Maximilian Salzmann.

The anatomy and histology of the human eyeball in the normal state, its development and senescence ; online

. (page 10 of 27)
Online LibraryMaximilian SalzmannThe anatomy and histology of the human eyeball in the normal state, its development and senescence ; → online text (page 10 of 27)
Font size
QR-code for this ebook

y Cajal or Held's stain.

The nerve-fibers do not go through the basal cones, as stated by Greeff,
but between them.

In the neighborhood of the fovea, the superficially perpendicular
course of the supporting fibers present in the extrafoveal territory of the
retina (perpendicular to the surface) gives way to an increasing obliquity
and, indeed, in the same sense as the clivus does, but it does not exactly
copy this declivity. The bending away from the surface is most marked
in the region of the outer fiber layer in the rest of the layers it is less so,
so that, in general, an 5-form curve is brought about.

(PI. IV, 3 , Li)

In the matter of the definition of this term two views are still opposed, as they were
in earlier periods. The one holds the membrana limitans interna to be the inner limita-
tion of the retina formed by the uninterrupted apposition of the basal cones of Mueller's
supporting fibers and logically considers the anatomically demonstrable membrane
lying inside the basal cones to be the border membrane of the vitreous, and calls it the
membrana hyaloidea.


The other calls the latter the membrana limitans interna retinae and, therefore,
denies the existence of a hyaloidea.

Against the first conception it is to be argued that a limitation of the retina is
probably effected by the totality of the basal cones of the supporting fibers, but only
so in the sense that it is a mathematical surface, not an actual border membrane. The
second definition of the m. limitans interna has for it that this membrane is anatomically
demonstrable, that is, has a measurable thickness and all the properties of a glass
membrane. In any case, one must say that this membrane has just as much relation
to the vitreous as it has to the retina, and that it looks like the inner glass membrane
of the retina in one preparation and like the outer border membrane of the vitreous
in another.

How Tornatola (224) comes to deny the existence of a membrane between the
retina and vitreous altogether cannot be understood without seeing the preparations
concerned. I have seen the membrane in many hundreds of eyes under normal as well
as under the most varied pathologic conditions, and so has everyone else seen it.
There has only been contention concerning what it should be called and to what it
belonged. I myself (184) have in my time called it the hyaloidea, following Retzius,
but I now prefer the name limitans interna, because it seems to me that the majority
of authors so designate it, and because an analogous membrane is present on the inner
surface of the ciliary epithelium.

So far as has yet been described, there lies between the retina and the
vitreous one (and only one) anatomically demonstrable membrane, the
membrana limitans interna retinae. It is a glass membrane of i to 2 mu
thickness, which in surface view sometimes shows the impressions of the
basal cones of Mueller's fibers as irregular polygonal fields. It continues
uninterrupted over ihefovea centralis, without any essential change; it is,
on the other hand, gradually lost at the optic-nerve entrance and also possi-
bly in many eyes at the ora serrata. (For more details see chap, ix, 8.)

b) Histologic and Functional Divisions of the Retina. Its Blood-
Vessels and Fovea Centralis

Following the detailed description of the structure of the retina, a
short survey of the significance of the individual elements and their
reciprocal relations is in order. In anticipation of some details of develop-
mental history of the eyeball, one may distinguish between elements of
ectodermal and of mesodermal origin in the retina.

The elements of ectodermal origin are partly framework elements,
partly nerve-cells.

The framework elements are partly of the same nature as the retina
proper (Mueller's supporting fibers and the two membranae limitantes),
partly of the same nature as in the optic nerve (neuroglia). The latter
is, however, found only in those layers which represent a direct expansion
of the optic nerve.


The nerve-cells of the retina are grouped in three superimposed planes.
Just how these planes are united opinions differ. A majority of the
authors hold firmly to the view of Ramon y Cajal that the cells subse-
quent to one another (in the sense of nerve conduction) (the neurons) are
not united but only lie in contact with one another, or are surrounded by
their branches. Another view, espoused by Apathy and Bethe, looks
upon the neurofibrilla as the essential structural element of the nervous
system and conceives of a continuity of the neurofibrillae.

