THE JOURNAL OF COMPARATIVE NEUROLOGY 309:250-260 (1991)
Neurogenesis in the Brain Auditory Pathway of a Marsupial, the Northern Native Cat (Dasyurus hallucatus) LINDSAY AITKIN, JOHN NELSON, MOYRA FARRINGTON, AND SUE S W A ” Departments of Physiology (L.A.,M.F.) and Zoology (J.N., S.S.), Monash University, Melbourne, Australia
ABSTRACT Neurogenesis in the auditory pathway of the marsupial Dasyurus hallucatus was studied. Intraperitoneal injections of tritiated thymidine (20-40 pCi) were made into pouch-young varying from 1 to 56 days pouch-life. Animals were killed as adults and brain sections were prepared for autoradiography and counterstained with a Nissl stain. Neurons in the ventral cochlear nucleus were generated prior to 3 days pouch-life, in the superior olive at 5-7 days, and in the dorsal cochlear nucleus over a prolonged period. Inferior collicular neurogenesis lagged behind that in the medial geniculate, the latter taking place between days 3 and 9 and the former between days 7 and 22. Neurogenesis began in the auditory cortex on day 9 and was completed by about day 42. Thus neurogenesis was complete in the medullary auditory nuclei before that in the midbrain commenced, and in the medial geniculate before that in the auditory cortex commenced. The time course of neurogenesis in the auditory pathway of the native cat was very similar to that in another marsupial, the brushtail possum. For both, neurogenesis occurred earlier than in eutherian mammals of a similar size but was more protracted. Key words: tritiated thymidine,cochlear nucleus, inferior colliculus, medial geniculate,auditory cortex
When marsupial embryos are born, they are exposed at a very early stage of development to environmental stimulilight, airborne vibrations and environmental chemicals. The fact that they begin pulmonary respiration at such an early time and that they absorb nutrition orally rather than via the placenta mean that the relevant body systems will be much more developed than those of Eutheria at a similar postconceptional age. Given the needs of a mobile, air-breathing, milk-drinking embryo, it seems likely that specific neural pathways must develop early relative to Eutheria, at least at a brainstem level. One sensory system with its first relays in the brainstem is the auditory system, and neurogenesis in this system in one marsupial, the brush-tailed possum (Trichosums uulpecula) (Sanderson and Aitkin, ’go), does occur early relative to that in the rat (Altman and Bayer, ’81). We are studying neurogenesis in a more “primitive” marsupial, the Northern native cat (Dasyurus hallucatus). In this species, up to 8 young are born after a gestation period of about 3 weeks and are very immature at birth, at a stage of development (stage 15 in the Carnegie series) equivalent to approximately 35-38 days postconception in humans (Crewther et al.,’88). Nelson (’88, and in prepara-
O
1991 WILEY-LISS, INC.
tion) has found that although some neural mechanisms have been developed to allow the newborn to locate and attach to a nipple and to suckle, most of the neural structures are at a level of development that allows the animal to be placed at stage 15. Pouch-young remain attached to the nipple for about 5 weeks and hence are unlikely to vocalize before this time. By this age they have reached the end of embryogenesis (stage 23, after which all Eutheria are born). Once they are able to remove their mouth from the nipple, there is a danger that animals might be lost. Their eyes do not open until 75 days (Crewther et al., ’881, but they can vocalise at 40 days (Nelson, unpublished observations) so that sound may be the only distance sense that could be used to maintain contact with the mother or other young during this period of pouch-life. Since the developing embryos are directly exposed to an acoustic environment and not shielded by the mother’s body or the walls of the uterus, it might be expected that Accepted March 25,1991. Address reprint requests to Dr. L. Aitkin, Dept. of Physiology, Monash University, Clayton, Victoria 3168, Australia.
AUDITORY NEUROGENESIS IN A MARSUPIAL
251
Fig. 1. Pouch-youngofDasyurus hallucatus aged 3, 13,23, and 45 days. Calibration: 5 mm.
such exposure could accelerate the development of the auditory system. As we show, this appears to be the case for Dasyurus hallucatus.
MATERIALS AND METHODS Pouch-young of various ages (1-16,18-20,22-24,26-28, 32,34,36-42,45,47, and 56 days) were injected with 0.5 pl
to 1 pl of tritiated thymidine (Amersham TRK 323). Examples of pouch-young at 3, 13, 23, and 45 days of age are shown in Figure 1. For each injection, 35-40 pl of TRK 323 were evaporated down to 0.5-1 p1, so that 2 0 4 0 pCi were injectedper animal. Injections were made with the mother anesthetized with the steroid anesthetic a l p h a olone (Saffan; 1ml/kg) and with the young still attached to the nipple. A Hamilton syringe (5 or 10 pl) was cemented to
252
L. AITKIN ET AL.
Fig. 2. Cells labelled in adults injected with tritiated thymidine as pouch-young of various ages (in parentheses) and counterstained with thionin, from the auditory cortex (A, 26 days); medial geniculate body (B, 7 days), inferior colliculus (C, 10 days) and dorsal cochlear nucleus
(D, 15 days). E. Inferior colliculus at 15 days. Heavy arrows: darkly labelled cells; light arrows; putative “daughter” cells. Calibration: 20 Fm for A-D, 1.26 mm for E.
an extruded capillary tube, thus providing a fine tip to the syringe. Injections were made into the abdominal cavity. All animals were killed as anesthesized adults (sodium pentobarbital, 60mg/kg) and perfused with saline followed by a 10% neutral-buffered formalin, Bouins, or a formalin/ glutaraldehyde solution. The brains were dissected out, processed through alcohols to paraffin, and cut in a transverse plane at 8-10 pm. Sections were mounted, defatted, then dipped in an emulsion (Kodak NTB-2 diluted 50/50 with water), and allowed to dry. They were stored in lightproof boxes wrapped in three layers of aluminium foil; all boxes were contained within a photographic lightproof bag, which was then kept in cool storage (4”)for about 6
weeks. Slides were developed in Kodak D19, stained with cresyl violet or thionin, and coverslipped. Sections were scanned at high power for thymidinelabelled cells. Although the maximum density of labelling varied between experiments, it was possible to identify a population of darkly labelled neurons and glia. We consider that these most densely labelled cells had undergone their last mitosis on the day of the injection, and examples of densely labelled neurons and putative “daughter” cells are shown for the cochlear nucleus, inferior colliculus, medial geniculate body, and auditory cortex in Figure 2A-D. Sections were projected by a Leitz Micropromar projection microscope at a magnification of 114 times, and all
253
AUDITORY NEUROGENESIS IN A MARSUPIAL
Fig. 3. Lightfield (A$) and darkfield (B,D) photomicrographs of a frontal section through the cochlear nucleus (A,B) and superior olive (C,D) of adults labelled with tritiated thymidine injected in pouchyoung aged 15 days (A,B) and 7 days (C,D). Counterstained with thionin. Arrow in B points to a patch of labelled granule cells. AVCN,
visible labelled cells were counted. At this magnification only the most densely labelled neurons were visible in this Nissl-stained material; glial cells and lightly labelled neurons (open-headed arrows in Fig. 2 ) could not be distinguished. Some idea of the ease in making the distinction between densely labelled neurons and other cells can be gauged from a low power view of the inferior colliculus at day 15 (Fig. 2E). However, this method makes quantitative comparisons between adjacent ages more difficult because of the variation in effectiveness of the injections in these very tiny embryos (Fig. 1).As a consequence, the maximum densities of labelling across the population differ. Thus we give greater emphasis to the changes in the locations of densely labelled cells than to their absolute numbers (other than zero).
