Somatosensation

Pinnipeds (walrus, sea lions, seals) are a group of carnivorous mammals that breed on land but feed in the water. These mammals have highly developed mystacial vibrissae, or whiskers (see Fig. 1), that they use in conjunction with acute color vision to navigate their environment and track prey (Griebel and Schmid, 1996; Dehnhardt et al., 2001). The whiskers can detect and analyze changes in underwater vortices to Dehnhardt et al. (2001) demonstrated that the harbor seal (Phoca vitulina) can use its whiskers to detect wakes caused by swimming fish, allowing them to track the fish and ultimately feed on them. The experiments placed a small submarine into a tank and propelled it through the water. It was then removed, and a seal was quickly introduced into the tank. The seal was able to follow the trajectory of the submarine utilizing vortices left in its wake. Furthermore, the seal was able to complete this task while blindfolded, but not while its whiskers were covered. Researchers concluded that seals use their highly evolved whiskers to compensate for the lack of underwater olfactory capabilities. As demonstrated in another study of similar design, the California sea lion (Zalophus californianus) also uses its mystacial vibrissae for hydrodynamic trail following (Gläser et al., 2011). Unlike in harbor seals, the sea lions' performance degraded with increased delay between removing the submarine and placing the animal in the tank.

Qualitative differences between the whiskers of sea lions and seals may explain variations in the animals' performance. Miersch et al. (2011) isolated single whiskers of both animals and tested the whiskers' response to underwater wakes. For both types of whiskers the researchers calculated a signal-to-noise ratio (SNR), or ratio of the power between ‘wanted’ signals and background noise. Results showed that while both whisker types responded accurately to vortices in the water tank, as supported by the aforementioned studies of these species, harbor seal whiskers significantly reduced background noise compared to California sea lion whiskers (Miersch et al., 2011). Walruses are the other members of the pinnipeds, but the role of their whiskers is largely unexplored. Future experiments should examine whether walruses use their whiskers to respond to underwater vortices and ultimately locate food sources.

Fig. 1.

A close-up picture of a sea lion (Zalophus wollebaeki) reveals a dense matrix of vibrissae. These tactile end-organs function to detect particles and vortices in the underwater environment. Photo by RM Hallock.

Little research has focused on the foraging techniques of sirenians, but mammals of this order have a highly developed somatosensation system that may aid in the foraging process. As with pinnipeds, sirenians have sensitive vibrissae that help these animals navigate the environment, with the aid of color vision (Griebel and Schmidt, 1996; Newman and Robinson, 2006). The West Indian manatee has over 5300 vibrissae covering its entire body; about 2000 of these vibrissae are located in the facial region (Reep et al., 1998). Manatees use their facial hairs for different functions (Bachteler and Dehnhardt, 1999) and both facial and postfacial vibrissae help the manatee detect hydrodynamic stimuli and navigate its environment (Reep et al., 2011). However, only facial vibrissae are used for tactile exploration. When a manatee investigates a potential food source or novel object, it extends oral vibrissae to increase tactile acuity and analyze the object's qualities. The manatee's vibrissae are extremely sensitive compared to the nasal appendage of the star-nosed mole or the trunk of an elephant (Bauer et al., 2005). Sarko et al. (2007) investigated the brain anatomy of the manatee, and found that nuclei devoted to somatosensation (particularly stemming from facial vibrissae) were disproportionately large compared to nuclei of other sensory inputs. These findings further support the notion that the manatee's primary adaptation is their advanced vibrissae. Interestingly, perioral vibrissae (thick pairs of vibrissae around the mouth) are also used to grasp plants during feeding.

Unlike pinnipeds and cetaceans, sirenians are herbivorous and thus tracking is less of the focus than the proper identification of stationary food sources. Future research should test the exact roles of vision and somatosensation in the foraging process. This may be tested using a mask that disrupts the vibrissae from receiving information, or alternatively a blindfold that masks the animal's vision. Manatees can be observed grazing along the bottom of the water, so it is also possible that contact between postfacial vibrissae and food sources are the animal's primary food detection mechanism. Yet experimental masking of postfacial vibrissae is unnecessary, as the role of facial vibrissae and color vision can be tested with greater ease.

Echolocation

Dolphins do not have a sense of smell to aid in the food foraging process (Cahill, 2000), and although dolphins do have an acute visual sense, their primary foraging adaptation is of course a highly specialized echolocation system. Echolocation is the production of sound waves and the subsequent obtainment of information from the waves' reverberations. The returning sounds are variations of the initial sounds produced, and through comparison of these returning signals, dolphins effectively perceive the topology of their surroundings (Jones, 2005).

