EXPERIMENTS ON THE FUNCTION OF THE EYE IN LIGHT MICROSCOPY*

• Document Type: CREST
• Collection: NGA Records (Formerly NIMA)
• Document Number (FOIA) /ESDN (CREST): CIA-RDP78B05171A000600070045-8
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 14
• Document Creation Date: December 28, 2016
• Document Release Date: April 7, 2003
• Sequence Number: 45
• Publication Date: January 7, 1966
• Content Type: OPEN

Approved4ror Release 2003/05/14: CIA-RDP78B %71A000600070045-8
0 Journal of the Royal Microscopical Society, Vol. 85, Pt. 3,,June 1966. Pp. 231-254.
Received 7,January, 1966.
Apoved For Release 2003/05&4cFa~!Q9 j7
Journal of
The. Royal Microscopical Society
Experiments on the function of the eye in light
microscopy*
by JOHN R. BAKER, Cytological Laboratory, Department of Zoology, Parks Road,
Oxford
n
diagrams showing the eye as well as the microscope, the rays from a point in the
primary image produced by the objective should diverge between the eye-lens and
h
SYNOPSIS
Since the rays of light can be traced from the lenses of the microscope to the retina,
and since discoveries by light microscopy are nearly always made directly by the eye,
this organ may be regarded as an integral part .of the instrument. Nevertheless, it
would not appear that experiments have previously been performed to establish
certain important facts about the function of the eye in microscopy.
The experiments described in this paper have given the following results.
In microscopy the eye is generally focused for closer-than-distant 'vision I
L e eye.
The author questions the validity of the method commonly used to determine
visual acuity. His experiments suggest that in conditions simulating those of micro-
scopy, visual acuity is only about I (i.e. lights subtending an angle of 2' at the eye
can usually be seen separately, but those subtending 1' usually cannot). It follows
that an eyepiece magnifying at least 8 times is required to render visible all the
details in the primary image produced by a first-rate oil-immersion objective.
Visual acuity does not rise progressively, as is usually supposed, with increase in
the luminance (brightness) of the object viewed. It is highest with quite moderate
The mean diameter of the pupil of experienced microscopists
while actuall
i
,
y us
ng
the microscope, is about 2.8 mm.
* The substance of this paper was delivered as the Presidential Address on 5th January, 1966.
 light ? ~, ^'1 U wu ILL nlul.U !L iiuy UP- Sd1U UiaL inc eye 15 an uiLe'grdl pdfl UI me' Pala= LUUISC 15 1Cp11CSCuLCU. iviuuiWg `17UJ) SayS LLldl ULLUUSLUpLS .LIL WLU1Ut,u
ghmicroscope. First, the rays of light may be traced all the way from the object to for the normal eye focused at infinity, and he gives a ray-diagram in accordance
the retina. Secondly, discoveries are nearly of { ~g d not a ear, however, that anyone has ever published a
p
except when cinematography is used. As a g A e t e p' fiotomicroggrapb 105114 : CIATRDP78B05 `rn on oanpe t designed to discover what iactually the focus of the
only used by the discoverer to exhibit his findings to others. In both these respects eye in microscopy.
h
t
ere is a strong contrast with electron microscopy, in which a new set of rays
originates in the phosphor particles in the screen, and discoveries are nearly always
made by the study of micrographs.
Although such a wealth of knowledge exists about the optics of microscopy and
about ophthalmology, little has been done to bring the two subjects together by
investigations of the function of the eye while it is engaged in microscopy, or is
acting in circumstances simulating those of microscopical vision. Problems of
considerable interest have hitherto remained unsolved, and indeed it would seem
that no one has ever previously performed any experiments in an attempt to solve
them. Experiments on four such problems will be described in the present paper.
These are (1) the focus of the eye, (2) visual acuity, (3) the relation between the
luminance (brightness) of the object and visual acuity, and (4) the diameter of the
Some of the best-known studies of visual acuity have been carried out on very
small numbers of persons. For instance, K6nig (1897), Wilcox (1932), and Shlaer
(1936) each investigated the eyes of only two persons; Berger (1942) worked with
three. Hecht (1928) based his calculations on Konig's data. In the present study
100 were subjected to the principal experiments, namely those on focus, acuity, and
the effect of luminance on acuity. A smaller group (15 persons) was used for the
study of the size of the pupil.
In the present paper the persons who were subjected to the tests will be called
the "subjects". Sixty-two of them were male, 38 female. For the sake of brevity,
all of them will be referred to by the use of the words "he ", "him ", and "his ".
They were a selected group in the sense that they were all (with the exception
mentioned on p..251) persons who did not wear spectacles or contact lenses,
whether for distant or near vision or to correct astigmatism, though a few had done
so in the distant past. The ages of the subjects varied from 9 to 50 years. When
their ages were arranged in 5-year groups, 5-9, 10-14, 15-19, etc., the mode fell in
the 20-24 year group. For certain purposes the subjects were classified as non-
microscopists (24 persons), occasional microscopists (54), and experienced micro-
scopists (22).
In a first attempt of this kind it seemed right to simplify the experiments so far
as possible, and for this reason only monocular vision has been studied.
THE FOCUS OF THE EYE
Introduction
In text-books of optics and microscopy, diagrams are often shown in which the
virtual image of the object is seen to be situated at or near the minimum distance for
distinct vision (fig. 1). The rays of light from a point in the object, having converged
from the objective to a point in the plane of the eyepiece diaphragm, must therefore
diverge between the eye-lens and the eye, and the latter must be accommodated for
close vision. In diagrams in other books, however, the rays from a point in the plane
of the eyepiece diaphragm pursue a parallel course between the eye-lens and the eye,
and it follows that the ciliary muscle must be relaxed to give a focus for distant
vision. In Michel's (1964) work on microscopical optics there are 7 ray-diagrams
232
observer's eye
mirror ; rays from centre
`aY5 from lower
-point of source
Fig. 1. Diagram illustrating the opinion that the eye of
the microscopist is focused for close vision. (From
Marshall and Griffith (1928). Reproduced by per-
mission of Messrs. George Routledge and Sons Ltd.
New lettering has been provided.)
Method
In the method here adopted to find the degree of accommodation of the eye in
microscopy, the subject was required to focus a microscopical preparation by
altering the distance between the objective and the eyepiece, as though the instru-
ment were a telescope.
The principle of the method adopted is indicated in fig. 2. The object throws an
image at the plane indicated by the broken line at the bottom of the diagram. The
eyepiece, here represented as a single lens, is placed with its lower focus at the plane
of the image thrown by the objective (fig. 2A). It follows that parallel rays pass from
a point in the image in this plane to the eye. If the eye is a normal one, with the
`ciliary muscle relaxed for distant vision, a sharp image will be formed on the retina.
If the focus of the objective remains unchanged and the eye is accommodated for
nearer vision, it will be necessary to lower the eyepiece, so that the rays from a
Approved For Release 2003/05/14: CIA-RDP78B05171A000600070045-8
 1 a? ^V IUY iLwu. IUL ulstauce y Ttpr cents the necessary, the eyepiece diaphragm would therefore be unlikely to attract the attention;-and
downward movement of the eyepiece (fig. 2B). hence to affect the eye-focus, of the subject. A Ramsden x 10 eyepiece was used
It is convenient to Ieaard an eve nrrnmm,A7, --1 F -
-- - __....+.+a..vuaL~u iui wsLanr vrsron, with a converging lens in front The lower focal plane of the eyepiece was accurately determined. It was found to
of it. The converging lens converts the rays diverging from the eyepiece into lie 65 mm (to the nearest 0 1 mm) below the flat lower surface of the lower lens
of
ll
l
para
e
rays (fig. A-4. The degree of accommodation of the eye may then be
represented by stating the power of the converging lens (e.g. in dioptres)?
