The seven families of Insectivora recognized at the present time (Erinaceidae, Soricidae, Talpidae, Tenrecidae, Solenodontidae, Chrysochloridae, Potamogalidae) are a zoological order that remains a highly rewarding field of investigation not only in communication problems. Their behavior patterns and the physiology of the senses of many of them—for example, most of the African and Asiatic shrews, the southern and east African golden moles, the Echinosoricini, etc.—have not yet been examined at all. These research complexes rather have been neglected in favor of mere taxonomic or anatomical studies. At present, only the modest beginnings of a comparative insectivorology are available, presumably since our knowledge of many of the species of Insectivora can be supported only by fragmentary reports and papers that are scattered throughout a diffuse literature on the subject.
One of the reasons for the current gap in our knowledge is that the Insectivora, which are placed at the root of the mammals' genealogical tree, are by no means an homologous group. Their phylogenetic age leads us to suspect an especially large number of long-extinct families and offshoots, whose fossil remains are more difficult to find and interpret than are those of larger animals, simply because of their small size. This also explains the differences noticeable among present-day insectivores. Many intermediary forms that might offer an explanation for and insight into modern species have not been discovered or are as yet incompletely known. Single, sharply distinguishing features have evolved at different times among insectivores, though they are considered the least specialized order of mammals.
The overall picture is, therefore, that of a mosaic: progressively differentiated, archaic, and conservative characteristics can be stated and have developed independently at various times (Thenius and Hofer, 1960). This of course is valid not only for anatomical but also for etiological details. It leads us to propose two objectives: parallel to a phenomenological presentation of the single species we should also undertake a program of more extensive criticism and survey. Admittedly this is difficult with Insectivora, which are tiny, often very fast-moving, and for the most part nocturnal animals. They live a great deal of their lives under cover and can only be observed with extreme difficulty. Our attempts to explain and understand the extraordinary importance of acoustic communication; their capacity for perceiving ultrasonics, noticeable within the whole order; the presence of numerous actively working glands that produce their own kind of communication; the sense of smell, illustrated anatomically by an enormously well-developed bulbus olfactorius; and finally the fact that almost all insectivores are amazingly well equipped tactilely reveal just how difficult a study is. In addition, whereas we humans tend to rely heavily on visual signs, these stimuli are only very poorly developed and appear to have the least importance in the communication systems of insectivores.
For this reason any study should be based to a far greater extent on the second objective stated above; the ensuing multiplicity of problems and varying points of view must simply be accepted as unavoidable. Another factor that makes our task so difficult is that most insectivores are solitary animals, which means we cannot assume with certainty that interspecific communication exists except during certain isolated periods in the animal's lifetime: e.g., among mother-young units, among members of the same litter, in courtship behavior, in mating behavior, and in encounters between conspecifics. Active intraspecific communication confines itself to defense against or warding off of predators; the ability to recognize allomones, the communication signals of non-conspecifics—especially when they are transmitted by predators—is known with some insectivores and can be presumed with others.
Happily we possess sufficient facts about Erinaceidae and Tenrecidae. The knowledge we have of solenodons is relatively encouraging when one considers the rarity of the material available for examination; whereas almost no such examinations have been conducted on Echinosoricini, Potamogalidae, and Chrysochloridae. A few good but in no way exhaustive works exist on some Talpidae and even on the semi-aquatic desmans; but the greatest part by far of the shrews, the family with the largest number of species, remains virtually unknown apart from a few excellent detailed works.
Since the smell epithelium has the same basic structure in all mammals, it had been assumed that because of their relatively small size, insectivores are not able to smell well (Müller-Velten, 1966). This assumption has been disproved to a large degree by the discovery, in a few insectivores at least, of the vomeronasal organ (Jacobson's organ), which has been recognized as an actively functioning and extremely precise sensory organ and which apparently corresponds to or derives from the phylogenetical old watertasting organ of fishes (Poduschka and Firbas, 1968). The ability to echolocate, which was proved to exist among insectivores in shrews and tenrecs and suspected in hedgehogs, Echinosoricini and solenodons, does not have any immediate relevance to communication in itself. We can, however, discuss it here since it has indirect relevance to the physiological and behavioral phenomena that may be important not only in echolocation but also in communication (Griffin, 1968). Echolocating animals must at any rate be capable of complex types of communication.
Because, apart from flying and gliding forms, the order Insectivora shows a radiation that is remarkable in its completeness, certain quantitative studies of the brains of the insectivores that have adapted to semi-aquatic life (Limnogale, Neomys, Nectogale elegans, Potamogalidae, Chimmarogale, and both Desmaninae) can also help us when we are studying higher forms. The comparison of results with those from more evolved species shows that this type of adaptation leads to the following modifications: regression of the olfactory centers, enlargement of the auditory centers and of those of the tactile trigeminal system, enlargement of the centers that are related to mo- tricity and correlation of motricity, and enlargement of neocortical regions, especially of the centers of association; on the whole, there is an increase in brain weight (Bauchot and Stephan, 1968).
The present lack of any comparative insectivorology seems to justify an attempt at a survey of the hitherto known patterns of communicative behavior divided into the four communication mechanisms: chemical, acoustic, visual, tactile. A survey of communication functions taking place for territorial, sexual, social, submissive, or aggressive motives would necessitate, in order to form any kind of firm basis for the arguments, many more specialized studies, the essence of which can only be hinted at here. Because of the phylogenetic age of insectivores, and because some of their features, as a result of their age, have remained largely unaffected by time, a thorough investigation of communicative ability in insectivores and the means of communication they have at their disposal ought to provide us with valuable clues in the study of certain higher orders of mammals. Finally, because of the central phylogenetic position occupied by the insectivores, this study may even serve as a basis for research in primatology.
FAMILY: ERINACEIDAE; SUBFAMILY: ERINACEINAE (SPINY HEDGEHOGS)
5 genera. Europe, Asia, Africa.
All hedgehogs are either nocturnal or crepuscular animals. Except for those forms that are capable of running very fast (Hemiechinus) they are seemingly slow-moving, cryptophile, solitary creatures. Group tendencies are nonexistent, except in mothers with litters. Males and females remain together only during the copulation period, about ten to twelve days.
The effectiveness of this form is questionable. When hedgehogs are slightly aroused, we notice a slow, continuous raising of the spines that normally lie flat when the animals are undisturbed. When aroused more intensely, the spines are raised with a jerk so sudden that a photograph taken with a shutter speed of 1/125 sec can only produce a blurred picture. It is possible that the erection of the spines may serve as a visual warning, analogous to the attempts of certain animals to make themselves appear as large as possible to impress adversaries—a very common phenomenon in animals. I have the notion, however, that this is not the case among hedgehogs. Rather, the erection of the spines is a means of enabling this attentive and cautious animal to adopt a defensive state. The mere increase in size of the silhouette never results in the retreat of another hedgehog; only taking a fighting position (after first pushing the erected spines on the head forward over the eyes and pressing the head against the substratum) and simultaneously emitting sharp snorts can cause a retreat. If the erection of the spines can be regarded as an optical threat at all, it is only in combination with acoustic signals.
Solitary females in estrus use the shoe of a human foster parent as a substitute for a mate when mounting; this seems to indicate that as far as mating behavior is concerned the appearance of the partner can be considered an optical signal, at least initially. There is, however, never more than the mere hint of mounting, which would mean that optical discrimination is very poor and that probably olfactory and/or auditory stimuli from the partner are necessary to trigger the ritually fixed behavior patterns of mounting proper. We should probably not consider the opalescent secretion from the eye glands as an optical signal; it is far more likely to be a chemical communication act which occurs in hedgehogs in only a rudimentary way; by contrast, tenrecs show the complete behavior pattern.
Threatened hedgehogs show threats of their own with a gaping mouth and thereby reveal their intention of biting, an optical signal that has been proved to exist in all insectivores studied so far.
The subfamily spiny hedgehogs (Erinacei- nae) has a vast repertoire of sounds at its disposal, ranging from quiet snorts to loud screams. They vary interspecifically, however, within the subfamily. A comparison reveals a concomitant series, not only of "dialects" with basic similarities that are comprehensible to all hedgehogs but also of completely different signals that are peculiar to and comprehensible to certain genera only, e.g., "schnalz" and squeaking noises in Algerian hedgehogs (Aethechinus algirus).
The contact or hunger signal of neonate hedgehogs is a quiet squeaking with high ultrasonic components, very low motorlike "tuckering," and smacking of the lips and/or tongue, the highest frequency components of which reach 37 kHz. Hungry young hedgehogs about three or four weeks old, having lost their own mother, run after strange females, squeaking in a similar way. Whether this is understood as a stimulus for adoption has not yet been tested fully.
The commonest acoustic signal in hedgehogs is a sharp, rapidly repeated snorting. When behavior turns aggressive, this normal quiet sound turns into a vigorous staccato and, when the animal is aroused further, into piercing cries very like those of hedgehogs in great pain or in the throes of death.
All acoustic signals in hedgehogs have strong ultrasonic components, which show easily visible harmonics on the sonagram (Poduschka, 1968, 1969). Frequencies lying within the human hearing range appear to be of little interest to hedgehogs, as filtering them out and playing them back to hedgehogs has proved. For technical reasons it has not yet been possible to discover exactly which harmonics present in the ultrasonic signals are actually relevant for the animals. Experiments indicate that the optimum hearing range for the Egyptian eared hedgehog (Hemiechinus auritus aegypticus) is about 40 kHz (Poduschka, unpubl.). Ravizza, Heffner, and Masterton (1969) have proved that this species is able to detect signals up to 45 kHz, presumably even up to 60 kHz.
This ability to perceive ultrasonics is of great advantage to hedgehogs, as indeed it is to all insectivores when searching for prey because they can hear and interpret the mechanical signals of insects, which form their main diet. Common European hedgehogs (Erinaceus europaeus), eared hedgehogs (Hemiechinus auritus), and long- eared hedgehogs (Hemiechinus megalotis), are also capable of roughly locating the source of ultrasonics emitted by insects.
Evidence has been found for echolocation in total darkness. An especially vigorous exhalation of breath occurred when the hedgehog was confronted closely with solid objects. It is still not clear whether the "rusty hinge creaking" emitted by young hedgehogs in dense cover, a sound used by many echolocating animals, which appears as a rattle on the sonagram, is also used by them as a means of echolocation.
Until they are about four weeks old, baby hedgehogs seek contact with the mother as often as possible, trying continually to crawl underneath her. When they touch her stomach hairs, which protrude from under the spiny coat of her sides, the mother reacts by raising her body slightly to allow the babies to reach her teats or her body's protecting warmth. If the baby hedgehogs are left alone, they crawl underneath one another; in the resulting pyramid formation, which is also common to many more highly evolved mammals, the most coveted position is at the bottom. This crawling underneath appears to be an intentional tactile stimulus, aimed at inducing the mother to raise her body; baby hedgehogs never react by lifting the body when touched in this way, but, on the other hand, they do not attempt to avoid their brothers and sisters when they do it.