Dolphins' vocalizations are produced by air movements in the nasal passage and not the larynx. Recently it has been proposed that the structural “monkey lips”/dorsal bursae (MLDB) complexes on either side of the head are crucial for sound production (Cranford and Amudin, 2004). The MLDB includes phonic lips, which vibrate in response to an increase in air pressure, generating signals upwards of 130 kHz (Cranford and Amundin, 2004). The signals then pass through the melon on the forehead, a structure consisting of special fats that serve to focus outgoing sound waves and project the waves forward (see Fig. 2); interestingly, the melon may filter out frequencies higher than 160 kHz (Madsen et al., 2010; Jones, 2005). Recent research demonstrates that in short-distance prey tracking, the harbor porpoise only produces clicks with the right pair of phonic lips, despite possessing a pair on the left side as well (Madsen et al., 2010). Using just the right pair, dolphins are still able to vary sound output and direction of echolocative clicks (Madsen et al., 2010). Future research should describe the function (if present) of the left pair of phonic lips.

In one of the first studies to suggest that dolphins use echolocation, Kellogg (1958) demonstrated that these animals are able to distinguish between different sizes and types of fish. Though initially making some errors, over time the dolphins were able to accurately choose target fish, even in darkness and in turbid water. The dolphin emitted sounds under the water that increased in frequency and intensity when a target fish was introduced. The author concluded that dolphins are able to distinguish various sizes of stimuli, and navigate the surrounding environment by using echolocation.

Norris et al. (1961) also studied bottlenose dolphins to investigate echolocation and the ability to discriminate stimuli. Though the dolphins were blindfolded with latex cups, the animals were still able to locate targets suspended randomly below the water surface. In a subsequent test, a maze was created in the dolphin's tank using poles, and fish were dropped into the water near the poles. The blindfolded dolphin was able to get to the fish while avoiding the poles in all but one case. The combined results from these experiments helped demonstrate that echolocation, rather than vision, is primarily responsible for the bottlenose dolphin's ability to avoid obstacles and locate food sources. Research has subsequently demonstrated that the amplitude of the emitted sound waves depends on the range to the target, the ambient noise level in the environment, and the difficulty of the task (Au, 1993; Li et al., 2011).

Fig. 2.

Right lateral view of Tursiops skull showing current model of echolocation. Figure kind courtesy A.T. White.

Herman et al. (1998) used object pairings to determine if the dolphins use vision or echolocation to match novel stimuli. The results indicated that bottlenose dolphins are able to generate a direct shape percept from echolocation, which allows for the accurate matching of various objects. They suggested that echolocation yields a representation of an object that directly corresponds to a vision-derived percept. Additionally, these experiments demonstrate that bottlenose dolphins do not seem to rely on specific details or features of the objects in question, but rather perceive the objects holistically (Herman et al., 1998). Similar findings by Kellogg (1958) and Norris et al. (1961) provide evidence that echolocation in bottlenose dolphins seems to replace sight and smell in the water and allow these mammals to encode their surroundings.

While bottlenose dolphins are used as model organisms for echolocation research, the ability to use echolocation for feeding and navigation is seen in other mammals of the order Cetacea, including various species of toothed whales (Barrett-Lennard et al., 1996; Simonis et al., 2012; Johnson et al., 2004). As in dolphins, toothed whales have two pairs of phonic lips capable of producing vocalizations (Cranford et al., 1996). Lammers and Castellote (2009) suggested that toothed whales actuate both pairs of phonic lips to produce one strong, directional echolocative click. However, Madsen et al. (2010) demonstrated that the harbor porpoise only utilizes its right pair of phonic lips for echolocation, establishing this species as an exception. Unlike toothed whales, baleen whales do not use echolocation. Advanced adaptations past the baleen filter system are presumably unnecessary for foraging. Though echolocation is an effective tool for navigation and locating prey, it is unable to detect pheromones or other semiochemicals in the marine environment.

Although it has been shown that cetaceans rely on echolocation rather than a sense of smell to navigate their habitat and forage for food, the relationship between the acquisition of echolocation and the loss of olfactory capabilities is not well documented. Two hypotheses have been proposed to explain the loss of smell in cetaceans that utilize echolocation verses those that rely on filter-feeding (including the mysticetes): the echolocation priority hypothesis, which suggests that the acquisition of echolocation caused the reduction in importance of olfaction, and the filterfeeder hypothesis, which proposes that the filter-feeders maintain an olfactory system so that they can locate prey based on the particular odor given off by clouds of plankton (Kishida and Thewissen, 2012). These latter authors' analyses of the selective evolutionary pressures on the olfactory marker protein (OMP) genes in various species of cetaceans supported the disparate foraging methods (echolocation versus filter-feeding), providing evidence for the filter-feeder hypothesis over the echolocation priority hypothesis.