RESULTS At 3 days of age, the pouch-young of Dasyurus has only two prominent external features-forelimbs and nares, in addition to its mouth (Fig. 1,3d). Over the next 4 weeks, the embryo grows from a rump-snout length of about 7 mm to about 30 mm at 45 days, at which age eyes, ears, hindlegs, and tail have become prominent (Fig. 1).
PVGN, DCN: anteroventral, posteroventral and dorsal cochlear nuclei; RB: restiform body; MSO, LSO: medial and lateral superior olivary nuclei; SPN; superior paraolivary nucleus; MTB: medial nucleus of the trapezoid body.
Brainstem auditory nuclei The cytoarchitecture of the cochlear nuclei and the superior olivary complex of Dasyurus hallucatus have been described in detail in a previous report from this laboratory (Aitkin et al., '86a); only a brief account is given here. The cochlear nuclear complex of Dasyurus, like that of other marsupials, has a medial disposition relative to the restiform body. At the transverse plane shown in Figure 3A, at approximately the midrostrocaudal center of the complex, all three subdivisions are visible. The dorsal cochlear nucleus (DCN) forms the lateral part of the floor of the fourth ventricle and has a clearly defined layer of pyramidal cells interspersed with granule cells. The anteroventral cochlear nucleus (AVCN) lies immediately medial to the restiform body; its cells are generally spherical, in contrast to the more common multipolar celis of the posteroventral cochlear nucleus (PVCN). The latter merges with the cochlear nerve root nucleus (CNR) at the point of entry of cochlear nerve fibers (CNF). The superior olivary complex has four major subdivisions. A triangular-shaped lateral superior olive (LSO) has its apex pointing laterally (Fig. 3C). The medial half of the main complex is composed of a column of fusiform cells, the medial superior olive (MSO), and an ovoid mass of multipo-
254
Fig. 4. Darkfield photomicrographs of frontal sections through the inferior colliculus of adults labelled with tritiated thymidine as a result of injections made when pouch-young were at the stated ages in days.
L. AITKIN ET AL.
Sections are selected at the frontal level where strongest labelling occurs, and show collectively a general ventral-to-dorsomedialprogression from 7d to 16d.
255
AUDITORY NEUROGENESIS IN A MARSUPIAL
A
1 0 0 ] ~ ~ ~ Y s
50 \
' 0 OR
50 loot OR 50 PERCENTAGE OF ROSTROCAUDAL DISTANCE
100 c
Fig. 5. The numbers of darkly labelled cells at various rostrocaudal levels of the inferior colliculus are expressed as percentages of the maximum number for m y one brain. The absolute number is shown adjacent to the peak for that brain, and the animal number below the
age of injection in days. The rostral limit of the inferior colliculus is defined as the region where the densely packed central nucleus is replaced entirely by the loosely packed rostral external nucleus. The caudal limit is that of the entire inferior colliculus.
lar cells, the superior paraolivary nucleus (SPN). The dark staining cells of the medial nucleus of the trapezoid body (MTB) form lines connecting the superior olivary complex to the midline of the brainstem. Relative to the midbrain, thalamus, and cortex, very few thymidine labelled cells were observed in adult animals in the two ventral nuclei of the cochlear nuclear complex or in the MSO or SPN when injections were made in pouchyoung older than 3 days. However, scattered pyramidal and deeper cells were present in the DCN at all injected ages examined (e.g., Fig. 3B), suggesting a prolonged period of neurogenesis without any clear-cut peak for this nucleus. In contrast neurogenesis in the MTB and LSO were completed in a short period, between days 5 and 9, peaking at day 7 (Fig. 3D). There was no evidence for any gradient in the MTB; cells were born at the same time along the entire rostrocaudal extent of this structure. The very low cell numbers in the medial superior olive and ventral cochlear nucleus at any of the injected ages studied, combined with the clear-cut later peaks in higher nuclei, suggests that the peak of neurogenesis for these brainstem nuclei has occurred earlier than 3 days of pouch age. However, in both ventral and dorsal cochlear nuclei, granule cells were labelled in increasing numbers as postnatal age increased (to day 56).
Like that of other mammals, the inferior colliculus of Dasyurus hallucatus has a cell-dense central nucleus surrounded by a thin mantle of cells and fibres (Aitkin et al., '86a). Caudally the structure is free-standing, whereas rostrally it gradually merges into the tegmentum beneath the superior colliculus. Thus the caudal boundary of the central nucleus is well defined, whereas its rostral limit is less clear. As with other mammals, the central nucleus has a systematic cochleotopic organisation (Aitkin et al., '86b). Some examples of frontal sections through the inferior colliculus in which the greatest density of labelling was found for a given animal are shown in Figure 4 for adults, following thymidine injections at 7, 10, 13, and 16 days. Labelled cells first appear in adults injected on day 5 of pouch-life, in the rostral half of the central nucleus, at its ventral margin. By day 7 numbers have increased considerably and labelled cells are concentrated in the rostral and ventrolateral half of the central nucleus (Figs. 43).In older injected pouch-young,labelled cells occupy the rostrocaudal and dorsoventral centers of the central nucleus (days 9-13; Fig. 4) gradually shifting caudally as increasingly older pouch-young were studied (Fig. 5). By days 15 and 16, labelled cells are only found in the caudomedial part of the
Inferior colliculus
L. AITKIN ET AL.
256
Fig. 6. Darkfield photomicrographs of frontal sections through the medial geniculate body (MG) labelled with tritiated thymidine injected when animals were pouch-young of 3 and 9 days of age. The internal
boundary of the MG is shown by a dashed line. Note that the 3d section is taken at a level corresponding to the caudal one-third, and the 9d section from the rostral one-third, of the MG.
