98
D.R. Ketten
of a T. truncatus signal (Type II dolphin; -21 dB re 1j/m
2 ) is greater than in
other dolphins and substantially different from that of the Type I, P. phocoena signal (-74dB re 1j/m
2 ) (Au 1993). Among bats, Eptesicus fuscus, the
big brown bat, is a T. truncatus parallel with an efd (-66.4dB re 1j/m
2 ) only
slightly larger than that of the Type I dolphin but substantially larger than
that of other bats. T. truncatus is primarily an open water forager; E. fuscus
(FM bat) is an open field forager. Both use comparatively high-energy,
lower range ultrasonic signals tolerant to Doppler shift in an open environment. By comparison, both P. phocoena and its parallel, Rhinolophus
ferrumequinum (the horseshoe bat, CF/FM) , have low-energy: highfrequency, narrow band signals. Both also have good discrimination and
deal primarily with imaging small objects in "cluttered" habitats that
acoustically are filled with time-smeared echoes from twigs, leaves, etc. and
their submerged, shallow water counterparts. Structurally, P. phocoena and
R. ferrumequinum both have highly specialized basilar membrane structures with foveal regions and high ganglion cell densities. This is consistent
with the conclusion that habitat and task-dependent signal characteristics
are tied to species-specific inner ear filter and response characteristics.
These comparisons are tenuous and are brought forth here primarily to
engender discussion. The similarities in relative signal parameters and
common cochlear formats between bats and dolphins raises interesting
questions about how overtly different habitats may have had common
selection pressures that led to parallel echolocation strategies. They also
suggest that cross-species hunts for task-related auditory adaptations in
different habitats could be a useful tool for understanding fundamental
auditory mechanisms.
The structural commonalities between CF/FM bat cochleae and Type I
odontocetes suggest that parallel processing strategies may have evolved
across media, despite the differences in scale and signal characteristics. The
CF/FM bat auditory system is thought to be geared in large part to process
Doppler phenomena; there is no evidence for odontocetes that Doppler
shift analyses are employed, and because of the broadband nature of the
majority of dolphin sonar clicks, there is good reason to think that they,
like the signals of FM bats, are Doppler tolerant (i.e., Doppler insensitive)
(Au, Chapter 9). However, it is also worth noting that the majority of data
on odontocete signals that propel us to this conclusion comes from Type
II animals. Au (1993) makes the comment that P. phocoena and, indeed,
several related phocoenid species produce narrow band, low-intensity,
ultra-high signals (peak spectra >120 kHz) that are markedly different from
those of delphinids. Those observations do not mean that Type I animals
are signal and processing aquatic clones of CF/FM bats. However, if the
signal data are reviewed in the context of ganglion cell densities an intriguingly consistent picture begins to form. All relevant data are preliminary,
but ganglion cell spikes to over 1O,000/mm (Ketten 1998a) located in the
mid basal turn segment of the P. phocoena ear are coincident with the pro-
D.R. Ketten
of a T. truncatus signal (Type II dolphin; -21 dB re 1j/m
2 ) is greater than in
other dolphins and substantially different from that of the Type I, P. phocoena signal (-74dB re 1j/m
2 ) (Au 1993). Among bats, Eptesicus fuscus, the
big brown bat, is a T. truncatus parallel with an efd (-66.4dB re 1j/m
2 ) only
slightly larger than that of the Type I dolphin but substantially larger than
that of other bats. T. truncatus is primarily an open water forager; E. fuscus
(FM bat) is an open field forager. Both use comparatively high-energy,
lower range ultrasonic signals tolerant to Doppler shift in an open environment. By comparison, both P. phocoena and its parallel, Rhinolophus
ferrumequinum (the horseshoe bat, CF/FM) , have low-energy: highfrequency, narrow band signals. Both also have good discrimination and
deal primarily with imaging small objects in "cluttered" habitats that
acoustically are filled with time-smeared echoes from twigs, leaves, etc. and
their submerged, shallow water counterparts. Structurally, P. phocoena and
R. ferrumequinum both have highly specialized basilar membrane structures with foveal regions and high ganglion cell densities. This is consistent
with the conclusion that habitat and task-dependent signal characteristics
are tied to species-specific inner ear filter and response characteristics.
These comparisons are tenuous and are brought forth here primarily to
engender discussion. The similarities in relative signal parameters and
common cochlear formats between bats and dolphins raises interesting
questions about how overtly different habitats may have had common
selection pressures that led to parallel echolocation strategies. They also
suggest that cross-species hunts for task-related auditory adaptations in
different habitats could be a useful tool for understanding fundamental
auditory mechanisms.
The structural commonalities between CF/FM bat cochleae and Type I
odontocetes suggest that parallel processing strategies may have evolved
across media, despite the differences in scale and signal characteristics. The
CF/FM bat auditory system is thought to be geared in large part to process
Doppler phenomena; there is no evidence for odontocetes that Doppler
shift analyses are employed, and because of the broadband nature of the
majority of dolphin sonar clicks, there is good reason to think that they,
like the signals of FM bats, are Doppler tolerant (i.e., Doppler insensitive)
(Au, Chapter 9). However, it is also worth noting that the majority of data
on odontocete signals that propel us to this conclusion comes from Type
II animals. Au (1993) makes the comment that P. phocoena and, indeed,
several related phocoenid species produce narrow band, low-intensity,
ultra-high signals (peak spectra >120 kHz) that are markedly different from
those of delphinids. Those observations do not mean that Type I animals
are signal and processing aquatic clones of CF/FM bats. However, if the
signal data are reviewed in the context of ganglion cell densities an intriguingly consistent picture begins to form. All relevant data are preliminary,
but ganglion cell spikes to over 1O,000/mm (Ketten 1998a) located in the
mid basal turn segment of the P. phocoena ear are coincident with the pro-