It is certain that the individual layers possess a degree of independence
of one another, e.g., atrophic processes are often limited to one plane.
In any case the neuron theory of Ramon y Cajal contributes more to the
understanding of these conditions than does the neurofibrillar theory.

The retina contains three neurons: they are named in the sense of
their conduction, therefore, in this case in the centripetal direction.

The first neuron is represented by the rods and cones and the outer
nuclei belonging to them. Its elements are broadened out into a simple
surface layer; only the nuclear portions are superimposed in layers,
because of their greater volume. This neuron serves for the reception
of the individual light impressions and the mosaic arrangement of its ele-
ments makes possible a separation of the respective impressions in space
and the appreciation of the picture projected through the optical system.
This is the sensory epithelium (neuroepithelium) of the retina (Schwalbe).
In the matter of its nutrition the first neuron is entirely dependent upon
the choriocapillaris, because it contains no blood-vessels. It therefore
happens that a circumscribed atrophy of the choriocapillaris leads to a
coextensive atrophy of the pigment epithelium and of the first neuron of
the retina, so that the subjective functional defect (the scotoma) and
the objective ophthalmoscopic change (the atrophic area) exactly corre-
spond in this instance.

The second neuron is represented by the nerve-cells of the inner
nuclear layer. These cells come in contact with several elements of the
first neuron. However, there are cells here which serve exclusively for the
union of the elements with one another. Finally, centrifugal nerve-fibers
come into contact with individual elements of this neuron. It is to be
conceived that nervous processes, even of a high order, take place in
this layer.

The third neuron is formed by the ganglion cells; it is the longest of
all, for its axis cylinders reach through the optic nerve, the chiasm and
tractus opticus as far as the brain (outer geniculate body, optic thalamus
and anterior corpora quadrigemina, Bernheimer, 23).

The second and third neurons possess a vessel system of their own in


the branches of the arteria centralis retinae and the veins of the same name.
These are the only elements of mesodermal origin in the retina. The
largest branches lie superficially in the nerve-fiber layer, yet the ganglion-
cell layers are usually absent beneath them and even the inner nuclear
layer shows a thinning; on the other hand, the inner surface of the retina
is bulged slightly forward. The finer branches pass deeper into the nerve-
fiber layer and the most extensive branching is found in the ganglion-cell

According to His (105), the capillary net of the retina consists of
narrow capillaries (5 to 6 mu, seldom wider), and is, naturally, most
developed in the posterior segment of the retina. Fine branches (art.
afferantes) are here given off at right angles from place to place; these
first break up into capillaries some 0.13 to 0.25 mm from the main
vessels. In this way a space free from capillaries arises on each side of
the main vessel. This first arterial capillary net lies in the nerve-fiber
layer; from it there ascend branches to the inner nuclear layer and each
of these forms a venous capillary net on the outer and inner surface of this
layer. The venous radicals (venae efferentes) then form from these,
descend again to the nerve-fiber layer, and empty into larger veins at
right angles.

The capillaries go as far as the outer surface of the inner nuclear layer
and no farther. The border between the first and second neuron (PI.
IV, 3, x) is, therefore, also the border between the avascular portion of
the retina (the sensory epithelium) and the vascular portion (cerebral
layers of Schwalbe).

Toward the ora serrata the vessel net becomes more and more simple.
The peripheral limit of the retinal vessel system does not, however,
coincide with the border of the retina, but lies about i mm farther back;
a few wide irregular projected loops form the border of the retinal vessel
system. These loops have a somewhat wider caliber than the rest of
the capillaries, and, if one so wishes, one may speak of a direct transition
of arteries into veins here. I have not been able to make out an out-
spoken circular course of the terminal venous branches in my preparations.

The fovea centralis, of which the gross anatomic relations have been
described above, arises mainly from a spreading apart of the cerebral
layers. For, aside from isolated cells of the second neuron and the very
sparse nerve-fibers found on the floor of the fovea, this area contains only
elements of the first neuron. Along with the cerebral layers, the retinal
vessels also fail in the center of the fovea. A capillary-free area of 0.4
to o . 5 mm size and irregular form is present in most cases. Yet this
area varies appreciably, probably with the size of the fovea in general.


