Three Dimensional Photography Principles Of Stereoscopy

3 1148 00865 8528

WSEPQ7199J

MOV 2 9 1985
140V 01,3981

NOVOSJ988

S3

=SS 7

JAM 02 199?

THREE-DIMENSIONAL

PHOTOGRAPHY

Principles of Stereoscopy

by
HERBERT C. McKAY

F. R. P. S. < .

AMERICAN PHOTOGRAPHY
BOOK DEPARTMENT

M/nneopo//s

Copyright 195] by the
AMERICAN PHOTOGRAPHIC PUBLISHING COMPANY

Printed in the United States of America
by the JONES PRESS INC., Minneapolis, Minn.

TO

my principal assistant
who is also my wife,

FRANCES A. McKAY

KANSAS CITY (MO) PUBLIC LIBRAE

TABLE OF CONTENTS

CHAPTER PAGE

1 Elementary Stereography 1

2 Stereoscopic Cameras 12

3 The Stereoscope 28

4 Correlation of Camera and Viewer 60

5 Mounting, Transposition and Spacing 75

6 Stereographic Technique 91

7 Flash in Stereo 104

8 Color in Stereo 119

9 Pictorial Stereography 131

10 The Nude in Stereo 154

11 Anaglyphs 167

12 Stereo Projection 175

13 Applied Stereoscopy . .201

14 Polarized Light Applied to Stereoscopy 213

15 Types of Stereoscopy 219

16 Trick Work 222

17 Close-up Stereography 241

18 Hypers tereo 262

19 Stereo Movies 274

20 Special Phases 292

21 Stereo Processing 305

Appendix A 314

Appendix B 319

Appendix C 326

Index 330

INTRODUCTION

DURING THE TWO YEARS since the first edition of this book
appeared, there has been a profound change in the status of
stereo in the United States. Then it was confined to a relatively
small number of specialists, today it promises to become the
number one medium for amateur photography. The change has
been aided by the introduction of small, convenient cameras
using 35mm film; it has been given great impetus by the intro-
duction of color; it has been made easier by the simplification of
mounting and the introduction of commercial mounting services.
However, the greatest force of all has been an aggressive publicity
campaign sponsored by the makers of the equipment used.

This does not mean that the popularity is artificial. Not at all.
A brief wave of success could be produced by such means, but the
campaign has served only to introduce stereo to the public. Once
that introduction has been made, stereo continues to advance by
its own merit.

Owners of stereo cameras are familiar with one expression
which, although it varies widely in actual wording, always remains
the same in meaning. It is heard from those who have been
familiar with photography all their lives in the way that every
American is familiar with it; printed reproductions, casual use of
an inexpensive camera, collections of family snapshots. When
such people see their first modern stereo they almost always say
(i) "I never saw anything like it!" (2) ''That's the only kind of
photography I'd be interested in." (3) "Where can I get a camera
to do that kind of work?"

I am not a salesman, and do not even try to interest people in
the purchase of equipment, yet I have "sold" over 100 stereo cam-
eras and almost half that many projectors of which I have defi-
nite record (although one dealer or another made the profit
incident to these sales!). The fact is mentioned only to show that
interest aroused is so deep that those interested are not satisfied
until they, too, are making stereos.

It is sometimes difficult for an oldtimer, steeped in the tradition
of the older stereo, to keep up with the modern pace. The old
rules and laws, except the few which are sound, have been thrown

viii INTRODUCTION

overboard and stereo does not seem to have been harmed by
the fact. Stereo used to be reserved for very carefully made expo-
sures of subjects selected with equal care, the new stereographer
blithely shoots 30 or 60 shots in an afternoon.

All of this has a most important bearing upon the development
of the new art, and certainly it must affect our approach to the
subject in any kind of discussion. Photography is one of the most
conservative of the common arts, foolishly so most of the time.
The writer can remember the senseless struggle of the glass plate
advocates (which means about 99 percent of the serious photog-
raphers, professional and amateur) against the encroachment of
the sheet film. Today we see the stereo Old Guard tearing their
hair and clad in sackcloth and ashes as one after another of their
cherished idols falls under the onslaught of the enthusiastic "new
stereographer." But with this Old Guard numbering considerably
under 1000 and with the new legion growing at about 20 to 30
thousand every year, the newcomers, fortunately for all con-
cerned, are going to have things their own way.

The stereographer was once very proud of his theoretical
knowledge, whether it was accurate or not, and he used stereos-
copy only as a stepping stone to elementary, very elementary
stereogrammetry, and to him "stereo" was far more grammetric
than scopic.

Stereoscopy as practised in stereography is a matter of produc-
ing a desired visual effect. The appearance is the ultimate goal of
the whole process. Therefore, as far as stereography is concerned,
errors, mistakenly called "distortions," if they are invisible simply
do not exist. I recall one enthusiastic young mathematician who
wrote about 20 pages pointing out that a certain stereo tech-
nique was not distortionless as claimed, but involved two serious
distortions. "Of course," he added, "these two distortions neutral-
ize each other and so are not apparent, but they are there as I
have proven!" He simply could not understand that stereo-
scopically, the invisible is non-existent. He was delving into the
realm of stereogrammetry.

Another, this time a stereogrammetric technician, produced an
involved mathematical dissertation proving the inevitability of
distortion produced by the "magnification" of the viewer. He

INTRODUCTION ix

could not understand that in ortho-stereo the final magnification
is unity, that is, it does not exist visually.

Therefore, in this volume we shall deliberately disregard all
of these fascinating mathematical complexities and treat our
subject from the only rational, stereoscopic point of view; that
is, with the visual appearance as the standard. If you can't see it,
it isn't there.

At the same time, most readers will have sufficient curiosity
to want to know "why" now and then. Also a superficial knowl-
edge of stereo theory will often be of assistance in producing bet-
ter work, so a certain amount of theory is necessary. However, it
will be kept as elementary as possible. In any event, just re-
member that you do not really have to know even this simple
theory to enable you to make excellent stereograms.

As a matter of fact the writer personally knows two stereogra-
phers who cannot even load their cameras, but have the dealer do
it, yet they both make perfectly satisfactory stereograms and often
produce really beautiful slides!

The extent to which you delve into theory is a matter of your
own personal choice. But above all remember that stereography
is the easiest type of photography, and it can be mastered with
little effort. In fact it is just as easy to operate the stereo camera
successfully as it is to use the familiar box camera.

Stereography is the automatic photography that produces the
kind of pictures you have always dreamed of but never really ex-
pected to see.

The word stereoscopy is compounded from the Greek stereos
(solid or firm) and skopos (the act of vision). Thus stereoscopy
means the act of seeing * 'sol id," or actually, seeing in all three
dimensions.

Stereoscopic photography is the only medium known by which
the appearance of an object or scene may be reproduced in every
detail so that the image appears to the eyes exactly as did the
original object. At first this does not seem to describe more than
the ordinary photograph until we stop to recall that the best of
conventional photographs falls far short of reproducing "every
detail." The ordinary or planar photograph does not reproduce
definite size or distance, it does not distinctly separate every
plane, it does not definitely exhibit the depth contour of an ob-

x INTRODUCTION

ject. In short, it does not reproduce more than a fraction of the
essential visual detail of any object.

When the stereoscopic photograph is a motion picture in color
and with sound, the picture will often deceive the spectator as
to its reality. One inherent factor of true stereo, or "orthostereo"
as it is called, is that all objects, regardless of the size of the
"print/' are reproduced in their full natural size and at their full
natural distance. Only those who have enjoyed looking at some
well-made stereograms can appreciate the startling realism of
these easily made photographs.

Any camera may be used for making stereographic negatives;
and a simple viewer may be made at home in a few moments by
using two inexpensive magnifying lenses. The negatives involve
nothing more than making two of them instead of one. That is
the sum and substance of elementary stereography. A great deal
of very beautiful and costly apparatus has been produced for
stereographic purposes, but it is not actually essential. These
things, both simple and elaborate, will be discussed each in its
proper place.

While a poor photographic print may pass muster, the poor
stereogram had better not be exhibited. The first requisite is that
the stereographer have the ability to make a really good photo-
graph. More than this, the fundamental aspects of stereography
comprise:

1. Two negatives are made of the same scene but from slightly
separated points of view.

2. The positives must be transposed, exchanging the right and
the left print.

3. The dual print or stereogram is produced by mounting
these two prints so that their bases as determined by the camera,
shall lie in a common line.

It is possible to make the two exposures freehand, but because
the positives must later be aligned according to the relative lens
positions during the two exposures, this method is one for emer-
gency only. The beginner will find it advisable to provide a
tripod with some kind of top which limits the scope and direction
of the camera movement. These may be obtained in elaborate
form, or you can make a simple sliding base in a few moments
from a board and three strips of wood. You will find herein in-

INTRODUCTION xi

structions for making even more elaborate shifting bases which
may be made at home.

Black-and-white stereo films may be processed by any com-
mercial finisher if you choose, but in that case you will have to
do the work of mounting for stereo viewing. On the whole it will
be found more satisfactory if you do all of the processing for
yourself. When color is involved, few amateurs will care to at-
tempt the processing.

Experienced stereographers often dispense with the viewer,
but the beginner will find it extremely difficult to obtain correct
visual fusion without it. The viewer need be nothing more elab-
orate than two of the simple magnifiers obtainable at most nov-
elty and stationery stores. These are attached to one another so
that the centers of the lenses are about two and a quarter inches
apart. Viewed through these lenses, the stereogram will fuse into
a single, solid picture.

Later you will want a better viewer and a better camera, but
these simple pieces of apparatus, to be more fully described later,
are the only actual essentials in addition to normal photographic
equipment.

The opinion has become widespread that some special knowl-
edge or experience is essential for making stereograms. This is
wholly erroneous. Anyone who can make a photograph can make
a stereogram. If his skill is such that he makes a poor photograph
he will make a poor stereogram, and if he can make a good
photograph he can make a good stereogram. The only thing in-
volved here is that stereoscopy enhances the effectiveness of any
photograph, so if it is poor it seems very poor indeed in stereo,
but if it is good then in stereo it becomes excellent.

Curiously enough, the very essence of nonentity in the form of
black shadows and white highlights forms a considerable part of
the modern planar photograph. Such elements seriously detract
from the quality of , the stereogram. Hence stereographic tech-
nique must be such that they are avoided. Ordinarily the stereo-
graphic negative will be better if the normal exposure is increased
and the negative development decreased.

The stereographic composition is not limited to one plane,
therefore classic pictorial composition loses much of its force
when applied to the stereogram, and often that which produces

xii INTRODUCTION

an excellent planar picture simply ruins the characteristic value
of the stereogram. A new form of composition must be used, in
fact it must largely be devised in the future, if we are to have
esthetically pleasing results.

The planar photographer need have no extensive knowledge of
the principles of human vision to produce pleasing pictures, but
the stereographer can never approach the ultimate possibilities
of his art unless he has a working knowledge of the principles of
stereoscopic vision. For that reason, it has been necessary to in-
clude somewhat more theory in this volume than should be found
in the usual photographic reference book. Much of this theory,
however, is subject to highly interesting experimental demon-
stration, so that the process of working through the theoretical
considerations will be found to provide quite as much pleasurable
activity as actually making the stereogram. The theoretical con-
siderations have in so far as possible been incorporated with the
practical procedure. Not only does this make the theory easier
to grasp, but at the same time it simplifies the process of acquiring
stereoscopic skill.

Stereography is now undergoing a revival of popularity, and
we are convinced that once the amateur has experienced the
deep pleasure of seeing his photographs in the incredible realism
of three dimensions, he will thereafter give to stereography his
full enthusiasm.

Three-Dimensional Photography

* ** *

CHAPTER 1

ELEMENTARY STEREOGRAPHY

WHAT Is STEREOGRAPHY? Stereography is a photographic
technique which enables you to realize your dream of
photography reproduction of the original just as you saw it.
You can easily imagine, when looking at a stereogram, that you
are looking through a window at the real scene. Color, size, space,
depth and distance are as clearly perceptible as when viewing
the original.

Is it difficult? It has been repeatedly demonstrated that a be-
ginner, knowing nothing whatsoever about photography, will
have a greater degree of success in stereo than in conventional
photography.

Is any special skill required? Definitely no! Anyone of average
intelligence can begin to make good stereograms without special
knowledge or the use of any special technique.

Is special equipment required? Strictly speaking, no, because
even a box camera may be used for making stereograms. But
practically it is highly advisable to use the regular two-lens stereo
camera which is in reality two cameras built into one body.

Is it expensive? No. The film roll that will provide 20 exposures
for the conventional miniature camera will provide 16 or 17
stereo pairs. The added cost will not exceed 10 percent.

With one camera (Personal), the 20 exposure magazine will
provide 36 or 38 stereo pairs.

Is color film essential? No. Color does add greatly to the value
of the stereogram, just as it does to any photograph. However
stereography has been practised for more than a century, so nat-
urally most of it has been in black and white. You can use any
sensitive media you choose.

What are the limitations? None. The normal stereo negative
may be used for making single enlargements, for example. Stereo
may be applied to photomicro, macro, telephoto, portraits, rec-
ords, figure study, landscape, flowers, and of course to motion
pictures. Stereo may be applied to any form of photography,
not even excepting stroboscopic studies and X-ray.

Color films are as limited in application as any color film.

1

2 THREE-DIMENSIONAL PHOTOGRAPHY

IV hy should stereo be chosen rather than conventional photog-
raphy? Much o this volume is devoted to the answers to this
question, but to sum it up, it may be said that the stereogram
accurately reproduces the original. In the stereogram you see
just what you saw with your naked eyes. In short, in conventional

Fig. 1-1 A. How It Works. An or-
dinary photograph of three posts
set up in a row as seen by the one-
lens camera set directly in front
of the row.

photography you must have years of experience before you can
predict the appearance of the final print; in stereo "What you
see, you get." Stereo never disappoints, it always gives you just
what you expect.

Moreover the utter, incredible realism makes conventional
photographs appear as artificial as a wood engraving.

How is it done? Load the camera, set the exposure, focus with
the rangefinder, sight the subject in the finder, press the button.
Repeat until the film is exposed. When the film is removed it is
factory processed as usual, and if you do not care to mount the
films yourself, they are mounted for you. You drop the stereo-
gram into the viewer (or place it in the projector) and enjoy it.
The headaches have all been removed. There is nothing more
than this that is absolutely essential. You will discover a great
deal that is fascinating in this renewed art, and while you will
never fully master all of its intricacies, you have the solace that
you will be making excellent stereograms right from the start.
That fact seems paradoxical, but it is the paramount reason for
the tremendous rise in stereo popularity during the past two years.

Stereoscopy is based upon the simplest of principles. We have

ELEMENTARY STEREOGRAPHY 3

two eyes which are separated by a distance of 65111111, more or less.
It is obvious that, inasmuch as the two eyes are in different posi-
tions, we have an individual point of view for each eye. It also
follows that the aspect of any scene must be slightly different from
these two points of view, even though the amount of difference
is so slight as to be hardly noticeable.

If two photographs are made from two points of view separated
by a distance similar to that between the eyes, two photographs
will be obtained, and these two will respectively correspond to
the visual image of each of the two eyes. If these two images are
then viewed in such a manner that each eye shall see its own
image and not the other, then we have the conditions of direct
stereoscopic vision duplicated, and instead of seeing the two
images as such, we shall see the single, three-dimensional image
of normal direct vision.

Fig. 1-1B. How It Works. The same three posts as seen by a
stereo camera. Note that the two lenses each being at one side,
show all three posts, but they "see" the two opposite sides of
the row. Viewed in a stereoscope, the posts are seen in a straight
line y but in depth.

Our problem then is two-fold. We must make the two photo-
graphs from separate points of view, and we must so look at these
pictures that each eye sees only its corresponding picture. To that
end the whole art of stereography is directed.

