Mean Soma Diameter Research Articles

Quantification of Retinal Ganglion Cell Morphology in Human Glaucomatous Eyes.

PurposeTo characterize retinal ganglion cell morphological changes in patients with primary open-angle glaucoma associated with hemifield defect (HD) using adaptive optics–optical coherence tomography (AO-OCT).MethodsSix patients with early to moderate primary open-angle glaucoma with an average age of 58 years associated with HD and six age-matched healthy controls with an average age of 61 years were included. All participants underwent in vivo retinal ganglion cell (RGC) imaging at six primary locations across the macula with AO-OCT. Ganglion cell layer (GCL) somas were manually counted, and morphological parameters of GCL soma density, size, and symmetry were calculated. RGC cellular characteristics were correlated with functional visual field measurements.ResultsGCL soma density was 12,799 ± 7747 cells/mm2, 9370 ± 5572 cells/mm2, and 2134 ± 1494 cells/mm2 at 3°, 6°, and 12°, respectively, in glaucoma patients compared with 25,058 ± 4649 cells/mm2, 15,551 ± 2301 cells/mm2, and 3891 ± 1105 cells/mm2 (P < 0.05 for all locations) at the corresponding retinal locations in healthy participants. Mean soma diameter was significantly larger in glaucoma patients (14.20 ± 2.30 µm) compared with the health controls (12.32 ± 1.94 µm, P < 0.05 for all locations); symmetry was 0.36 ± 0.32 and 0.86 ± 0.13 in glaucoma and control cohorts, respectively.ConclusionsGlaucoma patients had lower GCL soma density and symmetry, greater soma size, and increased variation of GCL soma reflectance compared with age-matched control subjects. The morphological changes corresponded with HD, and the cellular level structural loss correlated with visual function loss in glaucoma. AO-based morphological parameters could be potential sensitive biomarkers for glaucoma.

The Effects of De-Whiskering and Congenital Hypothyroidism on The Development of Nitrergic Neurons in Rat Primary Somatosensory and Motor Cortices.

Objective The aim of the present study is to investigate the effects of chronic whisker deprivation on possible alterations to the development of nitrergic neurons in the whisker part of the somatosensory (wS1) and motor (wM1) cortices in offspring with congenital hypothyroidism (CH). Materials and Methods In the experimental study, CH was induced by adding propylthiouracil to the rats drinking water from embryonic day 16 to postnatal day (PND) 60. In whisker-deprived (WD) pups, all the whiskers were trimmed from PND 1 to 60. Nitrergic interneurons in the wS1/M1 cortices were detected by NADPH-diaphorase histochemistry staining technique in the control (Ctl), Ctl+WD, Hypo and Hypo+WD groups. Results In both wS1 and wM1 cortices the number of nitrergic neurons was significantly reduced in the Hypo and Hypo+WD groups compared to Ctl and Ctl+WD groups, respectively (P<0.05) while bilateral whisker deprivation had no remarkable effect. The mean soma diameter size of NADPH-d labeled neurons in the Ctl+WD and Hypo+WD groups was decreased compared to the Ctl and Hypo groups, respectively. A similar patterns of decreased NADPH-d labeled neurons in the wS1/M1 cortices occur in the processes of nitrergic neurons in both congenital hypothyroidism and whisker deprivation. ConclusionOur results suggest that both congenital hypothyroidism and whisker deprivation may disturb normal development of the wS1 and wM1 cortical circuits in which nitrergic neurons are involved.

Morphometric Plasticity of Nitric Oxide Containing Neurons in the Barrel Cortex of De-whiskered Rats

Background: The rodent somatosensory barrel cortex is an ideal model to examine the effect of experience-dependent plasticity on developing brain circuitry. Sensory deprivation such as whisker deprivation may affect neuroanatomical aspects of the brain during developmental processes. The present study designed to investigate the possible effects of whisker deprivation on the morphometric characteristics of NADPH-d positive neurons in the barrel field cortex of adolescent rats.Materials and Methods: Pups were divided into the intact (n=4) and whisker-deprived groups (n=4). In whisker-deprived group, the total whiskers of subjects were trimmed every other day from postnatal day (PND) 0 to PND 60. NADPH-d histochemistry reaction was processed to quantitatively analyze the feature of NADPH-d containing neurons of barrel cortex.Results: Our results showed that the number of NADPH-d positive neurons remained unchanged in whisker-deprived group compared to controls. The mean soma diameter, dendritic length and the number of 3rd order processes were significantly decreased in the whisker-deprived rats (p<0.05).Conclusion: Our results indicate that postnatal whisker deprivation possibly alter NADPH-d/NOS neuronal features in the barrel cortex. The functional implications of these data may relate the plasticity of synaptic receptive field and developmental brain circuits.