It is now necessary to describe the method adopted in greater detail.
It was decided to use a Ramsden eyepiece, partly because the experiment would
be simpler if the image produced by the objective was outside the lens system of the
the eyepiece (fig. 3A).
A scratch was made with a diamond in a circular glass. plate, and arrangements
made whereby this could be readily slipped into the eyepiece in such a position
that the scratch on its upper surface lay 6.5 mm below the lower lens (fig. 3A), and
readily slipped out again.
A Watson "Van Heurck" microscope was used in the experiment, because it
has a rack-and-pinion by which the distance between the objective and the eyepiece
can be accurately controlled without danger of changing the focus of the objective.
A Watson "Holoscopic" 12 mm objective, of high N.A. (0.45) for its power, was
used. The condenser iris was set to fill the aperture of the objective with light. The
objective was intended for a mechanical tube-length of 200 mm, and the draw-tube
was set in exactly the correct position to give this tube-length. The plane of the
flat lower surface of the lower lens of the eyepiece, when the tube-length was 200 :..
B
C
Fig. 2. Diagram illustrating the principle underlying the author's method for deter-
mining the focus of the eye in microscopy. The eyepiece is represented by a single
lens (e). The dotted line at the bottom of the diagram represents the plane in which the
image is produced by the objective. In A the eyepiece is placed in the position it
would necessarily assume if the microscopist's eye were focused for distant vision. In
B the microscopist is supposed to have focused his eye for close vision, and has
therefore brought the eyepiece downwards through a distancey. In C the eyepiece has
been left in the same position as in B, and a lens (p) has been placed above it, of such
power that parallel rays are produced. The power of this lens (expressed in dioptres)
is the measure of the extent to which the microscopist has accommodated his eye for
closer-than-distant vision.
234
Approved For Release 2003/05/14: CIA
plane C-
plane A-
A
-plane C
-plane A
Fig. 3. Diagram illustrating, in greater detail than fig. 2, the author's method for
determining the focus of the eye in microscopy by finding the distancey. The diagram
is not drawn to scale; the numbers represent distances in millimetres.
 "' ? t"""` L lug. MAJ. i down the microscope. I he draw-tube was therefore put well above the focus before
After trials of various test-objects, a chromosome preparation was chosen for the the subject's first test, and well below before his second. He was allowed to move the
experiment, namely a section of the testis of tl FdvWt(tFW,el5WadinQ5114 CIA-kDP78B051 Ait Db0 ions until he was satisfied that he had obtained a sharp
fluid and dyed with iron hmmatein. In a particular first metaphase of maturation the locus on the tips of the two chromosomes.
tips of two chromosomes, beyond the centromeres, lay in exactly the same plane, The position of the draw-tube, when the subject had focused accurately, was
and these were selected as the objects to be focused (fig. 4). read off with a hand-lens, by estimation, to the nearest 0.1 mm. An error of 0.1mm,
Fig. 4. The object used in the experiments on the focus of
the eye in microscopy. It is a section of the testis of the
newt, Triturus sp., fixed in Bouin's fluid and dyed with
iron hmmatein. A side-view of the metaphase of the first
maturation division is seen. The arrows point at the ends of
chromosomes, beyond the centromeres. It was on these two
ends that the subjects were asked to focus. Both ends could
be brought into exact focus at the same time. The scale
represents 20 [.
The image thrown by the objective was brought exactly into the plane of the
diamond-scratch by use of the fine adjustment of the microscope. This was done by
myself. I was aged 64 and therefore had scarcely any power of accommodation, and
I wore spectacles designed for distant vision. As a result the scratch was sharply in
focus, and it was only necessary to bring the tips of the two chromosomes into the
same focus.
The glass plate was then removed, since the scratch would attract the focus of
the subject's eye. Allowance for the thickness and refractive index of the glass now
showed that the image would be thrown by the objective 0-4 mm higher, in plane D
on fig. 3B. If any subject focused his eye for distant vision, he would have to place
the draw-tube in such a position that the reading on the scale would be 200.4 mm.
If he focused his eye for nearer vision, he would have to lower the draw-tube from
Each subject focused the eyepiece twice. It. was thought possible that the
subject
236
or perhaps slightly more, was possible in the estimation.
The relation between y and the focus of the eye was determined experimentally.
It was necessary to know, for each distance of y, how powerful an accessory lens
placed in the eyepoint of the eyepiece must be, to cause rays diverging from a point
in plane "D" to be rendered parallel. This might have been done by use of a source
of light placed below the eyepiece, but in fact the experiment was done in reverse.
A parallel beam of light was directed along the optical axis of the eyepiece from its
upper side. Accessory lenses of various powers (spectacle lenses of +1, + 2, + 3,
etc., dioptres) were put successively in the plane of the eyepoint of the eyepiece,
and the distance from the flat surface of the lower lens to the focal point was meas-
ured. This distance was 6-5 - y mm. A graph was drawn, relating the power of
the accessory lenses to y. The graph, which only required slight smoothing to
eliminate experimental error, was a straight line. Thus for any value ofy the degree
of accommodation of the eye could be read off in dioptres.
Since it was possible that some subjects might be hypermetropic, the experiment
was extended by the use of negative lenses (-1, - 2, - 3, etc., dioptres). The value
ofy was now negative (i.e. the focal point was more than 6.5 mm from the lower
surface of the lower lens). The graph continued in the same straight line.
Results
The measurements taken in the tests, initially expressed as values ofy, were now
converted into dioptres, to the nearest dioptre in each case. The results; are shown
graphically in fig. 5. It will be noticed that in only 7 of the 200 tests did the subjects
accommodate their eyes for distant vision (i.e. to 0 dioptre, to the nearest dioptre).
One subject, in one of two tests, accommodated his eye for hypermetropic vision.
(In the other test he also showed himself hypermetropic, but the result fell into the
category of 0 dioptre, to the nearest dioptre.) In all the remaining 192 tests, the
subjects accommodated their eyes for closer-than-distant vision.
It had been supposed that the experienced microscopists might have developed
the habit of using distant vision in their work with the microscope. It will be noticed
that they showed a tendency to focus the eye for less close vision than the non-
microscopists. The latter showed a strong tendency to focus the eye for very close
vision. Many of them were young persons, easily able to obtain a sharp view at less
than 250 mm. It is probable that they were not accustomed to place the eye very
close to any such object as an eyepiece, and that the involuntary result was an
attempt to focus on it.
A remarkable feature of the results recorded in fig. 5 is that all three groups of
persons show a dip to a low figure in the histogram just where one would expect a
high one. It seems that the subjects tended involuntarily to make some attempt at
close or distant rather than intermediate vision.