There are refinements of technique, and there are limitations
which are rigidly enforced under certain circumstances, but for

4 THRl-.K-m.MENSIOXAL PHOTOGRAPHY

the present we shall ignore them and proceed to make the simple
stcreogram.

THE SIMPLE CAMERA. A stereoscopic camera or a stereoscopic
adapter for a standard camera serves but one purpose, that of
making the two exposures simultaneously. This permits stereo-
grams to be made of moving objects, but except for that purpose,
any camera, including the simplest box type, may successfully be
used for stereography. If you have a camera which you ordinarily
use, you can make stereograms of motionless objects with it.

THE POINTS OF VIEW. Theoretically, the two exposures should
be made with the optical axes parallel and with a distance of
65111111 between the points of view. We shall see that even in the
best of technical stereography these factors are often ignored, but
for the present we shall accept thefn as a working standard. In
this connection it is interesting to note that the interpupillary
distance varies with individuals, and that while 65mm has long
been the accepted standard in the metric system (with both 6omm
and 65mm as occasional competitors), 2i/ inches is the accepted
standard in U. S. measurement. Therefore, the two exposures
will be made by moving the camera approximately 21/4 inches
between exposures.

CONTINUITY OF BASE. The open frame in the rear of the cam-
era against which the film lies and which frames the negative, has
a position relative to the lens axis wiaich is mechanically fixed.
This means that the optical center of the image is always the same
distance above this frame baseline.

It is essential that in the completed stereogram the common
base of the two positives must be parallel to this camera base, and
that the two exposures must be made under the same conditions.

Thus if the two exposures are made freehand, as is not un-
commonly done by an expert stereographer, one of them with a
slight tilt to the right and the other with a similar tilt to the left,
there is no original continuity of base, and the prints must be
aligned by stereoscopic analysis, a task for the experienced stere-
ographer. Therefore, while freehand exposures are often made
successfully by the experienced stereographer, the beginner will
almost inevitably be disappointed if he tries it.

THE STEREO SHIFT-HEAD. To avoid the condition just de-
scribed, some kind of stereo tripod adapter or shift-head should

ELEMENTARY STEREOGRAPHY

be used. The simplest consists of a board with a raised border or
frame around three sides. (Fig. 1-2). Let us assume that the cam-
era to be used is a box camera three inches wide and five inches

D

BOX CAMERA

BASE FOR FOLD-
ING CAMERA

SLIDING

BLOCK

LEDGE*

O

BASE

"

FRONT VIEW

Fig. 12. Simple sliding base for making stereogrnm-s with any

camera.

long. A board o this size would just be covered by the camera. It
is desired that the camera be moved sidewise 21/0 inches. If we add
this to the width of three inches, we have 51^ inches, so the base-
board will be cut 5x514 inches. This board should be from a half
to three-quarters inch thick.

Along the two five-inch sides and along one 51^ -inch side, thin
strips of wood are nailed, rising perhaps an inch above the top
of the board.

It is of the greatest importance that the baseboard be cut ab-
solutely square, and that the two ends be accurately parallel.
Perhaps it should be stated that these terms of accuracy are
intended for the amateur workman. An error of one degree,
while undesirable, is not too serious.

This board may be provided with a screw socket so that it may
be mounted upon a tripod. In any event, some means should be
provided for holding the board motionless during the two expo-
sures. Otherwise the care spent in making the board accurately
will be wasted.

MAKING THE EXPOSURE. The board is attached to the tripod
and the camera aligned as in making an ordinary photograph,

6 THREE-DIMENSIONAL PHOTOGRAPHY

but care is taken to make sure the camera rests solidly against
one of the side rails and against the rear rail. In other words, the
camera is pushed into one corner of the rails. The exposure is
made as usual, erring if at all upon the side of overexposure.
The camera is then pushed sidewise until it rests firmly in the
opposite corner, and the second exposure is made.

As the negatives must later be cut apart and rejoined to estab-
lish the correct spacing, it is immaterial which exposure is
made first.

OTHER TYPES OF BASE. Stereo shifting bases are obtainable
from many dealers. They may be the simple sliding type, as for

Fig. 1-5. A Rolleiflex camera mounted up-
on a Rolleiflex stereo slide bar for making
stereo grams by successive exposure.

example that sold for use with the Rolleiflex camera (Fig. 1-3),
or it may be the parallel-arm type made popular in Germany.
An examination of the latter shows its construction so clearly

ELEMENTARY STEREOGRAPHY 7

that no further description is necessary. This type may also be
made at home, although more care is needed than in the construc-
tion of the board type.

STEREO REFLECTORS. The disadvantage of all '"successive"
types of stereo exposure is that the object must be quite motion-
less during the whole interval of both exposures. Because even
the slightest sway resulting from a breeze will effectually ruin the
result, this method is hardly practical for most exterior subjects.

The alternative of a stereoscopic camera is not alway feasible
because of the necessarily greater cost of such instruments. To '
meet this need, the stereo reflector was invented by F. A. P.
Barnard in 1853.

The stereo reflector is an arrangement of four reflecting sur-
faces which are set before the lens of any standard camera. This
divides the negative area into two portions, each of which receives
one of the stereo images. In the cheaper forms, the device is made
up of four mirrors, while the more elaborate types have two
prisms, each of which provides two reflecting surfaces. The prism-
type is fixed for one set of optical factors as embodied with some
given lens and camera, while the mirror-type may be so designed
that it may be adjusted for use with a variety of cameras and
lenses.

The Leica Stereoly is an excellent example of the prism re-
flector designed for use with the Leica camera, but easily adapted
to any miniature camera using a two-inch lens. The Stereotach
(Fig. 1-4) is an example of the mirror reflector of fixed adjust-
ment. Unfortunately there is no adjustable reflector available at
the present writing, although the Rexo stereo adapter which had
a full range of adjustment was sold about 1915.

These devices are attached to the camera lens in much the same
way that a filter or sunshade is attached. They require some
increase in exposure, the exact amount depending upon the
design of the reflector used. Ordinarily, a three- or four-times
exposure is about right. Care should be taken that the mirror
division is exactly vertical and that it exactly bisects the lens.
Other than this, operation is normal and the results are highly
effective.

There has been a great deal of criticism offered regarding
these attachments. Tests of all types mentioned have indicated

THREE-DIMENSIONAL PHOTOGRAPHY

, '^,;/ ,.;. T

^^Kv^o"#.

F/g. 14. For instantaneous stereograms, the Ster-
eotach is applied to the Polaroid-Land camera.

i

that when properly used, the results are satisfactory. Inasmuch
as the negatives may be printed or color films viewed just as they
come from the developing laboratory, without transposing, they
provide an ideal equipment for the beginner in stereography. It
is only fair to add, however, that the resulting quality is not as
good as that obtained with a conventional stereo camera.

When a high precision reflector is used, such as that advised
for motion picture work, the quality is better. But as such a
reflector costs as much or more than a good stereo camera, its
use is pointless for still work.

PHOTOTECHNICAL CONSIDERATIONS. (BLACK-AND-WHITE). In
making the exposure, give full attention to the brightness range
of the subject and do not attempt to make an exposure without
compensation if this range is excessive. It is desirable that the
meter be used to obtain a differential reading, and that the ex-
posure given be that demanded by the shadow areas. This is no
longer a question of the shadows "which contain desirable detail/'
but of the deepest shadow included, regardless of whether it is
important or not. Any shadow lacking detail will later be seen as

ELEMENTARY STEREOGRAPHY 9

a formless smudge. Therefore, give that exposure demanded by
the deepest included shadow of significant size.

This exposure will very often be from two to three times the
average exposure, (which is in addition to any factor used because
of the reflector, if used) and the differential reading will show
that the highlights are being seriously overexposed. It \vill be
necessary to compensate during development, often to the extent
of using only half the normal time. In stereography the best re-
sults are obtainable only by going back to the old principle of ex-
posing for the shadows and developing for the highlights.

It is difficult to imagine an exterior scene which can be stereo-
graphed successfully without a filter. An extensive series of filters
is not needed, but the set should contain the medium or dark
yellow, yellow-green, medium orange and tri-color red. Of these
the medium orange will probably be the most often used.

DEFINITION. As a rule the stereogram should be visually sharp
throughout. Any obvious unsharpness, or hint of softness, is
usually unacceptable.

It has been repeatedly suggested that the stereo depth be en-
hanced by softening the definition of the background. Such a
suggestion indicates a lack of knowledge of the most elementary
stereoscopic principles.

The stereogram duplicates in every way the appearance of the
original. When this stereogram is viewed, assuming it to be
needle sharp throughout, the eyes see sharply only that object
upon which they are focused, and as the vision shifts from object
to object, other objects in the field of vision take on exactly the
degree of unsharpness they would have in real life. This fact is
one of the fundamental principles of stereography. Therefore, any
attempt further to simulate the characteristics of direct vision only
introduces an unnatural effect and a loss of stereoscopic acuity.

It is therefore advisable always to use the smallest aperture
compatible with other essential factors. When a large aperture is
essential, try as far as possible to avoid any great depth in the
field, and focus carefully to include as many planes as possible
within the field depth.

MOVING OBJECTS. This, of course, refers only to the use of the
simultaneous exposure either with reflector or double camera.
Strangely enough, a slight blur caused by the moving object is

10 THREE-DIMENSIONAL PHOTOGRAPHY

less objectionable than loss of definition by poor focus. It is desir-
able to avoid all blur, of course, but if conditions necessitate a
compromise between small aperture and motion-stopping expo-
sure, it is advisable to favor the small aperture even at the cost of
a slight blur of motion.

On the other hand "follow shots/' in which the exposure is
made while swinging the camera to keep up with a moving ob-
ject, are often surprisingly successful. If the object is sharp and
the background is sufficiently blurred to rob it of identity, the
stereogram may be very good indeed. So if you blur the back-
ground, blur it a lot!

INTERRELATION OF FACTORS. Field depth demands a small
aperture. Motion demands fast shutter speed. Shadow detail
demands full exposure. Exterior scenes demand the use of a
filter. All of these combine to limit the possible exposure some-
what rigidly. Often compromises must be made. The first thing
is to use a lighter filter, or perhaps no filter, although this is
undesirable. Next sacrifice some speed even at the expense of
motion blur. Then manipulate the aperture, even to losing defi-
nition in the far background. Finally, rather than sacrifice shadow
detail, either go to a faster film or try the same subject later
when the light is more favorable. \Vith experience you w r ill learn
to compromise these factors so that with hardly any discernible
loss you will find it possible to make a good stereogram of any
normal subject.

When color film is used, its slow speed further complicates the
problem.

EMULSIONS. In stereography, differences which are so subtle
as to escape notice altogether in planar photography, become
painfully obvious. Therefore, it is desirable that the positives
have the greatest possible photographic quality. In fact, real suc-
cess will never be achieved by the amateur until he has learned
to make good transparencies. The print should be brilliant and
full toned, with detail in the darkest shadows and the brightest
highlights.

Ordinarily, while our most sensitive emulsions are perfectly
satisfactory for routine planar photography, it will be found that
emulsions of somewhat less than the greatest sensitivity will give
the best satisfaction. The ideal emulsion is a slow, fine-grained

ELEMENTARY STEREOGRAPHY 11

panchromatic, but because of the antagonistic factors we have
already discussed, a moderately high sensitivity is often very valu-
able in stereography. The popular panchromatic emulsions rated
at approximately ASA 50 are perhaps the best all-around stereo
emulsions. It is to be understood that like all generalizations, this
is subject to variation for any of the many special types of sub-
jects constantly encountered in stereoscopic work.

Of course the ideal material is color film. There is much to
be said in favor of monochrome planar pictures, nothing in
favor of monochrome stereo because color is an active stereo
factor aside from its esthetic weight.

When the 35mm camera is used, the shorter focal length makes
it possible to use larger apertures. Therefore the slower speed of
color film loses its significance, and in such miniature cameras
stereography with color film becomes as easy as, or in fact easier
than, black and white with a larger camera.

FACTORS OF DEPTH PERCEPTION. There are several factors
involved in our sense of visual depth perception, all of which
can be used to good advantage in stereography, but as these be-
long to the division of advanced work, we shall leave a discussion
of them for a later chapter.

CHAPTKR 2
STEREOSCOPIC CAMERAS

Two OR THREE YEARS AGO any discussion of stereoscopic cam-
eras necessarily included a number of obsolete models simply
because there were so very few current models available, and
among these only one of modern type. Today the situation is
wholly different. The cameras follow the modern trend of stereo-
scopic technique, namely, the use of 35111111, natural color film.
Of course there are hundreds of larger cameras in use, hundreds
which make black and white prints as a rule. Obviously any
camera can be used for color, but in practise the 35111111 camera
is largely devoted to color work; larger sizes to monochrome.

Nothing new has yet been introduced without meeting a host
of objections, a few justifiable, but most of them made simply
because the product is new\ The miniature camera had a difficult
time for years, but that precedent did little to win favor for the
35mm stereo camera, although the latter involves far less actual
change of existing conditions than did the former.

In stereo, the vital viewing factor is visual angle the angle
from the center of the pupil to the extreme corners of the positive
image. It is obvious that if a smaller picture is viewed nearer the
eyes, the same angle is filled; and assuming, as we may, that the
image structure (grain and the like) is such that the image defi-
nition is retained, there is no limit to the smallness of the image
which is satisfactory. The sole limitation is image breakdown.
This is more than being merely satisfactory. If the viewers are
masked so that only the lenses show, there is nothing to indicate
size. Under such conditions, spectators who have volunteered
for the experiment repeatedly state a 3x6 inch or 6x1 gem trans-
parency to be the smallest of a group which includes both 35111111
and i6mm films. Size per se offers no advantages, no disadvan-
tages in stereo. Naturally the one who uses some particular size
will rise in arms to defend his favorite. That is ahvays true in
photography, but disinterested experimenters have demonstrated
the impossibility of distinguishing among sizes by viewing the
images in normal viewers. The apparent size depends upon the
sole factor of visual angle.

STEREOSCOPIC CAMERAS 13

This point has been discussed at length simply because it often
becomes the most important consideration when buying a stereo
camera, when in fact, it is the least important. Certainly the
smaller the picture the cheaper it is to make. For example, if
you take a 20 exposure magazine of film, depending upon the
camera you use, you may obtain 10, 12, 15, 16 or 38 exposures
upon that film. If the operating cost is not a factor, then there
is a second one, namely, that at the present time there are certain
accessories which are available for the original 35mm size which
are not as yet available for some others.

For example, rigid cardboard mounts are available for the
five-perforation frame width, so wider films must be cut down
for use in such mounts. In the case of the European models this
amounts to a considerable loss of film area.

So the intending purchaser should study the salient points
of each camera available, and make his choice by deciding which
best meets his specific requirements. Among the three domestic
cameras offered, the choice will be so guided because in the matter
of results all three will afford performance beyond criticism.
The European cameras include the most costly of all, the Vera-
scope, which is truly de luxe, and the Continental wide frame
with the traditional French or German design details will appeal
to many users who are accustomed to Eureopean cameras.

Because of the wide use of 35mm (at least 100 to one), we
shall limit our description of "the larger cameras to the one which
is the prototype of all de luxe stereo cameras and which will serve
every need of the photographer who prefers the larger format.

There is no matter of preference involved in the order in which
the cameras are described. It is convenient to start with that mak-
ing the smallest picture, after which logical transition indicates
the order in which the remaining cameras are discussed.

PERSONAL CAMERA (Sawyer's Incorporated). This camera has
been introduced by the makers of the widely known "View-
master," and the finished films are mounted in the familiar
"reels" of seven pairs each, for viewing in the Viewmaster. It
is only fair to point out that inasmuch as the Personal films are
direct camera images, they are even more satisfactory than the
commercial reels which are, of necessity, copies.

