This April, our team, including Matthew Colbert and James Carew, published a paper announcing the discovery of Cotylocara macei, a 28 million-year-old species of fossil whale, and its importance for the evolution of echolocation. Echolocation is a behavior similar to sonar; an animal makes a sound and then interprets the delay in hearing the echoes to detect nearby objects and their locations. Bats are probably the best-known echolocators, but all living dolphins, porpoises, and toothed whales (classified in suborder Odontoceti), do as well. It is no coincidence that both bats and odontocetes use high frequency sounds to echolocate; the short-wavelengths of high frequency echoes enable them to produce a high resolution “audio-picture” of their surroundings. Filter-feeding whales, like the humpback, do not echolocate but instead produce low frequency sounds, which might reveal the presence of a submarine canyon but could not be used to locate individual fish. Non-echolocating whales, all bats, and other mammals, including us humans, produce sounds via vibrations of vocal cords in the larynx. Odontocetes are truly unusual in that they produce vocalizations in their face! Although now the scientific consensus, it took years of research to establish that fleshly folds in the nasal passages between the blowhole and the skull – called the phonic lips – are the source of odontocete clicks, whistles, and buzzes. The function of the phonic lips is supported by a strange cast of anatomical structures: cartilage that reinforces the lips, air sacs for reflecting sound waves, and large facial muscles to control the flow of air.
When confronted with the complex anatomy underlying echolocation, an obvious question arises: how did it evolve? It is my habit to look to the fossil record for answers, but here’s the problem. Soft tissues almost never fossilize, and there are, as yet, no fossil odontocetes that preserve soft tissues of the face. Thus at first glance, the evolution of echolocation appears to be a non-starter—a story of biological evolution that has been lost to the relentless process of decay. However there is hope; skulls of fossil odontocetes are not uncommon and the shape of the skull is strongly influenced by the soft tissues that originally enclosed it. Thus based on correlations between the anatomy of soft tissues and skulls in living whales, we argue in our Nature paper that some aspects of echolocation can be reconstructed from fossils. It is with this mindset that the very bizarre skull of our new fossil whale, Cotylocara macei, starts to make sense.
The only known skull of Cotylocara macei was discovered in a drainage ditch northwest of Charleston, South Carolina, in the early 1990s. Although more commonly associated with historic homes, the Civil War, and beautiful beaches, the Charleston area is also endowed with one of the richest and most important fossil whale faunas in the world. Thus although unique, the skull of Cotylocara is by no means an anomaly. Cotylocara belongs to the Xenorophidae, an extinct and diverse family that diverged early from the stock that eventually gave rise to all living odontocetes. The most salient feature of its skull is a pair of deep pits on the forehead that are separated by a narrow partition of bone. In this structure Cotylocara is unique among all living and fossil odontocetes and maybe all mammals (some tapirs and bats have a similarly placed depression, though not as deep). Another large pair of depressions occurs at the base of the snout. The skull is also quite asymmetric; when viewed from the front it is twisted counterclockwise and in several areas paired bones do not meet on the mid-line but are shifted toward one side. Although asymmetry is common in living odontocetes, the degree of asymmetry in Cotylocara is unprecedented for a fossil of its age. Even more surprising is the configuration of the main tooth-bearing bone of the skull, the maxilla. Like many living odontocetes, the maxilla has clearly adopted other functions as it extends back over the orbit and nearly reaches the back of the skull. The bones of the face are also unusually dense, including one that encircles the nose opening. Although the above might read like a list of anatomical oddities, we think that the presence of a relatively advanced sound producing apparatus, such as the phonic lips, can explain all of these features.