The possibility suggested itself that there might be a stronger tendency towards
distant vision if the microscope were placed horizontally. A small test of this was
made. Ten persons participated (5 experienced and I occasional microscopists, 4
non-microscopists). Each did the test four times with the vertical microscope and
four times with the horizontal. The mean focus of the eye was 4-0 dioptres with the
Approved For Release 2003/05/14: CIA-NDP78B05171A000600070045-8
 e above the focus b66 -
Xttiif trials of'various test-objects, a chromosome preparation was chosen for the d8~t ect's first test, and well. below before his second.tHe was allowed to move the
experiment, namely a section of the testis of the newt T the sub
fluid and dyed with iron haymatein. In a parti al &~q1 1 /14 : CIA-RDP78B051 ~f ~ Cl ons until he was satisfied that he had obtained a sharp
ti f e o mater Li
s o
t
h
f
p
on
wo c
the two chromosomes.
romosomes, beyond the cetntromeres, lay in exactly the same plane, ~cus a Lips o
and these were selected as the ohitrre The position of the draw-tube, when the subject had foci se l a~~?MrPi? was
Fig. 4. The object used in the experiments on the focus of
the eye in microscopy. It is a section of the testis of the
newt, Tritunts sp., fixed in Bouin's fluid and dyed with
iron hxmatem. A side-view of the metaphase of the first
maturation division is seen. The arrows point at the ends of
chromosomes, beyond the centromeres. It was on these two
ends that the subjects were asked to focus. Both ends could
be brought to exact focus at the same time. The scale
The image thrown by the objective was brought exactly into the plane of the
diamond-scratch by use of the fine adjustment of the microscope. This was done by
myself. I was aged 64 and therefore had scarcely any power of accommodation, and
I wore spectacles designed for distant vision. As a result the scratch was sharply in
focus, and it was only necessary to bring the tips of the two chromosomes into the
same focus.
The glass plate was then removed, since the scratch would attract the focus of
the subject's eye. Allowance for the thickness and refractive index of the glass now
showed that the image would be thrown by the objective 0-4 mm higher, in plane D
on fig. 3B. If any subject focused his eye for distant vision, he would have to place
the draw-tube in such a position that the reading on the scale would be 200-4 mm.
If he focused his eye for nearer vision, he would have to lower the draw-tube from
this position through the distance designated as y (fig. 3C).
Each subject focused the eyepiece twice. It was thought possible that the subject
236
----, ...7 u ., w n #LL4icaL u?t uuu. tin error or U'Imm,
or perhaps slightly more, was possible in the estimation.
The relation between y and the focus of the eye was determined experimentally.
It was necessary to know, for each distance of y, how powerful an accessory lens
placed in the eyepoint of the eyepiece must be, to cause rays diverging from a point
in plane "D" to be rendered parallel. This might have been done by use of a source
of light placed below the eyepiece, but in fact the experiment was done in reverse.
A parallel beam of light was directed along the optical axis of the eyepiece from its
upper side. Accessory lenses of various powers (spectacle lenses of +1, + 2, + 3,
etc., dioptres) were put successively in the plane of the eyepoint of the eyepiece,
and the distance from the flat surface of the lower lens to the focal point was meas-
ured. This distance was 6-5 -.y mm. A graph was drawn, relating the power of
the accessory lenses to y. The graph, which only required slight smoothing to
eliminate experimental error, was a straight line. Thus for any value ofy the degree
of accommodation of the eye could be read off in dioptres.
Since it was possible that some subjects might be hypermetropic, the experiment
was extended by the use of negative lenses (-1, -2, -3, etc., dioptres). The value
ofy was now negative (i.e. the focal point was more than 6.5 mm from the lower
surface of the lower lens). The graph continued in the same straight line.
Results
The measurements taken in the tests, initially expressed as values of y, were now
converted into 'dioptres, to the nearest dioptre in each case. The results are shown
graphically in fig. 5. It will be noticed that in only 7 of the 200 tests did the subjects
accommodate their eyes for distant vision (i.e. to 0 dioptre, to the nearest dioptre).
One subject, in one of two tests, accommodated his eye for hypermetropic vision.
(In the other test he also showed himself hypermetropic, but the result fell into the
category of 0 dioptre, to the nearest dioptre.) In all the remaining 192 tests, the
subjects accommodated their eyes for closer-than-distant vision.
It had been supposed that the experienced microscopists might have developed
the habit of using distant vision in their work with the microscope. It will be noticed
that they showed a tendency to focus the eye for less close vision than the non-
microscopists. The latter showed a strong tendency to focus the eye for very close
vision. Many of them were young persons, easily able to obtain a sharp view at less
than 250 mm. It is probable that they were not accustomed to place the eye very
close to any such object as an eyepiece, and that the involuntary result was an
attempt to focus on it.
A remarkable feature of the results recorded in fig. 5 is that all three groups of
persons show a dip to a low figure in the histogram just where one would expect a
high one. It seems that the subjects tended involuntarily to make some attempt at
dose or distant rather than intermediate vision.
The possibility suggested itself that there might be a stronger tendency towards
distant vision if the microscope were placed horizontally. A small test of this was
made. Ten persons participated (5 experienced and 1 occasional microscopists, 4
non-microscopists). Each did the test four times with the vertical microscope and
four times with the horizontal. The mean focus of the eye was 4-0 dioptres with the
Approved For Release 2003/05/14: C1PJ-RDP78B05171A000600070045-8 237
 20-
and the cornea, and (2) that the eyepiece diaphragm should be placed closer to the
Release 2003/05/14: CIA-IDP78B05171A000600070045-8
experienc
microsc
-1 0 1 2 3 4 5 6
D 1 0 P T R E S
ed
opists
Fig. 5. Histograms illustrating the focus of the eye in microscopy. The abscissa
represents the power (in dioptres) of the accessory lens that would be required to
convert a normal eye, focused for distant vision, into the eyes of the subjects when they
focused the microscope.
vertical microscope and 4.6 dioptres with the horizontal. Analysis showed that the
difference between the means was not statistically significant.
Comment
The results of this investigation of the focus of the eye in microscopy indicate (1)
that in diagrams showing the path of light from the object through the microscope
Introduction
The microscopist is primarily concerned with the resolving power of the integrated
instrument, microscope-plus-eye.
Berger (1942) sought to draw a distinction between the true resolving power of
the eye on one hand and visual acuity on the other. He evidently regarded the latter
as an ill-defined but empirically useful idea ("a complex function in which many
factors, at first not thought of, enter").
The mere recognition of the presence of a single point of light or of a dark dot
does not mean that it has been "resolved" by the eye, in the sense in which the
word "resolve" is used in optics; or, to put the same fact in other words, Punkten-
sehsharfe (Hofmann, 1919) is not the same as visual acuity. Hartridge (1923) claimed
what amounts to a visual acuity of about 6 (that is to say, the ability to distinguish
lights subtending at the eye an angle of about 1/6'), but this very high figure appears
to be open to the objection that he was not measuring visual acuity in the strict
sense (by determining the minimum separabile), but was relying instead on the
subject's Punktensekscharfe.
Hofmann (1919) based his conclusions about visual acuity on the ability to see
separately bright points, lines, or larger spaces, separated from one another by a
dark area. He remarked, "To avoid all misunderstanding it must be emphasized
that in what follows I use the convenient expression Sehschdrfe ... solely for the
capacity to separate points, and therefore naturally also lines or larger surfaces. The
limit of Sehschdrfe corresponds, then, to the minimum separabile." Irr the present
paper the expression "visual acuity" is used in exact accordance with Hofmann's
Sehsckdrfe.
The ability to see separately bright stripes or points on a dark background has
been used from time to time by various other authors as a test of visual acuity'
(e.g. by Lister (1913, in tests carried out in 1831-2), Wilcox (1932), and Berger
(1942)), but most workers on the subject seem to have regarded Landolt's (1904)
broken ring as the most critical test. It was used in the elaborate and well-known
experiments of Lythgoe (1932), and is widely quoted in authoritative text-books as
the best (see, e.g., Enoch, 1963). It is therefore necessary to consider in some detail
the reliability of Landolt's broken ring as a test-object for acuity of vision.