central nucleus, become progressively fewer in number by day 22 (Figs. 4,5), and are not observed in the central nucleus beyond this date. Variability in cell counts as a function of age and rostrocaudal location makes it difficult to identify sharp rostrocaudal gradients in these data. In this respect, the 13-day embryo was provided by a mother trapped in a year different to those illustrating the other ages, and the rostrocaudal cell count graph appears out of sequence with the others. The times of origin of cells in the lateral and dorsal cortices of the inferior colliculusmatch those for the central nucleus in relation to position; cells in rostral locations are generated between days 5 and 10, and more caudal cells are born between days 12 and 22. Thus the rostroventral to caudomedial gradient appears to occur through all subdivisions of the inferior colliculus.
appear to respect such boundaries. The internal boundary of the entire medial geniculate body (MG) is shown by the dashed line in Figure 6. Neurogenesis in the medial geniculate body has already begun on day 3 and is virtually complete on day 11.Labelled cells first appear at the caudal ventral margin and form a swathe in the ventral half of its caudal one-third (3d in Fig. 6). In adult animals injected as progressively older pouchyoung, labelled cells shift rostrally and dorsally and are present only in the dorsal half of the rostral medial geniculate by days 9-11 (9d in Fig. 6). Neurogenesis in the medial geniculate is thus complete when that in the inferior colliculus is only just approaching its peak.
Medial geniculate body The medial geniculate body of Dasyurus hallucatus is composed of a ventral division, which forms the major part of the nucleus, capped by a small sheet of cells-the dorsal nucleus-and flanked medially by the small medial nucleus (Kudo et al., '89). Boundaries, however, are difficult to discern in Nissl material, and neurogenetic gradients do not
The auditory cortex The primary auditory cortex (AI) is the only part of the auditory cortex that has been mapped convincingly in Dasyurus hallucatus (Aitkin et al., '86b). The mapped region extends for 5-6 mm in the rostrocaudal middle of the cortex and lies immediately dorsal to the rhinal fissure. It is characterised by a blurring of lamination in the middle part of the cortex (Fig. 7A) that beginsjust ventral (arrow in Fig. 7) to a shallow depression in the cortex, the lateral crease (LC in Fig. 7A). Above this crease a clear small-celled layer IV may be distinguished (Fig. 7A), and the weak auditory
AUDITORY NEUROGENESIS IN A MARSUPIAL
257
Fig. 7. A. thionin-stained frontal section through the auditory cortex of Dasyurus. The primary auditory cortex, AI,lies between the rhinal fissure (RF) and lateral crease (LC); its dorsal limit is marked in all sections with an arrow. The remaining, darkfield, sections are
graded in age (in days) when animals were injected as pouch-young. The boundary between cortex and white matter is shown by dashed lines in the dark field photographs. Calibration: 1.3 mm A; 0.5 mm for the darkfield sections.
responses in this region suggest that it is a secondary auditory area (Aitkin et al., '86b). No thymidine-labelled cells are ever seen in the AI in animals injected prior to day 7, but by day 9 labelled cells are found in the deepest part of the adult auditory cortex, adjacent to the white matter (9d in Fig. 7 ) .The cells of the deepest layer are born by day 15 (15d in Fig. 7 ) and a population has also been generated to form the middle layers of this particularly densely labelled preparation. The injection at 16 days of a pouch-young from a litter different to that at 15 days suggests that many neurons populating the middle layers have been born by this time (16d in Fig. 7). By 32 days labelled cells destined for the superficial one-third of the cortex are generated (32d in Fig. 7 ) and those of the most superficial layer are born by day 38. Beyond this age, labelled cell numbers simply diminish in layer I1 and are not observed when injections are made after day 45.
These qualitative observations are expressed in graphical form in Figure 8, in which the depth from the pial surface of the center of the swathe of labelled cells is plotted against the age at which the pouch-young was injected, in days. Three measures were taken from a section located in the rostrocaudal center of the AI, one close to the lateral crease, one near the point of maximum curvature, and the other midway between (Fig. 8). The accuracy of these measurements improved as the postnatal age of the injected young increased, because (as is obvious in Fig. 7) the first generated cells were scattered broadly across the deepest layers, whereas those born at a later time formed narrower bands in the AI. The measurements plotted in Figure 8 confirm for the marsupial native cat the inside-out development of the cerebral cortex described for Eutheria (e.g., Altman and Bayer, '81; Angevine and Sidman, '61). There is another gradient of neurogenesis in the cerebral cortex that influences qualitatively the AI gradient. At any
L. AITKIN ET AL.
258
2000
1500 m CK w t W
I
0
5I:
1000
z I c
a w
cl
500
0
0
20
10
30
40
INJECTED AGE IN DAYS Fig. 8. Data from the AI quantifying a trend suggested in Figure 7. The centers of the swathes of labelled cells in each brain have been measured at three dorsoventral positions and plotted as a function of depth, thus showing that cells in the deep and middle layers (i.e.,
greater than 50Qpm in depth) are “born” over a short period, between 9 and 16 days pouch life, whereas those populating the superficial layers are born over a prolonged period (16-40+ days). LC (lateral crease); RF (rhinal fissure).
given age during neocortical neurogenesis, cells at a given depth in dorsal parts of the neocortex of the native cat are born later than those ventrally (i.e., in the AI). This can be recognised in Figure 7 if the cortical regions above the arrows are compared with the AI, below. Similar gradients of neurogenesis have been observed in the possum neocortex (Sanderson and Weller, ’90). However, no consistent differences are seen in this regard within the AI. Measures taken in the dorsal, middle, and ventral A1 provide curves that are very similar in position as a function of injected age. Any differences can be accounted for by the different physical dimensions of the brains of different adult animals and by the sharp flexion of the temporal cortex as it folds into the rhinal fissure, rather than by any internal dorsoventral gradient.
