maybe composed of a simple, physically indivisibleelement called a phonatom (or pulse or noteaccording to different authors) which forms a sort of basic molecule. This structure may be repeated adetermined number of times in a given timeperiod. A rhythmic element thus appears. There may be relative amplitude variations in some oftheseelements. The phonatom itself may be composed of a preferential frequency or of groups of frequencies. The signal may be continuous; however, it always has a defined length. The simplest (although not simple) structures are those of invertebrate and lower-vertebrate signals. Bird signals are much more complex, the same message having an extremely varied and heterogeneous physical configuration. Moreover, individual variations appear which have been brought to light by systematic studies done by many authors on certain birds such as the finch, in which Borror (1961) finds 13 themes and 187 variations. These individual variations may be personifications of the message, although it is not yet well known which physical structure is related to individualization. One exception may be Laniarus erythrogaster described by Thorpe (1963). Inthis species the temporal parameter between themessages exchanged by two individuals (i.e., the rhythmic pattern of silences) carries the information, and not the acoustic part of the signal itself.

Taking into account individual variations, whether minor or large, physical structure of the message is organized followinga code particular to the species, and thus, from a purely zoological point of view, the structure has a specific character. Its physical variants may even characterize local populations (dialects). However, there may also be analogous structures in several species, thus inducing interspecific reaction as in the case of alarm signals in some Corvidae (Busnel et al., 1957; Frings et al., 1958) and sparrows (Marler, 1957; Marler and Tamura, 1964). Finally, certain animals are able, by memorization and imitation to copy the physical signal characteristic of other species. This has been shown to occur in birds (Armstrong, 1963) which imitate the songs of other species and even the voice of men and also in the porpoise (Lilly, 1962; Batteau, personal communication).

Graphic analysis of signals have not given bioacousticians as easily prehensible a result as they might have at first hoped. The spectrographic configuration, apart from artifacts which it sometimes induces, provides a general characterization of the phenomenon;however, in spite of near perfection in technique it is difficult to relate the image obtainedafter analysis to the semantic content of the different parts of the signal. Imperfect as it may be, it is certain that the tendency of the future will be to master all elements of the signal composition by using large computers (as isnow done) in a rather satisfactory manner, withword synthesizers which are being developed following the series of vocoders. This technology, applied to human acoustic communication, is theoretically easy to conceive and use, as the experimenter directly controls the phenomenon by himself. The approach is quite different for animal acoustics, for there, the animal's responses are the only objective proof of a positive result. This adds an extra difficulty, but it is almost certain that in the near future researchwill be orientated in this direction.

The notion of syntax, used here in reference tohuman language, is acceptableto the biologist because it can beassimilated to that of gestalt, which takes into account the temporal form perception obligatorily included in every signal. Theb sound signal is a form, modeled after the space-time relationship, and syntax is related to recognition of the organizational value of this structure. This is the notion of "pattern recognition." If the signal, as such, has a recognizable temporal structure and thus a composite one, its structure is syntactic. This aspect of the problem has been little studied up to the present. Research has beendone on the following species: insects (Busnel et al., 1956), amphibians (Capranica, 1964; Paillette, unpublished results), birds (Thielcke, 1962; Bremond, 1962; Falls, 1962), dogs (Busnel, unpublished results), and cats. Much new workis now under way in this field.

How can the notion of space configuration be approached? The answer is relatively simple in electroacoustics. The natural signal is taken as a starting point and is expanded, contracted, inversed, lengthened, shortened, etc. This type of transformation has many possibilities, as Bremond in our laboratory has shown, working with variable-speed tape records, heterodynes (for frequency transposition), ring modulators, frequency modulators, level modulators, and filters.

Message Content

An animal sound message contains a minimum of two types of information. The first intrinsic part indicates the presence of an individual of the species, his spatial position, and, in a number of species (birds and porpoises, among others), individualization—it is Peter or Paul. The hierarchical status is also present if a gregarious species is concerned. The second part of the message holds the semantic content which translates the transmitter's internal state in a given behavioral situation, that is to say, his motivation. It may also contain information relative to the milieu, such aslocalization of an individual, an object, a territory, or a predator. A thorough investigationof this subject has been done on birds in particular by Marier (1956) studying Fringilla coeles.

Signals are usually classed in function according to the behavior which theyelicit in the receiver. In the animal kingdom signals are principally related to sexual life (calling the partner, courtship, and rivalry), family life (contact and reciprocal parent-youngrelationship signals), and social life (hierarchy, group activity, alarm, predator signaling, alimentary behavior, territorial behavior, etc.). Vocabularies in the animal kingdom vary widely in their complexity.

In many species the male sexual signal is different from that of the female. As a matter of fact, in many invertebrates and lower vertebrates, females are often mute. Experimentation has shown that sexual signals are recognized by partners of the same or of the opposite sex, or by both. Information indicating the transmitter's identity has been indisputablyfound in many, if not in all birds and in some mammals. When both sexes emit sounds, partner or parent recognition is accomplished by very small variations in the signal's physical constitution. As already noted above, a curious example of this is the bird Laniarus, which does not use the signal itself as a means of individual recognition, but instead uses the interval of time between calls and replies in a duo (Thorpe, 1963, antiphonal singing). The transmitter's social situation may be equally transmitted in the signal, although dominance in a groupis a rather abstract notion. In reference to individual recognition, we may say there is parental, filial, and social information.

In social situations, we findcontact, anxiety, and distress signals and also aggressive motivation signals such as territorial, rivalry, food, alarm, attention, and flying-away signals. There are different sounds for different types of alarm, graduated according to dominant motivation. Many birds and mammals areable to express the degree of alarm, imminence of potential danger, or excitement by varying the speed or intensity of the emitted signal (Lorenz's geese). Some bird territorial defense signals vary according to the intruder's distanceaway, and bee signals vary according to sugar concentration.