In the courtship of Hemiechinus auritus the two partners nestle up against each other and touch each other's body with the side of the head by stretching it forward. This behavior pattern is largely identical with that of the tenrecs Hemicentetes and Microgale (Eisenberg and Gould, 1970) and that of Setifer and Echinops (Eisenberg and Gould, 1970; Poduschka, 1974a), whose numerous active gland areas also have communicative importance for the whole mating complex, but it is not like the courtship behavior of the more closely related common European hedgehog. It is not clear whether this can be explained by a mere retrogressive development of these glands in Hemiechinus, whereby only the tactile actions have remained. A female long-eared hedgehog in estrus rests her chin for a short period on the foot of the human keeper, which she accepts as a substitute, leading us to assume a tactile function of the intermandibular wart, from which sprout some sinus hairs. It is unknown whether there is an increase in the secretion of the glands at the roots of the intermandibular vibrissae which could possibly afford some kind of chemical communication in genuine mating behavior.
When the self-anointing process was understood, it was discovered that hedgehogs can "flehmen" and that they possess an active vomeronasal organ, which they use to define strongly irritant or new and unknown smells and tastes, particularly the individual stimuli encountered in their sexual lives (Poduschka and Firbas, 1968; Poduschka, 1970, 1973). The stimulating odors that enter the mouth while the hedgehog is flehming, as well as the tastes acquired by chewing or licking, are brought into contact with the spittle, which is increased in volume by chewing until it turns to foam. The odors and tastes travel via the ductus nasopalatinus to the sensitive epithelium of the vomeronasal organ, where they are registered and identified. Thus, both gaseous and solid stimuli induce involving a liquid medium. Olfaction in hedgehogs is therefore not merely a specialized form of chemical detection; the stimulus molecules do not necessarily have to reach the nasal cavity and the receptive epithelium through the air. In the vomeronasal organ the hedgehog has at its disposal a sensory organ of considerable communicative importance, which appears to function far more precisely than those sensory receptors used to identify lesser or more-usual smells and taste impressions. The externally visible process of self- anointing per se is not a form of communication but simply the last link in a chain of actions that has already reached its peak and fulfilled its purpose in the registration of the sense impression and its transfer into the central nervous system, and that now serves only to clean the vomeronasal organ and render it capable of further function (Fig. 1).
During courtship rutting males often leave scent marks and secretions on the substratum. In doing so, the hind legs are drawn close together and the spine is arched convexly upward, while the partially protruding penis exudes a sometimes whitish secretion, as the body is rocked gently to and fro and from side to side (Poduschka, 1969, 1976). This secretion probably is secerned by the accessory sexual glands, which are well developed in hedgehogs (Ottow, 1955), and seems to be diluted in urine. However, there is also the possibility that it may be some sort of innersecretory steroid diluted in the male urine. Without doubt this is a chemical form of communication that has a stimulating effect on the female. The smell of the secretion is also clearly discernible to the human nose and is completely different from the body and urine odors of a male hedgehog that is not in rut. The scent markings are always deposited behind the female, who usually allows herself to be driven around by the male in zigzag lines or in circles— the "hedgehog roundabout"—but always within a limited area of scarcely more than 100 square yards. The roundabout lasts several hours each night. The female crosses over the scent markings of the male again and again and by doing so comes into olfactory contact. Since before final intromission she allows herself to be mounted several times, from the side and head as well as in the usual manner, traces of the secretion on the underside of the male are deposited on the female's back. It can be safely assumed that this chemical message acts as the necessary release mechanism for her actual readiness to mate, i.e. as a true releaser pheromone. There are even hints that it may act additionally as a primer pheromone by modifying the estrous cycle (Poduschka, 1976).
Meaningful chemical signals derived from feces and urine of females not in estrus or showing mating appetite, as well as from customary defecation areas, have not been proved to exist among hedgehogs. They defend their normal hunting area against intruders, but do not set limits to their territory with scent markings. Juvenile hedgehogs, however, leave scent markings by occasionally pressing their feces against vertical objects at a height of about 2½ inches, even if they have been reared on their own. This is therefore an innate form of communicative behavior, which is not in any way triggered by the presence of a partner or a competitor. Presumably what occurs here is a combination of the chemical scents present in the feces and in the circumanal sebaceous glands, which, on the side toward the intestines, almost touch the especially large protodeal glands.
Individual scents in the nests of hedgehogs do not repel other hedgehogs. A strange nest will be occupied at any time, and the sex of the previous tenant is of no importance.
The ability, so noticeable in tenrecs, to produce a secretion from special eye glands is only weakly developed in hedgehogs and has only been observed so far in Erinaceus europaeus (Poduschka, 1969). On the other hand a rubbing of the oral angle (with its many glands) on the substratum by the Persian eared hedgehog (Hemiechinus auritus persicus) and the Syrian eared hedgehog (Hemiechinus auritus syriacus) has been observed, which can be assumed to be the placing of individual scent markings outside the mating season (cf. Quay, 1965).
SUBFAMILY: GALERICINAE; TRIBE: ECHINOSORICINI (HAIRY HEDGEHOGS)
5 genera. Indonesia, southeastern Asia, Philippines.
There is hardly anything known about the behavior of these animals. We can therefore only attempt to explain the sporadic observations made on living Echinosoricini—which are all from the largest genus, the moon rat (Echinosorex gymnura)—by drawing analogies with the behavior of their nearest relatives. Echinosorex gymnura is a strictly nocturnal animal, is not too rare but is locally spotty (Davis, 1962), and lives an ostensibly solitary existence (Lim Boo Liât, 1962). We can therefore only expect to find forms of communication between mother and young and during the mating season. Visual threatening consists of gaping with open mouth and producing a very low moaning sound (Davis, 1962). While doing this the animal adopts a crouching pose. Whether this can be called an optical signal is unknown. Acoustic emissions consist of snarls, growls, and groans when the animal is aroused or angry. Whether this "groaning" is identical with the well-known "rusty hinge creaking" referred to in connection with the hedgehogs is not known. Eisenberg (pers. comm.) stated that when Echinosorex gymnura explores a strange environment it emits ultrasonic clicks similar to those of the Haitian solenodon (Solenodon paradoxus). When searching for moving prey the moon rat moves its ears individually; this possibly means that it is able to locate the source of sounds through the sound waves that reach each ear independently.
The well-developed sinus hairs on the snout and a strip of hair on the lower side of the naked tail lead us to believe that the animals are very dependent on tactile stimuli. The moon rat catches small fish, but it is not known how it is able to locate them in the water—optically, by tactile stimulus through the vibrissae, or by changes in water pressure that reveal the position of moving prey.
The moon rat deposits fecal matter within one specific area, perhaps as a form of chemical communication. A very pungent odor emanates from the animal, apparently from the anal glands (Eisenberg, pers. comm.). It has none of the musky quality that is associated with the scent of shrews (Davis, 1962).
Nothing at all is known about any behavior or communication patterns among the other four genera: the lesser gymnure (Hylomys suillus), the shrew hedgehog (Neotetracus sinensis), the Hainan gymnure (Neohylomys hainanensis), and the Mindanao gymnure (Podogymnura truei).
FAMILY: TALPIDAE (MOLES)
5 subfamilies, which include 12 genera and 19 species. Europe, Asia, North America.
Observation of the mostly fossorial or semiaquatic Talpidae is especially difficult and timeconsuming. Existing reports probably differ so much because it is almost impossible to witness an entire behavior complex like reproduction behavior or ontogeny of the young. In communication we have to confine ourselves to forming hypotheses about the possibilities available for study, such as the moles' sensory powers or their vocal utterances, which have not yet been studied in detail.
Field observations confirm that the effective range of the mole's senses is very short (Godfrey and Crowcroft, 1960). On the other hand, the cutaneous senses are presumably more complexly developed in the moles than in any other animals, equipping them with an unusual tactile sense and possibly teletactile potential. In addition to the "special senses"—olfaction, taste, hearing, and vision (Quilliam, 1966)—the mole also has (for lack of a better term) the cutaneous senses at its disposal: touch, heat, cold, pain, and vibration. Quilliam emphasizes that the mole must also possess other sensory equipment, the location and structure of which remain a matter of conjecture; certain respiratory problems seem to indicate the presence of a well-developed baroreceptor system.
Most moles are nonsocial, mutually intolerant creatures. The common Eurasian mole (Talpa europaea) appears to know no exact limits to its territory. Encounters with other moles in the partly communally used tunnels occur over and over again. In most cases, if one of the animals does not immediately retreat, these encounters lead to serious fights that often end fatally. Talpidae apparently become aware of the presence of adversaries at the very last moment and blunder into them; this does not seem to say much for the presence or development of a communication apparatus.
If two moles in captivity are kept together in a space too confined or if they both fall into the same trap, as a rule one of them will be killed. Thus, a social life with communication can be expected to exist only between mother and young for a brief period, lasting from birth to a few weeks after weaning, and for the unknown but undoubtedly short period during which copulation occurs (Mellanby, 1971). At present we do not know if baby moles emit signals, or, if they do, what kind they emit in order to communicate with the mother. Talpa europaea breeds only once, in spring or early summer; the rest of the year it is in an asexual state (Matthews, 1935) in which interspecific communication appears to be even less distinct then during the mating season and rearing of the young.
An exception in puncto sociability is the American starnose mole (Condylura cristata), which cannot be considered solitary: it can be found in small groups (Hamilton, 1931), and we can therefore assume that it possesses a more distinct but unfortunately so far undefined communication system. Condylura cristata seems to be far better equipped for communication than most of the other Talpidae since it can see and hear better than they can. Another exception is the shrew mole (Neurotrichus gibbsi), the least fossorial of the five genera of American moles: it seems to be quite gregarious and apparently travels in flexible groups.
If any sort of acoustic communication does take place, it presumably does so within the range of ultrasonics. Although several components of the mole's signals can also be heard by the human ear, it can be assumed that the ultrasonic components serve as an interspecific means of communication: those ultrasonic elements that occur in the shrill squeaking sounds, the very harsh guttural squeaks, the short snorting sounds, and the noises made by the harsh grinding of the teeth, as described by Eadie (1939) in specimens of excited or frightened hairy tail moles ( Parascalops breweri). Godfrey and Crowcroft (1960) stated that their hearing is acute at close quarters: live earthworms were apparently detected by the sound of their movements. Just how difficult it is to determine acoustic communication in moles is shown by comparing Reed's discoveries (1944), which tell us that the shrew mole (Neurotrichus gibbsi) is for the most part mute; but it once emitted a faint, high-pitched, rather musical chattering, audible at short intervals for more than a minute. This is corroborated by Dalquest and Orcutt (1942), who were never able to observe a sound made by this species but suspected ultrasonic signals, since they were able to discern a reaction to noises having a frequency between 8,000 and 30,000 Hz. This is also confirmed by the results obtained by Quilliam (1966), who also noticed reactions in the common Eurasian mole within the normal hearing range and in ultrasonic frequencies. As a result of his anatomical investigations Quilliam suspects, however, that the ultrasonics are not registered via the cochlea, which in moles is the shortest in all of mammalia.
Since the bulbus olfactorius is very strongly developed in Talpidae (Godet, 1951; Stephan and Bauchot, 1968b) it can be assumed that moles are greatly influenced by chemical-olfactory stimuli—also having a communicative character—which is indicated by the frequent sniffing of moles with their snouts raised. Godfrey and Crowcroft (1960), however, suspected a range of detection by smell of only six to seven cm. The vomeronasal organ seems of questionable importance to me: Godet (1951) stated that it is clearly formed in the embryo, but degenerates later, and afterward it has no further function. This ought to be tested with modern means and methods of investigation; fresh studies of the vomeronasal organ in various animals during the past few years have resulted in a wealth of new facts and possibilities.