Spiral ganglion neuron quantification in the guinea pig cochlea using Confocal Laser Scanning Microscopy compared to embedding methods

Neuron counting in the cochlea is a crucial but time-consuming operation for which various methods have been developed. To improve simplicity and efficiency, we tested an imaging method of the cochlea, and based on Confocal Laser Scanning Microscopy (CLSM), we visualised Rosenthal's Canal and quantified the spiral ganglion neurons (SGN) within.Cochleae of 8 normal hearing guinea pigs and one implanted with a silicone filament were fixed in paraformaldehyde (PFA), decalcified, dehydrated and cleared in Spalteholz solution. Using the tissue's autofluorescence, CLSM was performed at 100fold magnification generating z-series stacks of about 20 slices of the modiolus. In 5 midmodiolar slices per cochlea the perimeters of the Rosenthal's Canal were surveyed, representative neuron diameters were measured and the neurons first counted manually and then software-assisted. For comparison, 8 normal hearing guinea pig cochleae were embedded in paraffin and examined similarly.The CLSM method has the advantage that the cochleae remain intact as an organ and keep their geometrical structure. Z-stack creation is nearly fully-automatic and frequently repeatable with various objectives and step sizes and without visible bleaching. The tissue shows minimal or no shrinking artefacts and damage typical of embedding and sectioning. As a result, the cells in the cleared cochleae reach an average diameter of 21 μm and a density of about 18 cells/10,000 μm2 with no significant difference between the manual and the automatical counts. Subsequently we compared the CLSM data with those generated using the established method of paraffin slides, where the SGN reached a mean density of 9.5 cells/10,000 μm2 and a mean soma diameter of 13.6 μm.We were able to prove that the semi-automatic CLSM method is a simple and effective technique for auditory neuron count. It provides a high grade of tissue preservation and the automatic stack-generation as well as the counter software reduces the effort considerably. In addition this visualisation technique offers the potential to detect the position and orientation of cochlear implants (CI) within the cochlea and tissue growing in the scala tympani around the CI and at the position of the cochleostomy due to the fact that the implant does not have to be removed to perform histology as in case of the paraffin method.

The size of motoneurons of the gastrocnemius muscle in rats with diabetes

Alterations in the number and size of motoneurons were studied in the medial gastrocnemius (MG) motor nucleus of diabetic rats (12 or 22 weeks after injection of storeptozotocin) and age-matched controls. Each group contained 6 animals. MG motoneurons were retrogradely labeled by dextran–fluorescein and the number and size of cell bodies were examined.Significantly fewer labeled MG motoneurons were found in the 22-week diabetic rats as compared with age-matched control animals. The mean soma diameter of MG motoneurons was significantly smaller in the 12- and 22-week diabetic animals. Furthermore the soma size for 22-week diabetic animals was smaller than for 12-week diabetic animals. The distribution of average soma diameters in the MG nucleus of control animals was bimodal; cells with larger average diameter were presumed to be alpha-motoneurons and those with smaller diameters were presumed to be gamma. Compared to control animals, the number of smaller MG motoneurons was reduced in 12 week diabetic animals. By 22 weeks, diabetic animals had no small MG motoneurons and the size distribution became unimodal.We conclude that there is a significant decrease in the absolute number and size of MG motoneurons in diabetic rats, with the possibility that the decrease occurred predominantly among the smaller gamma-motoneurons.

Interdigitation of nitric oxide synthase-, tyrosine hydroxylase-, and serotonin-containing neurons in and around the laterodorsal and pedunculopontine tegmental nuclei of the guinea pig.

