In a recent paper in The Journal of Experimental Biology, Madsen and his colleagues, ‘hypothesize that all toothed whale species only click with one set of their phonic lips at a time, and preferably their right pair’ [(Madsen et al., 2010) see p. 3110 of their article]. This conclusion was based on a limited data set from one odontocete species, and runs counter to established literature that may not have been adequately reviewed.
The authors of the paper in question:
offer a presumptuous hypothesis [proposing to know how all (73) odontocete species behave from their observational data for one species];
demonstrate overextended extrapolation (claiming knowledge of behavior for which they have no data);
provide an inadequate description of methodology (with no means to determine whether they actually studied echolocation);
present a functional proposal that does not match the morphological complexity in the sound generation apparatus;
fail to adequately consider contradictory anatomic evidence; (vi) omit contradictory reports in long-established literature.
In my view, it is somewhat presumptuous of these authors to posit the behavior of 70 or so species of mammals (the entire Suborder: Odontoceti) with any specificity, particularly when it is based on a single set of experimental observations using one small odontocete species.
Madsen and colleagues assert that the nasal apparatus on the left side can also be used to make clicks. But as their study never recorded a click from the left side, it is unclear how they could arrive at this conclusion. They then assert that the click generator on the left side is only used for communication, a function for which they also report no observations.
The methodological description by Madsen and his colleagues is inadequate to establish that they were, in fact, studying echolocation behavior. What exactly was the ‘echolocation’ task? For example, what was the size of the target and how far in front of the animal was the target placed? Was the fish only the reward, or was it also the target? How was the target presented? In any psychoacoustic experiment of echolocation behavior, safeguards must be in place to ensure that the animals are not using vision and are actually engaged in echolocation. There is no indication that the animals in Madsen and colleagues' experiments were blindfolded or otherwise prevented from solving the problem visually. Without these safeguards, they cannot claim to have studied echolocation.
There is also no indication of the level of difficulty of the task; was it target detection or discrimination? Madsen et al. report that their animals solved simple ‘short-range echolocation’. As a consequence, it is possible that they posed a problem so trivial that it did not require extraordinary skill or effort for the animals to solve.
Another primary objection is that their assessment of function does not match the anatomic complexity in the odontocete forehead. It would seem that the anatomic complexity of pertinent bilateral sound generation features (phonic lips and associated structures) is being maintained by natural selection, and this argues for functional complexity.
Anatomic structures tend to atrophy from disuse. Therefore, it is reasonable to expect that disuse of a biological structure would lead to inheritance of ‘architectural flaws’ that would accumulate over time if natural selection were not exerting pressure to keep the structural arrangement within some narrow range of tolerance. But despite their argument for disuse (see p. 3110 of their article), using sperm whale anatomy as a foil, Madsen and colleagues propose that the opposite is true in porpoises – that the structural complexity of both sets of phonic lips is conserved. It would appear that Madsen et al. have the worst case scenario to prove their case, because both sets of phonic lips and associated tissues in porpoises are intricate and virtually symmetrical (Cranford et al., 1996). The intricate bilateral sound production apparatus in the Odontoceti is nearly ubiquitous and therefore problematic for Madsen and colleagues' all-inclusive conclusion.
This conundrum is acknowledged in the last sentence of their paper: ‘Why they seemingly carry two identical phonic lip pairs while apparently only using one pair at a time for clicking needs to be addressed in future experiments...’
This statement reveals a weakness in their central conclusion. They are aware of the contradiction but may not have considered its implications or alternative explanations. Rather than address this contradiction directly, they suggest it is a prompt for additional research. No such catalyst is necessary; investigations of odontocete sonar signal generation have been a source of vigorous debate in cetology for almost half a century (Cranford and Amundin, 2003).
Madsen and colleagues have also omitted pertinent literature, including several studies that appear to contradict their primary conclusion: that all odontocetes click with only one set of lips at a time – preferably the right pair.
In 1962, John Lilly worked with bottlenose dolphins and reported, ‘Our animals tend to click only on the left side and whistle only on the right side and can do so simultaneously or separately’ [(Lilly, 1962) see p. 522 and fig. 8 of his paper]. Madsen and colleagues apparently also did not consider the results reported by Dr Lilly in 1978 (Lilly, 1978). Lilly described an experiment whose design is similar to that used by Madsen and colleagues. Lilly [(Lilly, 1978) see p. 68 and fig. 7] described dual (stereo) click sources (one on each side of the head). These studies directly contradict Madsen and colleagues' conclusion that all odontocetes preferably use the right side for clicking.
In addition, Madsen et al. did not cite the seminal work (in multiple papers) conducted by Dr William E. Evans, which can be directly compared on the issue of odontocete sonar signal generation. In 1973, Evans recorded, ‘click formation on either or both sides of the blowhole’ (Evans, 1973). Like Madsen and colleagues, Evans also used contact hydrophones and hydrophone arrays to record pulses emanating from the head of bottlenose dolphins (Tursiops truncatus). In calculations that must be similar to those of Madsen et al., Evans reported two distinct pulses being produced by two sources, one on each side of the midline [(Evans, 1973) see p. 197, paragraphs 1 and 2 of his paper), and that their dolphins could use ‘either or both’. This work also contradicts Madsen's conclusion because Evans indicates that dolphins can produce sonar clicks from both sources, separately or simultaneously.
In some respects, if Madsen et al. agree that all odontocetes (except sperm whales) have two click sources then there is only disagreement about function. The evidence that bilateral sources are being maintained by natural selection suggests that these sound generators are used when needed and probably for a multiplicity of functions. Contrary to the suggestion of Madsen and colleagues, it is unlikely that it can be known how the broad spectrum of odontocete species will behave from the paucity of evidence that now exists. The arguments put forth by Madsen and colleagues are in doubt because of deficiencies in the methodological description and contradictory evidence in the established scientific literature.
