Maximilian Salzmann.

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

. (page 5 of 27)
Online LibraryMaximilian SalzmannThe anatomy and histology of the human eyeball in the normal state, its development and senescence ; → online text (page 5 of 27)
Font size
QR-code for this ebook

cross-section. In its staining reactions it shows a distinct difference from
the corneal stroma. Even with eosin a slight difference is made out, and
this is still plainer by Van Gieson's stain : the stroma appears deep fuchsin-
red, the Descemet's membrane rose-red to orange-yellow, according to the


degree of differentiation. Most striking, however, is the difference after
staining for elastic fibers with orcein or resorcin-acid-fuchsin, for then the
stroma is without stain, and Descemet's membrane dense brown or vio-
let; yet the stain is by no means as intense as it is in the case of the
elastic fibers or in the intima of arteries.

In any case, there is a great difference between the Descemet's and
Bowman's membrane, and the name lamina elastica posterior is, therefore,
much better adapted to Descemet's membrane than is lamina elastica
anterior to Bowman's.

Pathology furnishes a further difference between the two membranes. Not the
slightest trace of regeneration is found in lesions of Bowman's membrane, while lesions
of Descemet's membrane are covered over after some time by a new glass membrane.

Descemet's membrane is 5-7 mu thick in the mid-portions; toward
the border the thickness increases to 8-10 mu. In the immediate
neighborhood of the border wart-like elevations appear along the inner
surface, as a rule (PL IV, i, w). These are almost semispherical or
semiellipsoidal, sharply demarkated projections between which the inner
surface is again smooth. The breadth of this wart zone (PL III, 2,
w-d) increases with age, as does the thickness of the whole membrane.
The border proper of Descemet's membrane and its relation to the scleral
meshwork will be spoken of in connection with the latter.

(PL II, 9, En)

This is a layer of single cells measuring some 5 mu in height and 18-
20 mu in width. Unlike other endothelia, these cells are rich in proto-
plasm and the nucleus is completely imbedded in it. Upon cross-section,
the cells, therefore, appear almost rectangular, with a slightly oval,
non-prominent nucleus; the protoplasm is often grossly granular
or shows normal granular vacuoles; the outlines of the individual cells
are sometimes oblique or curved.

In surface view the nuclei appear round (7 mu) and are quite regularly
arranged, so that most of the cells retain a rounded six-sided form (PL
III, 2, D}. The cell-outlines are not always entirely clear, because of the
oblique course already spoken of, so that by a change in the focus a change
in the form of the cell is produced. But if one compare these cells with
those of other endothelia, whose cell-outlines, as is well known, can only be
demonstrated by special methods of preparation, the cells of the corneal
endothelium must be said to be sharply defined.

From a purely morphologic standpoint, therefore, this layer is more like an
epithelium than an endothelium; yet on the basis of its developmental history one


must regard it as endothelium (cf. chap, xvi) and, furthermore, it is continuous with
endothelium of the usual appearance.

The endothelium extends over nearly the entire inner surface of
Descemet's membrane. Only in the region of the wart zone does it take
on another appearance, especially noticeable on surface view (PL III,
2); the cell-outlines disappear completely, the granulation of the proto-
plasm likewise, the nucleus becomes somewhat larger, oval, and goes back
into the interspaces between the warts, so that an irregular distribution
of the nuclei results. The height of the cells also decreases, especially
over the warts (PI. IV, i), where only a very thin scarcely distinguishable
protoplasmic covering is present.

b) The Limbus corneae

This, as already indicated, forms a transition zone between the cornea
proper on the one side and the adjacent tissues (conjunctiva and sclera)
on the other side. In the limbus there are combined certain characteris-
tics of the neighboring structures, e.g., the transparency which at least
a greater part of the limbus has in common with the cornea, the vasculari-
zation and the richness in special nerve-endings which characterize the
conjunctiva. The limbus, moreover, shows transitions, for in its territory
the stroma of the cornea goes over into that of the conjunctiva, of the
episcleral tissue, and of the sclera.