The topography of neurons containing nitric oxide synthase (NOS) and monoamines was investigated in the guinea pig mesopontine tegmentum. NOS-containing neurons were identified with NADPH-diaphorase (NADPH-d) histochemistry, and monoamine-containing neurons were identified with tyrosine hydroxylase (TH) and serotonin (5-HT) immunocytochemistry. The distribution of NADPH-d positive cells was centered on the laterodorsal tegmental (LDT) and pedunculopontine tegmental (PPT) nuclei. Diaphorase-containing cells had a mean soma diameter of 23.0 +/- 4.1 microns (n = 160) and were distributed inhomogeneously, with numerous cells found within densely packed clusters. A nearest-neighbor analysis revealed that these cells were closely spaced, with up to 20% within one cell diameter and more than 50% within two cell diameters of a neighboring NADPH-d cell. Within the LDT and PPT, NADPH-d positive cells were mixed with smaller, diaphorase-negative cells (diam: 12.8 +/- 3.3 microns; n = 182; P << 0.01). TH-containing cells were not organized into a compact LC as in rat and their distribution more closely resembled that observed in cat. On average, TH-containing cells (diam: 21.2 +/- 4.8 microns; n = 160) were smaller than NADPH-d cells (P < 0.01). 5-HT-containing cells were mainly located in the raphe nuclei, as in other species. 5-HT-containing cells (diam: 18.2 +/- 4.4 microns; n = 161) were smaller on average than both the NADPH-d (P < 0.01) and TH-containing cells (P < 0.01). An analysis of the overlap in soma distributions revealed that TH-containing cells were largely interdigitated with NADPH-d containing cells. As much as 78% of the area occupied by the NADPH-d cells of LDT was contained within the area occupied by TH cells. Substantial numbers of TH and 5-HT immunoreactive processes were seen in both LDT and PPT. Varicose 5-HT and TH-containing fibers, as well as thicker, possibly dendritic processes containing TH were often seen in close apposition to NADPH-d containing somata and proximal dendrites. These results support the hypothesis that NADPH-d cells of both the PPT and LDT receive input from TH and 5-HT cells. Moreover, the clustered substructure of LDT and PPT and the extensive overlap of NADPH-d and TH-containing somata raise the possibility that the membrane permeable messenger nitric oxide plays a role in modulating TH-containing somata and their processes as well as 5-HT-containing processes in the LDT and PPT.

Early postnatal changes in the somatodendritic morphology of ankle flexor motoneurons in the rat.

The development of locomotor function in the rat spans the first 3 postnatal weeks. We have studied morphological features of the soma and dendrites of motoneurons innervating the physiological flexor muscles of the ankle, tibialis anterior and extensor digitorum longus, by intracellular injection in vitro between the first and ninth postnatal days. We obtained serial optical sections of 96 adequately filled motoneurons in whole-mounted hemisected spinal cords by confocal microscopy, projected them onto a single plane and analysed them morphometrically. On the day after birth, the somatodendritic surfaces of most such motoneurons were covered in growth-associated spiny, thorny or hair-like appendages. These had disappeared from the soma by the fourth postnatal day and from most proximal dendrites by day 7, but were still common distally on day 9. During this period there was little or no net growth of either the soma (which was still much smaller than in the adult) or the dendritic tree. A dorsal dendritic bias was present and 'sprays' of long, loosely bundled dorsal dendrites were often seen. The mean number of primary dendrites remained constant at about eight, and their combined diameter was already significantly correlated with mean soma diameter, as in the adult cat. Thus, the critical neonatal period during which these ankle flexor motoneurons are known to change their electrophysiological properties and to be particularly sensitive to interference with neuromuscular interaction is characterized by major changes in the neuronal surface, presumably linked to synaptogenesis.

The morphology and electrical geometry of rat jaw‐elevator motoneurones.

1. The aim of this work was to quantify both the morphology and electrical geometry of the dendritic trees of jaw-elevator motoneurones. To do this we have made intracellular recordings from identified motoneurones in anaesthetized rats, determined their membrane properties and then filled them with horseradish peroxidase by ionophoretic ejection. Four neurones were subsequently fully reconstructed and the lengths and diameters of all the dendritic segments measured. 2. The mean soma diameter was 25 microns and values of mean dendritic length for individual cells ranged from 514 to 773 microns. Dendrites branched on average 9.1 times to produce 10.2 end-terminations. Dendritic segments could be represented as constant diameter cylinders between branch points. Values of dendritic surface area ranged from 1.08 to 2.52 x 10(5) microns 2 and values of dendritic to total surface area from 98 to 99%. 3. At branch points the ratio of the summed diameters of the daughter dendrites to the 3/2 power against the parent dendrite to the 3/2 power was exactly 1.0. Therefore the individual branch points could be collapsed into a single cylinder. Furthermore for an individual dendrite the diameter of this cylinder remained constant with increasing electrical distance from the soma. Thus individual dendrites can be represented electrically as cylinders of constant diameter. 4. However dendrites of a given neurone terminated at different electrical distances from the soma. The equivalent-cylinder diameter of the combined dendritic tree remained constant over the proximal half and then showed a pronounced reduction over the distal half. The reduction in equivalent diameter could be ascribed to the termination of dendrites at differing electrical distances from the soma. Therefore the complete dendritic tree of these motoneurones is best represented as a cylinder over the proximal half of their electrical length but as a cone over the distal half.