A test-object is shown in fig. 6A. The bright area enclosed by the black is seen to
be situated at the right side of the object. Suppose that at a certain distance, and
with a certain luminance (" brightness") of the bright area, the subject is just able to
recognize the fact that this bright area is to the right. Now present him with the test-
object shown in fig. 6B. The bright area is smaller, but provided that there is a
compensating increase in the luminance, there is no reason why the subject should
not still be able to recognize the fact that it is on the right side. It could be made
much smaller, and the subject would still be able to recognize its position in relation
to the whole object, provided that the luminance was sufficient. This test would not
provide evidence as to whether the subject could distinguish the light at the top of
the bright area from the light at the bottom of it. The angle subtended at the eye
by the top and bottom of the bright area would not be relevant to the problem of
visual acuity. Indeed, the test would not seem to solve any specific problem con-
nected with the subject's vision. The threshold of sensitivity of his eye to light would
2.3
Approved For Release 2003105114: CIAIRDP78B05171A000600070045-8 239
 way. One might possibly record the distance between the bright area T2PIldp8 We
centre of it) and the centre of the whole test-object, but a statement of the angle
s u b t e n d e d at t h e eye by t h e s e two points APPM OW RVM65F~OWS74
impression of visual acuity.
- It is true that one may be said to "see" a black (or very dark) object, if one has
just looked at a bright one (for instance, the bright surroundings), because of the
change in the nerve impulses from the suddenly-darkened area in the retina; and
in this sense it might be claimed that one could "see" the black limits of the gap,
so that the angle subtended by them at the eye would be relevant; but any such
"vision" of blackness would presumably be outlasted and outweighed by the real
vision of the light in the gap.
If the diameter of the black area is now progressively reduced and its centre
hollowed out (fig. 6C, D), the same arguments still apply. With sufficient luminance,
Fig. 6. Diagram illustrating the author's criticism of Landolt's broken ring as a
test-object for visual acuity. For full explanation see text.
the subject will be able to recognize the position of the bright area in relation to the
whole object. Next, present him with the test-object shown in fig. 6E. Further
reduction in diameter combined with the hollowing out of the centre has resulted
in the production of Landolt's broken ring, which is exactly represented in the
figure. The ring is shown to the subject in 8 positions separated from one another by
45?, in random order, and he is required to say whether the gap is to the N. (fig. 6F),
N.E., E. (fig. 6E), etc.
When Landolt's test is applied, the angle subtended at the eye by the gap is
measured in minutes or fractions of a minute, and smaller and smaller copies of the
broken ring are presented to the subject until he can no longer recognize the position
of the gap. The smallest angle at which he can still state its position is recorded, and
the reciprocal of the figure obtained is given as the numerical statement of his
visual acuity. Landolt himself and many others (e.g. Lythgoe,1932) have recorded
Approved For Release 2003/05/14:
Landolt's broken ring seems to be subject to the same adverse cn cis
I ected a ' st the test-objects shown in fig. 6, A-D. The angle subtended at the
CIA- P78BO51~(})~ cQ94&-8ppear to be relevant. No proof is afforded that the subject
can distinguish the light at one side of the gap from that at the other. The minimum
all could be recognized if
m
separabile is not measured. The position of a very s gap
the luminance were sufficient.
The familiar tests with printed letters as the test-objects fer visual acuity are
convenient for clinical use, but it was well understood long ago (e.g. by Guillery,
1891) that there are valid objections to them. The various letters differ in the ease
with which they can be read, and the recognition of their shapes involves not only
the act of seeing but also a mental process (Denkact). It is realized that the minimum
cognoscibile is not the same as the minimum separabile, which constitutes the true
test of acuity.
Method
For the reasons just stated, it was decided to discover the minimum separabile by
the use of test-objects consisting of bright parallel stripes on a dark background.
(fig. 7). The width of each bright stripe (left-hand pair of arrows in the figure) in
Fig. 7. Part of one of the author's test-objects (mag-
nified). The arrows indicate the widths of the bright
stripes and of the intervening dark area.
all cases subtended an angle of 1' at the eye. The distances between the bright
stripes (right-hand pair of arrows) were 4', 2', 1', and ?`. Fig. 8 represents one of
the test-objects. Four groups of bright stripes are seen. The subjects were asked to
count the number of stripes in each group.
To simulate as well as possible the usual conditions of microscopy, transmitted
light was used. The test-object was prepared by making a large drawing in negative
contrast: that is to say, the stripes were drawn in black ink on a white ground. The
drawing was then photographed to make a much smaller negative, in which the
stripes were nearly transparent on a nearly black background. The negative (a
2 in. x 2 in. plate) constituted the test-object.
It was thought undesirable to reduce the original drawing to such an extent that
the stripes would subtend angles of 4', 2', 1', and i' at the standard distance of
close vision, namely 25 cm. It would be difficult to make satisfactory photographic
negatives on such a minute scale and it would also be troublesome to hold the sub-
ject's head sufficiently firmly in the right position, for a small change in position
would have a large effect on his capacity to count the lines. For this reason every-
thing was done on a scale of x 10: that is, the test-object was placed 2.5 metres
2
8B05171A000600070045-8
 Fig. 8. Text-object Z. The bright
stripes subtend angles of 4', 2', 1' and
#' at the eye, if the test-object is held
21 metres away. (The small triangles
are included in the test-object to en-
sure that its size is correct. The dis-
tance between their upper angles
should be 20 mm.)
Release 2003/05/14: CIA-
away. Since the tangent of 1' at 2.5 metres is 0.727 mm, the test-object was made
approximately to this scale. Fig. 8 is made as exactly as possible of the correct size.
It is worth remark that Landolt (1904) made a serious error in stating that the
tangent of I' at 5 metres is 14.5 mm and at 33 cm is 0.96 mm. These figures for
tangents are approximately 10 times too great. They may have misled other workers,
who have repofted astonishingly high visual acuities.
The optical density (extinction coefficient) of the test-objects was measured in
both the bright and dark areas (see Appendix I).
The test-objects were illuminated from behind by a 12-volt, 100 watt Philips
"mirror-condensor" lamp, placed 12 cm behind the test-object. The lamp can be
obtained from photographic dealers. To diffuse the light, two pieces of ground glass
were fixed between the front of the lamp and the test-object, 28 mm and 4 mm from
the latter.
A variable resistor allowed the luminance of the ground glass behind the test-
object to be controlled by the subject.
There is evidence that the surroundings of a test-object affect visual acuity. The
test-object was therefore placed at the centre of a circular area of white paper,
surrounded by black (fig. 9). The circular area represented the microscopical field of
view, limited by the eyepiece diaphragm. It was made 1.54 metres in diameter, and
thus subtended the same angle at the subject's eye as the mean field of view of 8
typical eyepieces of medium power made by four well-known British and Continental
manufacturers. The white paper was illuminated by the daylight of the room.
It was so placed that sunshine could never fall directly upon it. Its luminance was
commonly about 1.1 log. foot-lamberts, but naturally this varied with the weather.
When the eye is placed at the eye-point of an eyepiece for microscopical vision,
light strikes the eye not only from the microscopical field of view, but also-very
obliquely-from the general illumination of the room, between the top of the eye-
piece and the eye. It was thought necessary to simulate the conditions of actual
microscopy in this respect.