The borders of the limbus have already been given, but since the
corneoscleral border is so difficult to recognize in stained specimens, the
following is given as an aid to orientation for the beginner (PL I). The
borders of Bowman's and Descemet's membranes can be found even with
moderately low power; if one think of the ends of these two membranes
as united by a line, this line is almost parallel to the outer part of the
corneoscleral border and lies 0.75 to i mm from it. The difference
which the corneoscleral border shows in the horizontal and vertical
meridian also comes out in the line of union between the borders of
the two basal membranes.

Accordingly, one can differentiate only two layers in the limbus:
the epithelium and the stroma.


Even over the marginal portions of Bowman's membrane the epi-
thelium is somewhat thicker than it is in the middle; over the limbus this
thickening increases and in the neighborhood of the corneoscleral border
attains its maximum (80 to 90 mu). The back border of the epithelium
(toward the stroma) loses its straight course and becomes wavy; indeed,



papillary forms appear at times. The thickening of the epithelium is
due to an increase in the number of layers; their number mounts up to
10 or more.

As in the cornea proper, the epithelium is stratified and squamous;
only the appearance of the basal cells is altered in the region of the limbus.
While the basal cells of the cornea proper are larger and rich in proto-
plasm and have nuclei which in size and staining are no different from
those of the next layer, the basal cells at the very limbus are at once
small, scant in protoplasm, and have small dense-staining nuclei. The
epithelium retains this kind of basal cells in its further course over the
conjunctiva and they are characteristic for the conjunctival epithelium.
The rest of the cells keep the peculiarities of the corneal epithelial cells,
and the low cylindrical cells which appear on the surface and are char-
acteristic for the conjunctiva, first appear beyond the corneoscleral border.
One sees, therefore, that the transition of corneal epithelium into con-
junctival epithelium does not take place throughout the entire thickness
of the layer all at once, but along the base before it does on the surface.

Since, too, the basal cells are more closely pressed together, the
epithelium of the limbus and of the conjunctiva shows a dark seam along
its base, even by low power (PL I). This seam is often found even at
the border of Bowman's membrane, but often first appears a consider-
able distance from it. It is not rare to find small islands of basal cells
strewn along and first united iato an unbroken line in the conjunctiva


The stroma of the outer layers loses its regularity even opposite
the marginal portions of Bowman's membrane. Beyond the limits of
this membrane it loses the regular arrangement of its spaces (parallel to
the surface) and takes on the appearance of an ordinary connective tissue
first in the outer layers only, then in the deeper ones as well. It seems
to be especially the appearance of blood-vessels which brings about this
alteration in appearance.

Occasionally one sees fine capillaries just beneath the surface immedi-
ately outside the margin of Bowman's membrane. Farther away the
lumina are more numerous, are encountered also in the deeper layers, and
are intermixed with larger vessels as well. This brings us into the territory
of the superficial marginal vessel-loops, the only vessel-net of the cornea,
its sole organ of nutrition. The area occupied by the marginal vessel-
loops shows, on meridional section, the form of a triangle with its apex
at the border of Bowman's membrane and its base continued into the
conjunctiva, sclera, and the episcleral tissue.


The marginal loop net is supplied by the terminal branches of the
anterior ciliary arteries. Leber has made a beautiful drawing of it (138) ;
according to him, when viewed from the front it is seen to be made up
of very fine, small straight arterial branches, which build bow-formed
anastomoses. From these arches the final branches are given off; these
are very fine (5 to 6 mu) ; they course on farther in a meridional direction
and then suddenly coil about into the venules. The venous trunks of
the loops are, for the most part, at least twice as wide as the arterial
ones and unite with their neighbors to form a finely subdivided net,
emptying into the episcleral venous net.

On the other hand, the layers lying deeper, on a level with the sclera,
are vesselless in the region of the limbus, or at the most provided
with a few vessel-loops scarcely going beyond the border of the sclera.
These layers go over into the sclera in such a way that at first the lamellae
preserve their transparency, but gradually they take on the form and
staining reactions of scleral bundles.

Elastic fibers are also found in the limbus but they are very sparse in
comparison with those in the adjacent tissues. 1 They are most numerous
and extend farthest centralward in the superficial layers of the limbus,
those layers continued into the conjunctiva and the episcleral tissue;
they only come out here and there in the deeper layers. Their frequent
appearance is a sure sign of scleral tissue.