With the ophthalmoscope this vessel-free area appears larger, because a
magnification of only 14 times in the direct image does not permit the
capillaries to be seen; the finest ophthalmoscopically visible vessels
extend only a little way over the wall about the depression in the retina
indicated by the macular reflex.

Since the histologic relations of the fovea centralis have already been
described somewhat more accurately in connection with the individual
layers, it seems superfluous to give a special histologic description of the
fovea. Still, certain peculiarities of the retina in this region may well be
again briefly pointed out. A circle, tangent to the temporal border of
the disc, drawn about the center of the fovea contains in its compass
(the area centralis) two remarkable histologic findings: the thickening
of the ganglion-cell layer and the outer fiber layer of Henle. The area,
naturally, has no sharp limits; therefore, its size can only be approxi-
mately given. The wall about the fovea is formed by the ganglion-cell
layer. In the middle of the fovea is an area scarcely of o . 5 mm, char-
acterized by extreme attenuation of the cerebral layer and by absence of
the retinal vessels, as well as by the presence, exclusively, of cones. But
the slender foveal cones are found only in the very center of this area.

Here everything is arranged to increase functional capacity as much
as possible. The fovea lies in the optic axis (although not exactly so),
where the picture projected by the optical system is the sharpest. The
inner layers of the retina are spread apart, for they can only obscure
the image; the shadow-producing retinal vessels fail entirely. Only the
elements of higher dignity, the color-perceiving cones, are present, and
these, moreover, possess a fineness found in no other part of the retina.
Each cone is united with but a single bipolar cell and, furthermore, with
but one ganglion cell possibly, so each individual light sensation is con-
ducted isolated from the rest. The center of the fovea centralis thus
becomes the seat of the highest function of the eye that of central vision.

The yellow color of the macula lutea is conditioned by a coloring mat-
ter distributed through all the elements of the cerebral layer, although
through the outer fiber layer to a lesser degree (Dimmer, 40).

c) The Extreme Periphery of the Retina

When the retina is viewed from its surface, its (anterior) border (ora
serrata retinae) appears more or less toothed (in biologic terminology),
i.e., sharp projections are directed toward the corona ciliaris and sepa-
rated from one another by rounded bays. The length of the teeth is
subject to great variation; in their maximal development they may
reach to the corona ciliaris.


As a rule the teeth are not uniformly developed in the entire circum-
ference of the border of the retina, but are plainer on those sides where
the ciliary body is shorter, i.e., especially, therefore, on the nasal side
(PL II, i); on the temporal side they often fail completely, and the
border is then only finely and irregularly wavy or angular.

The teeth correspond in position to the intervals between the ciliary
processes and all the irregularities of development in the corona ciliaris
are reflected in the ora serrata. Depending upon their length, the form
of the teeth varies from that of a triangle to an awl ; the bays are rounded
out as a whole, but often show several smaller projections and in this way
acquire an undulating or slightly jagged outline.

That which appears to be the border of the retina in a surface prepara-
tion by no means coincides with the corresponding border of the pigment
epithelium, for the border of the retina usually lies farther forward
(cf. p. 62). One can get the best general view of the topographic rela-
tions of these two borders after the retina has been detached from the
uvea in a hardened eye; the pigment epithelium of the chorioidea
(posterior zone of the pigment epithelium) remains with this, but the pig-
ment epithelium of the ciliary body remains attached to the retina, at
least far enough so that one can recognize its limits. The finer details
vary greatly. Schoen (193) differentiates four types, yet it is not always
an individual difference alone, for the relations in different parts of the
same eye are not always the same.

In many eyes one finds peculiar cavities in the tissues (Blessig's
cysts, Iwanoff's retinal oedema, cystoid degeneration) in the most periph-
eral portions of the retina. On surface view (PI. V, 5) they appear as
rounded pores, or, through confluence of adjoining cavities, as lobulated
or meandering or dendritically branched clear flecks, often separated by a
narrow partition only.