Interesting studies on sea gulls (Frings and Frings, 1956) and bees, Apis and Melipone (Wenner, 1962; Esch, 1962), have been made concerning transmitted signal information about food. The case of the honey-guide bird (Friedmann, 1955), holds special interest for its interspecificity. It conveys food-source information to any mammal showing an interest, such as the ratel or man, both of whom have been shown to integrate the messageand follow the bird to its food emplacement. Signals have also been described for bees and birds which give information concerning the location of nests or nest-building materials.

Information concerning predators has been well described for birds and for some mammals such as the prairie dog (King, 1955). The signal may contain data indicating the spatial position of a danger, whether it be on the ground or in the air. These signals may be physically quite different, however, without being specific to a special danger. They may be understood by several species (Marler, 1956; Busnel and Giban, 1958). Wenner (1962, 1964) has shown in bees that acoustic signals can, through combination of simple signals, transmit severalinformative parameters. In some birds, and in particular in the robin, Busnel and Bremond (1962) have found signals transmitting a minimum oftwo sets of information. One of these relates to the transmitter's situation, the other designates by a message a particular individual. Thisimportant fact has been proved by Weeden and Falls (1959) studying the ovenbird (Seiurus aurocapillus) and by Bremond (1966) studying the robin, the wren ( Troglodytes troglodyte), and the finch (Fringilla coelebs). The schamama (Copsychus malabaricus), studied by Gwinner and Kneutgen (1962), is particularly interesting in that it incorporates in its territorial defense signal an imitation of some of the motives of the song of the intruder, thus personalizing the threat. In the porpoise Tursiops truncatus Lang and Smith (1965) and Bastian (1967) also showed that acoustic signals could transmit complex data. Man excepted, these are so far the only known cases in the animal kingdom of a real combination of signals. This will pose anew the syntactic problem, as will the experiment with "crayfish counterpoint" (or cancrizan). The signal when readbackwards shows the semantic contents to be related to the order of sounds perceived either phonetically or prosodically. This means that, ingeneral, successive elements have a certain syntactic construction. In fact, if we consider the physical framework of the message and thus its organization, the semantic information is based on a structure which is sometimes simple, but more often complex. In many lower animals a very simple characteristic of the signal is reactogenic. The value of the signal's sharp wavefront has been shown to set off the attraction reaction in the female (Busnel and Dumortier, 1954). If a single artificial transient sound or sharp-wavefront sound, with a sufficient intensity is sent, it takes on the same value as the species' natural signal. This reaction to transients is interspecific.

On the contrary, pulse repetition rate is a very specific characteristic of a signal and may even serve in taxonomic determination (Dumortier, 1963a; Haskell, 1961). Variants of this character have been found in some amphibians (Busnel and Dumortier, 1955; Capranica, 1964). The entire signal is not always necessary as a support of information. In a fish, Bathygobius soporator, for instance, Tavolga (1965) found that the length of the signal could be reduced by 50 per cent and its rhythm and frequency modified in large measure withoutaltering the information value, which seemed toremain unchanged if judged from the point of view of the receiver's behavioral response. Thus, information is not carried by the entire signal, and the richness of most acoustic structures should probably be considered redundant. This redundancy acts as a protection against background noise and increases the signal's intelligibility. Redundance is an important notion in biology, and its value has not yet been fully understood.

Falls (1962) and Busnel and Bremond (1962), in their studies on some birds, have found several reactogenetic signal characters in which there is a certain syntactic organization. This organization has not as yet been found to be a general rule, and each species must be considered as a special case and studied separately. The structure and semantics of animal signals are, nevertheless, quite rigidly fixed, and in the communication of most known species today there are only invariants. Thus, the signal repertoire seems limited, and in a way, stereotyped, since in the entire animal kingdom(bees, some birds, and possibly porpoises excepted) semantics is limited to objective situations evoking an instanteous or immediate-future type of event. Modifications in rhythm and dynamics indicatean evolution in the degree of motivation, but in most cases do not show a change in semantics. However there may be in a message, at least two, if not more, semantically significant units such as distance and designation of an intruder contained in the territorial defense signal or the kind of danger and distance from it contained in the danger signal. We do not yet know howto relate a given unit to a distinct message structure; but it seems certain that there is no true signal combination, and thus there can be no language, in the true sense of the word, since signal combination forms one of the bases oflanguage (by signal combination is meant the possibility of associating basic acoustic structures, so as to form and transmit new information). There are probably degrees of complexity in the combinatory potentiality of an acoustic communication system which may be signs of complexity of the system itself. Animal vocabularies are rather poor and finite, the richest one having, as far as we know today, only thirty to fifty different signals.

Destination

The auditory system, which is the first physiological step in message reception, depends upon a series of functionally and anatomically complex organs. The ear does not present a unityof development. The external, middle, and innerears differ as much by their embryological origins as by their chronological formation, and the three elementary germ layers of the first embryonic stages all play a role.

In vertebrates, the bilateralorgans which perceive only excitations of a mechanical origin are always grouped in the head periphery, while, on the other hand, in invertebrates the system is composed of many receptors scattered over the body. These are mechanoreceptors, many of which are of the hair sensilla type. In vertebrates there is a pressure and timeanalyzer system which allows detection of vibrations as well as perception of spatial positionand inertia. The reception organs, called stato-acoustic, can reach a great degree of sensitivity in higher orders. We have in the animal kingdom quite complete morphological data for a large number of receptor apparatuses, their peripheries, the nerve channels, and the higher centers of the diencephalon.