I cannot refrain from once more remarking that in many particulars the micro-
scopic specimen does not appear to agree with the findings in life. The beginner most
readily falls into the error of considering that the sclera reaches farther central than
is actually the case, for he forms his judgment upon the staining of the tissues, and the
darker staining of the sclera as a matter of fact does extend somewhat beyond the
border into the limbus.

Another confusion which easily occurs is the following: the conjunctiva and
the episcleral tissue seem to be sharpened and to end in the limbus, and the sclera
to thicken the cornea to just this extent. In reality the course of the connective-tissue
bundles, parallel to the surface, is not essentially disturbed, as one can see best in those
pathologic cases in which inflammatory infiltration of the conjunctiva and the episcleral
tissues extends over onto the cornea.

It is, therefore, entirely justifiable to speak of a conjunctiva corneae, i.e., the super-
ficial layers of the cornea may be considered as a continuation of the conjunctiva sclerae.
However, this expression must be extended to include not alone the epithelium and
Bowman's membrane but the most superficial lamellae of the stroma, as well.

The nerve-endings found in the limbus have already been considered
with the nerves of the corneal stroma.

1 Although one speaks of elastic fibers, for short, only those which are positively stained by orcein
or resorcin-fuchsin are meant.



(PL III, i)

Close to the bottom of the scleral furrow (cf. p. 20), i.e., separated
from the scleral tissue by only a thin layer of tissue, one or more lumina
are seen; these have a closed endothelial covering and are usually con-
siderably larger and more prominent than the spaces of the meshwork of
the iris angle which lie farther inward. These lumina form the Schlemm's
canal, sinus venosus sclerae (Merkel), circulus venosus ciliaris (Leber) (Sch}.

The expression canal is incorrect for this structure in that it relatively
seldom is a single elongated lumen, and then for a stretch of only 0.2 to
o . 5 mm in a horizontal direction ; usually there are two or more lumina ;
they lie side by side or over one another in the scleral furrow. Seen from
the surface, Schlemm's canal is like the bed of a great stream which flows
along undivided for a stretch, then is divided into several branches for
a stretch. For this reason the form of the lumen changes in various
sections of the same eye.

The endothelial lining of Schlemm's canal has the same appearance
as in other vessels and forms an extremely thin membrane with nuclei
projecting inward. Aside from the endothelium, Schlemm's canal has
no real wall, at least none such as one finds in other vessels of the same
size; it seems to be simply entrenched in the adjoining tissue. On the
other hand, it is not correct to say that the endothelium lies immediately
upon the sclera, for a loose tissue, poor in fibers but rich in cells, is inter-
posed between the two, as a rule; sometimes this is only a thin layer,
sometimes it is quite well developed. One sees this layer best in sections
stained with Van Gieson; the tissue is then sharply set off from the deep
red of the sclera by its yellow color.

One encounters a similar tissue on the chamber side of the canal,
usually, only there it is less developed or is not present as a continuous
layer and is poorly differentiated from the neighboring trabecular mesh-
work. On the other side it extends along the veins (v) going off from
Schlemm's canal into the sclera.

Schlemm's canal is in any case closed off from the spaces in the
meshwork lying inward to it, i.e., no visible breaks in its wall are
present. For this reason only solutions, or the finest suspensions, such
as ink, pass into the canal; cells remain in the meshwork of the iris angle
and lie outside the wall of Schlemm's canal.

On the other side, however, the canal communicates freely with the
venous system by means of vessels given off here and there along the


scleral side of the canal; these go obliquely outward and backward into
the sclera and unite with the anterior ciliary veins (V) while still within
the sclera. An actual capillary net for the supply of the canal does not
exist, and it must be that some of the deeper limbus capillaries give off
blood to it; in any case it seems to connect laterally with the ciliary
venous system as a whole, if one may judge from the direction of its blood
stream (Leber, 1138).

In prepared specimens the lumen of the canal is usually empty or
contains only a few red blood cells. Complete filling of the whole canal
with blood only comes about in a stasis of the venous system, e.g., in
persons who are hanged.