The Personal camera is particularly attractive to those who are

14

THREE-DIMENSIONAL PHOTOGRAPHY

Fig. 2-1. The Sawyer Personal Camera.

unfamiliar with photographic technique, because it needs no
exposure meter, no calculating, no bother. The exposure con-
trol has been integrated with the camera controls so cleverly that
all you have to do is to set one dial to the speed you wish to use
and a second one to the light conditions, then press the button.
To those who are skeptical about the practicability of such a
device, I will only say that I made more than 100 exposures fol-
lowing the system and without benefit of meter and then gave
the camera to a novice who exposed some 70 frames. The nov-
ice's films exhibited a better average of good results than the
usual amateur photographer of normal competence will obtain
in conventional photography. I obtained 96 percent with an
additional 2 percent within consideration.

Unlike most cameras, the whole unit is self contained. The
lenses lie within the body, and the filters (when used) lie in the
body wall, retained by the standard Series V retaining ring. The
body is oblong with rounded ends, 6 inches long, 314 high and
2 thick. It weighs, loaded, one pound, nine ounces. The finish
is black and chrome, and is provided if desired with a good
eveready case.

STEREOSCOPIC CAMERAS 15

The lenses are of only 25mm focal length, a fact which makes
it possible to dispense with focusing adjustment. Satisfactorily
sharp pictures at a distance of 28 inches have been made, indi-
cating the entire practicality of the fixed-focus lens when the
focus is so short. The lenses have a maximum aperture of f/3-5
and are coated.

The shutter is of the time proven guillotine type which cannot

possibly get out of synchronization and is speeded 1/10 to 1/100

with Bulb. A flash synchronizer for five millisecond delay

(SF/S3M bulbs and most portable strobe units) is built in and

couples for all speeds.

The picture size is i^xigmm upon a 12x1 8mm film. The extra
6mm in length provides a tab which is indexed in the camera
to indicate right or left and also forms a "handle" so that the
film surface never need be touched in mounting. A 2O-exposure
magazine of standard 35mm film provides 38 stereo pairs. The
method of adapting the so-called i6mm size to the 35mm film
is unique. During the first 19 exposures, the lenses are opposite
the top half of the film and the exposures extend to the midline
of the film. When 19 exposures have been made, a dial on the
front of the camera is turned. Then the lenses are opposite the
lower side of the film, the film travel is reversed and 19 more
exposures are made along the bottom of the film. When the last
exposure has been made the film is already rewound ready to be
removed from the camera.

The release is located at the lower, right corner of the camera
(in operating position), and is pressed back into the camera.
This provides a natural squeeze motion which is unlikely to
produce camera bounce.

To make mounting easy, a special film punch is provided
which punches out both films simultaneously, thus providing
mechanically positive alignment between the two films. The
camera apertures have small notches in one edge. One is semi-
circular, the other square. These are the right-and-left indi-
cators. The blank reels have 14 pockets. One is numbered i, with
a semi-circle, the opposite one is numbered i with a square mark,
and so on for seven pairs. You do not even need to know which
is right or left. You simply take the punched film which has the
circular mark and place it in the pocket bearing the same mark.

1C THREE-DIMENSIONAL PHOTOGRAPHY

The companion picture with the square mark is placed in the
pocket bearing the same number but with the square index, and
the pictures are mounted correctly. The punch and the reels are
precision made so that correct alignment is assured with no pos-
sibility of obliquity.

The final measure of a camera, however, is the quality of re-
sult. In making well over 1000 exposures with the Personal
camera, over a period of alm6st a year, the writer found it to be
wholly satisfactory in every respect. For those who distrust their
ability to cope with the intricacies of photographic exposure, the
simplicity of this camera will prove most attractive. No focusing,
no rangefinder, no exposure meter. Preset one dial which is
rarely disturbed; set the other and shoot. It is as simple as that
yet quality has not been sacrificed to simplicity.

The finder is a direct optical finder. No rangefinder is used
with the fixed-focus lenses of course. The film transport and shut-
ter setting are linked to prevent double exposure.

STEREO VIVID CAMERA. This camera is sold by the makers of
the widely known Stereo Vivid projector. A glance at the two
illustrations will show a certain similarity between this camera
and the Personal. The same foolproof exposure control is used
on both cameras, while the built-in shutter and "squeeze" release
upon the front panel are common to both. In the Stereo Vivid,
the shutter is placed behind the lenses, instead of in front of
them, and lies almost in the focal plane. The shutter has the same
positive synchronization as the true guillotine shutter.

The Stereo Vivid is a 5.5-perforation camera. That is the film
is advanced 5.5 perforations for each picture instead of the five
normally used. This gives a picture roughly two millimeters
wider than the five-perforation advance, or a bit more. This gives
one less picture for each so-exposure magazine, but provides a
picture which is fully square instead of being somewhat narrower
than its height. This is a feature which will appeal to many stere-
ographers who have turned to the extra wide European frame to
avoid having a picture narrower than its height. This is a matter
of personal choice. There is no basis for saying that any of the
three is "best," but it does make it possible for the purchaser to
choose among three picture widths.

STEREOSCOPIC CAMERAS

17

Fig. 2-2. The Stereo Vivid Camera.

The lenses and filters are set into the wall of the camera so
there are no protrusions other than the filter retaining rings.
The exposure dials and film winding knobs are aligned to pro-
vide a smooth, almost flat top to the camera so it has remarkably
few projections.

Focusing is provided by moving the film aperture within the
camera, giving increased steadiness to the lens mounts, a desir-
able feature. The focusing knob revolves in a horizontal plane
upon a vertical shaft and is built back into the body w r ith only
an operating edge protruding. Focus is controlled by a range-
finder built into the top of the camera, immediately beneath the
exposure dials. Both finder and rangefinder have the same eye-
piece so shifting is avoided. The rangefinder windows lie at
approximately the same distance apart as the lenses so the range-
finder parallax is within the stereo parallax, while the finder
itself shows a field which lies within the stereo field. A level is
visible in the finder w r hen in use.

18 THREE-DIMENSIONAL PHOTOGRAPHY

The focusing dial Incorporates an easily read depth of field
scale, a feature of importance in all stereo work.

The coated lenses of f/g.s aperture and 35mm focus are cou-
pled with a shutter with automatic speeds 1/5 to 1/150. This
range is more than adequate for most stereo work.

The camera measures 614x31/^x214 inches and is provided
with neck cord loops and standard tripod socket.

Note; Neither the Personal nor the Stereo Vivid is available
at the time this is written, but it is planned to have both cameras
ready for the market in well under a year from the date of the
publication of this book (1951). In fact the Personal may be
available very shortly after our publication date.

STEREO REALIST CAMERA. This is the original, American, 35mm
stereo camera and in fact the only one to be described in the first
edition of this book. It is more conventional than the two models
just described in that it has the usual stereo coupled blade shut-

Fig. 2-3. The Stereo Realist Camera.

ters, and the shutter setting is independent of the film transport
so the camera may be used for space control, hyperstereo or (by
capping one lens) for single color exposures if desired.

The camera measures 694x21/6x2 inches exclusive of sunshades,
It is finished in black with chrome satin trim.

The two 35mm, 1/3.5 coated lenses are rangefinder controlled.
The finder is a brilliant optical finder with its objective set mid-

STEREOSCOPIC CAMERAS 19

way between the taking lenses. Both the rangefmder and finder
eye-lenses are set at the bottom of the rear side of the camera so
that the camera may be rested squarely against the forehead in
making the exposure. The actual focusing is achieved by moving
the film aperture within the camera, which results in greater
steadiness for the lens mounts.

The shutter has automatic speeds from one second to 1/150
and has Time and Bulb in addition. The timing ring forms a
rim for the central finder lens.

On top of the camera there is a small aperture under which a
red signal appears when the shutter is tripped, thus warning the
user that the film in the aperture has been exposed. This effec-
tively prevents double exposure despite the fact that the shutter
is not coupled to the film transport.

The Realist is a five-perforation camera, giving about 16 ex-
posures to a so-exposure magazine. It will be understood that
the actual number of exposures depends upon the care with
which the film leader is wound on. On the other hand too much
crowding often results in the loss of a frame in processing, so it
is best to be content with a smaller number of exposures and to
be sure of getting all of them safely back from the laboratory.

There are many accessories available for the Realist, and most
makers of mounts and the like have worked to the five-perforation
size up to the present time. These accessories include: flashgun,
for which the camera provides synchronization, sunshades to
take Series V filters, eveready case and the like. There is also a
complete mounting outfit which includes a film cutter, distribu-
tion box, mounting jig and heater for the irons which heat-seal
the films to the prepared masks. This makes stereo mounting
very easy.

Until the present year, the Realist has been the only domestic
35mm stereo camera available, and without it we should not
have had the present tremendous revival of interest in stereo
photography. \Vhen the camera was described in the first edition
of this book it was a new and untried instrument. Since then
many thousands of users have found it completely satisfactory.
The writer has used his first Realist for many thousands of expo-
sures and it has not failed once, a record of which any camera
should be proud.

jo I HR1- ^DIMENSIONAL PHOTOGR VPHY

Now before we go on to the imported cameras, there is one
inevitable question which must be answered here. There are now
three high-quality domestic cameras available. Which is the best?

The one which you choose and use.

Look at all three, handle them, examine their features. Then
choose the one which appeals to you the most. All will give excel-
lent results. No matter which you buy, by the time you have given
it a thorough tryout you will be ready to swear that it is by all
odds the best stereo camera ever made. The writer has used a
Realist for some three years and a Personal for almost a year,
and has no criticism to offer regarding either. Both are excellent
cameras. No doubt the same could be said of the Stereo Vivid
were it not for the simple fact that no advance model is as yet
available for test. By reason of its similarity in general design to
the Personal, the writer fully expects it to share the same high
quality of the two cameras which he has actually used over rela-
tively long periods of time.

Interlocking Images. Naturally, unless a great deal of film is
to be wasted, the stereo pairs cannot be kept separate when tak-
ing. In one sequence as used by Realist and Personal cameras,
they are intermingled in such a way that the film cannot be cut
at any point without dividing a pair. This is of no consequence
when the images are to be individually mounted, but for viewers
using strip films, such procedure means the loss of one picture
at each cut. It is also used in the Vivid.

This interlock can be easily understood when the "lefts" and
"rights" are considered separately as shown here:
Left Exposures, 12 345 678 9
Right Exposures, 012 345 67
Interlock

Sequence, 1-02-1-32-4-3-5-4-6-57-6-87-9

This normal interlock used in the three cameras just described,
is based upon the so-called normal sequence in which there are
two images between each stereo pair. European cameras includ-
ing the Iloca and the Verascope 40 make wider images and have
but one image between each pair, so their interlock sequence is:
i_2-i-2* 3-4-3-4* 5-6-5-6* 7-8-7-8* etc. It will be noted that
this film may be cut at any point marked * without injuring a
pair. This interlock is accomplished by having a film transport

STEREOSCOPIC CAMERAS 21

which moves the film a longer distance at one winding and a
shorter distance at the next.

VERASCOPE 1^40. This is a French camera made by the firm of
Jules Richard of Paris. The name of Richard has long been
known as a hallmark of quality in stereo equipment, but this is
their first model made in accord with modern miniature camera

Fig. 2-4. Busch Verascope F-40.

design. It was first imported sporadically, but recently the Busch
Camera Corporation started to import them as exclusive U. S.
distributors. The camera, while conforming to basic 35mm stereo
camera design has several unique features.

The camera is of the usual long shape, covered with black
leather with satin chrome top and bottom plates. The Busch
models have U. S, standard tripod sockets and are calibrated in
feet, while the original imports had the Continental tripod socket
and were calibrated according to the metric system. In giving this
description we shall deal only with the Busch imports. Any dis-
crepancy regarding Verascopes known to you who read this may
be attributed to the changes made for the U. S. market.

The film used is the standard jsmm cartridge as made for all

22 THREE-DIMENSIONAL PHOTOGRAPHY

miniature cameras, black and white or color. Loading and opera-
tion are similar to other 35111111 cameras.

The lenses are 40111111 focal length, f 3.5 aperture, coated. In
tests made in the Stereo Guild laboratories, definition was found
to be excellent yielding a good, crisp image. Focus is controlled
by coupled rangefinder. The control is a vertical disc set at the
top-rear of the camera. This dial has an engraved depth-of-field
scale for all apertures down to f 16.

The shutter is the usual French guillotine type, giving perma-
nent, positive synchronization and with automatic speeds from
one second to i '250 with Time and Bulb. The lenses and shut-
ter are mounted upon a movable lens board of the conventional
type which moves with the focusing operation.

The rangefinder and finder are located in the top plate of the
camera. The release is at the top-right-front of the body, and the
button is threaded for cable release. Just at the left of the lens
board are the two contact sockets for the internal flash synchro-
nizer.

So much is more or less conventional, but there are some novel
features.

The picture width is 30111111, although the height is the 24mm
normal to 35mm film. This format is directly related to two other
novel features. This width is considerably greater than the 22.5
and 24 of our domestic cameras. The choice, however, is a mat-
ter of personal preference.

This picture width makes it impossible to use the original inter-
locked sequence, so there is only one odd picture between pairs,
so the sequence is: 1-2-1-2*3-4-3-4*5-6-5-6*7-8-7-8* and so on. It
will be seen that the film may be cut at any point marked (*)
without interrupting a pair. This type of sequence makes neces-
sary a film advance which is long-short-long-short and so on, that
is the actual film winding knob is turned about 1/4 revolution to
wind the film, and the next film has to be wound % revolution
and so on.

The fundamental reason for the picture width is the fact that
this camera serves for both single and stereo exposures. By turn-
ing a small knob on top of the camera, one film aperture is ob-
scured and the film transport altered to a single frame advance.

STEREOSCOPIC; CAMERAS 123

To make the single picture operation comparable to that of stand-
ard miniature cameras (3601111 width), the frame width was made
a compromise between the normal stereo and the normal single
frame sizes. If, for any reason, it should be desired to make single
pictures, this feature enables one camera to serve both purposes.

This duality is reflected in the counter. The exposure counter
indicates single frames exposed. That is, it advances two units
for each stereo exposure made. It will be seen that all of the novel
features are therefore closely related to one another.

Accessories are being introduced for this camera. Case, filters,
sunshades and the like are now available, as well as flash.

Design, appearance and performance are those which might be
expected from a factory which has produced de luxe stereo equip-
ment for many years.

ILOCA CAMERA. The Iloca is the current German version of
the 35mm stereo. Its general appearance may be visualized by
imagining one of the original Leica models lengthened and with
a second lens added except that between-lens shutters are used.

The camera takes the standard 35mm film magazine, as do all
the cameras described before. Loading is done by removing the
bottom plate, just as in the Leica. The lenses are set off from the
relatively thin body by lens tubes, and carry the Prontor shutters
at the ends of these tubes in a manner familiar in many miniature
cameras using this type of shutter.

The camera is 6%x254xs% inches in size. It weighs ii/ 2
pounds loaded. The finish is black leather and chrome satin.

The lenses are brown coated, 45mm focal length, f/3-5- Focus-
ing is done by separating the elements, a familiar German design,
but as the two lenses are not linked for focus, an error might be
made. Iris and speed are linked, with a top speed of 1/300 plus
built-in flash synchronization. Focusing is by scale as there is
no rangefinder.

As we go to press we are informed that in later models the
lenses' focusing is connected. We are also told that this camera
is to be brought out in a stereo-reflex model.

The pictures are the usual wide format of the European film
and in both size and interlock resemble the Verascope.

Although the camera is of simplified design, lacking some
refinements w r hich we usually desire in a de luxe camera, the one

2J THRKE-DIMENSIONAL PHOTOGRAPHY

we tried out gave very good results, comparable to the work pro-
duced by other de luxe stereo cameras.

