440
J.L. Aroyan et al.
biosonar clicks was further illustrated by forward extrapolations of the farfield patterns produced by point sources placed at the tip of the larynx and
within the node of the right nasal plug in the complete model.
A surprising degree of acoustic asymmetry was also discovered. In the D.
de/phis specimen, forehead asymmetry is developed to a point where very
different beams are produced by left and right MLDB sources. The centers
of the left and right MLDB complexes are separated by about 3.4cm in this
specimen. Sources placed just below the right MLDB complex produced
well-focused forward-directed beams, while sources placed at the left
MLDB produced somewhat scattered emission patterns peaked roughly
26
0
to the right of center forward. These differences have a significant
bearing on the conjecture that both complexes may be used simultaneously
to produce coherently interfering beams (Cranford et al. 1996). Dual
MLDB source simulations above 50kHz (not illustrated here) have shown
that coherent forward beaming is constrained by the well separated pattern
peaks. Even if dual MLDB source coordination on a time scale of microseconds were biologically possible, phasing would have little effect on the
resulting multi-lobed emissions.
Perhaps the central conclusion of the current investigation was that the
forehead tissues of the D. de/phis model exhibit a simple focal structure for
forward-directed beams, with the right MLDB (or slightly below) as focal
center. While the behavioral evidence generally does not support the occurrence of multi-lobed emission patterns in delphinids (Dormer 1979; Mackay
and Liaw 1981; Au 1993), it is possible that previous experiments have not
been designed to encourage this behavior. Some dolphins appear capable
of selectively producing sonar pulses at either the left or right MLDB
(Cranford, Chapter 3). Nevertheless, it remains probable that the asymmetrically enlarged tissues of the right side are indeed specialized for click
production (Mead 1975). As suggested by Heyning (1989), "there may have
been selection for a single sound generating source in the narial region in
order to avoid interference generated from two sound sources." The development of separate motor control would also have contributed to the avoidance of sonar beaming problems.
The overall beam forming capability of the forehead tissues demonstrated here for the common dolphin is impressive. We have seen that the
complete forehead tissue model produces relatively uniformly directed and
well focused forward beams when appropriate source locations are used.
The -3 dB beam widths of approximately 12
0
(vertical) and 11
0
(horizontal) in the 100 kHz pattern are consistent with the experimentally measured
broadband (peaked near 120kHz) beam widths of 10.2
0
(vertical) and 9.7
0
(horizontal) for Tursiops truncatus (Au et al. 1978). The directivity indexes
computed here for the complete D. de/phis model, however, fall roughly 10
dB below those reported for T. truncatus (Au et al. 1978; Au 1980). Taking
both frequency and size differences into account for these animals, a disparity of roughly 5dB remains. In this context, it is important to note that
J.L. Aroyan et al.
biosonar clicks was further illustrated by forward extrapolations of the farfield patterns produced by point sources placed at the tip of the larynx and
within the node of the right nasal plug in the complete model.
A surprising degree of acoustic asymmetry was also discovered. In the D.
de/phis specimen, forehead asymmetry is developed to a point where very
different beams are produced by left and right MLDB sources. The centers
of the left and right MLDB complexes are separated by about 3.4cm in this
specimen. Sources placed just below the right MLDB complex produced
well-focused forward-directed beams, while sources placed at the left
MLDB produced somewhat scattered emission patterns peaked roughly
26
0
to the right of center forward. These differences have a significant
bearing on the conjecture that both complexes may be used simultaneously
to produce coherently interfering beams (Cranford et al. 1996). Dual
MLDB source simulations above 50kHz (not illustrated here) have shown
that coherent forward beaming is constrained by the well separated pattern
peaks. Even if dual MLDB source coordination on a time scale of microseconds were biologically possible, phasing would have little effect on the
resulting multi-lobed emissions.
Perhaps the central conclusion of the current investigation was that the
forehead tissues of the D. de/phis model exhibit a simple focal structure for
forward-directed beams, with the right MLDB (or slightly below) as focal
center. While the behavioral evidence generally does not support the occurrence of multi-lobed emission patterns in delphinids (Dormer 1979; Mackay
and Liaw 1981; Au 1993), it is possible that previous experiments have not
been designed to encourage this behavior. Some dolphins appear capable
of selectively producing sonar pulses at either the left or right MLDB
(Cranford, Chapter 3). Nevertheless, it remains probable that the asymmetrically enlarged tissues of the right side are indeed specialized for click
production (Mead 1975). As suggested by Heyning (1989), "there may have
been selection for a single sound generating source in the narial region in
order to avoid interference generated from two sound sources." The development of separate motor control would also have contributed to the avoidance of sonar beaming problems.
The overall beam forming capability of the forehead tissues demonstrated here for the common dolphin is impressive. We have seen that the
complete forehead tissue model produces relatively uniformly directed and
well focused forward beams when appropriate source locations are used.
The -3 dB beam widths of approximately 12
0
(vertical) and 11
0
(horizontal) in the 100 kHz pattern are consistent with the experimentally measured
broadband (peaked near 120kHz) beam widths of 10.2
0
(vertical) and 9.7
0
(horizontal) for Tursiops truncatus (Au et al. 1978). The directivity indexes
computed here for the complete D. de/phis model, however, fall roughly 10
dB below those reported for T. truncatus (Au et al. 1978; Au 1980). Taking
both frequency and size differences into account for these animals, a disparity of roughly 5dB remains. In this context, it is important to note that