In the question as to whether or not the canal contains blood or aqueous
during life, I am disposed to the view that it contains aqueous. It is
certainly demonstrated that its main role is to carry away the aqueous,
and its position to one side of the actual course of the blood-stream
warrants the supposition that it carries aqueous. On the other hand, a
slight circulation disturbance during life or hypostasis in the cadaver is
probably quite sufficient to fill it partially or entirely with blood owing to
its open communication with the veins.

Possibly it is not superfluous to call attention to the fact that a hypostasis, i.e., a
post-mortal sinking of the blood owing to the force of gravity, can play a certain role in
the fairly well closed vessel system of the eye, small as it is; this is true not only of
cadaver-eyes but also of eyes enucleated in life which have been placed in a slow-
working conservation fluid. The hypostasis is shown by the fact that the veins, or
even the capillaries, on one side of the bulb are completely filled with blood, on the
opposite side, empty. The matter of the side upon which the filling is found naturally
depends upon the position of the eyeball while the sinking is going on.

Schlemm's canal occupies only about the posterior half of the scleral
furrow, and not even all of this, since the scleral roll (Sw), springing axial-
ward from its back border, overhangs the furrow considerably. The
meshwork of the iris angle (H. Virchow) fills out the rest of the depression.

Considered in and of itself, this peculiar structure presents itself as a
three-sided prismatic band, of which the anterior edge is extremely sharp
and unites with Descemet's membrane (at D) and the most posterior
lamellae of the cornea (at T). Behind, it unites with the scleral roll, the
anterior surface of the ciliary body and, by a devious way, with the root
of the iris (i), as well. Its outer surface borders directly upon the corneal
and scleral tissue in front and upon the inner wall of Schlemm's canal,
or the loose tissue surrounding it, farther backward; its inner surface is
free and turned toward the chamber space.

Its union with neighboring structures can best be demonstrated by
pure anatomic preparations. If one detaches the ciliary body from its


insertion to the sclera from behind, one obtains quite a complete prepara-
tion of the meshwork. A seam, consisting of the meshwork and the
marginal portions of Descemet's membrane, of whitish color and more
than i mm in width, then remains clinging to the anterior end of the
ciliary body.

If one now separate the iris from the ciliary body, the innermost
meshwork, united with the iris root as well, stays attached to it, while
the main mass of the meshwork along with the border of Descemet's
membrane remains hanging to the ciliary body. In this way one sepa-
rates two portions, which, at least in the main, are different from one
another in their gross structure as well as in their histologic composition.
Seefelder and Wolfrum (205) have retained the old name ligamentum
pectinatum for the part remaining in union with the iris, and support
Rochon-Duvigneaud (182) in calling the rest the trabeculum sclero-
corneale. H. Virchow (234) has recently introduced the term uveal for
the former and scleral for the latter meshwork.

These two portions are of very unequal bulk; the uveal meshwork is
a delicate, thin structure which one can barely make out with the dis-
secting loupe. By far the greater part of the whole structure belongs
to the scleral meshwork.

When one studies a meridional section, the uveal meshwork (i)
appears to be made up of only a few obliquely or longitudinally cut sec-
tions of delicate trabeculae lying wholly superficial along the inner sur-
face; these do not give the idea of a special formation at all and are
entirely overlooked by the beginner. One may about as well say that
only scleral meshwork is visible in meridional section, for it makes up
so much of the whole mass.

The study of surface and teased preparation as well as of cut sections
is unconditionally necessary for a proper conception of the make-up
of the entire meshwork. The method of obtaining these was given above.
I begin with the description of the scleral meshwork and first its anterior

In the first place this border is united with the border of Descemet's
membrane. As already noted (pp. 37-38), the marginal portions of this
membrane are characterized by warts and changes in the endothelium.
The actual border (D) is, apparently, sharp and plain; the membrane
often maintains the same thickness to the border; in other cases it becomes
rapidly sharpened off near the border.

As a matter of fact the glass membrane does not stop here but con-
tinues over the trabeculae of the iris angle as a very thin layer. The
appearance of an ending is only brought about by the fact that the glass


membrane thins out so suddenly, and especially by the fact that as one
follows along Descemet's membrane toward the root of the iris one
necessarily at length comes upon a hole in the membrane in the situation
where naturally the membrane stops in the given section.

