FROM CURRENT LITERATURE

ULTRASONIC ECHOLOCATION IN BATS

The ability of bats to orient themselves unerringly while flying in complete darkness has long puzzled everyone who has carefully observed the behavior of these animals. Bats can fly, without touching with their wings, through a small opening in a partition. They fly around wires stretched across a room without touching them; they never bump into walls on the premises. While in flight, bats open their mouths wide, which helps them catch midges, flies, and butterflies (on which they mainly feed), but there is reason to think that they nevertheless in some way notice, in the dark, insects flying in the air and direct themselves toward them. It was shown long ago by the work of a number of zoologists that when one or both ears are blocked or tied up, bats lose the ability to orient themselves in flight and collide with obstacles.

In 1920 Hartridge¹ put forward the hypothesis that, in flight, bats emit “short-wave” sounds and, listening to the echo of these sounds, orient themselves while taking into account the interval of the echo reflected from objects.

This question was brought into complete clarity by the experimental work of Galambos and Griffin²˒³ in 1941–1942 in America and by Hartridge’s critical article in 1945. Galambos and Griffin, using special sound receivers, found that bats produce four kinds of sounds:

1. A buzzing—low-frequency sound that is audible if one is very close to the animal.
2. Separate “clicks,” which are also audible only at close range.
3. Tones with a frequency of about 7000 hertz, lasting approximately 1/4 sec and irregular, although often repeated.
4. Ultrasounds with frequencies from 30 to 70 thousand hertz; the most frequently encountered frequencies are 40–55 thousand hertz (wavelengths from 0.6 to 0.85 cm). Individual ultrasonic “cries,” or pulses, have a duration of only about 0.01 sec. At rest, a bat “cries” 5–10 times per second. When it begins to fly, the repetition frequency of the pulses increases to 20–30 per second; when approaching an obstacle, the pulse frequency rises to 60 per second.

The buzzing and clicks are always accompanied by ultrasounds emitted by the bat. Most likely, these low-frequency sounds are the result of a sharp interruption of the airflow when the glottis opens and closes at the moment of the ultrasonic “cry.” Anatomical studies of the larynx of bats indicate its small size and increased strength; the cartilages to which the vocal cords and the muscles that tense them are attached are, in bats, to a considerable extent ossified. This makes it possible to assume that very high tensions of the cords are possible, which are necessary for producing ultrasounds. It is possible, however, also to suppose that the ultrasonic oscillations are generated like a whistle, as in the well-known whistles of Galton or Hartmann; the resonating cavity may in this case be formed, for example, in the laryngeal

...ventricle between the true and false vocal cords. Galambos’s experiments with Griffin showed that when the mouths and nostrils were sewn shut, bats were not able to emit ultrasound: thus it is clear that the ultrasound generator is located in the cavity of the throat or mouth. Very important for clarifying the essence of the matter is the work of Galambos5, who proved that bats are indeed capable of perceiving ultrasounds. This author placed his experiments on the study of the so-called microphone effect of the cochlea in bats and detected it at frequencies from 10,000 to 100,000 hertz. The microphone effect consists in the fact that, under the action of sound, a healthy cochlea with undamaged fibers of Corti generates alternating electrical potentials (of the order of microvolts) that correspond exactly in frequency to the sound vibrations acting upon it; in this experiment the animal is put to sleep with a narcotic, its middle-ear cavity is opened, and a thin electrode is applied to the cochlea. The presence of the microphone effect indicates only the normal functioning of the cochlea and the presence of vibrations of the fibers of Corti, but it still says nothing about whether the animal hears the sound or not. It is known that when the auditory nerve is cut or damaged the microphone effect persists, but the animal, of course, hears nothing. Galambos showed that when ultrasound is amplified the microphone effect reaches a maximum and then begins to decline. This is explained by the action of the protective mechanism of the ear—the tension of the muscle (tensor tympani), which makes the eardrum more elastic and weakens the sensitivity of the ear. Consequently, the central nervous system reacts to the perception of ultrasound and produces a protective reflex that weakens the effect of overly strong ultrasounds. By inducing paralysis of the musculature that tightens the eardrum, through injection of curare poison, it is possible to raise the potentials of the cochlea. The microphone effect in the ear of bats was observed both under the influence of ultrasounds emitted by a special magnetostriction generator and under the influence of the “cries” of other bats. Ultrasounds with frequencies from 10,000 to 60,000 hertz produce a very strong effect; at higher frequencies the sensitivity of the ear weakens.