It has been assumed that the various scent glands play an especially large part in the communication ofTalpidae. The eastern mole (Scalopus aquaticus), the hairytail mole, the Pacific mole (Scapanus orarius), the Townsend mole (Scapanus townsendi), the California mole (Scapanus latimanus), and the starnose mole (Condylura cristata) have well-developed skin glands, which produce secretions that leave visible stains on the animals' fur. Of course, this cannot be regarded as an optical signal in an animal with such poor vision, but it could easily be a kind of olfactory signal. Eadie (1939, 1947, 1948) writes about additional glands on the ventral body surface of the hairytail mole and of a large medial, perineal gland having both holocrine and merocrine secretions in the starnose mole.
Although some moles are capable of using their eyes, sight does not seem to play a role in communication. Presumably Talpa europaea is not able to detect static objects (Quilliam, 1966), while the Mediterranean mole (Talpa caeca) is completely blind and cannot distinguish moving objects; the same is true of Neurotrichus, which shows not the slightest reaction to a sudden strong light (Dalquest and Orcutt, 1942).
Touch seems to be very highly developed. The vibrissae are arranged in rows and get longer the further they are from the tip of the proboscis; they are only slightly movable; each hair has its own innervation. Around the nostrils are numerous Eimer's organs for tactile and chemical use. Godfrey and Crowcroft (1960) assume that these organs are able to register mechanical pressure, temperature, humidity, and vibrations. The stiff hairs on the ears are also extremely sensitive, as are the bristles on the tail, which may be tactile but are less actively so than the vibrissae (Dalquest and Orcutt, 1942). Godfrey and Crowcroft (1960) assume that it is probably vibration that gives information about the general direction of a communication partner. The vibrissae on the face are so sensitive that they could probably be used as receptors for air pressure (Quilliam, 1966). Whether they also have a feasible function in a kind of echolocation process is still unknown, since the necessary experiments have not yet been conducted. To what extent the tactile hairs on the outer side of the digging paws of these nearly blind animals, described in more detail in Godet (1951), aid communication during an interspecific encounter is likewise still unknown.
SUBFAMILY: DESMANINAE (DESMANS)
The semi-aquatic Desmaninae occupy a special position in the family of moles. The modification of the sensory organs and the means of communication conditioned by life in another medium have resulted in a specialization that, at least from a cerebral-anatomical viewpoint, has been examined in a series of papers by Stephan and Bauchot (1959, 1968a, 1968b).
The genus Desmana moschata (western Russia) lives an apparently social life, which suggests some form of communication. In captivity the animals can be kept in pairs. Unfortunately I have not been able to locate any investigation of their ability to react to external stimuli of any kind.
The southwest European genus Galemys pyrenaicus, which in contrast to Desmana has never been seen in the wild with members of the opposite sex, appears to be very aggressive toward conspecifics. Only in recent years did Richard and Valette Viallard (1969) succeed in keeping specimens of the Pyrenean desman of both sexes together in captivity without mishap. This leads us to suspect the presence in these animals, too, of communication methods still unknown to us. Ritualized patterns of behavior that render fights harmless, at least between the sexes, and at the same time mechanisms that make possible successful courtship must exist. In the mother-young unit, too, there must be some kind of communication that we still know absolutely nothing about.
The only acoustic signal known in Desmana moschata occurs, when the animal is excited, as a metallic squeak (Ognev, 1928). Whether this is an active warning signal or a passive signal of fear is unknown. We know rather more about Galemys pyrenaicus: Niethammer (1970) states that it is for the most part silent and possesses only a limited repertoire of sounds—chirping and cheeping. When frightened or suddenly confronted by a conspecific, it emits a high-pitched, loud scream, which unfortunately has not yet been analyzed by an ultrasonic receiver. On the other hand, it does not react to loud noises within the human hearing range like, for example, a loud clapping of the hands, but it does show a strong reaction to infrasonics and noises with ultrasonic components (chirping, clicking, rustling of leaves) and especially whenever large or small objects—a drop of water is enough—fall into the current of the small, fast-flowing, and thus relatively loud Pyrenean streams in which it lives (Richard, 1973). This gives evidence of extraordinarily acute hearing or rather discriminating ability to hear noises and signals that occur within the animal's micro-habitat. Richard also suspects the possibility of echolocation, produced by a "tambourinage" (drumming or loud paddling) with the forefeet on the surface of the water, especially in the vicinity of unexpected obstacles. This action can, of course, also produce a movement in the water, the reflection of which helps orientation.
The only fact known at present is that the Pyrenean desman is able to register differences in light intensity.
Stephan and Bauchot (1968a) were able to discern an enlargement of certain brain structures in Galemys and Desmana and attribute it to the adaptation to semi-aquatic life. Unfortunately no one has yet investigated whether the ability of the Desmaninae to recognize the precise scent and taste of water (Richard and Valette Viallard, 1969) is connected with the activity of the vomeronasal organ, which is the mammalian equivalent of that special organ in fish that enables them to smell in water. Possibly we should look here for chemoreceptors for use in water, as described in the hypothesis of Bauchot, Buisseret, Leroy, and Richard (1973). As far as Galemys is concerned, the quality of the water is a matter of life and death, and precisely because of that the functioning of such an organ is vital to the preservation of its species. In addition, it could indicate, as a kind of communication organ, the presence of conspecifics, prey, or predators.
The anal and subcaudal glands are used as tools for chemical communication. Feces are deposited in specific areas, which are inspected olfactorily and tactilely with the vibrissae and the Eimer's organs of the proboscis. The same thing occurs after micturation. Whenever Galemys of different sexes are put together, each one inspects the feces of the other with great interest. This seems to indicate the presence of an intersexual, chemicoolfactory communication process. Chemical scent markings are left by the subcaudal glands, especially by males in spring, the time when reproduction takes place. These subcaudal glands are constantly in contact with the substratum. The smell of the individual scent markings is hardly detectable by the human nose. The nest and the immediate area around it are, however, so strongly impregnated with the gland secretion of the male that it is possible to detect a definite smell of "game." These gland secretions leave black, glistening streaks on the substratum (Richard, 1973).
Here, too, we can only draw conclusions from the development of the sensory tools and the functions that they presumably have. Those structures that serve the oral sense of touch (nervus trigeminus, sensitive trigeminal centers, and through them enlargement of the medulla oblongata) are more strongly developed in the Desmaninae than in Talpa. The vibrissae on the upper lip, which are unusually developed (Stephan and Bauchot, 1968a, 1968b), are innervated through the trigeminal system. The fact that Galemys possesses a proboscis extravagantly equipped with vibrissae and Eimer's organs, which underlines the importance of the tactile sense, was pointed out by Argaud (1944).
FAMILY: SORICIDAE (SHREWS)
3 subfamilies with 20 genera and more than 265 species. Europe, Asia, Africa, North America, northern South America.
Our knowledge of the behavior and particularly the communication patterns among most shrews is very fragmentary, if not nonexistent. It is therefore only possible to try to grasp the complexity of communication among shrews, using the results obtained so far. Further work to fill the gaps in our knowledge would not only contribute to answering many unsolved questions but also reveal that shrews, as far as communication patterns are concerned, in no way form an homologous group, so that considerable interspecific variations in communicative ability and the necessary anatomical requirements for communication have to be reckoned with. It would also help to explain, or rather clarify, the numerous contradictions that exist at present. In his excellent paper on communication in three genera of shrews Gould (1969) makes comparative observations on the communication complexes of shrews, using modern research techniques and technical apparatus for the first time.
The hitherto universally accepted idea that Soricidae were solitary animals, does not coincide with the facts, either under good conditions, where adequate food and sufficient space play a major role, or in captivity. So many exceptions are now known that the assumption of a general aggressiveness among conspecifics cannot be defended any longer. There are a great many graduated variations here, which are reflected in the numerous forms and means of communication. They make an overall survey extremely difficult.
The bicolour white-toothed shrew (Crocidura leucodon), the lesser white-toothed shrew (Crocidura suaveolens) the common European white- toothed shrew (Crocidura russula), the musk shrew (Suncus murinus), and the least shrew (Cryptotis parva) can even be regarded as partially social (semisocial) creatures (Vogel, 1969; Gould, 1969). On the other hand, the genera Soricinae (red-toothed shrews) and Neomys (European water shrew) are, according to Crowcroft (1955), definitely nonsocial. If several specimens of the least shrew are well looked after in captivity, they are compatible (Conaway, 1958), and as many as twelve sleep and even eat together (Davis and Joeris, 1936). According to Crowcroft (1955), shrews do not generally defend a specific territory but merely the spot where they happen to be at the time and the space as far as they can see. This is not very far, and is, in effect, even more restricted by the fact that they live under dense cover most of the time.
In Sorex vulgaris (= Sorex araneus) and Corcidura coerulea (= Suncus caeruleus ?) vision seems to be poor. The optic regions of the brain are small and poorly developed (Clark, 1932). The least shrew (Hamilton, 1944) and the shorttail shrew (Blarina brevicauda: Rood, 1958) have weak eyesight, but Blossom (1932) reports good eyesight in the masked shrew (Sorex cinereus).
If a male and a female shorttail shrew are put together, the hair above the area of the side glands in the male parts to reveal an apparently bare patch (Eadie, 1938). Since this species is noted for its poor vision, this change cannot be taken as an optical signal but rather as hypertrophy of a secerning gland—similar to the appearance of the swollen naked rings around the eyes in tenrecs, which become especially noticeable during the increase in secretion in the rutting period. Theoretically, however, it is possible that in shrews these patches function as a form of optical communication intended for the female in close proximity and supplement any olfactorychemical forms.
The common shrew (Sorex araneus) is also shortsighted, but the action of rearing up observed in this species is an optical threat recognized and understood by the adversary. This is also true of the action of throwing itself on its back and displaying the light-colored ventral surface, which occurs during an aggravation of the quarrel. Both movements are accompanied by an increase in the volume of screaming. We are able to observe here, therefore, an amalgamation of different communication systems, because screaming does not depend entirely on a visual stimulus: an excited common shrew will also scream in reply to a scream from another, unseen shrew in the immediate vicinity (Crowcroft, 1955). On the other hand, when Gould (1969) separated specimens of shorttail shrew and house shrew by placing a sheet of glass between them, he was not able to observe any signs of visual recognition. Only when they simultaneously placed their noses under the glass plate at the same point did each produce a highintensity chirp. This also led Gould to the conclusion that "either shrews [of these species!] do not see each other under these circumstances or visual imput must be coupled with tactile and/or olfactory stimuli before recognition will occur."
Many shrews do not react to conspecifics until they make vibrissal contact. We know that tactile communication is of great importance to the common shrew, northern water shrew (Sorex palustris), European water shrew (Neomys fodiens), shorttail shrew, masked shrew, smoky shrew (Sorex fumeus), least shrew, and Crocidura olivieri.
One of the few ethological examinations of African shrews showed that the African bicolor white-toothed shrew (Crocidura bicolor) is "very sensitive to the movements of insects" (Ansell, 1964). This species is reported to rely more on touch and hearing than on any other sense. However, it has not been studied whether this sensitivity to the movements of insects produces results with the aid of hearing or tactile orientation—possibly the teletactile reception of air- pressure waves caused by the wing movements of the insects.