Although of unfamiliar name, the moderate cost and the Con-
tinental si/e will undoubtedly appeal to many who like the wide
frame but who cannot afford to invest the rather considerable
amount necessary for the purchase of a Verascope.

These five cameras have been described in detail because this
si/e is relatively new in the stereo field, although rapidly becom-
ing popular. The five represent the American, the French, and
the German designs, in the 35mm field.

CONVENTIONAL SMALL CAMERAS. There have been literally
hundreds of models of stereoscopic cameras offered from time
to time, starting with the old Daguerrotype cameras and paral-
leling monocular design since that era. It would be useless to
try to describe all of them, so we shall limit our discussion to
relatively modern designs; to designs of some degree of origi-
nality; and to designs which are still available either new or used.
We are informed that the well-known Rolleiflex line is once
more available, and there are promises of additional domestic
cameras, but as details are not available at this time we must
pass over them.

FOLDING CAMERAS. Many of the well-known makes of folding
roll-film cameras were made at one time or another in the stereo
type. There was once a Stereo Kodak made in this country, also
the Stereo Hawkeye; while among the imported cameras, Ihagee
and the lea Ideal were both made in stereo models. These cam-
eras were typical of the cameras most familiar to us as amateur
film and filmpack models.

The press type of camera was also made in stereo. This camera
is typically a thin, flat instrument when folded and extends, not
by means of a hinged door, but by pulling the lensboard straight
out upon lazy-tongs supports. In this group the better known
stereo cameras included Ernemann, Zeiss (one model), Makina,
and Mentor. These cameras are superior to the conventional fold-
ing type because the lensboard is supported top and bottom, but
the ideal stereo camera is of rigid construction. A slight displace-
ment of the lens in a monocular camera is of little importance,
but in the stereo camera such displacement, because of the mov-

SI EREOSCOPIC GAMER \S 1>5

able lensboard, too often results in unequal displacement of the
two lenses, and the essential identity of base is destroyed. (Gen-
erally obsolete).

REFLEX CAMERAS. The reflecting principle has been popular
because the body is rigid, and the lens extension so short and so
sturdily built that there is far less danger of lens displacement
than in the folding types. One of the best of this type of camera,
although it does not belong to the small camera group, was the
Stereo Graflex using a 5x7 plate upon which the tw r o stereo units
were recorded in size suitable for direct printing in the American
3x6 inch paper print.

THREE-LENS CAMERA. These cameras are often said to follow
the Rolleiflex model, but that is reverse order. The Heidoscope
was originally built as a rigid, all metal, three-lens camera, the
center lens being the finder lens co-acting with a reflecting mirror
set in the center of the top of the camera. This camera proved so
successful that the manufacturers later brought out a tw r in-lens
monocular-type camera which they named the Rolleiflex.

The same design was adopted for the Yoigtlander Stereofleck-
toscope. Both the Heidoscope and the Stereoflecktoscope were
made in both the normal sizes.

ROLLEIDOSCOPE CAMERA. In this country we may regard the
plate cameras as obsolete, but the rollfilm version of the Heido-
scope, known as the Rolleidoscope is strictly modern and has all
of the quality and attractiveness of the plate model Heidoscope.
If you -are in the market for a 6x13 camera, you will do well to
give the Rolleidoscope very careful consideration. There is no
better camera made in this size.

The construction is conventional. Just imagine two Rolleiflex
cameras built side by side, with only one finder lens and that
situated midway between the two taking lenses instead of above.
The reflex focusing-and-fmding system is the same as in the
Rolleiflex.

The camera makes five stereo pairs upon a standard roll of
120 film, the negatives corresponding to the 6x13 stereo size.
The shutter is a Stereo-Compur of the usual set type. Focusing
and speed are controlled by knobs at the two ends of the lens-
board, and a moderate amount of rising front is provided. The
lenses are f /4-5 Tessars, 75mm focal length.

2t> THREE-DIMENSIONAL PHOTOGRAPHY

The general construction is the rigid box so desirable in stereo
cameras, but the camera is not inconveniently bulky. The carry-
ing case measures only 4x5x7.5 inches over-all while the camera
itself measures only 3.5x4x7 inches.

For color work, Ektaclirome 120 or Anscocolor provides excel-
lent results. For making stereos for reproduction either by normal
side-by-side halftone or anaglyph the camera is unexcelled.

REFLECTORS. At the present time the domestic market offers
the Stereotach, a fixed-adjustment reflector which may be used
with appropriate single-lens cameras for the production of satis-
factory stereograms. For the amateur interested only in the pro-
duction of a casual stereogram now and then, and for the
beginner who wishes to try out stereo, these reflectors are excel-
lent. The results are perfectly satisfactory within the limitations
of the instrument, but of course the serious stereographer will
wish to work with the double camera.

The stereo reflector does serve one purpose which cannot be
served in any other way at present. This is the production of in-
stantaneous pictures. The Stereotach does very well when used
in conjunction with the Polaroid Land camera, and makes pos-
sible the production of good stereograms in monochrome within
one minute after the exposure.

There are many technical applications particularly in the realm
of forensic photography, when this rapid availability is of great
value. Because stereo has met with such instant approval in the
forensic field, this special technique should not be overlooked.

Many years ago there was an adjustable reflector on the market
which was excellent in that it could be adapted instantly to any
negative size and any focal-length lens. Such reflectors were very
convenient, for while the stereographer will own and use the
double camera, there are many times when only the single-lens
camera is available, and if the reflector is carried habitually, it
will save the day when a good stereo subject presents itself.

It would be a mistake to say that the reflector replaces the
stereo camera, but at the same time every stereographer who owns
and uses a single-lens camera as well as those interested in making
stereo movies, will find it advisable to provide a reflector for
emergency use. Although the adjustable models are not available

STEREOSCOPIC: CAMERAS

at this writing, it is not difficult to make one, and it is to be hoped
that some manufacturer will, in the near future, provide such
a device.

SUMMARY OF $5MM STERFO CAMFRAS.

Personal

Stereo
Vi\ id

Realist

'Vera-
scope

Iloca

Length . .

6"

6.25"

6 -75"

6.50"

6.875"

Height

3-2.V

^.125"

2.50"

3.1 2.%"

2 625"

Thickness ...

2"

2-2.5"

2"

2.125"

2-375"

Weight loaded ....

2502

?

32 oz

34 oz

2402

Lens: Aperture . .
Focal length .
Separation
Coated

3-5
25mm
62mm
Yes

3-5
35mm

Yes

3-5
35 mm
69.5 mm
Yes

3-5
4omm
63mm
Yes

3-5
45mm
67mm
Brown

Between-lens shutter

Yes

Yes

Guillotine shutter

Yes

Mod.

Yes

Focusing

Fixed

RF

RF

RF

Scale

Max. speed ...

1/100

i 150

1/150

1/250

1/300

Approx. picture size (mm) . ...

12x13

24x24

22x24

30x24

30x24

Exposures (so-exp. mag.) . ...

38

15

16

"

11

Built-in synchronizer

Yes

Yes

Yes

Yes

Yes

Film-shutter interlock

Yes

Yes

No

Yes

No

Shutter coupling

Yes

Yes

Yes

Yes

Except
focus

Single picture . . .

No

No

With cap

Yes

No

Integral exposure meter

Yes

Yes

No

No

No

Integral level in finder

Yes

Yes

No

No

No

NOTE: All cameras take the standard 35mm magazines as used in conventional
miniature cameras. All have finders, tripod sockets, eveready cases and usual ac-
cessories. Picture sizes may be plus or minus a half millimeter. Number of pictures
per roll may vary with loading.

CHAPTER 3
THE STEREOSCOPE

MOST STEREOGRAMS MUST BE VIEWED with SOHIC device which
causes the two eyes to see their respective images optically
superimposed. This is true without exception of orthostereo-
grams. There are some types of photographs which incorporate
dual images which are seen in relief without any viewing device,
but this does not make them orthostereograms, because although
they exhibit relief they do not fulfill the fundamental require-
ment of exhibiting the object in full natural size at full natural
distance.

Many stereographers have developed the ability to see a stereo-
gram without a viewer. This is the result of so training the eye
muscles that accommodation and convergence are separated, so
that the eyes can focus upon the near-by stereogram while retain-
ing the zero convergence of infinity viewing. This type of stereo
vision approximates the ideal more closely than the free-view
devices, but it is still not orthostereo.

The device used for viewing is generally called a stereoscope.
There is a widespread but inexcusable error of calling it a stere-
opticon, which means simply a lantern used for projecting lantern
slides; specifically, the double-dissolving type now almost extinct.
Do not fall into this error.

The stereoscope may be a simple, folding arrangement of card-
board carrying two simple magnifying lenses, or it may be a
beautifully finished piece of furniture containing an elaborate
mechanism for displaying the stereograms in any desired order.
Between the two extremes there are many different forms.

The first difference is based upon size, of which we have far
too many. The sizes are divided into two great groups, those
which are spaced to correspond to the average eye spacing of
65mm, and those which are spaced farther apart than the normal
interpupillary distance.

NORMAL SPACING. Normal spacing starts with the i6mm film.
These are either printed upon a length of film with the units of
various pairs interlaced in sequence, or as in the Sawyer View-
master with seven pairs of pictures spaced about the periphery

28

THE STEREOSCOPE

of a cardboard disc and viewed in succession by moving a lever
in the device. Despite the absence of adjustments, the Viewmaster
is a highly satisfactory stereoscope.

Fig. 3-1. The Sawyer View master.

THE REALIST VIEWER (Fig. 3-2), companion to the camera of
the same name, is of polished black plastic material and contains
its own battery and lamp, thus making the unit self-contained
and ready for use regardless of external light conditions. This has
long been recognized as the desirable method of stereo viewing.
For home use a low voltage transformer is available so the light
can be maintained at a more uniform level, a most important
factor in color viewing.

Fig. 3-2, Stereo Realist viewer. Self-illuminated,
with focusing and interpupillary adjustments.

30 THREE-DIMENSIONAL PHOTOGRAPHY

The images are viewed through black channels or tunnels,

another instance of the most desirable condition. The lenses are
of adequate dimensions and are provided with motions for focus-
ing and for interpupillary adjustment. The focusing is by thumb
wheel, and the interpupillary by lever. This viewer thus includes
all of the features essential for stereo viewing at its best.

T.D.C. VIEWER. This viewer is also made of polished black
plastic, but more nearly resembles a pair of binoculars. The

Fig. 33. TDC stereo viewer, companion to the

Fhnd camera and one of the least costly of the

q ua lity viewers.

ocular lenses are mounted in the. binocular fashion and focusing
is done by the central wheel of binocular type.

It. is self illuminated with a switch which may be locked in
place for continued viewing. The translucent switch button serves
as a pilot to show when the lamp is burning.

Access to the interior for servicing is gained by simply raising
a hinged cover.

This viewer takes the American standard stereo slide which
measures iy 8 x4 inches, in either glass or cardboard mounts.

BUSCH-VERASGOPE VIEWER. In conformity with American prac-

THE STEREOSCOPE 31

tice, the U. S. distributors for the Yerascope have introduced a
normal, non-transposing viewer.

Like the two just discussed, this viewer is of black plastic ma-
terial, with self contained illumination. In general appearance
it resembles the two types described, but both focus and inter-
pupillary are provided with scales so that the viewer may be pre-
set for any individual needs once the setting has been determined.

Naturally, although this viewer takes the standard size mount,

Fig, 3-4. The Verascope viewer, provided with
calibrated scales for both focus and interocular.
It takes all standard 3^ mm stereograms includ-
ing the wide Yerascope type.

the apertures are designed to accommodate the wide frame which
is characteristic of the Verascope camera.

BRUMBERGER VIEWER. This viewer is similar in general details
to those which have been described, but does not have the inter-
pupillary adjustment. It is of plastic construction with self con-
tained illumination.

ILLUMINATION. The color characteristics of any slide are greatly
affected by the color of the viewing light. Viewers with.no self
illumination are decidedly unsatisfactory; even the battery pow-
ered viewers are not wholly so, as the color changes as the bat-
teries lose power. The most satisfactory illumination is that -ob-

32 THREE-DIMENSIONAL PHOTOGRAPHY

tained \vith the self illuminated viewer adapted for house current
illumination.

Conversion cords are available with a tiny transformer built
into the plug. These operate the standard bulb of 3 volt rating
normally used, so no bulb changes are necessary. Only with this
light is it possible to obtain and maintain the best effect from
color slides.

OTHER NORMAL SIZES. These include the 45x107111111 and the
6xi gem, both of which are normal dual images spaced by ap-
proximately Gsmm. Richard makes a 7x13, but its superiority to
the universal 6x13 is open to question. For these normal sizes the
range of available viewers is greatest.

Folding Types. The folding stereoscope is, at best, an emer-
gency instrument. It rarely maintains a level base, it has no field
mask, it has no interocular adjustment, and the focusing is often
of the most crude type. One of the surest ways to lose interest in
stereo is to habitually make use of a folding viewer.

Book Viewers. 'The one logical use for the folding viewer is to
accompany stereo-illustrated books. The German version is to
place a separate folding viewer in a pocket formed in the abnor-
mally heavy binding. One American viewer has a folding post
attached to the binding to support the viewer in correct position
above the page. Neither method is satisfactory to the stereog-
rapher, and they are important only in so far as their novelty
is concerned.

Box Viewers. This is the largest group of all. It starts with a
simple rectangular box equipped with lenses set in one side, and
slide grooves at the other side to accept the stereo diapositive.

The same box may be provided with a hinged lid in the top
which may be raised to admit light to the surface of a paper
stereogram. This is the basic form upon which the usual box
viewers are designed. All of the later forms consist in refinements
and elaborations of this form. Therefore, instead of repeating
the descriptions of the various models, we shall simply note the
refinement and its effect upon the whole instrument, bearing in
mind that all of those described may be, and are, applied sepa-
ately and collectively to the basic box form.

In some box viewers the lenses are set in sliding tubes not
unlike opera glasses, and are usually controlled by a button at the

THE STEREOSCOPE 33

side of the box. This permits the lenses to be focused for any
individual vision.

Variable Separation. The generally accepted standard stereo
separation is 651x1111, but human eyes may be separated by either
as little as 55 or as much as 75, with a variation of 5mm in either
direction relatively common. Anatomists usually accept 6gi/ as
the average interpupillary separation. Thus we find that while
most cameras and viewers conform to the stereo standard of
65mm, the Verascopes of Richard conform to the anatomical
standard of Gsi/^mm.

It is desirable, and almost essential, that the viewer lenses and

Fig. 5-5. Telebinocular. One of the most satis-
factoty of the Brewster type viewers. This is the
instrument used in the laboratories of the Stereo
Guild for all stereograms of abnormal separa-
ij such as the old "parlor** type of mew.

34 THREE-DIMENSIONAL PHOTOGRAPHY

camera lenses be of substantially the same focal length, and used
at the same separation. It is an ideal condition when one can look
through the centers of the lenses, but this is possible only when
the inter pupillary exactly corresponds to the separation of the
lenses. Although in theory the stereoscope lenses act as colli-
mators, making it unimportant to use the lens centers, it is visu-
ally more comfortable to have the lenses movable so that the
separation corresponds to the interpupillary rather than to con-
form to stereo theory and remain exactly opposite the image
centers. Normally the eyes converge and accommodate (focus)
at the same time. Thus when viewing images of near-by objects
the accommodation changes. Many persons, particularly those
not accustomed to stereo viewing, adjust the focus for a nearer
distance than infinity. Thus the true collimating effect is lost,
and interpupillary adjustment becomes more of a necessity.

Stereoscopes of the full lenticular type should have lenses not
less than i$mm in diameter to provide free vision for interpu-
pillary distances between 60 and 7omm. When the lenses are not
set at infinity focus, the adjustable separation becomes desirable.