What, then, are the physical foundations of the peculiar ultrasonic “echo sounder” or “radar” of bats? First of all, let us note that the animals emit ultrasounds with wavelengths on the order of 1 cm or somewhat less. Waves of this order can certainly give a distinct reflected diffraction pattern from small objects and even from wires several millimeters in diameter. Longer waves would, of course, make it possible to recognize only larger objects. Conversely, still shorter waves, although they would be better suited for a more detailed investigation of surrounding objects, are probably less advantageous for these purposes because of their strong absorption in air. It is interesting that hairs or fluffed-up cotton wool are not detected by bats. Sound is not reflected by them, evidently being completely absorbed in the porous mass.

Ultrasounds of such wavelengths as 1 cm can be emitted in sharply directed beams, like the light from a searchlight. The perception of ultrasound by a bat is carried out by means of two ears (with enormous auricles). It can also be sharply fixed in a definite direction, just as this is done by the sound receivers themselves. Thus, the time of arrival of the echo signal makes it possible to judge the distance of the object, and the binaural effect—its direction. The polar coordinates of the object are thereby found to be completely determined. The animal has the full possibility of localizing in darkness the obstacles surrounding it in flight. It has been noted that plugging one ear, which completely disrupts the ability to localize by the binaural effect, leads to a complete disruption of the ability of bats to fly freely in the dark. The ultrasonic “cry” of a bat lasts 0.01 sec. The echo from an object located at a distance of 1.7 m will return after...

...times, approximately after 0.01 sec. Consequently, objects lying closer than 1.7 m will give an echo that will be partly masked by the strong primary signal. More distant objects will give an echo only after the primary signal has ended and, consequently, they can be detected more distinctly.

With regard to the mechanism for determining distance, two hypotheses may be advanced. One may suppose that the bat emits a “cry” periodically, beginning it at the moment when it has finished perceiving the echo. In this case the repetition period of the “cries” \(T\) is related to the duration of the pulses \(\tau\) and the distance \(l\) to the obstacle by the relation:

\[ T = \tau = \frac{l}{c}, \]

where \(c\) is the speed of sound. One may also suppose that the animal achieves complete masking of the echo by the sound of the signal, selecting the pulse repetition frequency so that the echo sound is exactly overlapped by the signal sound. In the second case

\[ T = \frac{l}{c}. \]

Which of these hypotheses is correct is still difficult to judge.

The question arises whether a bat can recognize sounds reflected from a number of objects within its field of hearing. Here, possibly, it is helped by the ability to change the pulse repetition frequency. Thus, for example, suppose that there are two objects in the field of hearing—one at a distance of 1.7 m, giving an echo immediately after the end of the “cry,” and the other at a distance of 8 m, giving an echo after 0.05 sec. By selecting a pulse repetition frequency of 20 times per second, i.e., after 0.05 sec, the animal may arrange it so that the echo pulses from the more distant object fall exactly at the moments of the primary signals and are not noticeable; conversely, the echo pulses from the nearer object will be clearly perceptible. Let us note a remarkable property of the auditory apparatus: to reflexively reduce sensitivity under the action of a strong sound. This allows the animal to “block” the receiving (auditory) apparatus at the moments when powerful pulses are emitted by the voice transmitter (in exactly the same way as this is done in modern radiolocation installations—radars). Both here and there, probably, short waves, transmitter blocking, and pulse signals are used; moreover, in both cases a high pulse repetition frequency is used if it is necessary to locate closer objects (in radars, from 200 to 20,000 pulses per second are used).

With regard to signals with a tonal frequency of 7000 hertz, Hartridge suggests that bats use them for communication with one another; these are, so to speak, “conversational” frequencies.

S. N. Rzhevkin

LITERATURE

1. Hartridge, Journ. Physiol., 54, p. 54, 1920.
2. D. Griffin a. R. Galambos, Journ. Exp. Zool., 86 p. 481, 1941.
3. R. Galambos, a. D. Griffin, Journ. Exp. Zool. 89. p. 475, 1942.
4. H. Hartridge, Nature, Oct., 1945.
5. R. Galambos, Journ. Acoust. Soc. Am. 14 p. 41, 1942.