On the other hand, physiological theories of hearing are neither as definitive nor as satisfying. The phenomenon of nerve coding, by which information is transmitted to nerve centers, is still not clear. Besides, if hearing is taken as the basic physiological theme, message integration, as far as communicationis concerned, is still only a part of behavioral studies. We mean here to summarize only some of the many aspects of hearing. The comparison that generally can be made between the frequency hearing range of the animal (whether it be obtained by electrophysiology or conditioning) and the frequency spectrum of the sound signal itemits shows that, in most cases, hearing ability extends much beyond the frequency field belonging to the species' acoustic communication. Actually, focusing our attention strictly on the frequency field leads to a narrow conception ofauditory characteristics, since the hearing apparatus is a time gauge much more than it is a Fourier's series analyzer and especially measures the intensity-time relationship (Pimoniv, 1962).

Another essential point is the notion of feedback between the hearing system and the phonation system. Phonation controlby hearing exists in practically all the animalkingdom starting from the evolutionary stage when the animal is capable of choice in the motives of his vocalization. There is absolute proofof this in studies done on some deafened birds (Konishi, 1964; Mulligan, 1964) and in Kaspar-Hauser-raised birds. In comparing messagereception and spatial localization to emission systems, a second important point should be noted. "Binaurality" of the auditory system is currently found in the animal kingdom, but phonatory emission systems are often unpaired. Finally, it is to be remembered that message receptionthrough the auditory canal is possible only in the presence of a central motivation which permits an objective selection of a given signal from the background noise. This is even possible during a partial loss of consciousness, as during sleep. There are many classical illustrations of this fact: after copulation many female animals are no longer attracted by the male signal which they nevertheless perceive; a satisfiedanimal will not react to a food call.

I believe that lack of understanding of this problem is one of the main causes for the intellectual gap which exists between hearing physiologists and behavior specialists. Central motivation is thus one of the main keys to a semantic integration of the message incommunication. The starting point of the systemis, of course, audition, but at a sensitivity level controlled by a number of physiological functions. Mere observation of animal behavior, although useful, is not sufficient for a correctinterpretation of the message problem. It is necessary to be able to control the stimulus and thus follow eventual changes in the reactions, which can be the only criteria of the semantic value of the message. These reactions may be oriented movements, attraction, repulsion, flying, or running away and phonoresponse—all of which have been widely used by scientists experimenting on many animal species. One of the first experiments of this type which may be cited is that of Regen (1914) who, fifty years ago, attracted a female cricket to the telephone call of a male, and most recently the experiments of Lang and Smith (1965) and Bastian (1967), who, working along these lines, studied an eventual passage of information between two separated porpoises. It is in using such techniques that the semantic and syntactic value of a message or its structure may begin to be studied. This does not mean we should thus be able to understand all messages, because in any case the experimenter does not have access to that which we may term the innermost conscience of the animal. Thus, many real emotional factors are not perceived, since they are not manifested by coordinated motor reactions. Probably more specific information will be obtained in the future from chronic animals equipped with radio telemetric transmitters.

In some species reaction to amessage is innate, automatic, and often of the reflex type. This is true particularly in invertebrates and in relation to alarm and distress messages in almost the entire animal kingdom. For this reason, these messages are often referred to as stimulus-sign, releaser, etc. In otherkinds of messages, part of the information is learned. This is notably the case in individual recognition in some birds and mammals. This learning is associated with the receiver's generalability to learn the form of certain acoustic messages which he then integrates into his own communication vocabulary. The importance of signal-form learning in the acoustic signals of some species was first demonstrated by Koehler's German school (Sauer, Messmer, Thielcke), using the Kaspar Hauser technique of raising birds in acoustic isolation, and by Thorpe in England. According to Marler (in press), who has analytically reviewed this question, learning occurs at different ages or life periods, and isrelated in a certain way to Lorenz's imprinting. Learning seems to be important at differentlevels such as the group, species, family, couple. Although the signal may be locally modifiedin its physical characteristics, the semantic information usually remains unchanged. There maybe an exception to this in signals newly learned at a certain stage in the individual's life which were not part of his prior vocabulary. Nicolaï (1959) describes such instances in the female bullfinch who learned certainsignals from her first husband which were different from those she learned in her family surroundings. Another example is the robin's instantaneous imitation of an invader's signal. Thus, the invaded robin embodies its designation: "I am talking toyou, invader of the moment" (Bremond, 1966).

Codification

The study of animal acoustic communication may be considered as an outgrowth of semiology. Therefore, Sebeok's term, "zoosemiotics" (1965), taken from de Saussure's (1916) "semiotics," seems quite adequate, since it is a study of a system based on signs whatever their origin may be. In most of the animal kingdom, a particular signal corresponds to a definite situation or to a given or innate experience according tothe repertoire of the species. When there is a variant, it has not been proved whether it is only related to the physical aspect of the signal or whether it also concerns the semantic content. From a semiological point of view, if specific signals are total, complete messages, any mutilation destroys the meaning. It seems very probable, then, that according to the species the meaning or semantic content is rather strictly related to a series of phonic productions which would be distinctive, nonsignificant units. This makes most animal acoustic interaction truly semiotic, constituting a real sign communication system.

Language functioning is related to the fact that language may be decomposed into discontinuous, differential, or numerable discrete units. In the animal kingdom, can the total message be analyzed in smaller units, or is it itself the smallest communication unit? The message seems to lack first articulation units (since the smallest significant unit in eachcase is the indecomposable total message) and functions by second articulation units (since the least alteration of successive phonic productions changes, and probably destroys, the message in many species). Therefore, animal acoustic communication seems to be essentially a code composed of signals having as corollaries fixity of content, message invariability in relation to a single situation, irresolvability in the nature of what is said, and unilateral transmission. The significance of a statement is determined by the situation in which the speaker emits the statement, as well as by the auditor's behavioral response to that statement. Even in animals, relayed communications exist, at least at certain levels of communication such as those of birds and probably porpoises.