As might be expected among these small insectivores, the relevant acoustic communication signals lie partly in the ultrasonic range. According to Gould's observations one can detect, in addition to ultrasonics up to 107 kHz, low-intensity sounds of 500-1,200 Hz. Clicks have a high localization valence and seem to be used as contact calls. High-intensity chirps and buzzes repel an approaching conspecific; aggressive encounters evoke mixed and graded sounds (Gould, 1969). Buchler (in press) observed, however, that the sound pressure of a bat's signals is 4,000 times greater than that of a vagrant shrew (Sorex vagrans). The Herero musk shrew (Crocidura flavescens herero) threatens with "a single sharp metallic squeak" (Marlow, 1954/55). The description of this acoustic threat signal indicates the prevalence of ultrasonic elements.
Gould (1969) heard and investigated seven sound types of different intensity in the musk shrew, shorttail shrew, and least shrew: chirps, clicks, twittering, "put," buzzes, chirp buzzes, and putter twitter. These terms, as used by Gould, are an onomatopoeic approximation of the actual signals. He also detected the source of ultrasonic clicks in infant Blarina by pulling open the lower jaw, and he was able to see the forward movement of the tongue as it was pressed against the upper palate at the moment of the click emission. Infant house shrews emit a whistle, which is considered a possible variant of the twittering mentioned above. Gould elaborated a clearly arranged synopsis of sound patterns together with typical contexts in which they might occur. One of the most fascinating results is that sometimes completely different signals are emitted, despite the fact that the causes remain the same. Obviously it is not possible to classify the various possibilities of acoustic communication in animals as strictly separate types of sound, closely linked to predictable stimuli or situations.
Neonate Soricidae possess a repertoire of sounds that increases in variety from day to day. They are extremely sensitive to sounds with ultrasonic components, e.g., humans chirping with the lips, as soon as the meatus is open (Dryden, 1968). The signals given by young common European white-toothed shrews and shorttail shrews (Gould, 1969) stimulate the mother to start searching for her young. This cheeping, which can be regarded as a signal for being lost, can cause the father, who has remained with the litter—a fact that proves that the common European white-toothed shrew is by no means a solitary animal—to carry the baby back into the nest from which it has crawled. Interestingly enough, this rescue instinct triggered by the acoustic communication of the baby has no effect on nonpregnant females but does have an effect on suckling females; they will carry a baby that has been deserted, even though it is not one of their own, into the nest and adopt it (Vogel, 1969).
The shorttail shrew possesses a large repertoire of acoustic communication: sharp, high- pitched squeaks when irritated; shrill, piercing squeaks when frightened followed by loud birdlike chatter; and rapid squeaks when contented. As a response to a challenge or as a warning it emits rapid unmusical clicks, like the chatter of teeth. When two aggressive males were confined together, one approached the other and made this peculiar chattering sound, whereupon the latter retreated. Males are more aggressive than females, juveniles less so than adults (Rood, 1958).
As a result of Blossom's investigations (1932) we know that the masked shrew is also capable of producing different signals that can be used as a means of communication. However, the situation that causes them and the purpose they serve are not known. While eating or when searching for food, the masked shrew is reported to utter a succession of faint twittering notes. They are very soft and produced so rapidly that they have a quality somewhere between a purr and a soft twittering. Very similar signals with similar causes are known to exist among other insectivores, e.g., the Haitian solenodon and many of the tenrecs, and are presumed to be phylogenetically very old. Possibly the twittering is similar to the echolocating rattle produced by many bats and is indeed used by shrews for echolocation purposes too. We know from the excellent work by Gould, Negus, and Novick (1964) and by Buchler (in press) that shrews are very adept at echolocation. Pearson (1944) describes the sounds made by male shorttail shrews when pursuing estrous females as "a stream of dry, unmusical clicks like a chitter, similar to the sounds of a twig brushing against a bicycle wheel," which could just as easily be equivalent to the series of impulsive clicks used by bats for echolocating purposes.
When common shrews fight, acoustic signals play an extraordinarily large and complex role. The occupier of a territory raises its muzzle, opens its jaws, and screams. Squeaking contests may follow between the adversaries. Staccato squeaks identify the male sex. The female's voice has less of a barking quality; the individual beats are closer together and at a higher pitch, forming a more continuous scream. As a weapon, the female voice, or rather the manner of its use, seems more effective than that of the male. To human ears the male scream is the more violent and aggressive, but that of the female is more piercing and persistent (Crowcroft 1955, 1957).
If two or more Savi's pigmy shrews (Suncus etruscus) are kept together in a container too small, an acoustically varied communication pattern is released, coupled with strong aggressive tendencies (Vogel, 1970). This behavior becomes more intense in close quarters than when the animals have sufficient space.
Shrews are equipped with scent-producing glands that vary numerically and potentially according to species. Contrary to earlier assumption, olfactory stimuli are definitely not without a certain significance in such fast-moving animals as shrews. At any rate the degree of olfactory refinement in shrews apparently varies from species to species: it seems to be bad in the least shrew, smoky shrew, gray shrew (Notiosorex crawfordi), masked shrew, and shorttail shrew, but quite good in Crocidura olivieri and in the common shrew. Current information, however, is self-contradictory or merely descriptive (Buchler, in press).
There is still a great deal of disagreement about the function of the side glands of shrews. Marlow (1954/55) oberved that male Herero musk shrews were frequently seen to rub their sides against the walls of the cage, leaving traces of the oily secretions from their musk glands at various points. As the words "musk glands" show, Marlow believed the characteristic shrew smell was attributable to the secretion from these glands. However, we have since learned, through the work of Dryden and Conaway (1967)—at least for the musk shrew—that the side glands are not responsible for the characteristic musk production. Apparently, concentrations of sweat glands in the throat and behind the ears produce this odor. If a male musk shrew is pursued or otherwise disturbed, he discharges scent while rubbing his sides along the cage wall and his throat and belly on the floor. The characteristic musky odor remains on such objects for several days. This rubbing of the belly seems to be a behavior pattern parallel to that of disturbed Tenrecidae (Eisenberg and Gould, 1970; Poduschka, 1974e) and may be considered phylogenetically a very old behavior pattern in mammals. Pearson (1946) found that the scent glands decrease in size during captivity: after a month they become thinner than in wild animals, making an examination of these glands difficult. Thus a continuous observation of these animals with a specific study of these glands in mind, which is possible only in captivity, is just not feasible.
In addition to the side glands, the shorttail shrew possesses yet another large medial ventral gland, oval in shape, 30 mm long and 9 mm wide. The secretion it produces has a very distinct odor, yet it has not been proved that it is able to ward off enemies, as has sometimes been maintained. This gland is more strongly developed in males during the rutting season than in females. But if a female becomes pregnant, her ventral gland decreases in size (Eadie, 1938). The decrease in, or complete disappearance of, scent emanation due to the reduction in size of the gland shortly before parturition possibly protects the female from discovery by rutting males or even by predators. Another gland, which presumably has some communicative function too, was discovered beneath the tail in the common European white-toothed shrew by Niethammer (1962). Histologically it is very similar to the side glands of this species.
A white secretion that issues from the eyelids in moments of excitement, similar to the way it does among tenrecs, is known at present to exist in the bicolor white-toothed shrew (from a conversation with E. von Lehmann, 1973), in the African forest shrew (Crocidura giffardi = Praesorex goliath: Vogel, pers. comm.), and in the lesser shrew (Sorex minutus) (from a conversation with R. Hutterer, 1974). Whether the secretion also produces olfactory stimuli is still unknown but very probable.
The significance of fecal deposits as a means of chemical communication is still a subject of disagreement; some authors consider them to be communication media, others doubt their importance.
Since a better and more exhaustive presentation of the mating behavior of shrews, which includes an illustrative explanation of the closely interwoven tactile, acoustic, and chemical communication patterns, is hardly imaginable, I should like to take the liberty of quoting Gould (1969) verbatim:
During initial phases of courtship in both Suncus and Blarina, the male appears to play a passive role— approaching the female, rubbing the substrate, tolerating bites from the female without biting her—while rubbing and exuding odor in new areas and gradually increasing the receptivity of the female. The female repels the male with high intensity chirps and buzzes; the male is easily repelled by the female's loud vocalizations and bites. Some males emit frequent "put" while courting. Orientation of the female's head and body toward the male was particularly prominent when the male emitted frequent and loud "put." The male responds to bites and loud chirps by closing his eyes and ears and exposing his gland-covered neck. The male rubs his venter over the substrate and simultaneously over his body. His glandular odor, immediately detectable by the observer, is emitted after 2 or 3 minutes (Dryden and Conaway, 1967). The female's body is pervaded with the male's odor through the following means: rubbing of the substrate by the male followed by the female walking over the rubbed areas and toileting herself; occasional fights; the female rubbing her tail against the male's neck as he positions himself behind her. Female Suncus reduce biting after the male fur is covered by glandular secretions. (Distortions of lips and tongue-smacking after biting indicate that glandular secretions have a noxious taste to the female.) The male bites the female on the flanks, rump and tail and as she becomes more receptive he follows closely behind her, oriented in a manner similar to the caravan formation prevalent during infancy. Continual advances and increases in click rates by the male (Suncus only) are followed by a receptive chirp or twitter by the female Suncus and a series of clicks by the female Blarina. Copulation follows.
This report is valuable since it shows not only similarities but also differences among members of different genera.
2 subfamilies with 9 genera comprising 29 species. Madagascar and adjacent islands.
Most tenrecs do not display any deep-rooted grouping tendencies. They live a solitary existence, well dispersed within their environment. Single species, especially the lesser hedgehog tenrec (Echinops telfairi), can sometimes be found in twos and threes, but it is not certain whether these are merely the remaining members of mother-young units. The streaked tenrec (Hemicentetes) is the only tenrec genus in which we find colony formation. Hemicentetes, whose equipment for acoustic communication is the most specialized and best developed among tenrecs, has a special stridulation organ in the mid-dorsal region. Since a similar organ has not been developed elsewhere other than among juvenile specimens of the tailless tenrec (Centetes ecaudatus), the genus that produces the largest number of young among recent mammals (up to 32 in one litter), it is obvious that the acoustic interspecific method of communication is linked with the more gregarious way of life led by Hemicentetes and Centetes.
The perineal drag used for leaving chemical scent markings seems to be prevalent in all species of tenrec. The same is true for gaping with the mouth as an optical form of communication. Furthermore, the varied use of ultrasonics in communication is remarkable, as is the complex function of the white secretion produced from specially developed glands situated in the eyelids. This has been the subject of closer study in Setifer, Echinops, Microgale dobsoni, and Microgale talazaci and is adjudged to be, among other things, a form of chemical communication (Poduschka, 1972b, 1974b).
Whether the erection of the prickles, bristles, or quills, peculiar to all Tenrecinae when aroused, is an optical signal is not clear. It could be a sign of defense posture, similar to that among hedgehogs, especially since a conspecific only reacts to it when it is accompanied by acoustic signals. As a result, we can say with some degree of certainty that it is used as an interspecific optical threat only within a whole communication complex containing acoustic, possibly optical, and almost certainly chemical forms. Intraspecifically it can be the equivalent of a threatening gesture designed through the erection of the prickles or bristles to make the animal appear larger and, therefore, more capable of successful defense.
When a female greater hedgehog tenrec (Setifer setosus) is willing to mate she lifts the perineum from the ground while simultaneously curving her spine concavely downward (lordosis), thereby presenting an unmistakable optical mating signal, which could possibly be strengthened by the emanation of chemical stimuli from the perineal glands, but which is understood visibly by the male (Poduschka, 1972a, 1974a).