22. Neurogenesis begins in the auditory cortex on day 9 and is completed by about day 42. These time sequences suggest that neurogenesis occurs nearly simultaneously in two separate compartments-the brainstem (medulla,then midbrain) and forebrain systems (medial geniculate, then auditory cortex). The differencesbetween the time courses of neurogenesis in possums (Sanderson and Aitkin, ’90) and native cats are quite small and could be accounted for by individual variation in members of both species. In both, ventral cochlear nucleus cells were never labelled in the experimental material, suggesting that they are generated at or before the time of birth. In both species,auditory cortical neurogenesis occurs over a period of about 5-6 weeks and is complete, in native cats, at about the same time as that in the visual cortex-some 30 days before eye opening (Crewther et al., ’88). Comparisons between neurogenesis in marsupials and in eutherian mammals are complicated by the short gestation period in marsupials (2-3 weeks) relative to eutherian mammals of a similar size (cats 65 days, macaque monkeys 164 days). The time of eye opening is a convenient event to compare across species because it is associated with the beginnings of special sensory reception and occurs earlier in cats (about 73 days post-fertilization) than in native cats (95 days) or possums (117 days) (Sanderson and Aitkin, ’90). However, considerable visual development occurs over at least another 6 weeks postnatally in cats (e.g., Hubel and
DISCUSSION The time course of neurogenesis in the brain auditory pathway of the marsupial native cat, Dasyurus hallucatus, has been studied. There is no simple relationship between the level of an auditory nucleus and the genesis of its constituent neurons. Neurons are generated in the ventral cochlear nucleus prior to 3 days pouch-life, in the superior olive at 5-7 days, and in the dorsal cochlear nucleus over a prolonged period. Inferior collicular neurogenesis lags behind that in the medial geniculate, the latter taking place between days 3 and 9 and the former between days 7 and
-
AUDITORY NEUROGENESIS IN A MARSUPIAL
0
259
100%
1
CONCEPTION
EYE OPENING
NATIVECAT
I POSSUM
; MACAQUE MONKEY
RAT
I
1
I
MOUSE
Fig. 9. A comparison of t h e time periods over which cells in t h e inferior colliculus (above each line) and auditory cortex (below each line) are born, expressed a s a percentage of t h e period between
conception and eye opening. Data obtained from t h e present study: Sanderson and Atkin ('901, Cooper and Rakic ('811,Altman and Bayer ('81) and Taber-Pierce ('73).
Weisel, '701, and auditory thresholds continued to improve in kittens during the first 4 weeks after birth (Ehret and Romand, '81). The only marsupial for which behavioral evidence for the onset of hearing is available is the opossum (63 days postfertilization; Ehret, '83); eye opening in this species occurs at 73 days (Sanderson and Aitkin, '90). However, observation indicates that pouch young of Dasyum s begin vocalizing at about 40 days (Nelson, unpublished observations); this could correspond to the onset of hearing in this species. Even though eye opening is not likely for any species to signal the end of visual development, it is a convenient index for comparison between those species for which neurogenetic information about auditory development is available. Figure 9 expresses the postconceptual periods during which neurons of the inferior colliculus (above the lines) and auditory cortex (below the lines) are generated, to the times of eye opening. It is clear that neurogenesis begins very much earlier in the two marsupials but the process is, relative to Eutheria, prolonged. The last cells destined for the marsupial auditory centers are generated at a time corresponding to the beginning of neurogenesis in the auditory centers of the three eutherian species. It will be interesting to see when the auditory system begins to function, given the fact that neurogenesis occurs early in auditory structures in native cats. Preliminary observations from this laboratory reveal that some weeks may elapse between neurogenesis and the appearance of cytoarchitecturally distinct auditory nuclei. Thus although the medullary auditory nuclei are distinct at day 13, the central nucleus of the inferior colliculus and medial geniculate body are not easy to recognise until about day 23. Nothing is known of the time of arrival of connections to the brainstem from the cochlea, but the optic nerve reaches its target sites in the thalamus at about day 25 (Crewther et al., '88), so it is possible that the auditory connections may
form at a similar time. This, and the appearance of spontaneous vocalizations at 40 days, would concur with the onset of hearing some time midway during pouch-life in Dasyurus.
ACKNOWLEDGMENTS The authors thank Dianne Clare and Jill Poynton for preparing the illustrations, and Dr. Ken Sanderson for valuable critical comments. This study was supported by grants from the National Health and Medical Research Council of Australia. Animals were collected under Conservation Commission of the Northern Territory permit SL8/84 and held under Victorian Department of Conservation permit 84-26.
LITERATURE CITED Aitkin, L.M., M. Byers, and J.E. Nelson (1986a) Brain stem auditory nuclei and their connections in a carnivorous marsupial, the Northern native cat (Dasyurus hallucatus). Brain, Behav. Evol. 29:l-16. Aitkin, L.M., D.R.F. Irvine, J.E. Nelson, M.M. Merzenich, and J.C. Clarey (198613) Frequency representation in the auditory midbrain and forebrain of a marsupial, the Northern native cat (Dasyurus hallucatus). Brain, Behav. Evol. 29:17-28. AItman, J., and S.A. Bayer (1981) Time of origin of neurons of the rat inferior colliculus and the relations between cytogenesis and tonotopic order in the auditory pathway. Exp. Brain Res. 4 2 4 1 1 4 2 3 . Angevine, J.B., and R.L. Sidman (1961) Autoradiographic study of cell migration during histogenesis of cerebral cortex in the mouse. Nature 192:766-768. Cooper, M.L., and P. Rakic (1981) Neurogenetic gradients in the superior and inferior colliculi of the Rhesus monkey. J. Comp. Neurol. 202:309334. Crewther, D.P., J.E. Nelson, and S.G. Crewther (1988) Afferent input for target survival in marsupialvisual development. Neurosci. Letts. 86:147154.
260 Ehret, G. (1983) Development of hearing and response behavior to sound stimuli: behavioral studies. In R. Romand (ed): Development of Auditory and Vestibular Systems. New York Academic Press, pp. 211-237. Ehret, G., and R. Romand (1981) Postnatal development of absolute auditory thresholds in kittens. J. Comp. Physiol. Psychol. 95304-31 1. Hubel, D.H., and T.N. Wiesel (1970) The period of susceptibility to the physiological effects of unilateral eye closure in kittens. J. Physiol. (Lond). 206;419-436. Kudo, M., L.M. Aitkin, and J.E. Nelson (1989) Auditory forebrain organization of an Australian marsupial, the Northern native cat (Dusyurus hallucatus).J. Comp. Neurol. 27928-42.
L. AITKIN ET AL. Nelson, J.E. (1988) Growth of the brain. In C.H. Tyndale-Biscoe and P.A. Janssens (eds): The Developing Marsupial. Models for Biomedical Research. Berlin: Springer-Verlag, pp. 86-100. Sanderson, K.J., and L.M. Aitkin (1990) Neurogenesis in the visual and auditory pathways of a marsupial: The brushtailed possum (Trichosurus uulpecula). Brain, Behav. Evol. 35:325-338. Sanderson, K.J., and W.L. Weller (1990) Gradients of neurogenesis in possum neocortex. Dev. Brain Res., 55:269-274. Taber-Pierce, E. (11973) Time of origin of neurons in the brain stem of the mouse. Prog. Brain Res. 40;53-65.