It must be remembered that the richness of a communication code is not composed exclusively of acoustic signals, and that is why a large number of animals have a more complete, complex code constituting a large numberof other signal forms. It is in combining theirusage that information corresponding to the species' social relationships can be transmitted. The more evolved an animal species, the more complex is its code of acoustic signals. However, some of these species—birds, bees, and porpoises— merit special attention in the light of recent studies. It is yet too early to judge the importance of results obtained from experiments on these animals, but it is possible that, in view of future experiments, the idea of communication systems, which is at present a little too mechanical, may have to be revised. Even if in the animal kingdom an ontogenesis and anatomical phylogeny of the phonatory and auditory organs can be found, thus increasing our knowledge, the enormous differences concerning acoustic communication systems between species must be sought essentially at the level of the functioningof the brain, which, after all, remains the principal organ directing the essential psychological activity of all animals.

References

Akermann, B., B. Anderson, E. Fabricius, and L. Svensson. Observations on the central regulation of body temperature and of food and water intake in the pigeon, Columba livia. Acta Physiol. Scand., 1960, 50, 328.

Annandale N., J. Coggin Brown, and F. H. Gravely. The Limestone Caves of Burma and the Malay Peninsula, Myriapoda. J. Asiat. Soc. Bengal (Calcutta), (N.S.), 1913, 9, 415.

Armstrong, E. A. A. Study of Bird Song. Oxford Univ. Press, 1963, London.

Aronson, L. R., and G. K. Noble. The sexual behavior of Anura. II. Neural mechanisms controlling mating in the male leopard frog, Rana pipiens. Bull. Amer. Nat. Hist., 1945, 86, 87-139.

Bastian, J. Social communication content in the pulse outside the echolocation (in press).

Bastian, J. The transmission of arbitrary environmental information between Bottlenose Dolphins. In Les systèmes sonars animaux, R.-G. Busnel, ed., Frascati (Italy), Gap, France, 1967, pp. 803-873.

Blair, F. W. Call difference as an isolation mechanism in Southwestern toads (genus Bufo). Texas J. Sei., 1956, VII, no. 1, 87-106.

Blair, F. W. Mating call and stage of speciation of two allopatric populations of spadefoots (Scaphiopus). Texas J. Sei., 1958, X, no. 4, 484-488.

Blair, F. W. Isolating mechanisms and interspecies interactions in anurans. Proc. XVI Int. Congr. Zool. (Washington, D.C.), 1963, 315-319.

Borror, D. J. Intraspecific variation in passerine bird songs. The Wilson Bull., 1961, 73, no. 1, 57-78.

Borror, D.J. Songs of finches (Fringillidae) of Eastern North-America. The Ohio J. Sei., 1961, 61, no. 3, 161-174.

Bremond, J.-C. Parametres physiques du chant de défense territoriale du rougegorge (Erithacus rubecula L.) C R. Acad. Sa., 1962, 254, 2072-2074.

Bremond, J.-C. Thesis: Recherches sur la sémantique et les éléments physiques déclencheurs de comportements dans les signaux acoustiques du rougegorge (Erithacus rubecula L.), Fac. Sei. Paris, OA 1120, 1-112, Fig. 1-24, 1966.

Brown, J. L. Vocalization evoked from the optic lobe of a songbird. Science, 1965, 149, no. 3687, 1002-1003.

Brown, J. L., and R. W. Hunsperger. Neurothology and motivation of agonistic behaviour. Anim. Behav., 1963, II, no. 4, 439-448.

Brückner, J. H. Untersuchungen zur Tiersoziologie, insbesondere zur Auflösung der Familie. Z. Tierpsychol., 1933, 128, 1-105.

Bühler, K. Sprachtheorie. Fisher-Verlag, Iena, 1934.

Busnel, R.-G., ed. Colloque sur l'acoustique des Orthoptères. I.N.R.A., Jouy-en-Josas (5-8. avril 1954), I.N.R.A., ed., Paris, 1955.

Busnel, R.-G., ed. Acoustic Behaviour of Animals. Elsevier, 1963, Amsterdam.

Busnel, R.-G., and J.-C Bremond. Etude préliminaire du décodage des informations contenues dans le signal acoustique territorial du rouge-gorge (Erithacus rubecula L.). C. R. Acad. Sei., 1961, 252, 608-610.

Busnel, R.-G., and J.-C. Bremond. Recherche du support de l'information dans le signal acoustique de défense territoriale du rouge-gorge (Erithacus rubecula L.). C. R. Acad. Sei., 1962, 254, 2236-2238.

Busnel, M.-C., and R.-G. Busnel. La directivité acoustique des déplacements de la femelle d'Occanthus pellucens Scop. C. R. Soc. Bio., 1954, CXLVIII, 830-833.

Busnel, R.-G., and B. Dumortier. Etude des caractères du signal du sifflet de Galton provoquant la phonotaxie de la femelle d'Ephippiger bitterensis. C. R. Soc. Bio., 1954, CXLVIII, 1751-1754.

Busnel, R.-G., and B. Dumortier. Phonoréactions du mâle d'Hyla arborea à des signaux acoustiques artificiels. Bull. Soc. Zool. France, 1955, LXXX, no. 1, 66-69.