The eye-gland secretion, described in more detail in several specific papers, could in certain circumstances act as an antipredator mechanism: the dazzling white patches that suddenly appear when the animal is excited or afraid, in place of the tiny and inconspicuous eyes, could possibly aid a nocturnal or crepuscular creature to scare off a predatory enemy.
As nocturnal or crepuscular animals, Tenrecidae possess long and numerous vibrissae (Fig. 2). They appear, however, to be used only for orientation and not in tactile communication. In an encounter other forms of communication are used. In the marsh tenrec (Limnogale mergulus), the only semi-aquatic form of Tenrecidae, we find—presumably as a kind of practical adaptation for life in water—the same numerous vibrissae on the snout as among the equally semi-aquatic Talpidae Galemys pyrenaicus and Desmana moschata; these vibrissae increase uniformly in length the further they sprout from the tip of the snout, and they are innervated by unusually strong nerve cords. There is no doubt of their primary use as sensory tools (Bauchot and Stephan, 1968). Whether they have a communication value we cannot say, since nothing is known about the ethological details of this species, presumably the rarest of tenrecs.
During the courtship behavior of all the Tenrecidae studied so far (tailless tenrec, streaked tenrec, greater hedgehog tenrec, lesser hedgehog tenrec, microgale dobsoni tenrec, microgale talazaci tenrec), there occurs a ritual nestling up to each other with those parts of the body that have concentrations of glands. Here the nose-tonose, nose and eye area-side, nose-anal/genital contact is merely the beginning of the ritual, or rather the most stereotyped form. Following it, the male and female tenrec rub their sides against each other or crawl over and under each other. All this indicates a combination of tactile communication with stimuli from gland areas in the partner; in addition there is an individual olfactory-chemical stimulus caused by the animal's own secretions, which could be considered a feedback system.
The mounting that follows is soon accompanied by the male's scratching the female's sides with his hind legs, aimed at stimulating her, which acts as a release mechanism for the presentation of her perineum. The male avoids a premature release from the mating "tie" (lasting in Setifer for more than two hours, presumably because of the strong accessory erectile tissues in the penis or by an intravaginal retention of the vine-shoot-like glans penis: Poduschka, 1974a) by allowing his hands to rest on the female's sides during the whole tie period. If he gets tired and slides off, he rests his hands on her back. Whenever the female tries to break away from the tie position, the mere hint of a grasp from the male, which is never so firm that the female cannot escape, is enough to indicate that the tie between the genitalia is not yet broken.
During courtship the male lesser hedgehog tenrec repeatedly bites the spines on the female's side. By doing so he seems to stimulate her readiness to mate. This pattern is even more clearly and efficiently developed among the greater hedgehog tenrec: every time the female tries to escape—which is part of her mating ritual—she is held back by the male's energetic biting into the sagittal lower region of her back and drawn back to him again (Poduschka, 1972, 1974a). This bite is not identical with the copulation bite, which, among Setifer at least, is not meant to keep the female still so that mating can take place but is only delivered when intromission has already taken place, in order to improve the lordosis of the female. This bite is therefore a tactile signal, to which the female responds by trying to evade the copulation bite. As she slips away, she raises the perineum, thus improving presentation and the tie of the genitalia (Fig. 3).
All tenrecs possess a large and varied repertoire of signals, which seems all the more remarkable when tested by an ultrasonic receiver. Throughout its whole lifetime the streaked tenrec (Hemicentetes) even has a special, mid-dorsal stridulation organ consisting of several rows of specially formed quills, which are moved by special muscles in such a way that signals are produced by the resulting friction. These stridulation sounds show little harmonic structure. They form a broad band from about 2-200 kHz in adult animals. Young Hemicentetes produce stridulation sounds of lower intensity when they are between eleven and seventeen days old. When they are about seventeen days old the intensity of stridulation is very near adult level (Eisenberg and Gould, 1970). This is a phenomenon parallel to that observed in young Setiferes, where the signals of the young become stronger and higher in the ultrasonic range the older the babies get (Poduschka, 1974c); and certainly very dissimilar to that in young rodents, whose ultrasonic signals are noticeable just after birth, but gradually become softer or even disappear completely after a few days as soon as the meatus acusticus and the eyes are open, but, on the other hand, will continue for several more days if the animals are handled (Noirot, 1968; Noirot and Pye, 1969; Sales, 1972).
Hemicentetes stridulates almost continuously when active and on the move. Stridulation occurs while feeding, during social contact, during courtship and mounting, during exploration, when escaping pursuers, or when merely moving away. Generally speaking, low stridulation occurs when the animal is normally active. When the animal is aroused intensely, crest erection occurs (presumably more as a sign of readiness for defense than as an active optical signal) together with an increase in stridulation, which does not remain at a peak but decreases in waves after the excitement is over. It is difficult to decide what can be considered communication with a partner, with young, or with threatening predators, and what can be considered a reaction to the stimulus that caused the excitement.
When foraging, young Hemicentetes move about nine to ten feet away from the mother, who keeps them close to her by stridulation. Thus, stridulation serves in the mother-young unit not only to identify the female's whereabouts (Eisenberg and Gould, 1970), but also to indicate to the mother where the young are, since they too are able to stridulate.
Young tailless tenrecs also possess a stridulation organ in the mid-dorsal region. It is not such a specialized one, however, and it disappears during individual ontogenesis. It consists of two rows of white spines, which by means of a special dermal musculature vibrate together and produce sounds. Some of these spines may remain in the subadult animal and still produce signals but they are gradually lost and do not get replaced (Gould, 1965). Gould was also the first to find out that stridulation among young tailless tenrecs, a pulsating sound varying in intensity between 12 and 15 kHz, is associated with high levels of excitement, and keeps mother and young together.
It is probable that stridulation was originally a warning signal to a predator but later became an interspecific signal indicating position. Stridulation in Centetes is different from that among Hemicentetes. Here stridulation occurs in conjunction with crest erection and erection of the center quills and a simultaneous hiss from a half-open mouth. Stridulation seems to occur when there is the inclination to attack coupled with an equally strong inclination to withhold. It was also found that stridulation serves as a warning signal to other members of the group, resulting in arousal and attentiveness. It may also indicate the identity and position of a juvenile that has been startled. It could also help in the location of the young by the mother and/or the location of young by other young. The exact function is unknown (Eisenberg and Gould, 1970).
Echolocation in tenrecs has been proved to exist among Hemicentetes, Echinops, and Microgale dobsoni. Contrary to the obvious assumption, stridulation is not essential for echolocation; the species studied by Gould (1965) echolocate by means of clicks produced with the tongue.
The vocal emissions of Tenrecidae take place mostly within the range of ultrasonics—at least so far as they are relevant for the animals. When they were filtered out, it was noticed that sounds within the human hearing range were of little interest to the animals.
During mating Echinops females utter snapping signals with a frequency of up to 51 kHz and twittering noises up to 37 kHz. These are used not only as a defensive signal in response to the continual insistence of the male but also as an aural threat: Sometimes an unwilling female leaves the nest where she has been urged to mate by the stimulating scratches and mounting attempts of the male; she turns toward him energetically with gaping mouth, and emits these same twittering noises. We can assume, therefore, that these twittering noises serve a number of purposes. Echinops also possess other belligerent sounds, including an unmistakable hiss. Echinops also reacts to the high-frequency signals emitted by its prey, which are thus recognized as allomones. Mealworms crawling over each other produce a sound with a highest frequency of 42 kHz, which is recognized by Echinops as a signal from a well-known prey and which results in a direct search for it.
During mating the greater hedgehog tenrec produces snapping signals up to 83 kHz and squeaks whose strongest sound pressure is between 50 and 60 kHz.
In Gould's (1965) opinion the clicks are produced among Echinops, Hemicentetes, and Microgale by the lips or the tongue; according to my studies of Setifer and Echinops, however, they seem to be produced more by the root of the tongue or the soft palate or in the larynx.
The small, soft-furred Tenrecidae (Oryzorictinae) emit acoustic signals far less frequently. When defending themselves, they are mostly silent and threaten by gaping with the mouth. Many emit squeals or a long squealing trill, a scream, a wail, or a buzz. When an encounter with a strange conspecific occurs, they utter a soft squeaking sound, which in Eisenberg and Gould's opinion (1970) is meant to prevent any aggressive behavior in the possible adversary.
Chemical communication in tenrecs is a subject that has hardly been studied exhaustively. Until now only one aspect has been examined in any detail: the white secretion from the lid glands that was described so far among Setifer, Echinops, and the two Oryzorictinae, Microgale dobsoni and Microgale talazaci. This phenomenon can without doubt be observed in most, if not all, Tenrecidae, possibly in various forms or stages of development; its existence is even more probable in the light of our knowledge of at least the anatomical features necessary for communication that are present in several more species than those mentioned above (Cei, 1946). The chemical examination of this secretion, exuded from special eyelid glands, which exist in addition to the lacrymal glands, is extremely difficult since only a very small quantity can be obtained, but some preliminary results are available.
Eisenberg and Gould (1970) have described additional gland areas in Hemicentetes, Setifer, Echinops, and Microgale that emit olfactory stimuli. Their exact nature as far as communication is concerned has still to be investigated in detail. The glands in question are in the axillary, inguinal, head, ear, and caudal areas. A further study of the sternal and possible ventral gland areas is already under way (Poduschka, 1974e, 1974f).
The secretion from the lid glands also has communicative character in certain circumstances, since the exuded secretion gives off a strong and long-lasting smell. The odor is understood by conspecifics, which respond by getting very agitated and eventually rubbing off the secretion now being exuded from their own eyelids and nostrils: quite often they also rub the area around their own eyes on vertical objects but never on the substrate. These scent markings are not to be regarded in the same way as territorial markings used in the defense of the area in which an animal lives, since tenrecs, as far as we know, do not have any territorial possessions. They can serve the purpose of self-assertion on the spot where the animal happens to be at the time, a phenomenon already observed among many mammals and well documented by Eisenberg and Kleiman (1972) and Kleiman (1966).
Even if a male Echinops in an enclosure that is strange to him has neither smeared eye-gland secretion on the wall nor attempted to place sternal markings (see below), has neither defecated nor micturated, another male put into the enclosure with him will become very excited by scent deposits made by the first male—a phenomenon which has not been explained. The second male will attack as soon as he sees the first male, but will not attack a female in the enclosure or one placed there simultaneously with him. This proves the existence of sexually differentiated scent emanations. Besides the eye-gland, sternal, or ventral secretions, others that could help determine the presence of a male conspecific are pheromones in the spittle, the sweat glands, and other outlets in the skin; the breath; or digestion gases. It must be emphasized that it is not just dominant males that mark in this way, a behavior that Ralls (1971) assumes to be the general norm in mammals.
As a form of chemical communication the eye-gland secretion has various purposes and/or functions:
1. Active: marking behavior, warding off adversaries, suppression of own unease in strange surroundings, possible stimulation of the female during courtship through a smell that acts as an olfactory signal. Whether the eye secretion acts as a primer pheromone and induces ovulation by altering the physiology of the reproduction system is still unknown but not unlikely. Induced ovulation is presumed to exist in tenrecs; fertilization takes place within the ovary (Strauss, 1939, 1942). An optical significance of the white secretion among conspecifics cannot be said to exist, since the female does not look at the male during the preliminary and actual mating behavior; it is just as possible to assume that the male is unable to see the female during the actual mating activity since his eyes might be completely covered by the white secretion (Fig. 4).