Rather than the development of echolocation causing the sense of smell loss, it may be that the evolutionary pressures led to a decrease in importance of olfaction in those cetaceans that found echolocation more adaptive to their aquatic habitat. Furthermore, Kishida and Thewissen (2012) suggested that the cetacean sense of smell decreased gradually, and that it is possible that the olfactory marker protein may still have some functionality. Thus, the acquisition of echolocation may not have caused the loss of smell in cetaceans, but instead the evolutionary pressures of the habitat may have led to both. The relationship between the development of echolocation and the loss of a sense of smell in fully aquatic mammals is an important evolutionary question that warrants further research.

Underwater sniffing

Intriguingly, the semiaquatic mammals Condylura cristata (star-nosed mole) and Sortex palustris (water shrew) have adapted a process that allows underwater detection of semiochemicals. These animals exhale air bubbles onto underwater objects or scent trails and then inhale them through their nose, allowing diffusion of the newly-airborne odorant molecules into the olfactory epithelium (Catania, 2006). This underwater sniffing behavior is known as the bubble technique. A sniff is defined as a sequence of nasal inhalations that increases the velocity and alters the duration of airflow in the nose to obtain olfactory information (Kepecs et al., 2006; Scott, 2006). Experimentation on rodents indicates that only a single sniff is necessary for animals to discriminate between odors and to accurately sample the olfactory environment (Kepecs et al., 2006). Indeed the kinematics of underwater sniffs parallel the kinematics of terrestrial rodent sniffs, in terms of the volume of air (corrected for body weight) exhaled and inhaled, rate of airflow, and frequency (Catania, 2006, 2009). Such similarities support the notion that the underwater sniffing behavior is in fact an olfactory process.

The star-nosed mole is recognized for its ornate somatosensory organ (see Fig. 3), but video recordings have also suggested that the animal uses the bubble technique to forage for food underwater (Catania, 2005). In order to determine if the pattern of exhalations and re-inhalations observed underwater were a means of odor sampling, Catania et al. used a forced choice paradigm in which moles were trained to follow a trail of earthworm scent laid on one of two underwater paths (Catania, 2009). To ensure that only olfactory cues were used to choose/follow the path, trails were laid in a channel and covered with a steel grid that allowed the air bubbles to pass through freely, but prevented contact with the mole's nose. Moles chose the correct trail with 85% accuracy. When a finer mesh grid prevented bubbles from contacting the scent trail, performance dropped to chance levels. Such results indicate that the starnosed mole uses the bubble technique to locate and track prey underwater.

Fig. 3.

The star-nosed mole's unique somatosensory organ, composed of 11 pairs of fleshy appendages for tactile perception. The image on the left features the star-nosed mole (Condylura cristata) in the midst of an underwater sniff, with bubbles being re-inhaled by the nose. The image on the right zooms in on the nose and the air bubble. Photo courtesy of Kenneth Catania.

Catania et al. (2008) also recorded water shrews exhibiting the bubble technique. Two shrews were trained to follow an underwater fish scent trail and were able to follow the path in 80% and 85% of the trials, but performed at chance when a grid blocked their bubbles. Interestingly, when a terrestrial shrew (Blarina brevicauda) was trained to retrieve food from a shallow well, they were able to retrieve the reward but did not use underwater sniffing (Catania, 2009). This evidence suggests that the bubble technique is an olfactory strategy that has been exclusively adopted by semiaquatic species. Given that the water shrew and starnosed mole do not share a common semiaquatic ancestor, and seem to have evolved this mechanism of olfaction independently (Catania, 2009).

A recent BBC program, “Halcyon River Diaries,” showed footage of river otters exhaling and then re-inhaling bubbles that they expelled directly on the river substrate (James, 2010) (see Fig. 4). Future research should use controlled feeding experiments to test whether these otters use the bubble technique to sniff underwater. The independent acquisition of the technique in different species suggests that it is an efficient and effective means of underwater olfaction. This technique has been found in a range of animals and further investigation may discover additional semiaquatic mammals that have adapted the bubble technique for underwater foraging.

Fig. 4.

This image shows a small bubble (white arrow) released from the nostril of a river otter (Lutra lutra) that comes into contact with the river substrate. The bubble was re-inhaled, and may represent usage of the bubble technique to sample particles from the underwater milieu. Photo used with kind permission from Charlie Hamilton James.