Stereoscopes whose lenses are gomm or more in diameter and
of variable separation may be used either as full lenticular or
semilenticular (Brewster) instruments. They may be used with
stereograms of widely varying image separations.

It Is an advantage to have cups for shielding the eyes when the
viewer has no self-contained illuminating system. Achromatic
lenses are advisable in all types of viewers, and are well worth the
slight extra cost. Both full lenticular and Brewster instruments
are available so equipped.

Variable Fociis.Some elaborate European viewers are equipped
with interchangeable lenses of various focal lengths. Although
these are sold for the purpose of obtaining variable magnification,
the feature has value only when a single viewer is to be used with
stereograms made with different cameras having different focal
lengths. Other than this, the use of interchangeable lens viewers
is to be avoided.

Automatic Hand Viewers. The automatic viewer is simply a
box viewer of the better class, fitted with a simple mechanism into
which several views are loaded. Pressure upon a convenient
lever, or turning a button, causes the views to be presented in

THE STEREOSCOPE r>

succession. Usually this viewer has an auxiliary table stand, al-
though it may be held freehand. It is intermediate between the
box viewer and the cabinet type.

Cabinet Viewers. These are table instruments, and while all
models vary somewhat, it may be stated that as an average they
measure about 10 inches square and stand about 15 inches high.
They are usually in the form of a rectangular wooden cabinet
with the lenses set near the upper end of the front side.

The views are ordinarily loaded in trays of twelve or twenty-
five. An indexing mechanism on the outside permits the selection
of any desired view. If all the views are to be observed, pressure
upon the operating lever will bring the views into position in
succession without any manipulation of the index lever.

These cabinets are made for use with transparencies only, as
the viewing space is a black tunnel with the view masked to the
actual picture area. In the case of cabinets for the smaller 45x107,
which have the two picture images separated by a space in which
the title is ordinarily written, an auxiliary lever is provided which
lifts a shutter and inserts a prism in the optical path to make
the title visible.

The cabinet viewers usually have both focusing and interpu-
pillary adjustments, and most of them are provided (as extras)
with illuminating boxes for artificial light and with single or
double projection systems. Some of them (e.g., Mattey) are inter-
changeable for both 6x13 and 45x107 sizes. While there are many
very good cabinets, or classifiers as they are called, the Richard
Taxiphot may be regarded as typical.

The Mattey dual size cabinet instrument just described has
been converted, very easily, to take tti 35mm slides in both glass
and cardboard mounts. This type of stereoscope is often called a
"classifier."

Chain Viewers. The chain viewer resembles the classifier ex-
ternally, but usually has no interpupillary adjustment. The views
are carried upon a pair of endless chains which hang upon sprock-
ets attached to the operating knob. The knob is turned to bring
the views into position successively. In the simpler type, the chain
is built into the cabinet, and views must be removed from the
chain and replaced when a change is desired. The better type has

3(i I HREK-UIMENSION VL PHO LOGRAPHY

interchangeable chains so that any number of loaded chains may
be kept in readiness ior use.

VIEWERS FOR THE AMERICAN SIZE. This is the old parlor size
in which the stereogram measuring approximately 3x6 inches is
mounted upon a card measuring 31/^x7. It is the size most familiar
to American and British amateurs and to those who are not
stereographers, but the size most rarely used by the European
stereo enthusiast. The views are almost universally paper, al-
though a few transparencies are encountered now and then.

Hand Viewer. The common viewer is the old parlor hand
viewer known variously as the American,, Mexican, Holmes, or
Parlor viewer. It belongs to the same class as the folding viewer,
and is to be regarded with the same suspicion. The wood construc-
tion and the inaccurate fitting produce a shaky contraption which
is more often out of adjustment than in. Even those made by
reputable firms are acknowledged to be designed only for occa-
sional or demonstration use.

Table Viervers.The table viewer is the Holmes instrument
made of metal and carefully fitted. The whole is mounted upon
a sturdy base and has in general those physical characteristics ex-
pected of an optical instrument of practical quality. These instru-
ments are inexpensive in the simpler models, and should always
be used for the American size when no better instrument is
available. Inasmuch as these viewers are optically of the Brewster
type, focusing is accomplished by moving the view rather than
the lenses, an arrangement which is convenient.

Therapeutic Models. One of the most practical of all Holmes
viewers is that made for use in ophthalmic treatment. These in-
struments are similar to the table models, but they are heavier
and more rugged. The infinity homologous distance is greater, the
lenses are of better quality, the focusing scale is accurately cali-
brated, the body is fully adjustable for height and angle to
provide the most comfortable viewing, and the eyepieces are
grooved to take correction lenses so that even those with defective
eyesight may use the instrument without wearing spectacles.

Inasmuch as the therapeutic viewers cost about the same as
the automatic box viewer and much less than the cabinet classi-
fiers, they are well within the reach of every serious amateur who
works in this size, or those who use a small camera and then make

THE STEREOSCOI>K 37

use of enlarging to produce the final American size. The Keystone
Ophthalmic Telebinocular is an excellent example of this type
of viewer and one which can be recommended without reserva-
tion (Fig. 3-5). I use this instrument exclusively for this size of
stereogram.

Cabinet-Type Viewers. The classifier is not available for the
American size print (although it is available for transparencies
in the same size), but chain viewers with reflectors for illuminat-
ing the paper print are made. These are so exactly similar to the
transparency chain viewer that nothing further need be said
about them.

Pantoscope.The Pantoscope was once a great favorite but is
now rarely found in this country although still available in
Europe. It consists of a folding wooden frame, sometimes elabo-
rately carved, made in three sections. The base rests upon a table.
The center section rises at the front end to slope from the user
down toward the table. The third section rises at right angles to
the middle section. At the rear of the middle section two rods rise
at right angles to support the view.

The top or front section contains a large single lens from four
to seven inches in diameter. Below this lies a pair of stereo lenses.
The large single lens is for viewing ordinary photographs to give
them a "plastic" appearance (a phenomenon to be discussed
later), while the two lower lenses are provided for normal stereo
viewing.

MISCELLANEOUS VIEWERS. Coin-operated viewers are often
found in Penny Arcades. Highly elaborate instruments have been
designed for mapping, for therapeutic use, and for other special
purposes. As these are of little interest to the amateur, we shall
mention none of them except for the Wheatstone viewer, which
is a basic type widely copied for general purposes.

The Wheatstone viewer (Fig. 3-6) consists of a pair of mirrors
set at right angles and placed immediately before the eyes so that
the left visual axis is deflected to the left and the right axis to
the right. The two views are separate and are placed In suitable
supports facing each other. The operator must adjust ttie posi-
tions to produce final stereo fusion. The advantage lies in the fact
that the size of the prints used is unlimited. This viewer is largely
used in aerial survey work and in X-iay stereography. IB the latter

38

THREE-DIMENSIONAL PHOTOGRAPHY

Fig. 3_6. The Ryker, a small version of the Wheatstone' stereoscope

used for viewing prints up to six inches square ,, is an aerial photo type

of viewer. Courtesy J. P. Medders.

application, the unit pictures may be as large as 16x20 inches.

A small Wheatstone viewer with prismatic reflectors was made
by Zeiss under the name of Pulfrich. Improved prismatic viewers
have been repeatedly designed for industrial and survey work in
which the pictures are laid side by side upon a table, but the
fundamental principle is the same.

IMPORTANCE OF VIEWERS. We have spent a considerable
amount of time discussing viewers, for a very good reason, even
though most of them are types which seem to be growing obso-
lete. Success in stereoscopy depends upon the perfection of the
image, so in stereography satisfaction is derived only when the
presentation of the image is flawless. The fact that cheap viewers
can, and do, ruin the work of skilled stereographers using fine
cameras, is well known. If the beginner wants to derive the ut-
most satisfaction from stereography, let him first buy the best
viewer he can afford and then obtain the best camera he can for
what remains of his funds. A $30 box camera and a $100 viewer

THE STEREOSCOPE 39

will provide far more satisfaction than a |goo camera and a $5
viewer! Practically every disappointed beginner can trace his
disappointment to the use of a cheap or faulty viewer.

OPTICAL CHARACTERISTICS OF VIEWERS. The function of the
stereo viewer is not difficult to understand, but once understood,
the necessity for the best possible instrument will be appreciated.

Lenticular Viewers. The normal viewer, that is, the viewer
used with stereograms whose I-H separation (homologous sepa-
ration for objects at infinity) is Gsmm, is usually called a len-
ticular viewer. The lenses are used as simple magnifiers, but not
for the purpose of making the image appear larger.

The lenses serve a dual purpose. First they magnify the images
to just that degree needed to compensate the diminution or nega-
tive magnification of the short focus camera lens, and second to
bring the image into sharp focus while leaving the eyes in the
conditicfn of rest. (Infinity focus = zero accommodation).

Thus while there exists the magnification incident to viewing
anything through a positive lens, this magnification, under or-
thostereo conditions has no significance because it only brings
back to normal dimensions an image which has been made small
in the camera.

Fig, 3-7. A large lens focuses parallel rays a to g at F.
If lens is split and the halves transposed, the upper half
focuses rays drg at Fl; lower half focuses rays a-d at F2.

40 THREE-DIMENSIONAL PHOTOGRAPHY

Thus,, when a magnifying glass is used, we may assume that we
see a very large image at an infinite distance or we see a smaller
but still large image at the normal reading distance of (usually)
16 inches. In the stereoscope we use the lenses to show us a "life
size" image at (a maximum of) an infinite distance, and other
similarly life size images at their correct, relative positions in
space. This orienting of objects in space does not conform, sub-
jectively, to magnification as usually understood.

However, lenses bend rays of light according to known laws.
Simple lenses do not exert a uniform influence upon the rays, and
if the axis of the lens does not coincide with the original axis of
the image, the rays are not bent to the correct position. In short,
for results which are fully satisfactory, an achromatic lens should
be used and the instrument should provide such mounting that
the optical axes are maintained in their correct positions relative
to one another and relative to the optical axes of the stereoscope.

The Brewster Stereoscope. In the Brewster instrument we
have another set of conditions. To understand the origin of the
lenses used, imagine a large lens whose diameter is twice that of
the lenses to be used in the stereoscope (Figs. 3-7 and 3-8). From
this large lens two smaller lenses are cut, one from each half as

/<Vg. 5 tV. A, B and C show how lens is divided to cut out two smaller
lenses which are transposed and shown at D in edge view, and at E

in plane view.

shown in the illustration. These two derived lenses each have the
optical center of the original lens at their extreme periphery.
These two lenses are then mounted with their positions reversed,
that is with their thin edges adjacent. Thus for ideal results the

THE STEREOSCOPE 41

I-H separation of the stereogram should be equal to the distance
between the optical centers, or the extreme outer edges of the
two small lenses.

When adjusted lor normal vision, the stereogram is supported
in a plane whose distance from the lens is equal to the focal
length of the original lens. Now we shall see why this viewer re-
quires no interpupillary adjustment.

If the instrument is set up and a plain card placed in the sup-
port and a parallel beam of light is then projected into the lenses,
they will bring the beam to two foci which are separated by a dis-
tance equal to the distance between the optical centers of the
lenses and equal to the normal I-H distance. All parallel rays
falling upon any part of the lens are brought to the focus. This
is a characteristic example of the use of a collimating lens, or
using a lens to produce a parallel beam from a point source, or
vice versa.

As the human eye at rest is in a condition of zero convergence
and accommodation, the visual axes are parallel. Thus no matter
what the interpupillary, these parallel visual axes will coincide
with a pair of rays in the parallel beams, and the visual axes
will be projected to the optic foci by the collimating action al-
ready described. Thus, regardless of interpupillary, all eyes will
react in a similar manner.

The choice between the full lenticular and the Brewster (or
semilenticular) instruments has led to bitter argument, both
schools having adherents, but a little thought will show that they
are optically very similar, and that when the interpupillary is less
than normal, the viewing conditions of both types are identical.
The real choice lies in stereograin size, or rather homologous
separation. If the separation is greater than the normal 65mm,
the Brewster is used. The whole poiot is that the optical centers
of the lenses should be separated by the distance equal to the
homologous separation of the images. The Brewster type would
work just the same if full lenses were substituted for the half
lenses, but as the external halves could never be used, why
waste glass?

Thus, any full lenticular viewer may t>e used as a Brewster by
separating lenses and stereogram mages to any distance greater
than the normal interpupillary of the observer, Licewfce, IB a

4L> THREE-DIMENSIONAL PHOTOGRAPHY

Brewster instrument of 85111111 separation, anyone so abnormal as
to have this as their interpupillary would look through the outer
edges of the Brewster lenses, that is, through the optical centers
just as through a full lenticular. The Brewster is a full lenticular
stereoscope in which the optical axes are separated by more than
the interpupillary and in which the unused half lens is not physi-
cally present,

CHOICE OF A STEREOSCOPE. The choice lies primarily in the
size of the stereogram. For all sizes with normal interpupillaries,
the full lenticular type of viewer is used, including those for 35
and i6mm film stereograms. This refers to all stereograms, regard-
less of the dimensions of the print, as long as the homologous
separation is substantially 65111111.

For perfect results, the stereoscope should theoretically remain
with fixed separation and fixed focus. Any compensations made
for focus should be made by the use of supplementary ophthalmic
lenses rather than by altering the distance between lens and
stereogram. However, as this is not often practical, focusing by
the conventional method is acceptable, and supplemented by
variable separation does not produce obvious distortion. For the
most comfortable use, experience has shown variable focus to be
essential in all but the smallest sizes, and variable separation
often convenient.

The stereoscope should have a rigid mechanical structure,
should be equipped with achromatic lenses, and if possible, it
should be equipped with some kind of illuminator to provide uni-
form illumination to both images of the stereogram.

THE STEREOSCOPE IN USE. The stereoscope is too often re-
garded as simply a peephole device for looking at stereograms, no
thought being given to the optical and neuro-optical phenomena
involved. Because the stereographer can do much better work by
understanding the fundamental principles of stereoscopic vision,
we shall devote some space to this consideration, even though
every attempt has been made in this book to eliminate theoretical
discussion.

Cyclopean Vision and Stereo-projection. Stereoscopic vision is
based upon the perception of an image which has no physical
existence. It is synthesized within the brain from two dissimilar
sphere-planar images, and then projected into space. This involves

THE STEREOSCOPE 43

a process which seems at first glance to go back to the early days
of black magic and superstition, until we differentiate between
the projection of a real beam, such as of light, and psychological
projection which deals purely with the subjective. All of the clear-
ly visible space in a stereogram is projected psychologically from
the stereo images. Psychological projection can be a very real
thing!

In Fig. 3-9 there is a hexagonal object, the subject of vision.
The two retinas receive images which are dissimilar. These unit

PRGEPTION

Fig. 3-9. Schematic illustration of stereoscopic perception.
Retinal rivalry slide. Courtesy of Marel Laboratory*

images consist of an image of the front surface of the hexagon
combined with one side, but in the two unit images, different sides
are perceived. These images are formed upon the retina accord-
ing to conventional optical laws, but from this point onward optics
cease to be the controlling factor, and we enter the realm of sub-
jective perception.

The retina responds to the impact of the image, and the two
dissimilar images are perceived in the brain centers, as a result of

44 THREE-DIMENSIONAL PHOTOGRAPHY

normal sensation. At this point the less tangible reactions take
place. The two sensations are synthesized by a process which we
do not fully understand, to produce in the Cyclopean center a
perception of the three-dimensional image corresponding to the
visible surface of the solid hexagonal object. However, this image
is created and exists only in the brain, hence the necessity for the
final step, the projection. The image which is synthesized is then
psychologically projected out into space until it occupies the same
identical position in space occupied by the original solid body,
and we have the sensation of seeing into space for a definite dis-
tance. A more detailed discussion of this "projection" of the im-
age from the Cyclopean center into space may be found in any
authoritative reference book of physiological optics.