Morphology and projections of myenteric neurons to colonic fiber bundles of the cat

Regulation of colon function depends on the location of nerve cell bodies and the distribution of intrinsic nerve fibers in the myenteric plexus. The morphology and projections of myenteric neurons through colonic fiber bundles in cat colon were determined using in vivo retrograde transport of HRP and Fast blue. Myenteric neurons were found to project from at least 5 to 59 mm orad (mean: 42 mm) or aborad (mean: 54 mm) through colonic fiber bundles. Approximately 73% of labelled cells were in ganglia within 2.8 mm of colonic fiber bundles in the axis of circular muscle fibers; none was beyond 7.7 mm. There were 2 soma morphologies. One type (Dogiel type I) had a mean soma diameter of 40.5 μm and had a rough somal surface. There were few if any short, broad dendrites, but its one long process extended to a branch point of an adjacent colonic fiber bundle. The other type (Dogiel type III) had a mean soma diameter of 26.4 μm, had a smooth somal surface and had few if any fine dendrites. It also projected a single long axon to colonic fiber bundles. There were twice as many Dogiel type III neurons. We conclude that myenteric neurons in the cat colon project both orad and aborad over relatively long distances through colonic fiber bundles where they form another intrinsic neuronal connection for the myenteric plexus.

Physiological and morphological analysis of cat digastric motoneurons after intracellular staining with horseradish peroxidase.

Horseradish peroxidase was injected into the somata of cat digastric motoneurons. Twelve motoneurons for which the spatial distribution of dendritic arborization and axon trajectory were iden-tified physiologically were studied with reference to the location of cell bodies in the trigeminal motor nucleus. It was revealed that, irrespective of the rostrocaudal level of the soma, dendrities of digastric motoneurons were present mainly in the medial half of the nucleus as well as in extranuclear regions. Interestingly very few dendritic branches were observed in the lateral half of the motor nucleus, this “unoccupied” region corresponding to the masseter subdivision of the nucleus. The axons ran a dorsomedial course and descended to the motor root without giving rise to collaterals. The average size of the soma was 44.2μm with a range of 36.9-50.1μm. A linear correlation was observed between the total stem dendrite diameter and the mean soma diameter, giving a coefficient of 0.534. No excitatory postsynaptic potential was recorded from the soma upon stimulation of the digastric nerve with a current below the threshold, indicating a lack of muscle spindles.

Musculotopic organization of the facial motor nucleus in macaca fascicularis: A morphometric and retrograde tracing study with cholera toxin B‐HRP

Morphometric and retrograde tracing methods were used to determine the location and number of motoneurons innervating individual facial muscles in Macaca fascicularis. Intramuscular injections of the cholera toxin B subunit-horseradish peroxidase conjugate produced discrete labeling patterns in the ipsilateral facial motor nucleus with good definition of somata and their processes. The facial nucleus extended rostrocaudally in the pons for about 2 mm, varying in shape and cross-sectional area along this axis. Motoneurons were clustered in subnuclei, but their boundaries were not sharp and they were not segregated by fiber bundles. The length, number, and area of subnuclei varied with rostrocaudal location. Retrograde labeling patterns revealed that individual muscles were innervated by longitudinal columns of motoneurons with each muscle region represented at all rostrocaudal levels of its column. The columns began at different rostrocaudal levels and varied in length. Columns for closely related muscles, such as the orbicularis oris and mentalis of the lower lip, tended to overlap, whereas columns for disparate muscles, such as the perioral and orbital, did not overlap. The dendritic processes of most motoneurons branched extensively among several different columns or subnuclei. Some dendrites extended outside of the nucleus into the surrounding tegmentum. Mean soma diameter (10.4-42.2 microns) was distributed unimodally, reflecting the absence of gamma motoneurons and lack of muscle spindles in the facial muscles. Large and small motoneurons were found in all regions of the nucleus, but the largest ones were located caudally and innervated muscles of the upper and lower lip. The perioral muscles also had more neurons, longer columns, and a lower cell density than the other muscle groups examined. These features may reflect the functions of the perioral muscles in facial expression and vocalization.