Neurogenesis in the Brain Auditory Pathway of a Marsupial, the Northern Native Cat (Dasyurus hallucatus) LINDSAY AITKIN, JOHN NELSON, MOYRA FARRINGTON, AND SUE S W A ” Departments of Physiology (L.A.,M.F.) and Zoology (J.N., S.S.), Monash University, Melbourne, Australia
ABSTRACT Neurogenesis in the auditory pathway of the marsupial Dasyurus hallucatus was studied. Intraperitoneal injections of tritiated thymidine (20-40 pCi) were made into pouch-young varying from 1 to 56 days pouch-life. Animals were killed as adults and brain sections were prepared for autoradiography and counterstained with a Nissl stain. Neurons in the ventral cochlear nucleus were generated prior to 3 days pouch-life, in the superior olive at 5-7 days, and in the dorsal cochlear nucleus over a prolonged period. Inferior collicular neurogenesis lagged behind that in the medial geniculate, the latter taking place between days 3 and 9 and the former between days 7 and 22. Neurogenesis began in the auditory cortex on day 9 and was completed by about day 42. Thus neurogenesis was complete in the medullary auditory nuclei before that in the midbrain commenced, and in the medial geniculate before that in the auditory cortex commenced. The time course of neurogenesis in the auditory pathway of the native cat was very similar to that in another marsupial, the brushtail possum. For both, neurogenesis occurred earlier than in eutherian mammals of a similar size but was more protracted. Key words: tritiated thymidine,cochlear nucleus, inferior colliculus, medial geniculate,auditory cortex
When marsupial embryos are born, they are exposed at a very early stage of development to environmental stimulilight, airborne vibrations and environmental chemicals. The fact that they begin pulmonary respiration at such an early time and that they absorb nutrition orally rather than via the placenta mean that the relevant body systems will be much more developed than those of Eutheria at a similar postconceptional age. Given the needs of a mobile, air-breathing, milk-drinking embryo, it seems likely that specific neural pathways must develop early relative to Eutheria, at least at a brainstem level. One sensory system with its first relays in the brainstem is the auditory system, and neurogenesis in this system in one marsupial, the brush-tailed possum (Trichosums uulpecula) (Sanderson and Aitkin, ’go), does occur early relative to that in the rat (Altman and Bayer, ’81). We are studying neurogenesis in a more “primitive” marsupial, the Northern native cat (Dasyurus hallucatus). In this species, up to 8 young are born after a gestation period of about 3 weeks and are very immature at birth, at a stage of development (stage 15 in the Carnegie series) equivalent to approximately 35-38 days postconception in humans (Crewther et al.,’88). Nelson (’88, and in prepara-
O
1991 WILEY-LISS, INC.
tion) has found that although some neural mechanisms have been developed to allow the newborn to locate and attach to a nipple and to suckle, most of the neural structures are at a level of development that allows the animal to be placed at stage 15. Pouch-young remain attached to the nipple for about 5 weeks and hence are unlikely to vocalize before this time. By this age they have reached the end of embryogenesis (stage 23, after which all Eutheria are born). Once they are able to remove their mouth from the nipple, there is a danger that animals might be lost. Their eyes do not open until 75 days (Crewther et al., ’881, but they can vocalise at 40 days (Nelson, unpublished observations) so that sound may be the only distance sense that could be used to maintain contact with the mother or other young during this period of pouch-life. Since the developing embryos are directly exposed to an acoustic environment and not shielded by the mother’s body or the walls of the uterus, it might be expected that Accepted March 25,1991. Address reprint requests to Dr. L. Aitkin, Dept. of Physiology, Monash University, Clayton, Victoria 3168, Australia.
AUDITORY NEUROGENESIS IN A MARSUPIAL
251
Fig. 1. Pouch-youngofDasyurus hallucatus aged 3, 13,23, and 45 days. Calibration: 5 mm.
such exposure could accelerate the development of the auditory system. As we show, this appears to be the case for Dasyurus hallucatus.
MATERIALS AND METHODS Pouch-young of various ages (1-16,18-20,22-24,26-28, 32,34,36-42,45,47, and 56 days) were injected with 0.5 pl
to 1 pl of tritiated thymidine (Amersham TRK 323). Examples of pouch-young at 3, 13, 23, and 45 days of age are shown in Figure 1. For each injection, 35-40 pl of TRK 323 were evaporated down to 0.5-1 p1, so that 2 0 4 0 pCi were injectedper animal. Injections were made with the mother anesthetized with the steroid anesthetic a l p h a olone (Saffan; 1ml/kg) and with the young still attached to the nipple. A Hamilton syringe (5 or 10 pl) was cemented to
252
L. AITKIN ET AL.
Fig. 2. Cells labelled in adults injected with tritiated thymidine as pouch-young of various ages (in parentheses) and counterstained with thionin, from the auditory cortex (A, 26 days); medial geniculate body (B, 7 days), inferior colliculus (C, 10 days) and dorsal cochlear nucleus
(D, 15 days). E. Inferior colliculus at 15 days. Heavy arrows: darkly labelled cells; light arrows; putative “daughter” cells. Calibration: 20 Fm for A-D, 1.26 mm for E.
an extruded capillary tube, thus providing a fine tip to the syringe. Injections were made into the abdominal cavity. All animals were killed as anesthesized adults (sodium pentobarbital, 60mg/kg) and perfused with saline followed by a 10% neutral-buffered formalin, Bouins, or a formalin/ glutaraldehyde solution. The brains were dissected out, processed through alcohols to paraffin, and cut in a transverse plane at 8-10 pm. Sections were mounted, defatted, then dipped in an emulsion (Kodak NTB-2 diluted 50/50 with water), and allowed to dry. They were stored in lightproof boxes wrapped in three layers of aluminium foil; all boxes were contained within a photographic lightproof bag, which was then kept in cool storage (4”)for about 6
weeks. Slides were developed in Kodak D19, stained with cresyl violet or thionin, and coverslipped. Sections were scanned at high power for thymidinelabelled cells. Although the maximum density of labelling varied between experiments, it was possible to identify a population of darkly labelled neurons and glia. We consider that these most densely labelled cells had undergone their last mitosis on the day of the injection, and examples of densely labelled neurons and putative “daughter” cells are shown for the cochlear nucleus, inferior colliculus, medial geniculate body, and auditory cortex in Figure 2A-D. Sections were projected by a Leitz Micropromar projection microscope at a magnification of 114 times, and all
253
AUDITORY NEUROGENESIS IN A MARSUPIAL
Fig. 3. Lightfield (A$) and darkfield (B,D) photomicrographs of a frontal section through the cochlear nucleus (A,B) and superior olive (C,D) of adults labelled with tritiated thymidine injected in pouchyoung aged 15 days (A,B) and 7 days (C,D). Counterstained with thionin. Arrow in B points to a patch of labelled granule cells. AVCN,
visible labelled cells were counted. At this magnification only the most densely labelled neurons were visible in this Nissl-stained material; glial cells and lightly labelled neurons (open-headed arrows in Fig. 2 ) could not be distinguished. Some idea of the ease in making the distinction between densely labelled neurons and other cells can be gauged from a low power view of the inferior colliculus at day 15 (Fig. 2E). However, this method makes quantitative comparisons between adjacent ages more difficult because of the variation in effectiveness of the injections in these very tiny embryos (Fig. 1).As a consequence, the maximum densities of labelling across the population differ. Thus we give greater emphasis to the changes in the locations of densely labelled cells than to their absolute numbers (other than zero).