2. Passive: excitement during mating due to chemical stimuli from the secretion of the partner, also accompanied by tactile and acoustic stimuli. Several other stimuli, e.g., strong pungent odors or the occurrence of secretion during the very last minutes of a tenrec's life are passive reactions, which have no real bearing on communication.
I have been able to observe and film a special kind of communication or marking among the greater hedgehog tenrec. In strange surroundings or when the animal is unsure of itself because of the presence of several strange conspecifics, the upper part of the body is pressed down flat on the substratum and the forelegs are stretched out passively sidewards so that only the hind legs push the body forward (Poduschka, 1974e). To judge from appearance, it is the sternal glands that are being used here (a behavior pattern similar to those of some more-evolved mammals) or possibly the ventral glands, the presence of which has also been discovered in Soricidae and suspected in solenodons. This apparent method of marking can also take place on a three-dimensional object, for example, on a piece of wood in the enclosure: The animal slides over the top so that its ventral surface is pressed firmly on the object. If the lower side of the object is not lying on the ground (in the case of a large branch or bough of a tree) the animal crawls underneath it, and, lying on its back, the animal presses its ventral area against the object to mark it. The suspected secretion must exude from the gland areas in the breast, possibly from those in the throat or the ventral region. The communicative function of this act is not yet clear. Presumably it is similar to that produced by the rubbing off of the eye-gland secretion, but I have not been able to detect any olfactory stimulus. According to Eisenberg and Gould (1970), the streaked tenrec also leaves scent signals not only by means of perineal drag but also through rubbing the venter by extending and flexing the body on the substratum and by twisting the body when lying on its side. All this has definite potential significance in chemical communication.
Chemical communication also takes place among the two Oryzorictinae, Microgale dobsoni and Microgale talazaci, as it does in Setifer and Echinops among the Tenrecinae through markings that exude from the cloacal region. Markings are made by pressing the perineal area on the substratum: while moving forward, the cloaca is repeatedly pressed down on the substratum. In captivity lactating female Setiferes deposit feces and urine in one place in the enclosure. This action is then copied by the young, who also use this spot for defecation. The chemico-olfactory stimuli released by the mother's excrement can thus be considered a kind of communication leading to a closely related imitation (Poduschka, 1974c).
The salivating of the lesser hedgehog tenrec, which Eibl-Eibesfeldt (1965) has interpreted as a form of marking behavior, should not be considered as such, but rather as the last link in a chain of actions that reaches its peak in the registration and identification of smells and tastes in the vomeronasal organ. It is thus an equivalent to the self-anointing of the hedgehogs, which, like Echinops and Setifer, possess an actively functioning vomeronasal organ. Similar to that of hedgehogs, this behavior pattern takes place only when a strange smell or taste has been detected by the animal, and therefore it has no relevance to any active communication process of its own. This salivation also occurs among Setiferes, and as far as communication is concerned, it is only a reaction to an extraordinarily strong and stimulative allomone (Poduschka, 1974c).
FAMILY: TENRECIDAE; SUBFAMILY: POTAMOGALIDAE (OTTER SHREWS)
3 genera. West to East Africa.
There is hardly anything known about the communication behavior of these animals. The lesser otter shrew (Micropotamogale lamottei) in captivity sometimes emits a high-pitched, sharp, loud scream at intervals of about 2 sec (Kuhn, 1964). These acoustic signals have unfortunately never been measured or recorded. Judging from the results obtained from the study of other insectivores, especially the closely related tenrecs, we can assume that these sounds contain ultrasonic components that are important for interspecific communication.
The eyes are remarkably small and presumably have little communicative significance. On the other hand, the vibrissae are very numerous and extraordinarily well developed. Even in the newborn we can clearly see the sinus hair warts from which the strong vibrissae protrude. When the animal is resting, these vibrissae point backward, but they can be spread out sideways and forward when the animal is attentive, even when it is just a few days old (Vogel, pers. photos). Among adults rhythmical movements of the vibrissae backward and forward can be detected (Kuhn, 1964).
Among the Ruwenzori otter shrew (Mesopotamogale ruwenzorii) the use of specific defecation areas has been reported (Rahm, 1961), which could possibly have communicative importance.
According to Cei (1946) the big otter shrew (Potamogale velox) possesses special glands in the lids, which are presumably equivalent to those studied in Tenrecidae (Poduschka, 1974b). It is questionable, when one considers the otter shrew's semi-aquatic way of life, whether they have similar functions. Among Tenrecidae their communicative functions are limited to the emanation of individual- or at least sexually specific smells and to possible visual signals aimed at warding off inter- and intraspecifics.
FAMILY: SOLENODONTIDAE; GENUS: SOLENODON
2 species. Hispaniola: Haitian Solenodon (Solenodon paradoxus); Cuba: Cuban Solenodon (Solenodon cubanus [Atopogale cubana, sensu Cabrera syn.])
At present we are still not in a position to distinguish between the behavior patterns of the two species. The Cuban form was considered extinct several times during this century, but according to the latest reports, this is untrue.
Thanks to the work of the late Erna Mohr, who was lucky enough to be able to keep and observe more living specimens of solenodons than anyone else (fifteen in all), we know relatively much about the behavior of these extremely rare animals. Of course, modern demands for more detailed information extend beyond the scope of the reports she produced at the time. They were restricted for the most part to a phenomenological inventory of behavior patterns. This inventory has in the meantime been complemented by a few ethological works, which I have specified here.
Solenodons are not solitary animals. They have been found in groups of up to eight animals sleeping in the same hole.
Mohr (1936a) has already described the varied repertoire of sounds that the Haitian solenodon (Solenodonparadoxus) possesses. It consists of a "mournful sounding tone like that of kittens prior to the opening of the eyes"; gurgles; shrill screams lasting up to five seconds (?); and a melodious "Strophe," which also lasts a good five seconds, resembling that of a robin. It is often repeated using the same pattern of notes. Möhr assumed at first that this was produced only by young, unweaned animals but later revised her opinion (Mohr, 1936b) when she again heard this Strophe, this time during courtship and mating of sexually mature Haitian solenodons.
Using modern methods of detection, Eisenberg and Gould (1966) registered some chewing and digging sounds also. These were emitted while the animals were walking or running, and were used in particular by the young animals as a source of sound or as a signal to approach. These authors also carried out the first successful ultrasonic tests. My own studies in this field have so far produced varying results, insofar as I have been able, with the aid of a Holgate Ultrasonic Receiver, to detect both clicks with frequencies up to 74.8 kHz and noises produced by the exhalation of breath up to 73 kHz. Of course, this does not prove that these particular signals or their highest components have any relevance for the animals, but it does give some indication: By producing strange noises of a high frequency, which cause solenodons to get very frightened, it was possible to show that this animal is very sensitive to ultrasonic signals between 65 and 75 kHz. The highest components of the emitted signals do not remain constant.
It is also possible to record similar signals of up to 40 or 50 kHz, leading us to suspect varied communicative content. Solenodons are just as sensitive to shrill human voices and loud laughter (Mohr, 1936a) as to mechanical noises with high ultrasonic components (Poduschka, 1974d). Solenodons are able to detect lowpitched ultrasonics well and move toward them, indicating that the ultrasonic signals emitted by insects or small rodents, for example, have interspecific communication value for them. The highest frequencies recorded in the "tuckering" (motorlike) "schnalz" sounds, also mentioned by Möhr, were 28 to 40 kHz. I believe that these sounds are produced mechanically in the larynx or the oral cavity.
The clicks mentioned above can be heard especially when the Haitian solenodon is confronted with a strange conspecific or during the exploration of strange territory. It is not possible to say with certainty whether they are used in echolocation, or as a signal (Eisenberg and Gould, 1966), or could in strange territory be an acoustic parallel, aimed at self-assertion, to the olfactory-chemical behavior pattern of marking already known. These vocalizations of Haitian solenodons are similar to the echolocation pulses of shrews (Gould, Negus, and Novick, 1964).
The Haitian solenodon shows a welldeveloped appetence for nooks and crannies and employs a special kind of breath exhalation when exploring impenetrable cracks and crevices, which could very well be used as a form of echolocation similar to that detected in hedgehogs.
According to Mohr (1936b), the Haitian solenodon reveals a pattern of marking during mating or courtship that includes the use of ventral glands. Unfortunately these glands have not yet been studied. The solenodon slides on his ventral area around and alongside the female, pushing himself along with his forefeet and dragging his hind legs behind him. This reminds us strongly of a special kind of marking behavior observed in the tenrec Setifer setosus (Poduschka, 1974e, 1974f), but the solenodon does not press the pectoral area on the substratum; he presses down the venter.
The function of the secernent side glands, which are said to produce odorous substances, has been subject to interpretations that are partially contradictory. So rarely have a male and a female been kept together in captivity that it has not yet been possible to prove that these secretions have any communicative value. According to Mohr (1936b) the side glands only begin to secrete when the animals are six to eight months old; in her opinion the secretion indicates sexual maturity. On the other hand, I have not yet been able to detect any secretion from these glands in a male that was put in with a female after spending five and a half years as a solitary. It could be that the secretion from these glands only indicates sexual activity, which was no longer the case in this particular male—perhaps because of his long period of abstinence or perhaps because solenodon males are sexually active only during the first few years of their lives. The male and female lived together for almost three and a half years, yet never mated. This may indicate that the secretion from the side glands is of essential communicative importance, necessary for the release of a complete pattern of mating behavior in this species.
According to Mohr (1936b) the axillary and ventral areas of adult Haitian solenodons are continually moist. Therefore, glands that could have communicative character must be present. Ignoring for the moment the more abundant material available to Möhr for observation, I must report that I have never been able to detect such an unmistakable secretion in the male and female solenodons I have studied in the past three years. Variations must therefore exist among individuals, conditioned by age or the environment in which the animals are kept.
Fecal deposits are made at random while the animal is on the move. Even in the wild no specially reserved defecation areas have been reported. Eisenberg and Gould (1966) noticed perineal drag immediately after defecation. I have been able to observe and film this only after micturation.
The only optical signal so far observed among these animals, which live together peacefully with members of their own species and whose tiny eyes and nocturnal habits preclude vision as an important means of communication, is a threatening gesture with gaping mouth. The solenodons were very frightened by an electronic flash. A reduction in the intensity of the flash and the less fearful reaction from the animal that resulted could mean that the startle reaction was indeed the result of an optical stimulus and not of an acoustic one caused by the noise of the camera shutter. I suspect, however, a combination of both sensory impressions (Poduschka, 1974d).
Since the solenodon leads a nocturnal life, the vibrissae presumably act as tactile organs. They are found in abundance on the head (Fig. 5), where they sprout in especially large numbers on the sides and on the mandible pointing downward, and are also found on the ventral surface, where they stretch from extended embryonic Milchleiste up into the axillary regions. There are also carpal vibrissae, like those found in ground squirrels (Poduschka, 1971). The facial vibrissae can be moved in the skin by muscular movements, which are also referred to by Gundlach (quoted by Barbour, 1944) in the Cuban solenodon.