The ' stereoscope is the physical means which we use to make
possible a projection into space identical with that which would
have existed had we viewed the original object. It is well known
that "free- vision" stereograms, such as Vectographs, anaglyphs,
bar-screen stereograms, and the like, do not even approximate the
original distances of depth, as does the true orthostereogram, and
in that respect they are inferior to the viewer type of stereogram.
The projection of the free vision image is strictly limited.

The stereoscope provides a set of physical conditions which
make possible exactly the same projection which would exist in
directly viewing the original object. Thus it is possible for us to
sit in a room, facing a wall only a few feet away yet truly see
hundreds of yards out into space directly through the wall. How-
ever this may suggest fantasy, every competent stereographer
knows it to be a very real fact. That is the basis, not only of the
charm of stereo, but for some very genuine beneficial effects which
result from the consistent use of the stereoscope.

Stereoscopic Idiosyncrasy. Stereoscopic perception is a charac-
teristic of marked variability. Like those who are color blind,
there are those who have perfect binocular vision (that is, they
have two eyes which are synchronized and which produce perfect
visual images) yet who have little or no stereoscopic perception.
The condition itself is somewhat analogous to color blindness, biit
unlike that defect of vision, in many instances stereoscopic per-
ception can be developed by consistent exercise.

Some people have such a marked stereoscopic sense that they

THE STEREOSCOPE 45

are always painfully aware of the flatness of the ordinary photo-
graph; there are others who have so little stereo perception that
the perception of depth depends upon such extrinsic factors as
the diminution of size due to perspective, and similar effects. In
fact, this has led in the past to the use of the word "perspective"
when stereo depth is meant. The words, far from being synony-
mous, are complementary. These people see depth in a good
planar photograph or when viewing a motion-picture screen.

Such people see nothing unusual when viewing their first stereo-
gram. They see just what they do when looking at a transparency
in an illuminated viewer for example. This not uncommon lack
of stereopsis is one reason that stereo has not become even more
widely popular.

Several optometrists and ophthalmologists, who specialize in
stereoscopic treatment, were asked about the average conditions.
These men are located in the East, Southeast, Southwest, and
West, fairly representative of the nation. The consensus was that
about one-third of all adults have low stereoscopic perception,
and of the remaining two-thirds, from 12 to 15 percent have fully
developed stereopsis. How much of this is due to physical or or-
ganic defects in the visual apparatus is open to question, but all
are agreed that thorough training will show a marked increase in
stereopsis in the case of anyone who has the slightest degree of it
with which to start; that those who practice stereoscopic pho-
tography are usually in the high bracket; and most significant of
all, that the use of stereograms in the home and office results in a
marked improvement in vision. Incidentally, this answers those
who refuse to work with stereo for fear of injuring their eyesight.

Stereoscopic Exercise. It is well known that exercise, either
physical or mental, is of value in sustaining health and in develop-
ing ability. Visual exercises are often regarded as purely subjec-
tive, but the visual organs, the eyes, operate through the agency
of a complete system of delicate and highly coordinated muscles.
That these muscles will respond to exercise Jijst as will any other
muscle is not only logical, but also a proven fact.

With each succeeding generation more of our major eljEort is
expended within doors, and the maximum range of vision is per-
haps 50 feet, with 10 to 15 a more common average. With the
eyes adjusted for such distances, the position of rest (the

4<> IHREE-DIMENSIONU. PHOTOGRAPHY

of the eyes when we are looking at a far distant object) becomes
really an abnormal condition. Vision at close distances also neces-
sitates convergence of the eyes, and as a consequence we are tend-
ing toward national nearsightedness and convergent squint or
"crossed eyes." Optometrists find a surprisingly great number of
people who have a tendency to squint which is not manifest in
their ordinary vision, or in their appearance.

It has been found that even a quarter of an hour each day spent
in looking at stereograms will do much to offset this tendency and
give everyone a chance to look into the far distance. The use of
the stereoscope not only provides for the complete relaxation of
the ocular muscles into the condition of rest or infinity vision,
but it also provides for a period of complete relaxation of con-
vergence. While any good stereogram will approximate these con-
ditions, a standard therapeutic stereoscope and a set of special
exercise slides prepared for use with it, will provide exactly the
relaxation desirable and will also, through the use of other slides,
provide for flexing the muscles or alternating from convergence
to relaxation through specific angles and through specific series.
Thus the therapeutic stereoscope provides a visual gymnasium
which can become one of the most valuable possessions of the con-
sistent user. The reward for consistent exercise is better vision,
freedom from eyestrain and its attendant headaches, greater acuity
of vision, and retardation of the effects of .presbyopia or "old age"
vision.*

Non-stereoscopic Stereograms. While discussing the subject of
stereograms for visual training, the type of slide which is binocu-
lar but not stereoscopic should be considered. These are of several
types.

The simplest is a picture slide such as is shown here (Fig. 3-10).
The dog is seen by one eye and the pig by the other. The dog
will appear to be jumping over the pig, but it may be just starting
the jump, it may be midway over the pig, or it may be past the
pig and ready to land. The relative positions reveal any tendency
toward abnormal convergence or divergence. Similar slides con-
taining an arrow pointing toward a series of numerals, serve a
like purpose but provide a more exact result.

*Such stereqgraras are made by Bausch & Lorob, American Optical Co., and
Kcvstone View Co.

THE STEREOSCOPE

Fig. 3-10. Courtesy Keystone View Co.

Another type is the split slide which is actually two separate
slides, each having a portion of a picture upon it. Thus there may
be a star upon one slide and a circle upon the other. In this in-
stance the star is supposed to be seen within the circle, and to
achieve this result the two cards are moved farther apart or nearer
together in a calibrated holder.

In still another type, a special desk form of stereoscope is used
and the cards have a simple outline picture such as a flow r er, vase,
boat, or the like, printed upon one end only. The subject places
this in the stereoscope and then proceeds to draw the second image
upon the blank end of the card. When this can be done success-
fully the eyes are exceptionally well co-ordinated. The use of this
stereoscope develops both co-ordination and stereopsis to a marked
degree.

Although these slides are know r n generally as therapeutic ones,
they are of interest and of value to every stereographer. Unfortu-
nately space does not permit a detailed account of the experiments
and exercises which anyone may use at home, but the information
is readily available to those interested.

Similar therapeutic slides are made which do not use the stereo-
scope but which make use of Vectograph slides, such as the Bausch
&: Lomb Orthofusor.

CHARACTERISTICS OF THE STEREOGRAM. As we shall see in the
next chapter, the use of the stereoscope is essential for repro-
ducing the orthostereoscopic (true stereo) effect, and to produce
this effect the PePax principle is fundamental. This principle

48 1" H R EE-DI M ENSIGN AL PHOT OCR A PH V

states that true stereo is possible only when the ratios of perspec-
tive and parallax are kept constant. The principle is considerably
more complex than this simple statement indicates, but as it is
largely a matter of theory I shall not go further into it here.

It is necessary, though, to remember that perspective and paral-
lax are complementary, and in one sense they are opposed terms.
Parallax pertains to distance in depth, away from the point of
observation. Perspective pertains to dimensions in planes which
are perpendicular to the visual axis.

We ordinarily consider perspective to be essentially the appear-
ance of depth or relief, but those familiar with its principles will
recall that perspective is primarily the art of reproducing a series

Fig. 3-11. Divided word. In the stereoscope, normal vision will show
the complete word, "LONE"

of objects at various distances, as they appear in a single plane.
The ordinary planar photograph, for example, exhibits a perfect
example of perspective, yet it is the antithesis of the stereogram
in which depth is directly and unmistakably apparent.

Parallax Controlling Factor. The parallax, that is, the amount
of difference between the two unit images, is the sole controlling.
factor in presenting the appearance of definite depth or distance.

If an object, for example a man six feet tall, is photographed at
a distance of, let us say, 20 feet, he will appear to be 20 feet dis-
tant in the resulting stereogram and will appear to be about six
feet tall. However, if between two successive parallactic exposures

THE STEREOSCOPE .

49

the man be moved (without changing his posture) so that the
parallax is diminished by 50 percent, he will appear to be 40 feet
away and to be about 12 feet tall. If the motion is in a direction
to increase the parallax by 50 percent he will appear to be 15 feet
distant and to have a height of approximately four and a half feet.
This principle was once widely used in making fantastic stereo-
grams such as a dancing girl inside a wine bottle, and the like;
and is again becoming popular among those who like to experi-
ment with this so-called Space Control.

Size Indefinite. Our perception of absolute size is most un-
reliable. The reason the controlled parallax just described is op-
erable is that the parallax definitely places the image at a fixed
distance, and we estimate the size by comparison with that of
known objects at the same distance. Thus when the image plane
is artificially moved nearer, by increased parallax we still have the
image size as originally diminished by perspective, so the object

Fig. 3-12. One of the most notable of stereo illustrated books, with
231 three-dimensional pictures in full color.

50 THREE-DIMENSION VL PHOTOGRAPHY

appears to be smaller by comparison with other objects at the
same apparent distance.

The same thing applies to the use of hyperstereo. If we have a
vast expanse in a mountainous country, the distances are so great
that the major portion of the scene lies beyond stereo infinity and
we might as well use a planar camera. However, we can introduce
true stereo relief by making use of an increased base. In short, we
increase the degree of parallax beyond the natural. Thus by every
applicable law of stereo, we should have a small sized image at
an apparently near distance. In fact this is the argument ordinarily
chosen to combat the use of the technique. Unfortunately the
argument is not tenable because of the fact that size perception is
always relative, and if we have no other objects in the same stereo-
gram which exhibit a basically different degree of parallax for a
fixed degree of perspective, there is nothing to show that the sub-
ject is a "miniature view."

Pseudo-Stereograms.The pseudo-stereogram, or false stereo-
gram, must not be confused with the pseudoscopic stereogram.
The latter is simply a stereogram which has not been transposed
and which shows the relationships of depth reversed. This is often
used in the study of small, complex solid bodies, the reversal
bringing out points which in the ordinary view are overlooked.

The pseudo-stereogram is simply a slide made of two identical
images. That is, two prints from the same negative are mounted
side-by-side. It is obvious that the result will be flat because it
lacks parallax. However, it is also known that looking at any pho-
tograph through a lens system gives to it a vague appearance of
solidity. This appearance is enhanced in the stereoscope and the
result is that we have every factor of depth perception present
except parallax. Usually the result is the indefinite roundness sup-
plemented by the specific characteristics of perspective, so that
many who are not experienced in stereo are deceived by a well
made pseudo-stereogram. At the same time there is something
lacking, and usually the comment is that the stereogram "isn't
very good/'

Stereoscopic Diplopia.VJz may well ask what makes stereo
work. There is no conclusive answer to this, but the fundamental
mechanism of stereoscopic vision may be glimpsed when we con-
sider a factor or two which are ordinarily ignored.

THE STEREOSCOPE 51

The activating stimulus of stereo vision is undoubtedly that of
dynamic stereo diplopia. Diplopia means seeing two images of one
object, and the diplopia normal to stereoscopic vision is known
as stereo diplopia to differentiate it from the pathological form.
Hold one finger vertically about a foot in front of the bridge of
the nose and in line with a window. Look at the window. You
will see two ghost images of the finger, one at each side of the
window. Look at the finger and you will see two windows, one at
each side of the finger. These are examples of stereo diplopia.
Look from the finger to the window. During the act, the images
of the finger separate into two and the two images of the window
merge to form one. In the midway period you see two images of
both objects. This motion of the dilopic images toward or away
from one another is dynamic stereo diplopia. It will also be seen
that the nearer any secondary object is to that which is the center
of visual attention, the smaller is the separation between the
ghosts. This diplopia is a function of dynamic parallax.

When we view an open scene, not only does the visual attention
shift constantly from object to object, but with each shift there is
a complementary alteration of the distances separating each pair
of ghosts, so that stereo vision embraces an intricate, interweaving
dance of ghostly pairs. It is this contraction and expansion of the
degree of stereo diplopia which provides the stimulus to depth
perception. The amount of distance between two ghost images
which must be overcome to fuse the images is the subjective key
to its position in space. This must not be confused with the physi-
cal, muscular tension involved in converging the eyeballs. The
latter is of extremely minor importance in producing stereoscopic
vision.

This fact is subject to the most striking proof. Take up a com-
fortable position which can be maintained for several minutes
without discomfort. Select such a position that two objects, such
as two trees, are aligned in such a manner that footti may be seen
simultaneously without shifting the eyes. It is t>etter If the dis-
tance separating the two objects in the line of vision be not more
than half the distance separating the nearest from the eyes of the
observer, although this is not an essential condition.

Select some fairly prominent spot in the nearer object, perhaps
a broken twig, a dead leaf, or the like. Keep the eyes fixed upon

52 THREE-DIMENSIONAL PHOTOGRAPHY

this one spot. Do not allow the vision to shift for a second. This
will be extremely difficult, for the eyes normally shift all the time
from object to object. While maintaining the visual focus upon
the one spot, remain conscious of the images of both objects. At
first they will appear in normal relationship, but after an interval,
perhaps 15 seconds, perhaps a minute, they will be seen to be ap-
parently in one plane, just as in an ordinary photograph. The
specific stereo differentiation has been lost. This occurs very
quickly, but at first it takes a few T seconds for us to fully realize
the change. Once the loss of depth has been recognized, shift the
eyes, even as little as to the next bough of the tree, and instantly
the images jump apart into their original stereo separation. This
second change from the non-stereo to the stereo phase is marked
and unmistakable. Thus it seems to be the continual shifting of
the degree of stereo diplopia which is directly responsible for
stereo perception.

To support the Wheatstone theory that static parallax is the sole
stimulus of stereoscopic vision, an experiment is cited in which
a subject is permitted to view a scene "for a very short period
of time" with the (claimed) result that full stereo depth is per-
ceived. In the laboratories of the Stereo Guild several subjects
were used in a similar experiment. The slides were artificial stereo-
grams of graded depth but containing no familiar figures, no ex-
trinsic factors of depth. Illuminated with an electric flash whose
duration was extremely short (so that the period of vision should
be limited to the normal visual persistence), it was found that not
one subject could identify the relative spatial positions in any
slide, although all subjects could correctly classify all slides when
permitted a period of vision of only a few seconds. This seems to
support the theory of dynamic parallax rather than of static.

Parallax in the Stereoscope. As we have seen, the lateral dis-
tance separating the two images of the same object in the stereo-
gram is known as the homologous distance. In a stereogram the
homologous distance decreases as the object occupies planes near-
er the eye. Thus it is obvious that when the visual attention is
focused upon any one object in the stereogram, the images of all
other objects will fail to superimpose, but will produce stereo
diplopia or ghost images.

The angular separation of these images is, in the orthostereo-

THE STEREOSCOPE 53

scopic reproduction, exactly the same as that which separated the
similar images when the scene was viewed directly. Not only is this
true, but as the attention of the eye changes from object to object,
there is exactly the same continuous variation of diplopia which
would have been present in the real scene.