242
Approved For Release 2003/05/14: CIA-
diameter of top of eyepiece 28 mm
diameter of aperture for eye-lens 8.6 mm
DP78B0517AVMg6 I QQ,4Se-1?tesent the top of the eyepiece, 10 times this size; it was
blackened on the side on which the subject sat- The subject looked through the
hole with one eye (chosen by himself), the other eye being covered by an eye-shield
or hand (fig. 9). It was necessary to place the eye at the correct distance from the
board. The mean height of the exit-pupil (Ramsden circle) above the top of the
eyepiece, in the 8 examined, was 8.1 mm: it was therefore necessary to place the eye
about 8.1 cm from the board, since everything was on the scale of 10:1. This was
achieved by asking the subject to hold his head still in such a position that two white
dots, one of which is marked by an arrow in fig. 9, formed the limit of his field of view.
Fig. 9. A subject undergoing the test of visual acuity. She sits in front of a black ring
representing the top of the eyepiece, magnified 10 times. The test-object is in the
centre of the white circle that represents the microscopical field of view. The arrow on
the left points to one of the two white spots that serves to ensure that the subject's
eye is 2.5 metres from the test-object and about 8.1 cm from the black ring. For full
explanation see text.
It was arranged that when this was so, his eye was as nearly as possible 8.1 cm from
the circular board representing the eyepiece, and 2.5 metres from the test-object.
The test was carried out as follows.
The subject was shown one of the test-objects from close up, and the nature of
the test explained. He was briefly shown the 9 test-objects, without being given an
opportunity to study them; he was told that they were all different from one another
in the arrangement of the lines. He was told that there would be four tests, and that
one of the test-objects might be shown twice. He had no means of knowing that in
fact every one of the 100 subjects was shown the same three test-objects, designated
DP78B05171A000600070045-8 243
 ? ??e?? uu,Fw uu11 uULJ. d was slluwn oil the first occasion with the
stripes horizontal and on the second vertical, to allow astigmatism to reveal itself.
In the description of the results, the s'~N6sdur&tR&lre 57 ~1
placed with the stripes vertical. ~ "" &4eZW 41:
In all three test-objects there were 3, 4, or 5 bright stripes in each of the four
groups of stripes, but the subject had no means of knowing this, apart from his
ability to count them correctly during the test. Details of the three test-objects are
given in A
e
di
T
pp
n
x
The subject took up his position as described above, with one hand on the control
of the variable resistor. He was allowed to move the control as often as he wished. He
counted the easiest group of stripes first (4' apart), and then proceeded to the more
difficult ones (2', then 1', then I'). He was allowed as long a time as he wished. The
position of the pointer on the scale of the resistor was noted when he had finished
counting all the stripes on a particular test-object, so far as he was able. All his
counts were recorded.
Results
Since each subject was presented with four tests (test-objects X, Y, Z, and Xv), and
in each test there were four groups of lines, it was theoretically possible to count
correctly 16 groups. The maximum possible score, 16, was not obtained by anyone.
A girl aged 11 and a man aged 20 achieved the highest score, 14. A man aged 50
counted 13 groups correctly. It will be remembered that the subjects were a selected
group, since all of them were persons who did not use spectacles or contact lenses
for any purpose.
Comment
The visual acuity revealed in this experiment is much less than has been recorded
by several other investigators. This may be due in part to the simulation of the
circumstances of microscopical vision, but the main cause must be that nothing was
sought in the test except the discovery of the minimum separabile. The reasons for
rejecting the validity of the Landolt and similar tests has been considered in some
detail on pp. 239-241.
The results obtained in the test of visual acuity are briefly summarized in ~. 2 0 0
Appendix II.
e
i
~
proport
on of correct counts was.lower at all four
separations of the bright lines (4', 2', 1', and when the test-object was Y than
when it was X, Z, or Xv. Test-object Y has 5 bright lines in each group, whereas
the others have 4 or 3. The facts summarized in statistical form in Appendix III
leave no doubt that there is a highly significant difference between the results
obtained when there are 5 bright stripes and those obtained when there are 4 or 3.
The experiment was not planned in such a way as to allow a definite answer to the
question whether it was easier to count 3 than 4 stripes, but the figures obtained do
not suggest that this was so.
The fact that a small number of lines is more easily resolved by the eye than a
larger number was attributed by Conrady (1913) to circumstances arising from
the undulatory theory of light, but the possibility that the higher centres of the brain
may be concerned must not be overlooked. One might suppose that difficulty in
making the necessary eye-movements to cover the larger number of bright stripes
was the cause, but if this were so, one would expect that the difficulty in counting 5
stripes would be more evident with the more widely-spaced stripes. With the stripes
at separation of }', little or no movement of the eye might be necessary to count 5
stripes. Whatever the explanation may be, it seemed best to exclude the results
obtained with test-object Y from the graphical representation of the results (fig. 10),
since the results with test-objects X, Z, and Xv appear to be more reliable as indi-
cators of visual acuity unaffected by other factors.
Fig. 10 reveals that a group of bright stripes separated by 4' can nearly always be
counted accurately, and the 2' group in 81 p.c. of tests. The 1' group can usually
not be counted, and when the separation is only z` the number of correct counts is
less than 9 p.c. It seems right to put the visual acuity at z (in the sense that reliable
244 Approved For Release 2003/05/14: CIA
it is I`}.
Several subjects (who were presumably astigmatic) found it much easier to
CIA-RpP78B05wt**o o? 4oo44tsibject X when it was turned through 90? into the Xv
position, but no general tendency of this sort was observed.
test- objects
X, Z, and X
4 2 1 112
ANGLE SUBTENDED (MINUTES)
Fig. 10. Diagram illustrating the results of the tests of visual acuity with test-objects
X, Z, and Xv.
 L - to u..t,muuuu ui L), it luuuws Utdt fit! MIgnt 156 expected to p, apart at a distance of 25 cm. There would, therefore, be no neces0- J
sity for
count them at the standard distance of near vision (25 cm) if they were separated him to use an eyepiece magnifying more than 155 times with a x 100 objective of
by 145 p.
pQ ved For RRelea a 2003/05/14: CIA-FbP78BO51 045-
rho _
The
stance apart (d) of two points of li'l i~ at can just be resolved as separate k~
di 23.5 = I ) ? It is unlikely that a single microscopist could be found
by a microscope objective is given by the familiar equation
who would
accept such
a proposition.
d 0.61A
nsin.a
where A is the wave-length of the light
a is the half-an
le of ac
b
,
g
ceptance
y the
objective, and n is. the-refractive index of the medium in which the angle is measured.
An. ordinary oil immersion objective of N.A. 1.30
used with blue-
li
h
f
,
green
g
t o
wave-length 500 rnp, is therefore just capable of resolving as separate two points,
0.235 p, apart. If the objective magnifies 100 times, the corresponding points in
the primary image will be 23.5 ? apart. In order that these may be seen by the eye,
it follows that the eyepiece must magnify at least 2145 = 6.3 times. Somewhat
3.5
greater magnification would probably be justifiable to relieve the "strain" that is
supposed to exist when the eye is used at or near the limit of visual acuity.
The equation given above shows that an objective of N.A.1.40, used with violet
light of wave-length 415 mp, resolves 0.181 p,. It follows from the results recorded in
the present paper that if the objective magnifies 100 times, a x 8.0 eyepiece -would
be required to resolve every detail in the image thrown by the objective. Once again,
it would presumably be justifiable to use an eyepiece of somewhat higher power.