While observing the communal existence of the male and female solenodon, I have never been able to describe with any certainty a particular behavior pattern as an interspecific act of tactile communication. The animals are quite uninhibited in touching each other with various parts of the body, sometimes vigorously sometimes gently, and seem not to be influenced by the partner while absorbed in their remarkably energetic, continual activities, except that they understand the actions of the partner as an optical signal inviting them to join in or to imitate. They do, however, snuggle up to each other, a behavior pattern common in Tenrecidae and many Erinaceidae; and sometimes they push the nose and cheek area along the partner's body, an action that could be considered the beginning of a tactile communication pattern that changes into a chemical one influenced by the gland concentrations of the partner. It is, of course, possible that this is an action combining both kinds of communication, an assumption made even more feasible by the fact that when solenodons meet they nudge each other with the point of the nose. One pushes its nose into the ears or the axillary areas of the other; that is, into parts of the body that give off strong olfactory stimuli (Eisenberg and Gould, 1966). I have experienced this particular behavior on my own person by a specimen that was especially familiar with me. It seems to me, therefore, to be an integral part of their ethogram.
5 genera. Southern and eastern Africa.
I have not been able to locate any papers on items of the behavior and/or communication of golden moles. It seems, therefore, that nothing at all is known about these very interesting animals, which, because of their solitary and cryptophile way of life, are very difficult Because of adaptation to to study.
So far only anatomical or taxonomic studies have been produced. The bulbus olfactorius is very large (Stephan and Bauchot, 1960), indicating the importance of the sense of smell. It is questionable, however, whether one can expect to find chemical forms of communication as a result, since olfactory stimuli soon disappear in the dry air of the extremely arid habitat of these animals and cannot therefore be detected easily. Such forms are at least fairly probable, however, within the mother-young unit and indeed are very necessary for these completely blind animals as an indication of readiness to mate.
In the following attempt at a comparative survey of the presently known communication systems of insectivores, I have listed two subfamilies under headings of their own. I want to emphasize that I am quite aware of contravening the normal practice of systematic zoological observation in doing so. This survey, however, is concerned with only one section of ethology: communication, along with the anatomical features necessary for it to take place. It must therefore deal separately with these two subfamilies, since the one differs so completely in its behavior patterns, depending on conditions in its Umwelt, and the other's behavior is virtually unknown.
The first subfamily is the semi-aquatic Desmaninae. Because of adaptation to another medium, the Desmaninae brain differs greatly from that of other Talpidae: modifications include regression of the olfactory centers and enlargement of the auditory, trigeminal, and motor centers. There is also enlargement of neocortical regions, especially of the centers of association. On the whole, there is an increase in brain weight and of the index of encephalization (Bauchot and Stephan, 1968). These differences make it impracticable to compare the communicative behavior and relevant sensory powers of the Desmaninae to those of the other Talpidae.
The second subfamily taken out of its normal systematic context is made up of the five genera of Gymnures (Echinosoricini). So little is known about them that practically nothing can be said about their ability to communicate. Because of the dearth of reports on their behavior, we can merely draw analogies from the study of other insectivores. Nevertheless, we know more about Echinosoricini than about Chrysochloridae. The latter is a zoological family in its own right and ought therefore to be given a rubric of its own, even if it remains unstudied.
When studying the four systems of communication—chemico-olfactory, visual-optical, acoustic-auditory, and tactile—the means by which they are produced, the manner in which they are used, and the anatomical features necessary for their existence, we come across several forms that have parallel, convergent, or analogous functions. In mentioning them we find that a division into completely separate systems is a limitation that cannot be maintained. The discovery, for example, of an actively functioning organ that reacts to two completely different stimuli (the vomeronasal organ in Erinaceidae and Tenrecidae) or of the hitherto unforeseeable complex significance in a few insectivore families of supplementary eye-gland secretions and their functional importance as a means of communication indicates that a combination of various sensory functions can often occur even within the field of communication. This would corroborate the well-represented view that, as a rule, whole complexes of methods are employed in order to achieve communication inter- and intraspecifically. The diversity of these methods and the number of possibilities in combination with others cannot be foreseen. Moreover, we cannot exclude the future discovery of other functioning organs or abilities in insectivores that have long since been discarded by higher orders of mammals or are now present only in rudimentary form and thus extremely difficult to interpret.
In the following tables are listed some of the individual abilities or behavioral patterns that have been found to exist in at least a few genera of the families concerned. The question then arises whether signs of these abilities can be found among other families. Since the data available to us at the moment are still very incomplete, the result is merely an approximate survey of those communication methods common to all insectivores, using the few details that our present knowledge affords us. Because of lack of space some of the details listed in the following tables have not been described in the preceding survey, but they are to be found in various other works on the subject, which in most cases I have listed here.
All the insectivores studied so far emit and react to ultrasonic signals. Five of the nine families (and/or subfamilies) listed in Table 1 emit ultrasonic clicks; four have not been examined. Four use other ultrasonic signals or signals with strong ultrasonic components; in three of them the use of ultrasonics is suspected; two have not been examined. A positive answer to the question of reactions within the human hearing range can be given with certainty only among Potamogalidae, Soricidae, and solenodons; among the others these frequencies seem to be of lesser significance. Echolocation has been observed in Tenrecidae and Soricidae and suspected in Erinaceidae, Desmaninae, and Solenodontidae. To sum up we can say that insectivores are a group of animals that have adapted an ability to detect and emit ultrasonic signals and thus possess the necessary anatomical requirements for echolocation. The question of receptors for the reflected signals has not yet been clarified.
We have not yet exhausted the whole repertoire of signals. Their vocalizations seem—generally speaking—to fall into three groups: some combined with inhalation or exhalation through the nose; others derived from clicks with the tongue (Gould, 1969); and others probably originating in the larynx. These three types of sounds are known at present to exist in Tenrecidae, solenodons, and hedgehogs. The continuously emitted sounds, combined with a changing state of agitating and locomotive activity, are rather similar in Hemicentetes (Gould and Eisenberg, 1966; Eisenberg and Gould, 1970), in Suncus and Blarina (Gould, 1969), as well as in Setifer, which emit a series of "put" signals. Tenrec ecaudatus, too, emits a variety of respiratory sounds that are comparable to "puts" (Eisenberg and Gould, 1970). "Put" sounds can also be heard in Echinops and may have the same significance as the low sniffing sounds and low chuckling of a hedgehog that is only slightly agitated.
The broad perspective of chemical communication (Table 2) is too large a field to be treated exhaustively here, since present knowledge has to confine itself, for the most part, to single aspects and single observations of a relatively small number of species.
The deposition of feces as a communication signal is not common to all families and/or subfamilies. On the other hand, the emission of chemico-olfactory stimuli from various glands occurs in all families studied so far. Especially significant seems to be the eye-gland secretion, which apart from its other functions also has communicative character. As far as we now know, it could be regarded as a fully developed behavioral complex; at least the anatomical and histological requirements for such a complex do exist. The exact function of the evolutionary archaic vomeronasal organ has still to be tested in most insectivores. The salivating of Echinopswhich Eibl-Eibesfeldt (1965) has described as marking behavior, is not to be considered as such, but as the last link in a chain of actions that reaches its peak in the registration and identification of a sensory impression of taste or smell in the vomeronasal organ. Just as in hedgehogs, it only takes place after the animal has detected a strange smell or taste and, therefore, it is in no way an active communication signal in its own right. It also occurs among Setifer, where it can be regarded only as a reaction to an extraordinarily strong stimulant or completely new allomone (Poduschka, 1974c).
As predominantly nocturnal or crepuscular creatures with, for the most part, very poor eyesight, insectivores depend a great deal on their tactile abilities (Table 3), which—apart from the vibrissae that are well developed and numerous in all families—have led to the development in Talpidae of especially effective, and apparently extremely versatile, tactile organs on the proboscis (Eimer's organs). Roughly speaking, we can say that the importance of tactile communication is indirectly proportional to the optical ability of insectivores. Investigations have shown that tactile communication is common to all insectivores during courtship. The Tenrecidae occupy a somewhat special position with their unique use of stimulant scratching. Stimulant bites during courtship and mating have been observed in Tenrecidae, Erinaceidae, and Soricidae. The male Soricid offers the female some of his own gland areas for her to bite and thus succeeds in getting her to detect the secretion or at least in transferring some of it onto her body.
Visual gestures or changes in appearance among insectivores (Table 4) are to be regarded for the most part less as active communication systems than as passive reactions to the active stimuli of other systems, since eyesight among insectivores is generally poor. Gaping with the mouth seems to be an action peculiar to all insectivores. In Soricidae, however, we know that it is accompanied by ultrasonic emissions. Whether the remarkable eye-gland secretion of the tenrecs is in fact understood by conspecifics as a visual signal has not been proved either way, but it is certainly of less importance than the chemical message thus conveyed. Lordosis, as a signal inviting intromission of the penis, is common to all insectivores, but presumably even this action is accompanied by chemico-olfactory exudation. Since these animals have such poor eyesight, the odor is much more effective than a visible change in appearance or body position.
Ansell, W. F. H., 1964. Captivity behaviour and postnatal development of the shrew Crocidura bicolor. Proc. Zool. Soc. London, 142:123-27.
Argaud, R., 1944. Signification anatomique de la trompe du Desman des Pyrenées. Mammalia, 8:1-6.
Barbour, T., 1944. The Solenodons of Cuba. Proc. New Engl. Zoölog. Club, 28:1-8.
Bauchot, R.; Buisseret, C.; Leroy, Y.; and Richard, P. B.; 1973. L'équipement sensoriel de la trompe du Desman des Pyrenées (Galemys pyrenaicus, Insectivora, Talpidae). Mammalia, 37:17-24.
Bauchot, R., and Stephan, H., 1968. Étude des modifications encéphaliques observées chez les insectivores adaptés à la recherche de nourriture en milieu aquatique. Mammalia, 32:228-75.
Blossom, P. M., 1932. A pair of long-tailed shrews (Sorex cinereus cinereus) in captivity. J. Mammal., 13:136-43.
Buchler, E. R., in press. Experimental demonstration of echolocation by the wandering shrew (Sorex vagrans ).
Buchler, E. R., in press. Echolocation by the wandering shrew (Sorex vagrans): Discrimination and potential adaptiveness.
Cei, G., 1946. Morfologia degli organi della vista negli insettivori. Archivio ital. Anat. Embriol., 52:1-42.
Clark, W. E. Le Gros, 1932. The brain of the insectivora. Proc. Zool. Soc. London, 1932:975-1013.
Conaway, C. H., 1958. Maintenance, reproduction and growth of the least shrew in captivity. J. Mammal., 39:507-12.
Crowcroft, P., 1955. Notes on the behaviour of shrews. Behaviour, 8:63-81.
Crowcroft, P., 1957. The Life of the Shrew. London: Max Reinhardt.
Dalquest, W. W., and Orcutt, D. R., 1942. The biology of the least shrew-mole. Neurotrichus gibbsi minor. Amer. Midi. Nat., 27:387-401.
Davis, D. D., 1962. Mammals of the lowland rain-forest of North Borneo. Bull. Nat. Mus. Singapore, 31:1-130.
Davis, W. B., and Joeris, L., 1936. Notes on the lifehistory of the little short-tailed shrew. J. Mammal., 26:136-38.
Dryden, G. L., 1968. Growth and development of Suncus murinus in captivity on Guam. J. Mammal., 49:51-62.
Dryden, G. L., and Conaway, C. H., 1967. The origin and hormonal control of scent production inSuncus murinus. J. Mammal., 48:420-28.
Eadie, W. R., 1938. The dermal glands of shrews. J. Mammal., 19:171-74.
Eadie, W. R., 1939. A contribution to the biology of Parascalops breweri. J. Mammal., 20:150-73.
Eadie, W. R., 1947. The accessory reproductive glands of Parascalops with notes on homologies. Anat. Rec., 97:239-52.