Thus we find that although the individual pictures of the stereo-
gram are sharply defined throughout, when viewed in the stereo-
scope there is the same "out-of-focus' 1 effect in objects in remote
planes that is observed in real life. The diffuse definition observed
in direct vision in all objects not the center of attention, is not
actually a matter of the focus of the crystalline lens, but is due to
the doubling of the stereo images, an effect which is exactly re-
produced in the stereogram. Hence, it follows that those who sug-
gest using wide apertures when making stereograms to differenti-
ate planes as in planar photography, fail utterly to understand
the extent to which the stereogram duplicates conditions of direct
stereoscopic vision,

Accommodation and Convergence. Accommodation refers to
the actual alteration of focus of the crystalline lens of the eye. The

Fig. 3-13. Individual mergence. In the stereoscope, one numbered circle
will have a dot over its center. This shows the individual tendency to
. convergence or divergence.

ability of the individual to adjust his vision for both near and
distant objects is known as his "accommodation range." At the
same time, when nearby objects are viewed, the axes of the eye-

34 THREE-DIMENSIONAL PHOTOGRAPHY

balls con\'erge toward each other. It is obvious that both actions
are definite, and that for any specific accommodation there must
be an exactly corresponding convergence. There are many people
who believe the two actions are inseparably combined by func-
tional apparatus. This is not at all true. The two actions are
wholly independent and only because long habit has caused them
to synchronize, are they apparently inseparable.

The normal stereoscope, lenticular or Brewster, used with the
correctly made normal stereogram, is so adjusted that the viewing
lenses act as collimators. That is, the eye itself is in the condition
for infinity viewing or zero accommodation. The lens is complete-
ly relaxed. At the same time convergence is also at zero for those
objects lying beyond stereo infinity, but for all nearer objects con-
vergence is normal. Thus in any stereoscope, convergence is di-
vorced from accommodation.

Naturally it is not easy to overcome a lifelong habit, and those
unaccustomed to the instrument will unconsciously accommodate
as they converge. As a result they 'find it impossible to hold sharp
focus, and as a compromise will almost always select a point mid-
way between far-point and near-point focus. As a result they see
nothing sharply outside the one plane of compromise focus. This
is literally true. In viewing a stereogram which is needle sharp
throughout, these people will see both distance and near fore-
ground as out of focus if the compromise convergence focus point
lies in the mid distance. This is because their eyes constantly
change focus to keep up with the actual convergence changes in-
duced by the variable separation of the stereo images. The condi-
tion does not, as a rule, last very long however. It is often over-
come within 15 or 20 minutes, and rarely lasts through as much
as two hours of actual viewing. Such a person finds the use of the
stereoscope uncomfortable, and not until by continued use the
two functions are separated, does he really enjoy stereoscopy.

In some instances, the coupling works the other way, and focus
is maintained but convergence insists upon remaining with focus,
and the two images cannot be fused. No matter what is done, the
subject sees two distinct images. This, too, will readily yield to
practice. Just as soon as the two functions are separated there is
usually experienced a sense of increased visual comfort, not only
in viewing the stereogram but in normal, direct vision as well.

THE STEREOSCOPE 55

The writer has never experienced either of these types of cou-
pling, and for many years the failure of tyros to see the stereo
image was a puzzle. There was always the possibility of the ab-
sence of stereopsis, but this was not really a satisfactory explana-
tion. Only after some period of time spent in study did the facts
reveal themselves. Unfortunately for our pet slogan, the stereo-

Fig. 3-14. Drawing stereoscope used to develop better stereopsis.

56 THREE-DIMENSIONAL PHOTOGRAPHY

gram does not in this one respect duplicate direct vision; but this
fact is fortunate as far as the stereographer is concerned inasmuch
as it provides him with increased visual skill after a little practice.

It will be noticed that the foregoing remarks assume the exist-
ence of correctly made and correlated stereoscope and view. It is
of interest to know that the therapeutic stereoscope is so accurate-
ly calibrated that the operator can introduce any desired accom-
modation and any desired convergence, both independently ad-
justable. Note however, that parallax is not adjustable. It is a
characteristic of the stereogram. This makes it possible to measure
the degree of both functions, to study the results of the habitual
bond between the two, and to treat certain tendencies toward ab-
normal vision. "The Telebinocular Manual," Keystone View Co.,
Meadville, Pa., provides detailed instructions for this manipu-
lation.

Trick and Puzzle Stereograms.'The stereoscope is also a source
of enjoyment for a group. Those present need not have any
knowledge of stereo. In fact, when they do not, the results are
even more baffling. For example through the control of parallax
(space control) see Chapter 16, a stereogram can be made of
someone well known to those present, but the figure is obvi-
ously only six or eight inches tall and standing upon a table at
which another familiar person is seated. The method has already
been described.

The pseudoscopic stereogram may also be made a source of
pleasure when well done. Suitable subjects are rarely found in

Fig. 3-15. Iridescent drawing.

THE STEREOSCOPE 57

landscapes, owing to the confused overlapping contours, but a
rounded, solid object makes an attractive slide. (See Pictorial
Stereography.)

Iridescent slides are ordinarily hand drawn, one of black lines
upon white and the other of white lines upon black. The binocu-
lar mingling of the black and white produces a scintillating, iri-
descent effect.

Similarly, a pseudo-stereogram of some subject such as a foun-
tain or a waterfall, one image made in dull light, the other in
direct sunlight, will have no real stereo relief. However, due to
retinal rivalry the sparkling image will conflict with the dull one
and often the effect is that of moving water.

Retinal Rivalry. It is impossible for two widely different im-
ages, transmitted by the two retinas independently, to be simul-
taneously perceived. This fact should not be confused with stereo

Fig. 3-16. Retinal rivalry. The eye does not see a true Crosshatch,, but
alternating vision in small patches of first one set of lines, then the

other.

diplopia, for there the dual images are seen by oblique vision. But
when two different objects are the subject of direct macular
vision, first one image will be seen, then the other. The successive
phases may last for several seconds, or for only a fraction o a
second. This is best illustrated by a dual slide in which there are
oblique lines at right angles in the two units (Fig. 3-16). Instead
of a perfect Crosshatch pattern, the area will show small patches

->8 THREE-DIMENSIONAL PHOTOGRAPHY

of left lines mingled with patches of right lines, and the patches
will continually shift.

The old color anaglyph was based upon the assumption that a
red perception in one eye and a simultaneous green perception in
the other would produce white by the addition of color. As is well
known this method of stereoscopy proved a failure because of the
visually painful "bombardment." This effect is nothing more than
retinal rivalry which transmits first the green, then the red. Many
people however see only the color which results from the additive
mixing of the two colors, i.e., red and green (of correct hues) pro-
duce white.

Stereo-Cryptograms. The stereo-cryptogram is another source
of amusement. In this, a series of letters are shown in solid block

Fig. 317. Cryptogram Christmas card. The stereo cryptogram adapted
to a Christmas card. Courtesy Robert Hayes.

form in each unit (Fig. 3-17). Examined in the stereoscope certain
letters stand out before the mass or lie behind it, and so form a
message. Another form consists of an innocent message, certain
words or syllables of which are stereoscopically displaced to pro-
duce a hidden message.

Such cryptograms are made through parallax control. The sig-
nificant letters or words are written with different homologous
separation from the rest of the message. When the messages are
typewritten, the first message, the left for example, is written
straightforwardly. Then in the correct position, the second mes-
sage is written with the significant letters or words omitted. When
this draft is complete, the paper is shifted carefully in the carriage.

THE STEREOSCOPE

The slightest possible motion is enough. One fourth a millimeter
displacement will produce well defined stereo relief. The use of
the typewriter maintains both type alignment and degree of stereo
relief, so there is no sign of the irregular relief seen in hand-
lettered messages.

When such cryptograms are printed in type, the characters are
separated by "thin spaces/' and for the second impression the ap-
propriate spaces are withdrawn, thus shifting the position of the

WHATEVER WORDS
YOU WRITE YOU
MUST STAND BY
YOUR STATE-
MENTS WITHOUT
C E A S I N G THERE-
IN YOU WILL
CONFRONT YOUR
OPPONENTS WHO
OFFER THEIR
OBJECTIONS TO
EACH SENTENCE
YOU COMMIT TO
PAPER

WHATEVER WORDS
YOU WRITE YOU
MUST STAND BY
YOUR STATE-
MENTS WITHOUT
CEASING THERE-
IN YOU WILL
CONFRONT YOUR
OPPONENTS WHO
OFFER THEIR
OBJECT! ONS TO
EACH SENTENCE
YOU COMMIT TO
PAPER

Fig. 3-18. Regular spaced cryptogram.

character. If ordinary 24 pound paper is used for spacing, the
difference in the two imprints will not be obvious to the unaided
eye, but the stereoscope will reveal a decided relief.

These are but a few suggestions as to the ways in which the
stereoscope may be used by the interested stereographer. But in
any experimental w r ork, one thing must be kept in mind; in fact it
must be kept in mind at all times. The stereoscope and the stereo-
grain must, independently and jointly, be in perfect adjustment.
This means a rigid, accurate stereoscope, preferably of metal con-
struction with lenses correctly aligned, and if it is of the lenticular
type the interpupillary must be correctly adjusted. The stereo-
gram must be exactly aligned along the base, the homologous
separation must be that correct for the stereoscope used and pref-
erably matched to the interpupillary of the spectator, although
this is hardly practical when more than one person is to view it.
The stereogram should be set at the infinity focus of the instru-
ment, and the viewer should have lenses accurately matched in
focal length to those of the camera. When such conditions are ob-
served, even the experienced stereographer will find new pleasure
in the complete realism of the subject revealed.

CHAPTER 4
CORRELATION OF CAMERA AND VIEWER

ORTHOSTEREO. Orthostereoscopy (true stereoscopy) is that
stereoscopy in which the objects when viewed appear in
their full life size and at their full natural distance. Direct vision
is automatically orthostereoscopic. It is also a result of a specific
method of stereogram making and viewing combined; and cannot
exist other than as a subjective phenomenon. There is no such
thing as an "orthostereogram," simply because a stereogram which
will provide ortho results with one viewer will not necessarily do
so with another.

In speaking of "life size," the term is used literally, but it
should be understood. Suppose you look at two men, each six feet
tall, one twice as far away as the other. You know that by the rules
of perspective, the actual retinal image of the nearer is twice as
tall as that of the farther. You can prove this by holding a ruler
at arm's length and measuring the image height of each man
against it, but in actual experience both men appear to us to be
equally tall, to have "life size," in other words. By experience we
see the distance, and from that unconsciously compensate for the
size discrepancy. Thus life-size means a size appearance which is
identical to that which the original object would have presented
had we been enabled to see it by direct vision.

To attain this objective it is essential that the camera and the
viewer have identical bases (lens separation); that this base equal
the interpupillary of the user; and that both camera and viewer
be equipped with lenses of identical focal lengths. In actual prac-
tice the deviation from ortho is not obvious until the viewer lenses
differ from the focal length of the camera lenses by some 30 per-
cent. The use of the standard base of 65mm does not introduce
any visible distortion even though the actual interpupillary is not
6510111. However, these are practical tolerances whose existences
have no bearing upon the validity of the theory. Similarly, in prac-
tice the camera base is often greater than that of the viewer, but
no visible distortion results.

It is obvious that to maintain orthostereo conditions renders
any attempt at conventional telestereography impossible, for the

60

CORRELATION OF CAMERA AND VIEWER 61

gain resulting from using long-focus camera lenses is exactly neu-
tralized by the loss resulting from the long-focus viewer lenses.
Thus no matter what the focal length used, the results are identi-
cal, with the sole exception that the shorter the focal length the
greater the area embraced within the field. Otherwise every object
has that size-distance character possessed by the original as viewed
from the camera position.

Inasmuch as both stereomicrography and telestereography are
of importance, it is highly desirable that some such techniques
may be used. Various methods have been suggested in the past,
but only parastereoscopy fully meets the demand for telestereogra-
phy. Parastereoscopy is substantially the same as orthostereoscopy
with the exception of one condition whose importance is far great-
er as a matter of theory than as a matter of practical application.
In fact, parastereograms have no distortion which is discernible
to vision. Measurement alone will reveal the difference. Para-
stereoscopy is made by strictly observing the PePax relationship,
which is done by altering the stereo base and focal length in exact
ratio.

If the camera and viewer are normally equipped with three-
inch lenses, then that is the normal for that particular pair of
instruments. If it is desired to use six-inch lenses, then the base
is doubled by moving the lenses to igomm separation. The re-
sulting stereogram will be viewed without perceptible distortion
in the normal viewer.

The use of long bases is difficult with the ordinary camera,
particularly when simultaneous exposures are desired* My first
PePax reflector was made upon the sliding tube principle, and
later I substituted one designed along the lines of the periscopic
stereobinocular, with barrels rotating in an arc to provide the
desired separation. This, however, was a purely experimental
device and has not been commercially manufactured. These re-
flectors are in every way similar to the conventional stereo reflector
with the single exception that the internal reflectors instead of be-
ing positioned immediately adjacent, are separated by the normal
stereo base. This construction makes the reflector applicable to the
normal stereo camera and provides any desired base up to the max-
imum, which may reasonably be tea times iKraaal, or 65 centi-
meters.

liU THREE-DIMENSIONAL PHOTOGRAPHY

Because so many inquiries have been received about this re-
flector it should be repeated that it is not available commercially.
It can be made by altering a war surplus instrument, but it is
subject to the very real disadvantage that the field is extremely
narrow. It is more suitable for experimental purposes than for
the production of satisfactory pictures.

Keeping in mind the fact that the orthostereogram reproduces
the original in full life size and at full natural distance, we shall
now consider those variations in manipulation which result in the
stereoscopic distortion of either size or distance, or both. This will
serve to indicate those errors of technique of which the stereogra-
pher should not be guilty, and even more to the point, it will in-
dicate some violations of traditional stereo rules which may be
made to great advantage.

In practice, orthostereoscopy is rare because the average viewer
lenses exceed camera lenses in focal length by amounts varying
between 25 and 50 percent. A 25 percent degree of discrepancy
yields an image in which the distortion is not readily apparent,
but a 50 percent difference will produce a distortion of spatial
dimension, perceptible to the expert stereographer.

FIXED AND VARIABLE FACTORS. Although there is an infinite
variation in the interpupillary distance of human beings, these
variations are within such close limits that we may establish an
arbitrary standard such as 65mm and henceforth regard that as a
fixed factor.

Once the exposure has been made, the parallax of that stereo
image is fixed.

The perspective dimension of the image (dimensions in the
picture plane) is subject to alteration in making the negative by
changing the focal length of the lens. It is subject to further alter-
ation in the process of projection printing by either enlargement
or reduction, and it is subject to final variation by the focal length
of the viewer lenses.

To sum up, we may say that, generally speaking, parallax is
fixed while perspective dimensions are variable. Hence it shall be
our goal to control the perspective so that it will always match the
parallax correctly or otherwise as we desire. It is true that in
parastereoscopy we do this indirectly by deliberately introducing
an abnormal base in making the negative, but as this has already

CORRELATION OF CAMERA AND VIEWER

been discussed we shall confine ourselves largely at this time to
the effects of abnormalities of perspective and parallax without
giving undue regard to balancing such abnormalities.

One important phase of this discussion is that of pointing out
the serious errors which arise, for example when a club or society
holds an exhibition or sends out a postal portfolio composed of
slides made to conform to a variety of standards. This inevitably
means that many stereograms will be examined in stereoscopes
wholly unsuited to the view. As a result, the stereograms are sub-
jected to criticism which is manifestly unfair. These criticisms
should have been leveled at the lack of stereo technique displayed
by the critic who failed to observe one of the great fundamental
laws of stereoscopy. 35mm viewers have different focal lengths
but the differences are slight, hence the modern amateurs can ex-
change slides freely. This is far different from the time when
the camera might have had g-inch lenses and the stereogram be
viewed with 8-inch ones.

Variations of Focal Length. 'Figure 4-1 shows the visual rays

Fig.

64 THREE-DIMENSIONAL PHOTOGRAPHY

when the two eyes view the objects A, B, C and >, as well as an
object at Infinity. It is well to note that although we speak of an
''object/' we actually refer to some prominent point in the visible
surface of that body. Thus the significant ray path is, in each in-
stance, truly represented by a single straight line from the eye
(or lens) to that point in the object. As we have already seen,
orthostereoscopic results will not be achieved unless these signifi-
cant ray paths are duplicated when viewing the positive stereo-
gram.