Busnel, R.-G., B. Dumortier, and M.-C. Busnel. Recherches sur le comportement acoustique des Ephippigères (Orthoptères, Tettigoniidae). Bull. Bio. Fr. et Belg, 1956, XC, fasc. 3, 219-286.

Busnel, R.-G., A. Dziedzic, and S. Andersen. Rôle de l'impédance d'une cible dans le seuil de sa détection par le système sonar du Marsouin Phocaena phocaena. C. R. Soc. Bio., 1965a, 159, 69-74.

Busnel, R.-G., A. Dziedzic, and S. Andersen. Seuils deperception du système sonar du Marsouin Phocaena phocaena L. en fonction du diamètre d'un obstacle filiforme. C.R. Acad. Sei., 1965b, 260, 295-297.

Busnel, R.-G. and J. Giban. La protection acoustique des cultures et autres moyens d'effarouchement des oiseaux. C. R. de l'I.N.R.A. (26.-27. nov. 1958) I.N.R.A., ed., Paris.

Busnel, R.-G. and J. Giban, ed. Le problème des oiseaux sur les aérodromes. C. R. de l'I.N.R.A., Nice (25.-27. nov. 1963), I.N.R.A., ed., Paris, 1965.

Busnel, R.-G., J. Giban, Ph. Gramet, H. Frings, M. Frings, andj. Jumber. Inter-spécificité des signaux acoustiques ayant une valeur sémantiqué pour des Corvidés européens et nord-américains. C. R. Acad. Sei., 1957, 245, 105-108.

Busnel, R.-G., and W. Loher. Sur l'étude du temps de la résponse du stimulus acoustique artificiel chez les Chorthippus et la rapidité de l'intégration du stimulus. C. R. Soc. Bio., 1954. CXLVIII, 862-865.

Capranica, R. R. Evoked vocal response of the Bullfrog. J.A.S.A., 1964, 36, no. 10, 2007.

Capranica, R. R., M. Sachs, and M.J. Murray. B. Auditory discrimination in the Bullfrog. XVI. Comm. Biophys., 1964, no. 71, 245-249.

Cherry, C. On Human Communication. Wiley, New York, 1957.

Cloudsley-Thompson, J. L. A new sound-producing mechanism in Centipedes. Entom. Month. Mag., 1961, 96, 110-113.

Curio, E. Verhaltensstudien am Trauerschnäpper. Z. Tierpsychoi, 1959, suppl. 3, 1-118.

Delgado, J. M. R. Free behaviour and brain stimulation. Int. Rev. Neurobiol, 1964, 6, 349-449.

Dijkgraaf, S. Over een merkwaardige functie van den gehoorzin bij vleermuizen. Ned. Akad. v. Wetensch., Afd. Natuurk., 1943, LII, no. 9, 3-8.

Dijkgraaf, S. Die Sinneswelt der Fledermäuse. Experientia, 1946, II, no. 11, 1-31.

Dijkgraaf, S. Sinnesphysiologische Beobachtungen an Fledermäusen. Acta Physiol. Pharmacol. Neerl., 1957, 6, 675-684.

Dumortier, B. Etude expérimentale de la valeur interspécifique du signal acoustique chez les Ephippigères et rapport avec les problèmes d'isolement et de maintien de l'espèce (Orthopt., Ephippigeridae). Ann. Epiphyties, 1963a, 14, no. 1, 5-23.

Dumortier, B. Morphology of sound emission apparatus in Arthropoda. Acoustic Behaviour of Animals, Elsevier, 1963b, Chapter 11, 277-345.

Dumortier, B. The physical characteristics of sound emissions in Arthropoda. Acoustic Behavior of Animals, Elsevier, 1963c, Chapter 12, 346-373.

Dumortier, B. Ethological and physiological study of sound emission in Arthropoda. Acoustic Behaviour of Animals, Elsevier, 1963d, Chapter 21, 583-654.

Esch, H. Uber die Schallerzeugung beim Werbetanz der Honigbiene. Z. Vergl. Physiol., 1961, 45, 1-11.

Esch, H. Uber die Auswirkung der Futterplatzqualität auf die Schallerzeugung im Werbetanz der Honigbiene. Dtsch. Zool. Ges., 1962, 302-309.

Esch, H., and I. Esch. Sound: An element common to communication of stingless bees and to dances of the honey bee. Science, 1965, 149, no. 3681, 320-321.

Falls, J. B. Properties of bird-song eliciting responses from territorial males. Proc. XIII Int. Ornithol. Congr. Ithaca, 1962, June 17-24, 1, 259-271.

Friedmann, H. The honey guides. U.S. Nat. Mus. Bull., 1955, 208, 1-279.

Frings, H., and M. Frings. Auditory and visual mechanisms in food-finding behaviour of the Herringgull. Wilson Bull., 1956, 67, 155-170.

Frings, H., M. Frings, J. Jumber, and R.-G. Busnel. Reactions of American and French species of Corvus and Larus to recorded communication signals tested reciprocally. Ecology, 1958, 39, no. 1, 126-131.

Fulton, B. B. Inheritance of song in hybrids of two subspecies of Nemobius fasciatus (Orthoptera). Ann. Ent. Soc. Amer., 1933, XXVI, no. 2, 368-376.

Fulton, B. B. Experimental crossing of subspecies in Nemobius (Orthoptera:Gryllidae). Ann. Entomol. Soc. Amer., 1937, XXX, no. 2, 201-207.

Fulton, B. B. Speciation in the field cricket. Evolution, 1952, VI, no. 3, 283-295.

Gould, E. Evidence for echolocation in the Tenrecidae of Madagascar. Proc. Am. Phil. Soc., 1965, 109, no. 6, 352-360.