RESULTS At 3 days of age, the pouch-young of Dasyurus has only two prominent external features-forelimbs and nares, in addition to its mouth (Fig. 1,3d). Over the next 4 weeks, the embryo grows from a rump-snout length of about 7 mm to about 30 mm at 45 days, at which age eyes, ears, hindlegs, and tail have become prominent (Fig. 1).
PVGN, DCN: anteroventral, posteroventral and dorsal cochlear nuclei; RB: restiform body; MSO, LSO: medial and lateral superior olivary nuclei; SPN; superior paraolivary nucleus; MTB: medial nucleus of the trapezoid body.
Brainstem auditory nuclei The cytoarchitecture of the cochlear nuclei and the superior olivary complex of Dasyurus hallucatus have been described in detail in a previous report from this laboratory (Aitkin et al., '86a); only a brief account is given here. The cochlear nuclear complex of Dasyurus, like that of other marsupials, has a medial disposition relative to the restiform body. At the transverse plane shown in Figure 3A, at approximately the midrostrocaudal center of the complex, all three subdivisions are visible. The dorsal cochlear nucleus (DCN) forms the lateral part of the floor of the fourth ventricle and has a clearly defined layer of pyramidal cells interspersed with granule cells. The anteroventral cochlear nucleus (AVCN) lies immediately medial to the restiform body; its cells are generally spherical, in contrast to the more common multipolar celis of the posteroventral cochlear nucleus (PVCN). The latter merges with the cochlear nerve root nucleus (CNR) at the point of entry of cochlear nerve fibers (CNF). The superior olivary complex has four major subdivisions. A triangular-shaped lateral superior olive (LSO) has its apex pointing laterally (Fig. 3C). The medial half of the main complex is composed of a column of fusiform cells, the medial superior olive (MSO), and an ovoid mass of multipo-
254
Fig. 4. Darkfield photomicrographs of frontal sections through the inferior colliculus of adults labelled with tritiated thymidine as a result of injections made when pouch-young were at the stated ages in days.
L. AITKIN ET AL.
Sections are selected at the frontal level where strongest labelling occurs, and show collectively a general ventral-to-dorsomedialprogression from 7d to 16d.
255
AUDITORY NEUROGENESIS IN A MARSUPIAL
A
1 0 0 ] ~ ~ ~ Y s
50 \
' 0 OR
50 loot OR 50 PERCENTAGE OF ROSTROCAUDAL DISTANCE
100 c
Fig. 5. The numbers of darkly labelled cells at various rostrocaudal levels of the inferior colliculus are expressed as percentages of the maximum number for m y one brain. The absolute number is shown adjacent to the peak for that brain, and the animal number below the
age of injection in days. The rostral limit of the inferior colliculus is defined as the region where the densely packed central nucleus is replaced entirely by the loosely packed rostral external nucleus. The caudal limit is that of the entire inferior colliculus.
lar cells, the superior paraolivary nucleus (SPN). The dark staining cells of the medial nucleus of the trapezoid body (MTB) form lines connecting the superior olivary complex to the midline of the brainstem. Relative to the midbrain, thalamus, and cortex, very few thymidine labelled cells were observed in adult animals in the two ventral nuclei of the cochlear nuclear complex or in the MSO or SPN when injections were made in pouchyoung older than 3 days. However, scattered pyramidal and deeper cells were present in the DCN at all injected ages examined (e.g., Fig. 3B), suggesting a prolonged period of neurogenesis without any clear-cut peak for this nucleus. In contrast neurogenesis in the MTB and LSO were completed in a short period, between days 5 and 9, peaking at day 7 (Fig. 3D). There was no evidence for any gradient in the MTB; cells were born at the same time along the entire rostrocaudal extent of this structure. The very low cell numbers in the medial superior olive and ventral cochlear nucleus at any of the injected ages studied, combined with the clear-cut later peaks in higher nuclei, suggests that the peak of neurogenesis for these brainstem nuclei has occurred earlier than 3 days of pouch age. However, in both ventral and dorsal cochlear nuclei, granule cells were labelled in increasing numbers as postnatal age increased (to day 56).
Like that of other mammals, the inferior colliculus of Dasyurus hallucatus has a cell-dense central nucleus surrounded by a thin mantle of cells and fibres (Aitkin et al., '86a). Caudally the structure is free-standing, whereas rostrally it gradually merges into the tegmentum beneath the superior colliculus. Thus the caudal boundary of the central nucleus is well defined, whereas its rostral limit is less clear. As with other mammals, the central nucleus has a systematic cochleotopic organisation (Aitkin et al., '86b). Some examples of frontal sections through the inferior colliculus in which the greatest density of labelling was found for a given animal are shown in Figure 4 for adults, following thymidine injections at 7, 10, 13, and 16 days. Labelled cells first appear in adults injected on day 5 of pouch-life, in the rostral half of the central nucleus, at its ventral margin. By day 7 numbers have increased considerably and labelled cells are concentrated in the rostral and ventrolateral half of the central nucleus (Figs. 43).In older injected pouch-young,labelled cells occupy the rostrocaudal and dorsoventral centers of the central nucleus (days 9-13; Fig. 4) gradually shifting caudally as increasingly older pouch-young were studied (Fig. 5). By days 15 and 16, labelled cells are only found in the caudomedial part of the
Inferior colliculus
L. AITKIN ET AL.
256
Fig. 6. Darkfield photomicrographs of frontal sections through the medial geniculate body (MG) labelled with tritiated thymidine injected when animals were pouch-young of 3 and 9 days of age. The internal
boundary of the MG is shown by a dashed line. Note that the 3d section is taken at a level corresponding to the caudal one-third, and the 9d section from the rostral one-third, of the MG.