Although a person with normal sight would not actually need eyepieces of higher
Power than these, except to reduce "strain", and could not increase the resolving
power of his microscope by using such eyepieces, yet it would be perfectly justifiable
for persons with low visual acuity to use eyepieces of higher power. A curious out-
come of this line of reasoning may be mentioned. If animals were to become
sufficiently intelligent to be able to design and make optical instruments, they might
well construct objectives similar to our own, but very different eyepieces. The
square-lipped or "white" rhinoceros, which has very poor vision, might make
eyepieces of very high power, or even use an accessory compound microscope to
view the image thrown by the objective; but it seems almost certain that many
species of birds would use eyepieces of lower power than those needed by man.
There is little direct evidence on the visual acuity of birds (Rochon-Duvigneaud,
1943), but the slenderness and close packing of the cones in the fovea suggest high
acuity. Rochon-Duvigneaud counted 81 cones in a distance of 100 p across the
central fovea of the common buzzard, Buteo buteo (L.), in comparison with 39 to
50 in man. He considered that the short-toed eagle, Ciruetus gallicus (Gmelin), may
have even closer packing of the cones than other Accipitres. On the basis of the
number of cones in each square mm of the fovea, Walls (1942) claimed that some
members of this group must have a visual acuity at least 8 times that of man. If
this were true, such birds could discard the eyepiece without loss of resolving
power. They could use the objective as a simple microscope, though they would have
to re-design it slightly for the production of divergent or parallel rays (and they
would need to devise a new way of correcting apochromatic objectives and high-
power achromats for the chromatic difference of magnification).
It is perhaps legitimate to express the different eyepiece requirements of different
observers by the aphorism, "The objective is objective, the eyepiece is subjective."
If the conclusion of Lythgoe (1932) were correct, and were applicable to micro-
scopical vision, very surprising results would ensue. With a visual acuity of 2 (i.e.
246
Approved For Release 2003/05/14: CIA-
THE EFFECT OF LUMINANCE ON VISUAL ACUITY
Introduction
In this section the word "illumination", in inverted commas, will be used wherever
it is convenient to have a comprehensive word to include all the various technical
terms, with precise meanings, that authors have used to represent the amount of
light used in their investigations (illumination in the strict sense (lux), luminance
(foot-lamberts, etc.), luminous flux (lumens), luminous intensity (candela:), and
retinal illumination (photons)).
It has long been supposed that visual acuity rises with increased "illumination"
in such a way that if a graph be drawn with visual acuity as ordinate and the logarithm
of the "illumination" as abscissa, a straight line is produced. It follows, on this
supposition, that there is not an optimum "illumination" for maximum acuity,
beyond which'acuity falls off.
Konig (1897) was one of the first to put this belief in concrete form. His graph
consisted of three straight lines. There was first a gradual rise in acuity with increase
of "intensity ", but this part of the graph must have related to scotopic vision, which
is irrelevant to the microscopist. The line then turned sharply upwards till it reached
a maximum, beyond which it passed on parallel to the abscissa. Hecht (1928)
re-analysed Konig's data and showed that a somewhat S-shaped curve, with a long,
nearly-straight central portion, would represent the facts more exactly, but he
accepted Konig's main conclusions.
Lythgoe's graphs show the same general form, except that there is no terminal
part parallel to the abscissa. He stopped increasing the "illumination" because he
thought the intense light would damage his test-objects. Lythgoe's conclusions
have been accepted by most writers on the subject, including the author of the
standard work on photometry (Walsh, 1958). Reasons have already been given for
-supposing that Lythgoe's result was almost. inevitable from the use of Landolt's
ring as test-object.
Lythgoe lays great emphasis on his contention that in studies of this sort the
"illumination" should be presented on the abscissa in logarithmic form. He
reaches the surprising conclusion that those investigators who have reported an
optimum visual acuity with rather low "illumination", and a fall-off in acuity
beyond this point, have been led to incorrect results by using non-logarithmic
abscissa;.
Method
The experiment formed part of the test of visual acuity, which has already been
described (p. 241). It differed from those of other investigators in that the subject
was permitted to choose the luminance best suited to his vision, by control of the
resistor included in the electrical circuit supplying the source of light.
The variable resistor was provided with a arbitrary scale. To calibrate this
arbitrary scale, the luminance of the ground glass immediately behind the test-object
was measured by an S.E.I. photometer, with the pointer of the resistor in various
 ^ li , "J W UUnL u IL 1 ue 1eauitlgs of lU$imance,_
in tog. foot-lamberts, for a particular settin of A. 100
g
resisor, d1d not vary even it
considerable changes in distance were madd,e~ The rea ' v~g~p
intermediate values. The luminance could%R Pr c~~r k A 4 : CIA-R P78B05171A 00600070045-8
more than 4-U log. foot-lamberts. `V
It. is to be noted that the values obtained represent the true luminance of the
ground glass screen just behind the test-object. Reliable figures could not have
been obtained if the photometer had been held at the distance of the subject's- eye
(2.5 metres), since the area of the ground glass was not sufficient to give readings
unaffected by distance when the photometer was held so far away. It follows that
although the luminances were measured by the appropriate method for the purpose
intended, they do not provide a direct indication of the flux of light entering the eyes
of the subjects.
It was soon noticed that many of the subjects, after first experimenting with
various degrees of luminance, used a fairly bright light for counting the groups of
stripes separated by 4', but cut down the light when they were trying to count the
more difficult groups, especially those separated by only 1' or 1'. Indeed, a consider-
able number of them said aloud that they found it necessary to do this. The lumin-
ance was recorded (by noting down the position of the pointer on the scale of the
resistor) as soon as the subject had finished his attempt to count the bright stripes on
each test-object, i.e. when he had been trying to count the more difficult groups.
Care was taken to avoid error arising from thoughtless movement of the pointer by
the subject after he had finished announcing his counts.
Results
These are recorded in fig. 11. It will be noticed that, in accordance with Lythgoe's
strong recommendation, the abscissa is logarithmic. Nevertheless, the results
obtained do not bear out his contention that there is a continuous rise in visual
acuity with increase of luminance. On the contrary, the modal luminance chosen
by the subjects when trying to count the closer stripes was only 2-5 to 2.75 log.
foot-lamberts (i.e. from about 316 to 562 foot-lamberts), though by the mere
turning of a knob they could have used any amount of luminance up to 4 log. foot-
lamberts and beyond (i.e. to well over 10,000 foot-lamberts).
It might be supposed that the subjects chose the luminance more or less at ran-
dom. The experiment was conducted in such a way that this possibility could be
tested, although the subjects had no knowledge that this was being done. The
optical density of the bright stripes in test-object Z was twice that in test-object X,
though in all other respects the two test-objects were closely similar (see Appendix
I). It was impossible for the subjects to guess this, because they had their attention
fixed on test-object Y after attempting to count the stripes on X and before attempt-
ing to count those on Z. It was therefore relevant to notice whether the subjects
would use a stronger illumination with Z than with X.
The result of this investigation is set out graphically in fig. 12. The reader will
at once notice the marked difference between the two histograms. There was a
strong tendency for the subjects to use more light when the test-object was Z.
