On
the Caudal Extension
Background
Recording from auditory afferent
axons from the ear of the American bullfrog (Rana catesbeiana), several investigators, beginning with
Larry Frishkopf and Moise
Goldstein, found that the auditory responsiveness in each axon was sharply tuned, with a best
excitatory frequency (BEF) somewhere
between approximately 100 Hz and 1500 Hz (Frishkopf
and Goldstein, 1963; Frishkopf and Geisler, 1966; Liff et al., 1968). Based on their locations in the VIIIth nerve, axons with BEFs
below 1000 Hz were presumed to originate at the AP. Those with BEFs
above 1000 Hz were presumed to originate at the basilar papilla (BP). Later, in the Capranica
Lab, Albert Feng, Peter Narins
and Bob Capranica identified two
classes of afferent axon from the AP itself (AS Feng,
PM Narins & RR Capranica,
J. Comp. Physiol. 100: 221-229). One class, with BEFs
from approximately 100 Hz to approximately 700 Hz, exhibited two-tone
suppression. The other class, with BEFs from approximately 400 Hz to approximately 900 Hz did
not exhibit that phenomenon.
Bob Capranica
and Anne Moffat looked for this other class (higher-frequency axons, without
two-tone suppression) in
other anuran species. In
their earliest studies, they found none in Couch’s spadefoot
toad (Scaphiopus couchi) (RR Capranica & AJM Moffat, J. Comp. Physiol. 100:
231-245, 1975). In the green toad (Bufo debilis) they
found that a small proportion of the AP afferent axons belonged to this class,
and in the American toad (B. americanus) a larger proportion (personal
communication). The proportion in R. catesbeiana
had been even larger. I decided to
examine the AP morphology of the four species to determine if there were any
obvious differences that might be related to the Capranica-Lab
observations.
There were. R. catesbeiana, B. debilis and B. americanus all exhibited caudal extensions. S. Couchi did not.
The caudal extension had two features: (1) it was formed by recurving of the AP’s caudal patch, and (2) its hair
bundles were oriented in a way that would make them most sensitive to
transverse shear of the tectorial membrane (shear
normal to the long axis of the patch).
In the rest of the caudal patch, the hair bundles were oriented in a way
that would make
them most sensitive to longitudinal shear (shear parallel to the long axis of
the patch). The transition took place
under the tectorial curtain. The proportions of hair cells in the caudal
extensions of these three species were good matches to the proportions of
higher-frequency afferent axons lacking two-tone suppression (ER Lewis, Scanning Electron Microscopy 1977(II): 429-436, 1977).
These results strongly suggested that the
AP is tonotopically organized, with the
higher-frequency axons originating in the caudal extension. They also suggested that two-tone suppression
is associated with longitudinal motion of the tectorium,
and that the mechanical mode of tectorial vibration
(in response to sound stimuli) was different on the two sides of the tectorial curtain.
The results encouraged me to expand the
comparative morphological studies to other anuran species. They also added to the urgency of beginning
physiological studies of the bullfrog ear in the Lewis Lab—with goal of obtaining
functional overlays for the morphological maps we now had for that ear.
In the years that followed, Lewis-Lab
studies of bullfrog AP physiology were carried out by Ellen Leverenz,
Xiaolong Yu and Walter Yamada. Andrea Megella
Simmons collaborated on part of Ellen’s work (related to adaptation). Obtaining the functional overlays was
achieved by a team that included Ellen and Andrea (see the large section “The
Bullfrog Sacculus” in this website).
AP tonotopy was corroborated in the American bullfrog.
Based on the
observed tonotopy and the evidence from the Capranica Lab, it seems clear that the caudal extension
provides at least two things:
(1) extension of the range of AP frequency sensitivity—to higher
frequencies,
(2) a new response feature, namely lack of suppression by a second
tone.
During the 1980s, the Narins Lab added two important pieces of information. The range of best excitatory frequencies
among AP units of Bombina bombina,
which has no caudal extension, extends to 700 Hz (Hillery & Narins, 1987); and the
range of best excitatory frequencies among AP units of Eleutherodactylus coqui, which has the largest proportion of
caudal-extension hair cells we have seen so far, extends to 1400 Hz. And that same range applies to both male and
female coqui frogs, although female coqui frogs have a conspicuously larger proportion of
caudal-extension hair cells. If we
assume that the Bombina bombina AP
data show the greatest range achievable without the caudal extension, then
(1a) the
extension of the AP frequency range in Eleutherodactylus
coqui was one octave.
In the bullfrog it was less than ˝ octave (to
approximately 1 kHz.)
We
are left with some puzzles, however.
What selective advantages were provided by the conspicuously greater
proportion of hair cells in the female coqui frog?
Furthermore, in the bullfrog, the Lewis-Lab dye-tracing studies showed that
caudal extension included many units with best excitatory frequencies (450 Hz
to 700 Hz) in the range already available without such an extension. Was lack of two-tone suppression the only
selective advantage of these units?
From
the Lewis-Lab dye-tracing studies, we have two additional pieces of information
for the bullfrog. The typical afferent
axon of the caudal extension innervates one to three hair cells. The typical afferent axon of the rostral patch innervates six to fifteen hair cells. This implies that the neural image
transmitted to the brainstem from the AP caudal extension has much greater
resolution than that from the AP rostral patch. In Lewis, Hecht & Narins,
1992, we elaborated on this feature and its selective advantages. The second piece of information was provided
by the work of Ellen and Andrea-- weakly and moderately adapting AP afferent
axons innervate many hair cells; strongly adapting AP afferent axons innervate
only one or a few hair cells.
Based on these
results, it is clear that the caudal extension provides at least two additional
things:
(3) greatly increased resolution (far more afferent axons per
octave) in the spectrographic images sent to the brainstem,
(4) greatly increased proportion of a specific response feature,
namely strong adaptation.
Both of these
things should aid the frog in the task of discriminating acoustic signals from
one another and from background noise.
For further discussion of this topic, with
references to examples of enhanced discrimination, see Lewis, Hecht & Narins, 1992.