The first traces appear very early in life, between perhaps the years of
1 6 and 20, and at first immediately behind the teeth of the ora serrata.
From here the cavities broaden out backward and toward the sides, and
the originally isolated flecks flow into a closed zone, which always becomes
broader with the years. But the cavities increase not only in extent but
also in size ; the individual cavities enlarge and so merge with one another
in age that only column-like pieces of the separating walls remain.

Cystoid degeneration is, therefore, probably to be considered as
physiologic despite the great individual variations in grade and extent
of the change; it increases with age, somewhat as the far point always
moves away with the years. The cavities are not, however, peculiarly


senile appearances, for one may encounter them in beautiful formation
even between the ages of 30 and 40.

The thickness of the retina is markedly decreased by the formation
of cavities. In such eyes the minimum thickness is not just at the border
of the retina but behind the zone of the cavities (PL V, 6).

Great difficulties lie in the way of one who wishes to produce a picture of the
histologic structure of the extreme periphery of the retina, for light pathologic changes
can occur in this region which entirely escape the control of clinical observation.

The extreme periphery of the retina is not accessible to ordinary ophthalmoscopic
study. This region is only visible to the ophthalmoscope under especially favorable
circumstances, such as coloboma of the iris with aphakia (Reimar, 178), or in tumors
which press the ora serrata toward the optic axis. Traptas (225) has given a method
(pressing in the wall of the bulb with the finger), but we still lack a comprehensive
investigation of the ophthalmoscopic appearance of this region, and especially a control
of these findings through anatomy.

In addition, it happens that the most peripheral portions of the retina are blind
(Bonders, 45). The visual field, as well known, has an extent of only 60 at the most
on the nasal side; on the temporal side it has one of 90 or even more. If one construct
this angle in the schematically cross-sectioned eye (Text Fig. i) with one side along the
visual axis and the apex at the posterior nodal point (which falls at about the posterior
pole of the lens), one notes that a 4 mm broad zone of the retina, at least on the
temporal side, is devoid of a visual function. When a pathologic condition develops
in this zone it is, therefore, neither appreciable objectively (by the ophthalmoscope) nor
subjectively (by testing the function).

Yet the appearance of the outermost periphery of the retina is altered by age, alone.
It is, therefore, fundamentally incorrect to base the description upon the relations in
children's eyes only, for then all later changes are classified as pathologic, which most
certainly is not correct.

The fact that the outermost periphery of the retina is blind should
cause a high degree of astonishment, because the histologic structure of
the retina does not justify this observation. The organs for the reception
of light-stimuli (the rods and cones) reach, indeed, to the border of the
retina; the elements simply spread apart somewhat, becoming thicker and
shorter; therewith the number of the rods decreases much more than the
cones (Greeff, 75). With the cessation of the rod-and-cone layer the
limitans externa bends down toward the pigment epithelium and, accord-
ing to Wolfrum (242), goes over into the cement ridges lying between
the pigment epithelium and the non-pigmented ciliary epithelium. The
anterior border of the limitans externa coincide exactly with the border
in the pigment epithelium.

The two nuclear layers become correspondingly thinner toward the
border of the retina, but otherwise show no striking changes. At the
same time nuclei (outwardly displaced inner nuclei?) appear in the outer


plexiform layer; the border line between the two nuclear layers thereby
becomes indistinct, and finally they fuse into one at the border itself.

No changes can be made out in the inner nuclear layer. The ganglion-
cell layer, together with the nerve-fiber layer, cease 0.5 to i mm behind
the ora serrata. To the same extent in which these elements disappear,
the supporting tissue increases in mass; the outermost periphery of the
retina, therefore, contains only scattered neuroglia cells in place of
ganglion cells, only closely pressed basal cones of Mueller's fibers in place
of nerve-fibers, and, therefore, looks cross- or diagonally striated. The
limitans interna retinae often becomes notably thinner, and, so, often
indistinct toward the ora serrata (cf. chap, ix, 8).