Griffin, D. R. Echolocation by blind men, bats and radar. Science, 1944, Dec. 29, 100, no. 2609, 589-590.

Griffin, D. R. Bat sounds under natural conditions, with evidence for echolocation of insect prey.J. Exp. Zool., 1953a, Aug., 123, no. 3, 435-466.

Griffin, D. R. Acoustic orientation in the oil bird, Steatornis. Nat. Acad. Sei., 1953b. Aug., 39, no. 8, 884-893.

Griffin, D. R. Listening in the Dark. Yale Univ. Press, 1958.

Griffin, D. R., and A. D. Grinnel. Ability of bats to discriminate echoes from louder noise. Science, 1958, 128, 145-147.

Griffin, D. R., J. H. Friend, and F. A. Webster. Target discrimination by the echolocation of bats. J. Exp. Zool., 1965, March, 158, no. 2, 155-168.

Griffin, D. R., F. A. Webster, and Ch. R. Michael. The echolocation of flying insects by bats. Anim. Behav., 1960, VIII, no. 3-4, 141-154.

Grimes, L. Antiphonal singing and call notes of Laniarus barbarus. Ibis, 1966, 108, 122-126.

Gwinner, E., and J. Kneutgen. Uber die biologische Bedeutung der "zweckbedienlichen" Anwendung erlernter Laute bei Vögeln. Z Tierpsychol., 1962, 19, 692-696.

Haskell, P. T. Insect Sounds. H. F. Sc C. Witherby, Ltd., 1961, London.

Heinroth, O., and K. Heinroth. Verhaltensweisen der Felsentaube (Haustaube). Columbia livia livia L. Z. Tierpsychol., 1949, 6, 153-201.

Hess, W. R. Stammganglien-Reizversuche. Verh. Dtsch. physiol. Ges. (Sept. 1927). Ber. ges. Physiol., 1928, 42, 554-555.

Hess, W. R. Das Zwischenhirn. 2nd Edition, Schwabe, 1954.

Holst, E. von and U. von Saint Paul. Vom Wirkungsgefüge der Triebe. Naturwiss., 1960, 18, 409-422.

Holst, E. von, and U. von Saint Paul. On the functional organisation of drives. Anim. Behav., 1963, 11, no. 283, 1-20.

Hüber, F. Verhaltensstudien am Männchen der Feldgrille (Gryllus campes tris L.) nach Eingriffen am Zentralnervensystem. Verhandl. Dtsch. Zool. Ges., Freiburg, 1952, 138-49.

Hüber, F. Stiz und Bedeutung nervöser Zentren für Istinkthandlungen beim Männchen von Gryllus campestris L. Z. Tierpsychol., 1955, 12, no. 1, 12-48.

Hüber, F. Auslösung von Bewegungmustern durch elektrische Reizung des Oberschlundganglions bei Orthopteren (Saltatoria: Gryllidae, Acrididae). Verhandl. Dtsch. Zool. Ges., 1959, 248-269.

Hüber, F. Experimentelle Untersuchungen zur nervösen Atmungsregulation der Orthopteren (Saltatoria:Gryllidae). Z. Vergl. Phsyiol., 1960, 43, 359-391.

Hüber, F. Untersuchungen über die Funktionen des Zentralnervensystems und insbesondere des Gehirnes bei Fortbewegung und der Lauterzeugung der Grillen. Z. Vergl. Physiol., 1960, 44, 60-132.

Hüber, F. Central nervous control of sound production in crickets and some speculations on its evolution. Evolution, 1962, XVI, no. 4, 429-442.

Hüber, F. The role of the central nervous system in Orthoptera during the coordination and control of stridulation. Acoustic Behaviour of Animals, Elsevier, 1963, Chapter 17, 440-484.

Huchtker, R., and J. Schwartzkopff. Soziale Verhaltensweisen bei hörenden und gehölosen Dompfaffen (Pyrrhula pyrrhula). Experientia, 1958, XIV, no. 3, 106-111.

Hunsperger, R. W. Affektreaktion auf elektrische Reizung im Hirnstamm der Katze. Helv. Physiol. Pharmacol. Acta, 1956, 14, 70-92.

Kalmus, H. Analogies of language to life. Language and Speech, 1962, 5, no. 1, 15-25.

Kanai, T., and S. C. Wang. Localization of a mechanism for vocalization in the hypothalamus and its descending pathways in the brain stem of the cat. Feder. Proc., 1961, 20, 331.

Kanai, T., E. A. Day, and C. B. Weld. A brain stem mechanism for vocalization in the cat. Feder. Proc., 1964, 23, no. 2, 1 p.

Karplus, J. P., and A. Kreidl. Gehirn und Sympathicus. Pflügers Arch. ges. Physiol. Menschen u. Tiere, 1909, 129, p. 138, 1928, 219, pp. 613-618.

King, J. A. Social behavior, social organization and population dynamics in a black-tailed prairiedog down in the black hills of South Dakota. Contr. Lab. Vertebrate Bio., 1955, no. 67, 1-123. Univ. Michigan Press, Ann Arbor.

Konishi, M. Effects of deafening on song development in two species of juncos. The Condor, 1964, 66, no. 22, 85-102.

Kulzer, E. Flughunde erzeugen Orientierungslaute durch Zungenschlag. Naturwiss., 1956, 43, 117-118.

Landois, H. Thierstimmen. Herdersche Verlagsbuchhandlung, Freiburg, 1874.

Lang, T. G., and H. A. P. Smith. Communication between dolphins in separate tanks by way of an electronic acoustic link. Science, 1965, 150, no. 3705, 1839-44.

Lanyon, W. E., and W. N. Tavolga, ed. Animal Sounds and Communication. A.I.B.S., 1958, Washington, DC.