central nucleus, become progressively fewer in number by day 22 (Figs. 4,5), and are not observed in the central nucleus beyond this date. Variability in cell counts as a function of age and rostrocaudal location makes it difficult to identify sharp rostrocaudal gradients in these data. In this respect, the 13-day embryo was provided by a mother trapped in a year different to those illustrating the other ages, and the rostrocaudal cell count graph appears out of sequence with the others. The times of origin of cells in the lateral and dorsal cortices of the inferior colliculusmatch those for the central nucleus in relation to position; cells in rostral locations are generated between days 5 and 10, and more caudal cells are born between days 12 and 22. Thus the rostroventral to caudomedial gradient appears to occur through all subdivisions of the inferior colliculus.
appear to respect such boundaries. The internal boundary of the entire medial geniculate body (MG) is shown by the dashed line in Figure 6. Neurogenesis in the medial geniculate body has already begun on day 3 and is virtually complete on day 11.Labelled cells first appear at the caudal ventral margin and form a swathe in the ventral half of its caudal one-third (3d in Fig. 6). In adult animals injected as progressively older pouchyoung, labelled cells shift rostrally and dorsally and are present only in the dorsal half of the rostral medial geniculate by days 9-11 (9d in Fig. 6). Neurogenesis in the medial geniculate is thus complete when that in the inferior colliculus is only just approaching its peak.
Medial geniculate body The medial geniculate body of Dasyurus hallucatus is composed of a ventral division, which forms the major part of the nucleus, capped by a small sheet of cells-the dorsal nucleus-and flanked medially by the small medial nucleus (Kudo et al., '89). Boundaries, however, are difficult to discern in Nissl material, and neurogenetic gradients do not
The auditory cortex The primary auditory cortex (AI) is the only part of the auditory cortex that has been mapped convincingly in Dasyurus hallucatus (Aitkin et al., '86b). The mapped region extends for 5-6 mm in the rostrocaudal middle of the cortex and lies immediately dorsal to the rhinal fissure. It is characterised by a blurring of lamination in the middle part of the cortex (Fig. 7A) that beginsjust ventral (arrow in Fig. 7) to a shallow depression in the cortex, the lateral crease (LC in Fig. 7A). Above this crease a clear small-celled layer IV may be distinguished (Fig. 7A), and the weak auditory
AUDITORY NEUROGENESIS IN A MARSUPIAL
257
Fig. 7. A. thionin-stained frontal section through the auditory cortex of Dasyurus. The primary auditory cortex, AI,lies between the rhinal fissure (RF) and lateral crease (LC); its dorsal limit is marked in all sections with an arrow. The remaining, darkfield, sections are
graded in age (in days) when animals were injected as pouch-young. The boundary between cortex and white matter is shown by dashed lines in the dark field photographs. Calibration: 1.3 mm A; 0.5 mm for the darkfield sections.
responses in this region suggest that it is a secondary auditory area (Aitkin et al., '86b). No thymidine-labelled cells are ever seen in the AI in animals injected prior to day 7, but by day 9 labelled cells are found in the deepest part of the adult auditory cortex, adjacent to the white matter (9d in Fig. 7 ) .The cells of the deepest layer are born by day 15 (15d in Fig. 7 ) and a population has also been generated to form the middle layers of this particularly densely labelled preparation. The injection at 16 days of a pouch-young from a litter different to that at 15 days suggests that many neurons populating the middle layers have been born by this time (16d in Fig. 7). By 32 days labelled cells destined for the superficial one-third of the cortex are generated (32d in Fig. 7 ) and those of the most superficial layer are born by day 38. Beyond this age, labelled cell numbers simply diminish in layer I1 and are not observed when injections are made after day 45.
These qualitative observations are expressed in graphical form in Figure 8, in which the depth from the pial surface of the center of the swathe of labelled cells is plotted against the age at which the pouch-young was injected, in days. Three measures were taken from a section located in the rostrocaudal center of the AI, one close to the lateral crease, one near the point of maximum curvature, and the other midway between (Fig. 8). The accuracy of these measurements improved as the postnatal age of the injected young increased, because (as is obvious in Fig. 7) the first generated cells were scattered broadly across the deepest layers, whereas those born at a later time formed narrower bands in the AI. The measurements plotted in Figure 8 confirm for the marsupial native cat the inside-out development of the cerebral cortex described for Eutheria (e.g., Altman and Bayer, '81; Angevine and Sidman, '61). There is another gradient of neurogenesis in the cerebral cortex that influences qualitatively the AI gradient. At any
L. AITKIN ET AL.
258
2000
1500 m CK w t W
I
0
5I:
1000
z I c
a w
cl
500
0
0
20
10
30
40
INJECTED AGE IN DAYS Fig. 8. Data from the AI quantifying a trend suggested in Figure 7. The centers of the swathes of labelled cells in each brain have been measured at three dorsoventral positions and plotted as a function of depth, thus showing that cells in the deep and middle layers (i.e.,
greater than 50Qpm in depth) are “born” over a short period, between 9 and 16 days pouch life, whereas those populating the superficial layers are born over a prolonged period (16-40+ days). LC (lateral crease); RF (rhinal fissure).
given age during neocortical neurogenesis, cells at a given depth in dorsal parts of the neocortex of the native cat are born later than those ventrally (i.e., in the AI). This can be recognised in Figure 7 if the cortical regions above the arrows are compared with the AI, below. Similar gradients of neurogenesis have been observed in the possum neocortex (Sanderson and Weller, ’90). However, no consistent differences are seen in this regard within the AI. Measures taken in the dorsal, middle, and ventral A1 provide curves that are very similar in position as a function of injected age. Any differences can be accounted for by the different physical dimensions of the brains of different adult animals and by the sharp flexion of the temporal cortex as it folds into the rhinal fissure, rather than by any internal dorsoventral gradient.