At the border itself, the retina of the adult (and the description relates
only to such) is sharply set off against the ciliary epithelium (PL V, 6).
Since now this is much thinner than the retina, there arises a step at
the border of the retina, sometimes rounded, sometimes sharply angular,
sometimes falling abruptly at right angles, sometimes overhanging the
ciliary epithelium. Not only are individual variations found here, but
the form of the border of the retina changes in different sections from the
same eye. For example, the more the section approaches the apex of a
tooth of the ora serrata, the more the border of the retina overhangs. In
such sections one sees a sort of spur made up of a loose reticulum with a
few irregularly arranged nuclei projecting into the vitreous from the inner
surface of the retina. When the section goes exactly through the front
of a tooth, this spur lies upon the inner surface of the ciliary epithelium.
Usually there rules a relationship similar to that of the pterygium corneae,
which also is grown fast along the middle line while the marginal portions
are undermined. This comparison should, however, only serve for the
visualization of the relations; I am far from the intention of conceiving
of the origin of the ora serrata as similar to that of the pterygium. One
now and then sees, rather, a little group of elongated, fibrillated
ciliary epithelium cells at the border of the retina, ' after the manner
of a buttress.

The first traces of cystoid degeneration are shown in a rarification of
the interior of the retinal tissue close to the border and its transformation
into a widespread reticulum. The formation of actual smooth-walled
cavities, mostly rounded in form, takes place later. But numerous small
isolated cystic spaces are always found at the posterior border of
the degenerated zone in the far-advanced cystic degeneration of older

Such young cavities, so to say, lie in the outer plexiform layer, and
soon extend from there into the outer nuclear layer, often more into the


inner nuclear layers; they are rounded or oval, sharply bordered, mostly
empty, and more rarely divided by remnants of the outer plexiform layer
into planes. The immediate neighborhood of the cavities shows no strik-
ing changes.

The larger cavities (PL V, 7) reach so close to the m. limitans externa
that only a layer of outer nuclei remains; indeed, even this may dis-
appear. The larger spaces extend inward as far as the inner fiber layer,
yet a covering made up of remnants of the nerve-fiber layer and the m.
limitans retina almost always remains. The tissue intervening between
the spaces, which more and more back up against each other, thickens
itself into partitioning walls made up of tensely spanned protoplasmic
fibers provided with longish nuclei between which lie the spaced-apart
outer nuclei. This is associated with a very considerable increase in the
thickness of the retina to almost double the original thickness.

The highest grade of cavity formation consists in a further reduction of
the partitioning walls to individual columns and extensive confluence
of the cavities. Further histologic changes do not appear. This highest
grade is better recognized in surface preparations of the entire retina
than in individual sections. The cavities, however, never break through
the limits set for them by the two m. limitantes; even in their most
extensive development they remain intraretinal.

I should consider the following to be pathologic changes: complete
absence of rods and cones in the joutermost periphery and the fixation of
the limitans externa to the pigment epithelium, as well as the occasional
occurrence of high-grade thinning of the peripheral portions of the retina,
with loss of the regular stratification without cavity formation.

The peculiar morphologic relations of the ora serrata have been first thoroughly
studied in the last decades and various theories concerning their origin have been put

According to Schoen (193), in the newborn the border of the retina shows a straight
course and only microscopic little juttings, some 850 in number, are present. He holds
this condition to be the normal one; the larger microscopically visible juts first appear
later; they are, therefore, acquired and are, for Schoen, equivalent to a pathologic
condition. The cause of this diseased condition is the pull of the zonular fibers attached
to the juts, and this again arises from overstraining of the accommodation.

It is not difficult to disprove this theory, because the premises are not warranted.
E. von Hippel (102) found larger teeth at the ora serrata in the newborn. The zonular
fibers are not, in general, inserted directly into the border of the retina, and even if this
were so, the pull of the zonula could not bring about the jutting form but only a uni-
form displacement of the border of the retina, because zonular fibers are also richly
present in the situations corresponding to the bays of the ora serrata.

O. Schultze (198), on the other hand, proceeds from the development of the
retina and the ciliary epithelium. In the fourth month of pregnancy the border


between the two lies just behind the ciliary processes and extends into the ciliary

1 2 3 4 5 6 7 8 10 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

Online LibraryMaximilian SalzmannThe anatomy and histology of the human eyeball in the normal state, its development and senescence ; → online text (page 10 of 27)