To produce sound, odontocetes push air past the phonic lips, similar to the squeaks generated when slowly releasing air from a balloon. They tightly control the passage of air with very large and complex facial muscles that attach to and around the phonic lips. In living species, these facial muscles consistently originate on the expanded part of the maxilla that sits over and behind the orbits. Thus the very large and backwardly expanded maxilla in Cotylocara suggests that it too had large facial muscles that could modulate the passage of air past the phonic lips. Living odontocetes can also focus their vocalizations by reflecting “wayward” sound waves at boundaries between materials of different densities. For example, in a recent study on Cuvier’s beaked whale, Dr. Ted Cranford (San Diego State University) and colleagues showed that an expanded plate of dense bone situated behind and below the phonic lips helps redirect sound waves propagating backward so that they are moving forward instead. We think that the dense bones of Cotylocara had a similar function, particularly the flange that encircles the bony nostrils. It has also been suggested that air sinuses reflect sound waves because of the large density difference between air and tissue. Air sinuses are abundant in the heads of living odontocetes and appear to excavate the bone around them. The resulting deep cavities and the struts of bone that support them are clear indicators of soft-tissue-lined air sinuses. Thus we interpret the skull cavities in the base of the snout and on the forehead of Cotylocara as being formed by air sinuses, and that these sinuses, particularly those on the snout, would have reflected sound waves forward.
Artistic reconstruction of Cotylocara macei as it searches for fish in coastal waters of what is now South Carolina, 28 million years ago. Artwork by Carl Buell.
Another striking feature of the skull of Cotylocara is asymmetry, although one far more difficult to interpret. Asymmetry is common to all living odontocetes, and there are three different hypotheses for its importance. The first directly links asymmetry to echolocation; differences between the soft tissues on the right and left sides could allow for a more diverse acoustic repertoire or, alternatively, one side might specialize in vocalization whereas the other in breathing. The second hypothesis is that the asymmetry allows odontocetes to consume larger prey. The respiratory and digestive paths of odontocetes are completely separate, which means that they cannot choke on what they ingest. The downside to this arrangement is that the larynx divides the throat into right and left sides, and food must fit through the gaps on either side to be swallowed. Odontocetes that have asymmetric skulls tend to have the larynx shifted toward one side, which means that larger prey can be swallowed. Despite evidence for this hypothesis, the greatest asymmetry occurs well removed from the larynx, suggesting that asymmetry in the face and throat are only loosely linked. The third hypothesis is that asymmetry is related to high frequency hearing, based largely on the behavior of owls. In these birds, the ears are at different levels, and the difference in time between hearing a sound in the right versus the left ear allows owls to triangulate the source. Currently there is not enough evidence to determine which, if any, of these ideas applies to odontocetes. Thus asymmetry in Cotylocara is tantalizing, but uncorroborated, evidence for echolocation.
If Cotylocara macei did echolocate, as our study suggests, then the simplest hypothesis is that echolocation evolved very early in the history of odontocetes, sometime between 35 and 32 million years ago. In large part, our case relies on anatomical relationships and functional hypotheses for living species. One potential pitfall is that we have to make many assumptions, and there is a fine line between solid scientific inference and developing a story to fit known facts. One way, and probably the best way, to draw this line is testability; if you cannot test an idea, then there is no way to gain confidence in its truth. Given how challenging it is to reconstruct soft tissues in fossils, is there any way to test our hypothesis to see if it is correct? Fortunately there is, and these possible tests give clear direction to future research in this area. First, I have focused on the sound emission component of echolocation, but odontocetes must also be able to hear and interpret the resulting high frequency echoes for the entire system to work. Although the inner ear bones of Cotylocara are not well preserved, those of other fossil odontocetes are, and we predict that their ears will show evidence of the ability to hear high frequency sounds. Second, if echolocation evolved early, as we suggest, then some of the skull features that we found in Cotylocara should occur in other fossil species, albeit in more subtle ways. There are likely other tests that will only become obvious after additional discoveries are made.
Cotylocara macei has provided us some important clues for determining when echolocation evolved in odontocetes. My sense is that this is really just the beginning, and that we stand on the cusp of a new period for evolutionary studies where the merger of abundant data from living species with the unique insights provided by fossils will accelerate the pace of scientific discoveries. Some of these discoveries could challenge our hypothesis for the evolution of echolocation or lead to alternative explanations for our observations, yet this how science works. It is a price I am more than willing to pay for the privilege to study what can only be described as a wonderful history of anatomical adaptation.
Jonathan Geisler is an Associate Professor of Anatomy at the New York Institute of Technology. Find his study in Nature journal online.
The Paleontological Research Institution, Ithaca, New York, is pleased to sponsor Paleontology content for This View of Life. Founded in 1932, PRI has outstanding programs in research, collections, and publications, and is a national leader in development of informal Earth science education resources for educators and the general public.