Morphometric parameters of the superior colliculus of albino and pigmented rats.

The superior colliculus (SC) or optic tectum of mammals consists of seven layers, numbered I-VII from superficial to deep, each of which has distinct connectivity patterns and electrophysiological response properties. The present study is devoted to a morphometrical analysis of neuronal diameters, densities, and numbers in different layers and regions of the SC of albino as well as pigmented rats in order to present a quantitative characterization of the collicular neuronal population involved in the different connectivities and functions of these compartments. The morphometric parameters were calculated from tracings of nuclei and cell bodies by means of Kontron-Videoplan equipment and a Micro PDP 11/23 computer. The mean soma diameter per superior colliculus appears to be 12.0 microns, the average neuronal density 70 cells per 0.001 mm3, and the total number of neurons about 600,000. The mean soma diameter gradually increases from superficial to deep layers (i.e., from 10.0 to 14.0 microns). Cellular density is highest in layer III, the retinal afferent layer (90 cells per 0.001 mm3), and decreases both in more superficial layers (to about 80 in layer I) and deeper layers (to about 44 in layer VII). About 25% of all collicular neurons are situated in layer II whereas layer I contains the lowest percentage of cells (4%). Rostrally within each collicular layer, cellular volumes are about 25% larger than caudally. On the other hand, neuronal densities are rostrally about 38% lower than caudally in all layers except for layers VI and VII. We conclude that collicular neurons, in contrast to collicular axons, are not arranged in distinct layers or clusters but basically establish a random network with only gradual transitions. In this respect, no statistically significant differences were observed between albino and pigmented rats.

Brainstem projections to the phrenic nucleus: A HRP study in the cat

Brainstem neurones which project to the phrenic nucleus were identified using retrogradely transported horseradish peroxidase (HRP) as a marker. Following iontophoretic injection of HRP into the phrenic nucleus, labelled cells were encountered throughout large areas of the medulla and pons, but occurred with characteristic high densities in those regions known to contain phasic respiratory neurones: namely, the ventrolateral solitary tract nucleus (vl-NTS), known as the dorsal respiratory group (DRG), the ambiguus complex or ventral respiratory group (VRG) and the parabrachial pontine nuclei (BCM-KF). In 12 cats a total of 1540 cells was identified within these regions, the relative contralateral and ipsilateral contributions were respectively 72%:28% (vl-NTS), 65%:35% for the ambiguus complex, and 5%:95% (BCM-KF). In addition, labelled cells, predominantly ipsilateral, were observed in the pontine and medullary reticular formation and the vestibular nuclei. The labelled cells of the DRG had round, oval or triangular perikarya. Their mean soma diameter was 18.3 μm. The HRP-positive cells of the VRG had slightly larger somas (mean 21.2 μm) and they were fusiform and triangular. The neurones labelled in the BCM-KF nuclei were more heterogeneous with a mean soma size of 14.9 μm. The bilateral projections to the phrenic nucleus from the DRG and the VRG, and the predominantly ipsilateral projection from the BCM-KF are discussed in relation to current electrophysiological and autoradiographic findings.

Does α-motoneurone size correlate with motor unit type in cat triceps surae?

The cell bodies and first-order dendrites of α-motoneurones supplying different functional types of muscle units in the cat gastrocnemius (type FF, FR and S units) and soleus (type SOL-S units) muscles, were studied after intracellular injection of horseradish peroxidase. The SOL-S neurones had smaller values for cell body diameter in comparison with both the FF and FR neurones. The SOL-S neurones also had significantly thinner first-order dentrites than the FF, FR and S neurones. In the gastrocnemius pool the S neurones had smaller values for dendritic diameters than the FF and FR cells. The values for combined diameter of the first-order dendrites indicated that the dendritic trees of the FF and FR neurones are, on the average, larger than those of the S and SOL-S neurones. Furthermore, the relationship between the combined dendritic diameter and the mean soma diameter, indicated that a difference in relative scaling of soma and dendrites exists between the FF and FR neurones on the one hand and the S and SOL-S neurones on the other. Similar results were obtained also when relating the combined dendritic parameter εd 3/2 to the soma surface area. Although a certain statistical relation seems to exist between motoneurone size and motoneurone type, it should be emphasized, however, that the range of values for each parameter studied overlapped considerably between the different types of motoneurones.