Statistical study confirms this conclusion. The mean chosen luminance with X was
464 foot-lamberts; with test-object Z, 792. Thus the difference between the means
was 328. The standard error of this difference was 83-34. The difference between
the means was 3-94 times the standard error of the difference, and therefore highly
significant statistically. It follows that the subjects did not make random choices of
luminance, but adjusted the luminance to their needs.
248
1-75 2.0 2-25 2-5 2.75 3-0 3.25 3.5 3.75 4-0
LOG. FOOT-LAMBERTS
CHOSEN GRADES OF LUMINANCE
Fig. 11. Histogram showing the grades of luminance chosen by subjects in the tests of
visual acuity. (The luminance is that of the ground glass plate placed 4 mm behind the
test-object.)
Comment
The facts presented here prove that the subjects chose rather weak luminance of
the test-object when they were trying to count the bright stripes subtending small
angles at the eye. This result is contrary to the prevailing opinion, but the reason
for the difference has already been made clear.
It is a remarkable fact that three independent observers (besides myself) who
used transmitted light to illuminate their test-objects, found that the highest acuity
was not obtained with the strongest light. Lister (1913), who used light reflected by a
mirror from a white cloud on sunny days, sometimes found it necessary to interpose
thin "gauze paper" between the mirror and the test-object, to moderate the light.
Wilcox's (1932) bright stripes consisted of metallic mirrors, and although these were
illuminated from the subject's side, yet the virtual image of the source of light was
Approved For Release 2003/05/14: CIA-JtDP7813051714000600070045-8
 type of transmitted light, did not find any decrease in visual acuity with
increase in "illumination".
u 2 4 6 8 10.12 14 16 18 20 22 24
HUNDREDS OF FOOT-LAMBERTS 26+
Fig. 12 CHOSEN GRADES OF LUMIN
. Histograms showing that in the tests of visual ANC E
Z.
choose different grades of luminance according to whether the tffiest object wasX or
a
B
BL
erger cessart(y behind the stripes, so that in effect transmitted light was used. It was
B
in this (1942) who first noticed that transmitted light
respect. He did not use the
differed from incident light
"self-luminous objects". expression ,
Actually transmitted light", but wrote of
dark backgrounds from behind, however, he illuminated transparent spaces in
behind. It is scarcely necessary to point out the significance
microscopist, since transmitted is more commonly sedf these findings
250 Y than incident light ine
Approved For Release
10 - Appna lj gQrl Rejease 2003/05/14: CIA-RI P78B05171A000600070045-8
test-object X
Introduction
If the refractive properties of the component parts of the eye were such as to provide
a perfect image-forming system, it would follow that a large pupil would give high
resolving power. In fact, however, as is well known, the marginal rays are less well
corrected than the intermediate and paraxial, and as a result the visual acuity
scarcely changes when the diameter of the pupil increases beyond 2.0 mm (Sten-
strom,1964). It is perhaps for this reason that Michel (1964), in his discussion of
the visual acuity of the eye in microscopy, assumes a pupillary diameter of 2.0 mm.
The diameter is of interest, however, not only for its influence on visual acuity, but
also because the diameter of the Ramsden circle of the eyepiece should be related
to it. It was therefore regarded as important to obtain measurements of the pupillary
diameter in persons actually using the microscope, partly because the diameter might
be found to be less than 2.0 mm (in which case the visual acuity would be less),
partly so as to provide reliable information on which to base the design of eyepieces.
Method
The diameter of the pupil was found by reliance on the consensual reaction, which
ensures that in all normal persons the diameters of the two,pupils are the same, even
when they are differently illuminated (Steinach, 1887; Hedda:us,1904; Duke-Elder,
1932).
The microscope was set up in a vertical position. The object was a section of the
testis of the newt, Triturus sp., dyed with iron ha'matein. A 4 mm objective and x 6
eyepiece were used. Kohler illumination was provided. The iris diaphragm of the
substage condenser was set to give a cone of light filling about a of the aperture of
the objective.
Each subject chose for himself the intensity of the light by moving the control
wheel of the variable resistor forming part of the microscope lamp.
The subjects of these experiments were all experienced microscopists. This
limitation of the subjects to a particular group was imposed because others would
not have had sufficient experience in the adjustment of the intensity of the light
when making microscopical observations. In this set of experiments visual acuity
was not directly concerned, and experienced microscopists who sometimes wore
spectacles were therefore not excluded, provided that they were persons who were
not accustomed to wear them when using the microscope. No one wore spectacles
when the eye was photographed, since this would have altered the apparent size
of the pupil.
A Zeiss-Ikon "Cobra" 35 mm camera was used to photograph the eye that did
not look down the tube of the microscope. The lens of the camera had a focal length
of 45 mm. In front of it was placed a Zeiss "Proxar" double lens of focal length 10
cm. With the camera focused at infinity and the Proxar in position in front of it, the
object to be photographed must be 105 mm from the front of the mount of the
Proxar. A block of wood was provided, on which the camera was placed, facing
vertically upwards. The thickness of the block was such that the front of the mount
of the Proxar was 105 mm below the plane of the eyepoint of the eyepiece. The
camera was placed with its optical axis separated from that of the microscope by
2003/05/14: CIA-RDP78B05171A000600070045-8
 uu c11 calibrate the phot mP r- 1U1 ug 1] ]111 velure M8 photograph .. ~? uuwyiat a Pupil 11 atlt pegtaously been mea-
horizontally in the oghs obtained, an accurate sured). If the exit-pupil of the eyepiece were the smaller, some of the effective
plane of the eyepoint and photo a ruler was placed aperture of the e e mi h"t be wasted, to the detriment of the resolving power of the
~
same position as when used to photograp gr r5~34144 CIA-R~P78B051 7 ~ t eyepiece e and entrance-pupil of the eye were of exactly
The photograph when
scale was projected onto a screen placed at such a distance the same size, the slightest st movement of the head would cut off on one side some
that the magnification was exactly 10 rimes, The photographs of the eyes were of the a
the pupils aperture of the lens-system of the microscope. In the entrance pupil were
Projected with the same arrangement. The diameters of the images the
the smaller, some of the aperture of the lens-system of the microscope would
were read off on the screen with a ruler, and divided by 10. Since the Pupil a necessarily be wasted.
pproxi-
mates closely to a circle, the diameters were read in whichever direction the photo- graphic image was clearest. In some cases reflections from the bench interfered The diameters of the exit-pupils of the 8 eyepieces of moderate power, already
mentioned, were measured. The mean diameter was 1-7 mm (extremes 1.0 and
with part of the image; in others the edge was only s th2-0 e mm). It evident that with such eyepieces as these, no part of the aperture of
ge, ge of the iris direction, especially when it was dark brown. y harply defined in a the microscopical lens-system would be lost, but it is probable that with most of
Each recorded figure represents the diameter of the "entrance-pupil- them the full resolving power of the eye would not be used.
, tha is to
say, the apparent diameter of the pupil as viewed through theornea. Intwhat
follows, the expression "diameter of the pupil" must be taken to mean the diameter
of the entrance-pupil.
Sufficiently satisfactory photographs were obtained of the pupils of 15 subjects. ACKNOWLEDGMENTS
Three of them provided two photographs each. Eighteen measurements were thus The work more could not have been done without the unselfish co-operation of the
available. subjects, more than 100 in number.
If the head was not held in such a position that both eyes es were at approximately Mrs. Elizabeth G. M. Williams gave me a lot of help in different ways, as usual.