Leroy, Y. Etude du chant de deux espèces de grillons et de leur hybride (Gryllus commodus Walker, Gryllus oceanicus Le Guillon, Orthoptères). C. R. Acad. Sei., 1963, 256, 268-270.

Leroy, Y. Transmission du paramètre Fréquence dans le signal acoustique des hybrides F, et P X F_b de deux Grillons: Teleogryllus commodus Walker et T. oceanicus Le Guillon (Orthopt., Ensifères). C. R. Acad. Sa., 1964, 259, 892-895.

Leroy, Y. Essai de definition du comportement hierarchique chez les grillons (Insectes, Orthopteres). C R. Acad. Sei., 1966a, 263, 1752-1754.

Leroy, Y. Signaux acoustiques, comportements et systématiques de quelques espèces de Gryllidés (Orthopt., Ensifères). Fac. Sei., Paris, 1966b, Pierre Fanlac Perigneux.

Lilly, J. C. Correlations between neurophysiological activity in the cortex and short-term behavior in the monkey. Bio. and Biochem. Bases of Behav., Harlow 8c Woolsey, Ed., Univ. Wisconsin Press, Madison, 1958a.

Lilly, J. C. Some considerations regarding basic mechanisms of positive and negative types of motivations. Am.J. Psychiatry, 1958b, 115, no. 6, 498-504.

Lilly, J. C. Vocal behavior of the bottlenose dolphin. Proc. Am. Phil. Soc., 1962, 106, no. 6, 520-529.

Lilly, J. C., and A. M. Miller. Sounds emitted by the bottlenose dolphin. Science, 1961a, 133, no. 3465, 1689-1693.

Lilly, J. C., and A. M. Miller. Vocal exchanges between dolphins. Science, 1961b, 134, no. 3493, 1873-1876.

Lilly, J. C., and A. M. Miller. Operant conditioning of the bottlenose Dolphin with electrical stimulation of the brain.J. Comp, and Physiol. Psychol., 1962, 55, no. 1, 73-79.

Lorenz, K. Beiträge zur Ethologie sozialer Corviden. J. Ornithol., 1931, LXXIX, Heft 1, 67-127.

Lorenz, K. Der Kumpan in der Umwelt des Vogels: Der Artgenosse als auslösendes. Moment sozialer Verhaltensweisen.J. Ornithol., 1935, 83, 137-213; 289-413.

Lorenz, K. Vergleichende Bewegungsstudien an Anatiden. J. Ornithol., 1941, 89, no. 3, 194-294.

Lorenz, K. The comparative method in studying innate behavior patterns. Symp. N.Y. Soc. exp. Bio., 1950, 4, 221-268.

Marler, P. The voice of the chaffinch and its function as language. Ibis, 1956, 98, 231-261.

Marler, P. Specific distinctiveness in the communication signals of birds. Intern. J. Comp. Ethol., XI, Pt. I, 13-39, 1957.

Marler, P. Bird song and mate selection. Animal Sounds … Comm., Am. Inst. Bio. Sei. (A.I.B.S.), 1960, no. 7, 348-367.

Marler, P. Inheritance and learning in the development of animal vocalizations. In: Communicative Behavior in Animals (in press).

Marler, P., M. Kreith, and M. Tamura. Song development in hand-raised Oregon juncos. The Auk, 1962, 79, 12-30.

Marler, P., and M. Tamura. Culturally transmitted patterns of vocal behavior in sparrows. Science, 1964, 146, no. 3650, 1483-86.

Messmer, E., and I. Messmer. Die Entwicklung der Lautäusserungen und einiger Verhaltensweisen der Amsel (Turdus merula merula L.) unter natürlichen Bedingungen und nach Einzelaufzucht in schalldichten Räumen. Z. Tierpsychologie, 1956, 13, no. 3, 341-441.

Möhres, F. P., and E. Kulzer, Über die Orientierung der Flughunde. Z. Vergl. Physiol., 1956, 38, 1-29.

Mulligan, J. A. Physical analysis of variation and development in the song of Melospiza melodia. University Microfilm, Inc., Ann Arbor, 1964.

Nicolai, J. Familientradition in der Gesangentwicklung des Gimpels (Pyrrhula). J. Ornithol., 1959, 100, 39-46.

Nicolai, J. Der Brutparasitismus der Viduinae als ethologisches Problem. Z Tierpsychologie 1964, 21, no. 2, 129-204.

Norris, K. S., ed. Whales, Dolphins and Porpoises. U. of California Press, 1966, Berkeley.

Norris, K. S., and W. E. Evans. Variations in porpoise echolocation signals. Symp. Marine Bio-Acoustics, 1966, in press.

Norris, K. S., J. H. Prescott, P. V. Asadorian, et al. An experimental demonstration of echolocation behavior in the porpoise, Tursiops truncatus (Montagu). Bio. Bull., 1961, 120, no. 2, 163-176.

Perdeck, A. C. The isolating value of specific song patterns in two sibling species of grasshoppers (Chorthippus brunneus Thunb and C. biguttulus L.) Behaviour, Netherl., 1958, 12, no. 1-2, 1-75.

Phillips, R. E. Wildness in the mallard duck: Effects of brain-lesion and stimulation on "escape behaviour" and reproduction.J. Comp. Neurol., 1964, 122, 139-155.

Pierce, J. R. Symbols, Signals and Noise: The Nature and Process of Communication. Harper 8c Bros., N.Y., 1961.

Pimoniv, L. Vibrations en régime transitoire. Dunod, 1962, Paris.

Prevost, J. Ecologie du manchot empéreur. Actual. Sei. et Industr., Hermann Ed., Paris, 1961, no. 1291.