At the end of Descemet's membrane lies the anterior border ring
(Schwalbe). It sets immediately upon the outer surface of this membrane
or appears to be imbedded in its substance (PI. IV, i, vG). It is a
flat bundle of circularly fibrillated connective tissue supported by elastic
fibers which stain well with orcein. It varies in its position, thickness,
and breadth very much in different eyes as well as in different portions
of the same eye, yet one seldom misses it completely in meridional sections.

In surface preparations (PL III, 2, vG) it stands out owing to its
fairly compact and circular fibrillation, i.e., fibrillation parallel to the
margin of the cornea; it represents the most anteriorly placed (corneal)
area in which one finds circularly fibrillated connective tissue supported
by elastic tissue. The most anterior spaces opening into the anterior
chamber are encountered first posterior to it. When the meridional
section goes through one of these foremost spaces (as in PL IV, i, at d),
one sees how the Descemet's membrane, which is still thick at the anterior
surface of the border ring, bends about the margin of the space and con-
tinues over the outer surface of the border ring as a thin glass membrane ;
farther forward it again merges into Descemet's membrane or goes over
into the deeper portions of the meshwork.

The anterior border ring can to a certain extent be looked upon as the
superficial or circular root (with respect to the anterior chamber) of the
meshwork. In a histologic sense it is in no respect different from the
meshes lying immediately posterior to it. Since its fibers leave their
meridional course and spread apart from one another in bundles, the
thick, compact anterior border ring passes backward over into a mesh
of flat, compressed, thin bands (PL III, 2, right part of the drawing).
Glass membrane and endothelium, somewhat modified, continue over the
trabeculae and round out the spaces of the meshwork.

The most posterior lamellae of the cornea also go over into the mesh-
work, and this transition is completed even o . i to o . 2 mm in front of
or axial to the end of Descemet's membrane. In meridional sections one
sees at this point (PL IV, i, T) a group of longish nuclei between the
2 or 3 most posterior lamellae, i.e., those immediately bordering Desce-
met's membrane. These nuclei belong to the endothelium. With the
proper stain one also recognizes a lighter homogeneous seam (glass mem-
brane) interposed between the endothelial nuclei and the connective
tissue; the latter has a meridional fiber-direction, and yet does not show


any elastic tissue fibers by orcein staining. These first make their
appearance farther back, where the division and threading of the bundle
is richer, and the direction of the fibers more nearly circular. At last the
trabeculae take on wholly the same appearance as those coming out of
the anterior border ring. The border of Descemet's membrane and the
anterior border ring are, therefore, to a certain extent undermined.

Still clearer is the transition of the corneal tissue into the meshwork in
teased preparations (PL III, 2, left side of the drawing). The fiber mass
of the particular corneal lamella divides up into .narrower bundles, which,
however, are always very flat, and bend about into a meridional direc-
tion (T). These bundles contain a glass membrane and an endothelial
covering and become ordinary trabeculae (TV) in the course of further
subdivision and reticulated union. One may, therefore, look upon these
as constituting the deep or meridional root of the meshwork.

The trabeculum sclerocorneale, or the scleral meshwork, thereupon
comes out of both roots. Its trabeculae are pressed down flat (per-
pendicular to the surface of the bulb) and form a mesh in which a coursing
of the fibers parallel and circular to the surface predominates. One may
then just as well speak of a structure of fenestrated lamellae formed by
the union of obliquely coursing bundles. One obtains this impression in
meridional sections especially. The holes (windows) in these lamellae
are drawn out in a circular direction in the middle layers and are of varying
size and distance apart, so that the trabeculae lying between them are
sometimes broader, sometimes narrower. The spaces between super-
imposed lamellae do not lie over one another.

The bulk of the meshwork (or the number of the trabeculae) increases
from before backward, partly because new trabeculae radiate into the
meshwork from the floor of the scleral furrow, although this only holds

Online LibraryMaximilian SalzmannThe anatomy and histology of the human eyeball in the normal state, its development and senescence ; → online text (page 5 of 27)