Eadie, W. R., 1948. The male accessory reproductive glands of Condylura with notes on a unique prostatic secretion. Anat. Rec., 101:57-79.
Eibl-Eibesfeldt, I., 1965. Das Duftmarkieren des Igeltanrek (Echinops telfairi Martin). Z. Tierpsychol., 22:810-12.
Eisenberg, J. F., and Gould, E., 1966. The behavior of Solenodon paradoxus in captivity with comments on the behavior of other Insectivora. Zoologica, 51:49-58.
Eisenberg, J. F., and Gould, E., 1970. The Tenrecs: A Study in Mammalian Behavior and Evolution. Washington, D.C.: Smithsonian Institution Press.
Eisenberg, J. F., and Kleiman, D. G., 1972. Olfactory communications in mammals. Ann. Rev. of Ecology and Systematics, 31:1-32.
Firbas, W., and Poduschka, W., 1971. Beitrag zur Kenntnis der Zitzen des Igels ( Erinaceus europaeusLinne). Säugetierkundl. Mitt., 19:39-44.
Godet, R., 1951. Contribution à l'éthologie de la Taupe ( Talpa europaea L.). Bull. Soc. Zool. de France, 76:107-28.
Godfrey, G., and Crowcroft, P., 1960. The Life of the Mole. London: Museum Press.
Gould, E., 1965. Evidence for echolocation in the Tenrecidae of Madagascar. Proc. Amer. Phil. Soc., 109:352-60.
Gould, E., 1969. Communication in three genera of shrews (Soricidae): Suncus, Blarina, and Cryptotis. Communications in Behavioral Biology, part A, vol. 3:263-313.
Gould, E., and Eisenberg, J. F., 1966. Notes on the biology of the Tenrecidae. J. Mammal., 47:660-86.
Gould, E.; Negus, N. C.; and Novick, A.; 1964. Evidence for echolocation in shrews. J. Exper. Zool., 156:19-37.
Griffin, D. R., 1968. Echolocation and its relevance to communication behavior. In: Animal Communication, T. A. Sebeok, ed. Bloomington: Indiana University Press, pp. 154-64.
Hamilton, W. J., 1931. Habits of the star-nosed mole, Condylura cristata. J. Mammal., 12:345-55.
Hamilton, W. J., Jr., 1944. The biology of the little short-tailed shrew Cryptotis parva. J. Mammal., 25:1-7.
Kleiman, D. G., 1966. Scent marking in the Canidae. Symp. Zool. Soc. London, 18:167-77.
Kuhn, H.-J., 1964. Zur Kenntnis von Micropotamogale lamottei Heim de Balsac, 1954. Z Säugetierk., 29:152-73.
Liat, Lim Boo, 1967. Note on the food habits of Ptilocercus lowii Gray (pentail tree-shrew) and Echinosorex gymnurus (raffles) (moonrat) in Malaya with remarks on "ecological labelling" by parasite patterns. J. Zool. London, 152:375-79.
Marlow, B. J. G., 1954/55. Observations on the Herero musk shrew Crocidura flavescens herero, St. Leger, in captivity. Proc. Zool. Soc. London, 124:803-08.
Matthews, L. H., 1935. The oestrus cycle and intersexuality in the female mole (Talpa europaea Linné). Proc. Zool. Soc. London, 1935:347-83.
Mellanby, K., 1971. The Mole. London: Collins.
Mohr, E., 1936a. Biologische Beobachtungen an Solenodon paradoxus Brandt in Gefangenschaft. I. Zool. Anz., 115:177-88.
Mohr, E., 1936b. Biologische Beobachtungen an Solenodon paradoxus Brandt. II. Zool. Anz., 116:65-76.
Müller-Velten, H., 1966. Über den Angstgeruch bei der Hausmaus (Mus musculus L.). Z. vergl. Physiol., 52:401-29.
Niethammer, G., 1962. Die (bisher unbekannte) Schwanzdriise der Hausspitzmaus Crocidura russula (Hermann, 1780). Z Säugetierk., 27:288-34.
Niethammer, G., 1970. Beobachtungen am PyrenäenDesman, Galemys pyrenaicus. Bonner zool. Beitr., 18:157-82.
Noirot, E., 1968. Ultrasounds in young rodents. II. Change with age in albino rats. Anim. Behav., 16:129-34.
Noirot, E., and Pye, D., 1969. Sound analysis of ultrasonic distress calls of mouse pups as a function of their age. Anim. Behav., 17:340-49.
Ognev, S. I., 1928. Desmaninae. In: Mammals of Eastern Europe and Northern Asia. Vol. I, Insectivora and Chiroptera. Moscow-Leningrad: Glavnauka, 1928; Jerusalem: Israel Program for Scientific Translations Ltd., 1962.
Ottow, B., 1955. Biologische Anatomie der Genitalorgane und der Fortpflanzung der Säugetiere. Jena: G. Fischer.
Pearson, O. P., 1944. Reproduction in the shrew (Blarina brevicauda Say). Amer. J. Anatomy, 75:39-93.
Pearson, O. P., 1946. Scent glands of the short-tailed shrew. Anat. Rec., 94:615-29.
Poduschka, W., 1968. Über die Wahrnehmung von Ultraschall beim Igel, Erinaceus e. roumanicus. Z. vergl. Physiol., 61:420-26.
Poduschka, W., 1969. Ergänzungen zum Wissen über Erinaceus europaeus roumanicus und kritische Überlegungen zur bisherigen Literatur über europäische Igel. Z. Tierpsychol., 26:761-804.
Poduschka, W., 1970. Das Selbstbespeicheln der Igel. Film CT 1320 der Bundesstaatl. Hptst. f. wiss. Kinemat., Vienna.
Poduschka, W., 1971. Zur Kenntnis des nordafrikanischen Erdhörnchens, Atlantoxerus getulus (F. Major). Zool. Garten (Leipzig), N.F., 40:211-26.
Poduschka, 1972a. Das Paarungsverhalten des Grossen Igel-Tenrek, Setifer setosus (Froriep, 1806). Film CTF 1451 der Bundesst. Hptst. f. wiss. Kinemat., Vienna.
Poduschka, W., 1972b. Augendriisensekretionen der Tenreciden Setifer setosus, Echinops telfairi, Microgale dobsoni und Microgale talazaci. Film CTF 1520 der Bundesstaatl. Hptst. f. wiss. Kinemat., Vienna.
Poduschka, W., 1973. Erinaceus europaeus, Hemiechinus auritus (Erinaceidae): Selbstbespeicheln. Encycl. Cinemat. E 1962/1973, Göttingen.
Poduschka, W., 1974a. Das Paarungsverhalten des Gro βen Igel-Tenrek (Setifer setosus, Froriep, 1806) und die Frage des phylogenetischen Alters einiger Paarungseinzelheiten. Z. Tierpsychol., 34:345-58.
Poduschka, W., 1974b. Augendrüsensekretionen bei den Tenreciden Setifer setosus (Froriep, 1806), Echinops telfairi (Martin, 1838), Microgale dobsoni (Thomas, 1918) und Microgale talazaci (Thomas, 1918). Z Tierpsychol, 35:303-19.
Poduschka, W., 1974c. Fortpflanzungseigenheiten und Jungenaufzucht des Gro βen Igel-Tenrek, Setifer setosus (Froriep, 1806). Zool. Anz., Jena 193 (314): 145-80.
Poduschka, W., 1974d. Sinnes- und verhaltensphysiologische Beobachtungen an Solenodon paradoxus Brandt. Manuscript.
Poduschka, W., 1974e. Die verschiedenen Markierungsarten des Groβen Igel-Tenrek, Setifer setosus (Froriep, 1806). Manuscript.
Poduschka, W., 1974f. Markierungsverhalten des Grossen Igel-Tenrek, Setifer setosus (Froriep, 1806). Film CTF 1562 der Bundesst. Hptst. f. wiss. Kinemat., Vienna.
Poduschka, W., 1976. Das Paarungsvorspiel des Osteuropäischen Igels (Erinaceus europaeus roumanicus) und theoretische Überlegungen zum Problem männlicher Sexualpheromone. (Manuscript.)
Poduschka, W., and Firbas, W., 1968. Das Selbstbe-speicheln des Igels, Erinaceus europaeus Linné, 1758, steht in Verbindung zur Funktion des jacobson-schen Organes. Z. Säugetierk., 33:160-72.
Quay, W. B., 1965. Comparative survey of the sebaceous and sudiferous glands of the oral lips and angle in rodents. J. Mammal., 46:23-36.
Quilliam, T. A., 1966. The mole's sensory apparatus. J. Zool. London, 149:76-88.
Rahm, U., 1961. Beobachtungen an dem ersten in Gefangenschaft gehaltenen Mesopotamogale ruwenzorii (Mammalia—Insectivora). Rev. Suisse Zool., 68:73-90.
Ralls, K., 1971. Mammalian scent marking. Science, 171:443-49.
Ravizza, R. J.; Heffner, H. E.; and Masterton, B.; 1969. Hearing in primitive mammals. II: Hedgehog (Hemiechinus auritus). J. Audit. Res., 9:8-11.
Reed, C. A., 1944. Behaviour of a shrew-mole in captivity. J. Mammal., 25:196-98.
Richard, P. B., 1973. Le Desman des Pyrenées (Galemys pyrenaicus). Mode de vie. Univers sensoriel. Mammalia, 37:1-16.
Richard, P. B., and Valette Viallard, A., 1969. Le Desman des Pyrenees (Galemys pyrenaicus): Premieres notes sur sa biologie. La Terre et la Vie, 3:225-45.
Rood, J. P., 1958. Habits of the short-tailed shrew in captivity. J. Mammal., 39:499-507.
Sales, G. D., 1972. Ultrasound and aggressive behavior in rats and other small mammals. Anim. Behav., 17:542-46.
Stephan, H., and Bauchot, R., 1959. Le cerveau de Galemys pyrenaicus Geoffroy, 1811 (Insectivora Talpidae) et ses modifications dans l'adaption à la vie aquatique. Mammalia, 23:1-18.
Stephan, H., and Bauchot, R., 1960. Les Cerveaux de Chlorotalpa stuhlmani (Matschie) 1894 et de Chryso-chloris asiatica (Linné) 1758 (Insectivora, Chrysochloridae). Mammalia, 24:495-510.
Stephan, H., and Bauchot, R., 1968a. Gehirn und Endocranialausguβ von Desmana moschata (Insectivora Talpidae). Morphol. fb., 112:213-25.
Stephan, H., and Bauchot, R., 1968b. Vergleichende Volumenuntersuchungen an Gehirnen europäischer Maulwiirfe (Talpidae). J. Hirnforschung, 10:247-58.
Strauss, F., 1939. Die Befruchtung und der Vorgang der Ovulation bei Ericulus aus der Familie der Centetiden. Biomorphosis, 1:281-312.
Strauss, F., 1942. Vergleichende Beurteilung der Placentation bei den Insectivoren. Rev. Suisse Zool., 49:269-82.
Thenius, E., and Hofer, H., 1960. Stammesgeschichte der Säugetiere. Berlin, Göttingen, Heidelberg: Springer.
Vogel, P., 1969. Beobachtungen zum intraspezifischen Verhalten der Hausspitzmaus (Crocidura russula Hermann, 1870). Rev. Suisse Zool., 76:1079-86.
Vogel, P., 1970. Biologische Beobachtungen an Etruskerspitzmausen (Suncus etruscus Savi, 1832). Z.Säugetierk.,35:173-85