Randall, W. L. The behavior of cats (Felis catus) with lesions in the caudal mid-brain region. Behaviour, 1964, 23, 107-139.

Regen, J. Über die Anlockung des Weibchens von Gryllus campestris L. durch telephonisch übertragene Stridulationslaute des Männchens. Pflügers Arch. ges. Physiol. Menschen u. Tiere, 1914, 155, 193-200.

Rougeul et Assenmacher. Cited by Buser in Brain Mechanisms and Learning. Blackwell Ed., Oxford, 1961.

Ruben, R. J., D. Warfield, and R. Glackin. Word discrimination in cats. J.A.S.A. 1965, 1204-1205.

Sauer, F. Le développement des signaux sonores de la fauvette (Sylvia c. communis). Comparaison entre des individus sauvages et des individus élevés isolés en chambre sourde depuis l'oeuf ou plus tardivement. Z. Tierpsychol., 1954, 11, 10-93.

Sauer, F. Behaviour of the young garden warblers. J. Ornithol., 1956, 97, 156-189.

Saussure, F. de. Cours de linguistique générale. Paris, 1916.

Schleidt, M. W., M. Schleidt, and M. Magg. Störung der Mutter-Kind-Beziehung bei Truthühnern durch Gehörverlust. Behaviour, 1960, 16, no. 3-4, 254-260.

Schmidt, R. S. Central mechanisms of frog calling. Behaviour, 1965, 26, 251-285.

Scott, J. P., and J. L. Fuller. Genetics and the Social Behavior of the Dog. U. of Chicago Press, 1965, Chicago.

Sebeok, Th. A. Coding in the evolution of signalling behavior. Behavioral Sei., 1962, 7, no. 4, 430-442.

Sebeok, Th. A. Zoosemiotics: Juncture of semiotics and the biological study of behavior. Science, 1965, 147, no. 3657, 492-493.

Signoret, J. P., F. du, Mesnil du Buisson, and R.-G. Busnel. Rôle d'un signal acoustique de verrat dans le comportement réactionnel de la truie en oestrus. C. R. Acad. Sei., 1960, 250, 1355-1357.

Silva, C., J. Estable, and P. Segundo. Cited by Segundo in Brain Mechanisms and Learning. Blackwell Ed., Oxford, 1961.

Tavolga, W. N., ed. Marine Bio-acoustics. Proc. Symp., Lerner Marine Lab., Bimini, Bahamas. Pergamon Press, Ltd., 1964, Oxford.

Tavolga, W. N. Review of Marine Bio-acoustics (State of the Art: 1964), U.S. Nav. Traing. Device Center, 1965, Navtradevcen 1212-1, 1-100.

Tavolga, W. N., ed. Marine Bio-acoustics. Pergamon Press, Ltd., New York (in press).

Tembrock, G. Probleme der Bio-Akustik. Wiss. Zeitschr., Humboldt-Univ., Berlin, 1958/59, 4-5, 573-587.

Tembrock, G. Stridulation und Tagesperiodik bei Cerambyx cerdo. Zool. Beitr., Neue Folge (1960), 5, 419-441.

Tembrock, G. Tierstimmen. Die neue Brehm-Büchere: Ziemen Verlag, Wittenberg-Lutherstadt, 1959.

Thielcke, G. Stammesgeschichte und geographische Variation des Gesanges unserer Baumläufer (Certhia familiaris und c. braehydaetyla) Z. Tierpsychol. (1961), 18, no. 2, 188-204.

Thielcke, G. Versuche mit Klangattrappen zur Klärung der Verwandtschaft der Baumläufer (Certhia familiaris L.), (C. Braehydaetyla Brehm), und (C. Americana Bonaparte). J. Ornithol., 1962, 103, no. 2-3, 26-71.

Thorpe, W. H. The analysis of bird song. Proc. Roy. Instn., 1954, 35, no. 161, 1-13.

Thorpe, W. H. The process of song-learning in the chaffinch as studied by means of the sound spectrograph. Nature, 1954, 73, no. 4402, 465-469.

Thorpe, W. H. The learning of song patterns by birds, with especial reference to the song of the chaffinch (Fringilla coelebs Ibis), 1958, 100, 535-570.

Thorpe, W. H. Further studies on the process of song learning in the chaffinch (Fringilla coelebs gengleri). Nature, 1958, 182, no. 4635, 554-557.

Thorpe, W. H. Bird-song. Cambridge Univ. Press, 1961, New York.

Thorpe, W. H. Antiphonal singing in birds as evidence for avian auditory reaction time. Nature, 1963, 197, no. 4869, 774-776.

Thorpe, W. H. Aggression and fear in the normal sexual behavior of some animals. Pathol. Treatm. Sex. Deviation, 1964, 3-23.

Thorpe, W. H., and M. E. W. North. Origin and significance of the power of vocal imitation: with special reference to the antiphonal singing of birds. Nature, 1965, 208, no. 5007, 219-222.

Tinbergen, N. Comparative studies of the behaviour of gulls (Laridae): a progress report. Behaviour, 1959, XV; no. 1-2, 1-70.

Tinbergen, N. The evolution of behaviour in gulls. Scient. Am., 1960, no. 12, 118-130.

Weeden, J. S., andj. B. Falls. Differential responses of male ovenbirds to recorded songs of neighboring and more distant individuals. The Auk, 1959, 76, 343-351.

Wenner, A. M. Sound production during the waggle dance of the honey bee. Anim. Behav., 1962, 10, no. 1-2, 79-95.

Wenner, A. M. Sound communication in the honey bee. Scient. Am., 1964, 181, 1-7.