22. Neurogenesis begins in the auditory cortex on day 9 and is completed by about day 42. These time sequences suggest that neurogenesis occurs nearly simultaneously in two separate compartments-the brainstem (medulla,then midbrain) and forebrain systems (medial geniculate, then auditory cortex). The differencesbetween the time courses of neurogenesis in possums (Sanderson and Aitkin, ’90) and native cats are quite small and could be accounted for by individual variation in members of both species. In both, ventral cochlear nucleus cells were never labelled in the experimental material, suggesting that they are generated at or before the time of birth. In both species,auditory cortical neurogenesis occurs over a period of about 5-6 weeks and is complete, in native cats, at about the same time as that in the visual cortex-some 30 days before eye opening (Crewther et al., ’88). Comparisons between neurogenesis in marsupials and in eutherian mammals are complicated by the short gestation period in marsupials (2-3 weeks) relative to eutherian mammals of a similar size (cats 65 days, macaque monkeys 164 days). The time of eye opening is a convenient event to compare across species because it is associated with the beginnings of special sensory reception and occurs earlier in cats (about 73 days post-fertilization) than in native cats (95 days) or possums (117 days) (Sanderson and Aitkin, ’90). However, considerable visual development occurs over at least another 6 weeks postnatally in cats (e.g., Hubel and
DISCUSSION The time course of neurogenesis in the brain auditory pathway of the marsupial native cat, Dasyurus hallucatus, has been studied. There is no simple relationship between the level of an auditory nucleus and the genesis of its constituent neurons. Neurons are generated in the ventral cochlear nucleus prior to 3 days pouch-life, in the superior olive at 5-7 days, and in the dorsal cochlear nucleus over a prolonged period. Inferior collicular neurogenesis lags behind that in the medial geniculate, the latter taking place between days 3 and 9 and the former between days 7 and
-
AUDITORY NEUROGENESIS IN A MARSUPIAL
0
259
100%
1
CONCEPTION
EYE OPENING
NATIVECAT
I POSSUM
; MACAQUE MONKEY
RAT
I
1
I
MOUSE
Fig. 9. A comparison of t h e time periods over which cells in t h e inferior colliculus (above each line) and auditory cortex (below each line) are born, expressed a s a percentage of t h e period between
conception and eye opening. Data obtained from t h e present study: Sanderson and Atkin ('901, Cooper and Rakic ('811,Altman and Bayer ('81) and Taber-Pierce ('73).
Weisel, '701, and auditory thresholds continued to improve in kittens during the first 4 weeks after birth (Ehret and Romand, '81). The only marsupial for which behavioral evidence for the onset of hearing is available is the opossum (63 days postfertilization; Ehret, '83); eye opening in this species occurs at 73 days (Sanderson and Aitkin, '90). However, observation indicates that pouch young of Dasyum s begin vocalizing at about 40 days (Nelson, unpublished observations); this could correspond to the onset of hearing in this species. Even though eye opening is not likely for any species to signal the end of visual development, it is a convenient index for comparison between those species for which neurogenetic information about auditory development is available. Figure 9 expresses the postconceptual periods during which neurons of the inferior colliculus (above the lines) and auditory cortex (below the lines) are generated, to the times of eye opening. It is clear that neurogenesis begins very much earlier in the two marsupials but the process is, relative to Eutheria, prolonged. The last cells destined for the marsupial auditory centers are generated at a time corresponding to the beginning of neurogenesis in the auditory centers of the three eutherian species. It will be interesting to see when the auditory system begins to function, given the fact that neurogenesis occurs early in auditory structures in native cats. Preliminary observations from this laboratory reveal that some weeks may elapse between neurogenesis and the appearance of cytoarchitecturally distinct auditory nuclei. Thus although the medullary auditory nuclei are distinct at day 13, the central nucleus of the inferior colliculus and medial geniculate body are not easy to recognise until about day 23. Nothing is known of the time of arrival of connections to the brainstem from the cochlea, but the optic nerve reaches its target sites in the thalamus at about day 25 (Crewther et al., '88), so it is possible that the auditory connections may
form at a similar time. This, and the appearance of spontaneous vocalizations at 40 days, would concur with the onset of hearing some time midway during pouch-life in Dasyurus.
ACKNOWLEDGMENTS The authors thank Dianne Clare and Jill Poynton for preparing the illustrations, and Dr. Ken Sanderson for valuable critical comments. This study was supported by grants from the National Health and Medical Research Council of Australia. Animals were collected under Conservation Commission of the Northern Territory permit SL8/84 and held under Victorian Department of Conservation permit 84-26.
LITERATURE CITED Aitkin, L.M., M. Byers, and J.E. Nelson (1986a) Brain stem auditory nuclei and their connections in a carnivorous marsupial, the Northern native cat (Dasyurus hallucatus). Brain, Behav. Evol. 29:l-16. Aitkin, L.M., D.R.F. Irvine, J.E. Nelson, M.M. Merzenich, and J.C. Clarey (198613) Frequency representation in the auditory midbrain and forebrain of a marsupial, the Northern native cat (Dasyurus hallucatus). Brain, Behav. Evol. 29:17-28. AItman, J., and S.A. Bayer (1981) Time of origin of neurons of the rat inferior colliculus and the relations between cytogenesis and tonotopic order in the auditory pathway. Exp. Brain Res. 4 2 4 1 1 4 2 3 . Angevine, J.B., and R.L. Sidman (1961) Autoradiographic study of cell migration during histogenesis of cerebral cortex in the mouse. Nature 192:766-768. Cooper, M.L., and P. Rakic (1981) Neurogenetic gradients in the superior and inferior colliculi of the Rhesus monkey. J. Comp. Neurol. 202:309334. Crewther, D.P., J.E. Nelson, and S.G. Crewther (1988) Afferent input for target survival in marsupialvisual development. Neurosci. Letts. 86:147154.
260 Ehret, G. (1983) Development of hearing and response behavior to sound stimuli: behavioral studies. In R. Romand (ed): Development of Auditory and Vestibular Systems. New York Academic Press, pp. 211-237. Ehret, G., and R. Romand (1981) Postnatal development of absolute auditory thresholds in kittens. J. Comp. Physiol. Psychol. 95304-31 1. Hubel, D.H., and T.N. Wiesel (1970) The period of susceptibility to the physiological effects of unilateral eye closure in kittens. J. Physiol. (Lond). 206;419-436. Kudo, M., L.M. Aitkin, and J.E. Nelson (1989) Auditory forebrain organization of an Australian marsupial, the Northern native cat (Dusyurus hallucatus).J. Comp. Neurol. 27928-42.
L. AITKIN ET AL. Nelson, J.E. (1988) Growth of the brain. In C.H. Tyndale-Biscoe and P.A. Janssens (eds): The Developing Marsupial. Models for Biomedical Research. Berlin: Springer-Verlag, pp. 86-100. Sanderson, K.J., and L.M. Aitkin (1990) Neurogenesis in the visual and auditory pathways of a marsupial: The brushtailed possum (Trichosurus uulpecula). Brain, Behav. Evol. 35:325-338. Sanderson, K.J., and W.L. Weller (1990) Gradients of neurogenesis in possum neocortex. Dev. Brain Res., 55:269-274. Taber-Pierce, E. (11973) Time of origin of neurons in the brain stem of the mouse. Prog. Brain Res. 40;53-65.