I also paratus grateful i the
the would same ead not distance was record above accurately the
a asuring surig wing: uaMr. l James M. McCrae (for help in M
the bench a size of the the pupil; eyepoint but the the we depth of focus the
focus of photograph $
David appara used m visual acuity); Mr. John Haywood and Mr.
.
system of the camera (with Proxar lens) was very small at the apertures used, and
photographs that David Turner (photography, especially making the test-objects for visual acuity)
were obviously out of focus were rejected. M. Sheila Bradbury (measuring the optical density of the test-objects); my wife
(help in analysing the results); Professor E. B. Ford and Mr. G. J. Draper (advice
Results on statistical methods); Dr. A. Pirie (organizing a discussion with authorities on
The mean diameter of the pupil in the 18 recorded measurements was 2 84 mm vision); and Miss Gillyan Bennett (skilful secretarial help).
(standard deviation 0.35).
A d certain amount .3 doubt was felt about the reliability of the measurements
taken on the screen, because there were cases in which the exact limit of the pupil Appendix I. Details of the three test objects for visual acuity,
was not perfectly distinct, and an error of more than 1 mm (in the magnified ima e
was possible. All the g) for Number of bright stripes present optical density
photographs were therefore projected a second time, and
measurements were made without any reference to the direction or result of those (4' 1st group 2nd group 3rd group 4th grow
group of bright of back-
made earlier. The mean of the second set of measurements was 2.82 mm (standard ' apart) (2' apart) (1' apart) 1' apart) stripes ground
3
deviation 0-32). It is clear that the method of measuring the photographs was capable 3 i o
of Y 3 5 3 4 0.5 3.4
5 5 0 6 2 g
giving sufficiently accurate results, and that one may have some confidence in Z 3 4 3 3 10
saying that the u it
P p has a mean diameter of about 2-8 mm 3 5
microscopist, p of brwill be monocular vision, controls the illumination of a microscopical density of the bright stripes.
xtre se rage was 2 2 to 3 5 mm in the first set of ex The symbol Xv is used to mean the test-object X placed with the bright stripes vertical.
to 3,4 mm in the second. of a micr and 2 3
commmt
It is a familiar fact that emotional states may affect the size of the u il see Appendix 11. Test for visual acuity: summary of results.
1934). Even the sight of f an n interesting picture may cause achange in the Test- out 2-8 m
diameter of the
more interesting Stripes Stripes Stripes Stripes
than any other ordinary Pcriit did was 4' apart 2' apart 1' apart 1' apart
microsco state du
appear that any subject was in an emotional
pical ring the test. and it did not }
X 3 97 56 11
Presumably there must be some ideal relationship between the diameter of the Z 3 98 39 10
exit-pupil (Ramsden circle) of the eyepiece and that of the entrance-pupil of the eye; Xv 4 99 39 5 76 but it would not appea
r that any precise study of this subject has been published Z 4 58
Xv 4
80
252 Y ~ 87 22 5
Approved For Release 2003/05/14: C14RDP78BO5171A000600070045-8
 Approve olibr Release 2003/05/14: CIA-RDP78BU471A000600070045-8
John R. Baker
Appendix III. Comparison of the results of the test for visual acuity when the number of
bright stripes was 3 or 4 with the results when the number was 5.
Test- Separation Number Percentage Standard Differences
objects of bright of of correct error of between
stripes bright counts percent- percent-
stripes age ages
X, Z, & Xv 2' 3 81 P.C. 2.26
Y 2' 5 58 P.C. 4.94
X, Z, & Xv 1' 4 44J P.C. 2.87
Y 1' 5 22 P.C. 4.14
References
Standard Number of
error of times the
differences differences
between between
percent- the per-
ages centages
exceeds the
standard
error of the
differences
} 23 543 4.2
221 5.04 4.5
Berger, C. (1942).-A comparison, under different degrees of. brightness, of minimum visual
distances of two luminous points and of a broken circle. Amer. -7. Psychol., 55, 354.
Conrady, A. E. (1913).-Footnote to Lister (1913).
Duke-Elder, W. S. (1932).-" Text-book of ophthalmology", Vol. 1, Kimpton, London.
Enoch, J. M. (1963).-Article on "Vision" in "Modern ophthalmology", edited by A. Sorsby.
Butterworth, London.
Guillery, - (1891).-Ein Vorschlag zur Vereinfachung der Sehproblem. Arch. f. Augenheilk.,
23, 323.
Hartridge, H. (1923).-Visual acuity and the resolving power of the eye. J. Physiol., 57, 52.
Hecht, S. (1928).-The relation between visual acuity and illumination.,. gen. Physiol., 11, 255.
Hedda:us, E. (1904).-Article on "Semiologie der Pupillarbewegung" in "Graefe-Saemisch
Handbuch der gesamten Augenheilkunde ", 2 Aufl., IV Bd., I Abt. Engelmann, Leipzig. .
Hess, E. H. (1965).-Attitude and pupil size. Sci. Amer., 212, 46.
- & Polt, J. M. (1960).-Pupil size as related to interest value of visual stimuli. Science,
132, 349.
Hofmann, F. B. (1919).-Article on "Physiologische Optik (Raumsinn)" in "Graefe-Saemisch
Handbuch der Augenheilkunde", 2 Aufl., III Bd., XIII Kap. Springer, Berlin.
Konig, A. (1897).-Die Abhangigkeit der Sehscharfe von der Beleuchtungsintensitat.
Sitz. kon. Preuss. Akad. Wiss. Berlin (vol. not numbered), 559.
Kuntz, A. (1934).-" The autonomic nervous system ". London, Bailliere, Tindall & Cox.
Landolt, E. (1904).-Article on "Die Bestimmung der Sehscharfe" in "Graefe-Saemisch
Handbuch der gesamten Augenheilkunde", 2 Aufl., IV Bd., I Abt. Engelmann, Leipzig.
Lister, J. J. (1913).-On the limit to defining-power, in vision with the unassisted eye, the
telescope, and the microscope. (Papers dated 1842-3 on work done in 1831-2).)J. roy.
micr. Soc. (vol. not numbered), 34.
Lythgoe, R. J. (1932).-The measurement of visual acuity. M.R.C. Special Reports, 173, 1.
Marshall, C. R. & Griffith, H. D. (1928).-"An introduction to the theory and use of the micro-
scope". Routledge, London:
Michel, K. (1964).-"Die Grundzilge der Theorie des Mikroskops in elementarer Darstellung".
Wissenschaftliche Verlagsgesellschaft, Stuttgart.
Mdllring, F. K. (1965).-Methods of optical adaptation of the movie camera to the micro-
scope. Zeiss Information, 54, 126.
Rochon-Duvigneaud, A. (1943).-"Les yeux et la vision des vertdbrds." Masson, Paris.
Shlaer, S. (1937).-The relation between visual acuity and illumination.,'. gen. Physiol., 21, 165.
Steinach, E. (1887).-Vergleichend-physiologische Studien uber Pupillarreaction. Centralbl,
f. Physiol., 1, 105.
Stenstrom, S. (1964).-" Optics and the eye." Butterworth, London.
Walls, G. L. (1942).-" The vertebrate eye and its adaptive radiation." Cranbrook Institute,
Bloomfield Hills (Mich.).
Walsh, J. W. T. (1958).-"Photometry." Dover, New York.
Wilcox, W. W. (1932).-The basis of the dependence of visual acuity on illumination. Proc.
Nat. Acad. Sci. Wash., 18, 47.
Approved For Release 2003/05/14: CIA-RDP78BO5171A000600070045-8