The duration of after-hyperpolarization in hindlimb alpha motoneurones of different sizes in the cat

In the present intracellular investigation, direct measurements were made of the size of the cell body and the duration of after-hyperpolarization (AHP) among 76 hindlimb motoneurones of the cat. Following the penetration and intracellular recording of action- and after-potentials, each one of the cells was individually labelled by an injection of Procion yellow, Procion red or horseradish peroxidase from a microelectrode. In accordance with expectations, a significant negative correlation was found between the size of a motoneuronal cell body and the duration of its AHP. Motoneurones with a mean soma diameter exceeding about 50 μm usually had an AHP shorter than 110 msec. Cells smaller than 50 μm could, however, have AHPs of widely varying durations (range from 55 to 265 msec).

Localization and morphology of cat retractor bulbi motoneurons.

1. Motoneurons innervating the cat retractor bulbi muscle have been identified by retrograde axonal transport of horseradish peroxidase (HRP). Following injections of the four slips of the retractor bulbi muscle, labeled motoneurons were found in the abducens nucleus overlapping the distribution of lateral rectus motoneurons and in the oculomotor nucleus partially overlapping the distribution of medial rectus motoneurons. Retractor bulbi motoneurons also were found in the accessory abducens nucleus situated ventral and lateral to the abducens nucleus. 2. Retractor bulbi motoneurons varied considerably in shape and size, but in all instances contained similar cytoplasmic organelles. Quantitative analyses of mean soma diameter indicated that the average size of retractor bulbi motoneurons was larger than the average size of lateral rectus and medial rectus motoneurons. 3. Retractor bulbi motoneurons in the accessory abducens nucleus were identified electrophysically and stained by intracellular injection of HRP. Neuronal reconstructions demonstrated a dorsomedial axonal trajectory directed toward the abducens nucleus and elongated dendritic fields oriented in a dorsomedial-ventrolateral axis. Another major dendritic extension was directed toward the magnocellular division of the spinal trigeminal nucleus, a major source of excitatory input to these motoneurons. 4. Quantitative analyses of synaptic density indicated that the somata of retractor bulbi motoneurons were contacted by significantly fewer synaptic endings than the somata of motoneurons in the abducens nucleus. Retractor bulbi motoneurons in the abducens nucleus exhibited variations in synaptic density that were similar to the densities on lateral rectus motoneurons. 5. Given the morphological differences in location, size, and somadendritic extent between motoneurons in the accessory abducens, abducens and oculomotor nuclei, it is suggested that such features reflect functional differences between the motoneurons with respect to fiber composition of the muscles they innervate, and subsequently to the role each muscle plays in eye movement. 6. Since the morphological features of retractor bulbi motoneurons in the accessory abducens nucleus are quite different from those in either the abducens or oculomotor nuclei, it appears that each motoneuronal population may perform unique oculomotor functions.

Retinal ganglion cells in the crucian carp (Carassius carassius). I. Size and number of somata in eyes of different size.

Ganglion cell somata were drawn, measured and counted in flat-mounted crucian carp and goldfish retinas stained with cresyl violet or methylene blue. Some diameter histograms suggest that the ganglion cells can be divided into two populations with overlapping soma sizes: a large group of small cells and a small group of large cells, the latter constituting 2.5-5% of all ganglion cells. With increasing distance from the optic disc the mean soma diameter increases while the ganglion cell density decreases. In a peripheral growth zone close to the margin the ganglion cells become smaller again. The total number of ganglion cells in retinas of different size was calculated from the areas of the flat-mounted preparations and the cell densities in two representative regions. In the crucian carp population used in this work the total number of ganglion cells per retina was found to increase from roughly 140,000 (mean of 8 scattered value) to a full 200,000 between eye diameters 4 and 10 mm, this increase taking place mainly between eye diameters of 4 and 6.5 mm. Thus, due to a drastically decreasing cell density, the total number of ganglion cells increases only by a factor of about 1.5 while the retinal area becomes sixfold. During the same growth period the mean soma diameter increases by a factor of about 1.3 and the soma volume more than doubles. The optic nerve of a small crunated and myelinated axons were found. The axons in the optic nerve are, on an average, considerably thicker than the axons on the retinal surface.