Figure 4-2 shows the same field as it is photographed by two
lenses LL and LR. In both figures CC marks the position of the
Cyclopean Center, or the position of synthesized stereoscopic
vision.

If the negatives fp-fp are rotated vertically and horizontally
about the common center of the lenses, fp-fp will coincide with
p-p of Fig. 4-1. The focal plane of the camera becomes the picture
plane of the positive, with an inversion of image.

Figure 4-3 is the same as Fig. 4-2 except that the lenses have a
focal length which is one-third less than normal (considering the
conditions of Fig. 4-2 as normal). The direction of all ray paths
remains the same but fp-fp is now necessarily nearer the lenses,
making the images smaller throughout. If viewed with lenses of
similarly short focal length, it is obvious that the relationship be-
tween Fig. 4-1 and Fig. 4-2 would also exist between Fig. 4-3 and
any such condition. Therefore we pass directly to the viewing of
this smaller image in the normal stereoscope, as illustrated in
Fig. 4-4. The points of intersection remain fixed because they are
a part of the physical silver image. When this pair of images is
placed in the normal picture plane the results are as shown. The
points at which the rays terminate indicate the images as actually
seen; the points with no connecting rays show the true positions
of the objects where they should be seen orthostereoscopically.
The difference indicates the degree and kind of distortion (object
D is omitted on account of space limitations).

In each instance the object is seen at a greater distance than
normal, and all dimensions in the line of the visual axis (depth)
are increased. Objects appear elongated in depth and far distant.

To make these distortions clearer, we shall supplement the

CORRELATION OF CAMERA AND VIEWER

Fig. 4-3* Camera focus

less than normal.

Fig. 4-4. Viewing stereo-
gram made with camera
focus less than normal.

basic diagrams with others showing the actual structure of the
image.

Fig. 4-5 is a key diagram to the series. The triangular object
A rf B"G" is seen by two eyes or, photographed by two lenses, at
EE'.Pn is the normal picture plane for the normal image A"B"C".
Pd is the picture plane for the distorted image D /r F"G"

Rays A, B and C connect the eyepoint E with the three signifi-
cant points A", B" and C" of the original object and also of the
orthostereoscopic image A rf B"G fr . The small circles indicate the
intersection of these rays with the plane Pn. Points D, F and G
are disposed in the same plane symmetrically to A-BC.

If we assume the stereoscope to have lenses which are half the
focal length of those in the camera, we place the plane Pd half-
way between the plane Pn and the eye plane as shown. Then
from A, B y C 3 D } F and G perpendiculars are drawn from the
original Pn plane to the Pd plane, thus locating A', B' y C', D f , P
and G' in the latter. Because D, F and G exactly correspond to A,

66

THREE-DIMENSIONAL PHOTOGRAPHY

Pd

Fig. 4-5.

Fig.

B and C, we may use the right side of the diagram for the new
condition to prevent too great confusion in the drawing.

The pomts D 3 F and G are fixed in the image, so when the im-
age (print) is moved nearer the lenses as it must be when these
have shorter focal length, we find that the outer* rays D and G

CORRELATION OF CAMERA AND VIEWER

67

intercept the original Pn plane at D" and G" instead of at D and
G. As D"G" must equal -4"C" and occupy a corresponding lateral
position, we find side D"G" located just halfway between the
eyes and side A"C". Thus we locate the viewed image D".F"G"
which is seen to be nearer the observer than normal. Because the
lateral width is truly the same as the original but the distance
only half, the apparent size of the object is increased. As might
be expected with the general condensation of depth, the altitude
of the triangle is lessened and the degree of loss is one-half. Thus
the degree of distortion keeps step with the original differences
throughout.

In the following diagrams of the triangle series, the various
measuring points are not numbered, only the triangle and the
planes being indicated. The measuring lines are shown but not
indexed.

Figure 4-6 shows in the graphic triangle form the error which
was illustrated by Figs. 4-3 and 4-4. As it is just the reverse of the

Fig. 4-7. Camera focus
greater than normal.

Fig. 4-S* Viewing stereo-
gram made with camera
foci&s greater than nor-
mal.

68

THREE-DIMENSIONAL PHOTOGRAPHY

key plate, it may be easily followed. Here we see the elongation
in depth and the increased apparent distance caused by having the
camera lens of shorter focal length than the viewer lenses or
having viewer lenses of longer focal length than the camera lenses.
It is all the same, but at times we overlook this obvious inter-
change of terms.

When the focal length of the camera lens is greater than that
of the stereoscope lenses, the reverse is true. Having gone through
both types of diagrams in detail, we do not believe it essential to

Fig. 4-9.

CORRELATION OF CAMERA AND VIEWER

69

repeat the details in each instance. Figures 4-7 and 4-8 show the
significant ray paths for this condition, while the graphic diagram
is shown in Fig. 4-9. Incidentally, Fig. 4-9 is the complete draw-
ing for the condition used in the key plate, Fig. 4-5. Longitudinal
distances appear to be lessened and solid objects are thinner in
depth, tending toward the "silhouette" form.

Variations in &se. When variations in base are considered,
the matter becomes more puzzling because we have complications
arising from subjective interpretation to confuse the stereoscopic
reproduction.

Naturally, unless we use some kind of special reflector eyepiece,
the viewer interocular must conform more or less closely to the
anatomical interocular, so these variations of base will be assumed
to apply to the camera, being either greater or less than the
normal.

B

Fig. 410. Camera base
less than normal.

Fig. 4 11. Fiewing stereogram

made with camem bm$e less th&n

normal.

THREE-DIMENSIONAL PHOTOGRAPHY

Figure 4-10 illustrates the significant rays for the condition
wherein the camera base is less than normal. The broken line in
/p-/p indicates that portion where the two bases of the camera
overlap. In Fig. 4-11 note that the original position of the object
lies along the line which bisects the two visual axes. In short, the
direction of the image is identical with that of the original object,
but it seems to have been moved farther away. The net result is
that of a larger object removed into the distance. As D is beyond
the limit of the diagram it is not shown.

The greater the stereo base, the greater will be the parallax.
Also, the nearer any object is to the eye, the greater is its parallax,
and conversely. In Fig. 4-11 the decreased parallax has given the
appearance of an object farther away, while the undiminished
perspective gives to that apparent distance an apparently greater
object size.

Figure 4-12 shows a base which has been increased by 50 per-

fl

/r c

Fig. 4-12. Camera base greater
than normal.

Fig. 4-13. Viewing

stereogram made with

camera base greater

than normal.

CORRELATION OF CAMERA AND VIEWER 71

cent over normal Figure 4-13 shows the result of viewing such a
stereogram. The object displays greater parallax and it appears
to be nearer the spectator than the original, but having the origi-
nal perspective it appears to be smaller than normal.

Figure 4-14 adapted from Liischer, illustrates graphically both
these base conditions. Assuming the use of the greater than nor-
mal base, L-L' represent the two lenses, while -' are the two
eyes. Note that for simplicity U and E r are given the same posi-
tion. It would be equally correct to place E and E' equidistant
from L and L f , but would involve a far more complex construc-
tion in the diagram.

In taking, the lenses form the image of triangle ABC from the

Fig. 4-14.

Fig. 4-15.

72 THREE-DIMENSIONAL PHOTOGRAPHY

two points L-ZA The line wx passes through L and intersects the
picture plane xz. Draw yz parallel to wx, through E to the picture
plane. The intersection of the rays LA, LC 3 and LB with xz are
repeated upon xz starting at point z. The significant ray paths are
then drawn from E through these repeated intersections, to form
the triangle A'B'C'.

It will be seen that this corresponds to the former description
of an object without perspective or parallax distortion, but which
appears to be nearer and smaller than the real object.

It is obvious that if E and E' are assumed to be the lens posi-
tions, and L-L f the eye positions, the same diagram will illustrate
the use of a less than normal base.

It should be said that the expression "appears to be smaller"
should be regarded as highly theoretical inasmuch as we have no
absolute size perception, size being always estimated by a system
of comparison. Therefore, as a matter of fact the abnormal size
factor is rarely observed unless the subject is so familiar that we
know its approximate size, or unless there is some external object
(as in trick stereograms) by which comparison may be made. On
the contrary, the differential distance perception by parallax is
surprisingly accurate.

Finally we have in Fig. 4-15 the explanation of the PePax
principle of maintained perspective-parallactic ratio. The lenses
F-F' have twice normal focal length and are used at twice normal
separation. The viewer lenses E-E' are normal. By the same rea-
soning used heretofore it will be seen that the original triangle
ABC will be seen in the apparent position A', B', C'. Here the
object appears to be at only half the distance, it appears to be
undistorted, and it appears to have natural size. This is the para-
stereoscopic image, and differs from the orthostereoscopic only in
the fact that a given space contains a greater amount of solids than
is physically possible, but because of the sequence system of hu-
man vision which prevents us from looking at the whole of a
scene simultaneously, we are unable to grasp the physical impos-
sibility in the form of any perceptible visible distortion. There-
fore the distortion of parastereoscopy is only that of knowing the
reproduction to be unnatural, not one of seeing that it is.

Hyperstereo.No discussion of variable base would be complete
without the mention of hyperstereo. This is the making of stereo-

CORRELATION OF CAMERA AND VIEWER 73

grams with a greater than normal base. It is a practice which
has been almost universally condemned, simply because it is a vio-
lation of the principles of orthostereoscopy. However, as the image
is not proportionately distorted, the subject has a natural appear-
ance, thus making the technique one of the most potent available
to the stereographer.

Because of the arguments involved and the importance of the
subject, it will be fully discussed in a later chapter, instead of
trying to deal with the subject in the limited space available in
the present chapter.

Convergent Axes. The argument is often offered that the best
stereogram results when two cameras are used, converging in the
manner of the human eyes. This argument, although quite valid
in special fields such as photomicrography of objects in a single
plane, is a result of faulty reasoning. The stereoscopic camera is
not a substitute for our eyes. It is a tool, an instrument which pro-
duces an imitation of the original scene toward which our eyes
function just as they would with the real scene. Our eyes still con-
verge normally when the stereogram is made in a parallel axes
camera.

The objection to the convergent axes system is that a very
definite spatial distortion is produced, and the greater the degree
of convergence the greater the distortion. When the camera axes
are parallel, the homologous distances of all images are directly
related to their real distances, but when convergent axes are used
this essential relationship is lost, and apparent distances vary
widely from the original ones. Any degree of convergence intro-
duces distortions of distance, and hence is not satisfactory unless
the subject matter is confined substantially to one plane.

The convergent stereogram (and the divergent as well) can be
made and can be viewed up to a limit, but they are never truly
realistic.

Figure 4-16 shows two cameras converged upon object A. When
viewed, as shown in Fig. 4-17, we find A is satisfactory, but B and
C are apparently farther away than their true positions, b and c.

It should be noted that the farther the object from the cen-
ter of convergence, the greater the distortion. Thus convergent
axes result in a variable degree of spatial distortion.

Only parallel axes will produce unifooBly satisfactory

THREE-DIMENSIONAL PHOTOGRAPHY
C

Fig. 4-16.

Cameras with
convergent axes.

Fig, 4-17. Viewing stereo-
gram made with cameras
with convergent axes.

grams under all conditions and of all types of subject.

STEREO RE-CREATION. It is now possible to understand why
the stereogram viewed orthostereoscopically does actually so re-
create the original scene that all of the major visual phenomena
experienced in viewing the original are also experienced in view-
ing the stereogram.

Referring again to Fig. 4-1, as the eyes converge upon A both
eyes are equally converged to a marked degree. Shifting vision to
B, the convergence is less, and both axes incline to the left. Pass-
ing to C the convergence is still less, and the dual inclination is
to the right. In looking at D while the inclination is still to the
right, both the inclination and convergence are less than for C.
Viewing a distant object, the axes are parallel, that is, there is zero
convergence and zero inclination.

If the stereogram is mounted so that the centers of the unit
images are not spaced to the correct degree of separation, then
when looking at some object in the far distance the eyes will either
converge and assume the position of looking at a relatively nearby
object, or they will diverge, a condition wholly foreign to normal
eyesight and usually painful, if not impossible.

Thus the stereographer has the responsibility of seeing that his
stereograms are mounted with the correct separation, as deter-
mined for the stereoscope he uses.

CHAPTER 5
MOUNTING, TRANSPOSITION AND SPACING

So FAR WE HAVE CONSIDERED only the essential factors involved
in making the stereogram, but there are some factors involv-
ing the relationships of the two images which must be considered,
TRANSPOSITION. One of the most important things the stereo
beginner must learn is correct transposition. To do this it is neces-
sary to know why the operation is required, w r hat the operation
actually consists of, and what happens if it is ignored.

Figure 5-1 is a plan view of a stereo field as seen by the two eyes
L and R. The field contains the objects A, B, C and D as well as
an object at an Infinite distance. In front of the eyes lies the
picture plane p-p. The intercept of die ray path in this plane
from any point in the field determines the perspective o the field,
and, when viewing the stereogram, the stereo images actually
occupy this field whence they are visually projected to their cor-

75

76

'THREE-DIMENSIONAL PHOTOGRAPHY

rect position in space. If this plane Is physically fixed by placing
in it a sheet of glass, for example, and the apparent positions of the
objects marked upon it, the result will be similar to that shown in
Fig. 5-2. CC is the Cyclopean center or the synthetic, stereoscopic

point of view.

i

oo D

9

CO

.

*

B

1-

*B

"

r

*

o

c

*A

'A

Fig. 5-2.

*

A

A

c

*

C

-*. |

'8 .

o

CO

a* .

.

Fig. 5-3.

However, if we examine the ground glass of a stereo camera, or
if we make a stereo negative, we find that it has the appearance
shown in Fig. 5-3. There seems to be some similarity between Figs.
5-2 and 5-3, but they are noticeably different.

Reference to Fig. 5-4 shows the reason for the similarity and the
difference. The field of Fig. 5-4 is the same as that of Fig. 5-1 with

MOUNTING, TRANSPOSITION AND SPACING

77

objects A y B y C and D included in the same positions. However,
instead of two eyes, we have two camera lenses LL and LR. The
rays from the objects pass through these lenses and form a dupli-
cate of the picture plane behind the lenses at fp-fp. It will be
noticed however that, due to the crossing of the rays in the camera
lens, these images are reversed in relation to the images in the
picture plane. Right and left and top and bottom are reversed.
This is familiar in ordinary camera work, and the two cameras

Fig. 5-4. Fig. 5-5.

which make up the stereo instrument act individually just like
any ordinary camera.

In Fig. 5-5 we have the deer as an object. In the original we
have arbitrarily designated the left and right aspects with the
letters "L" and * ( R" The original is duplicated at the top of the
figure. At the left we see the visual sequence. The two eyes see
the two stereo aspects L and R, and each of these is independently
inverted upon the retina. The neuro-psychological mechanism of
stereo vision then synthesizes these aspects to form the final repro-
duction of the original.

78 THREE-DIMENSIONAL PHOTOGRAPHY

At the right the camera makes two negatives (even color film
is developed as a negative before being converted into a positive)
as shown at A. From this a print is made and turned right side up
(B). However when this is done the left image is at the right and
the right image is at the left. Therefore, for the time being we
move step B to the left of the main column as shown. Restoring
the print to the upside down position of A 9 w r e cut the two pic-
tures apart and rotate each one upon its own axis, through 180
and then \\ r e have the correct relationship as shown at C. As a
matter of fact, w T e do not do this. Instead we return to B at the
left of the column, cut the two pictures apart and simply lift the
right image over the left until it lies at the right as shown at C.
Thus the rotation and the position interchange both accomplish
the end result. Rotation is what is done theoretically, the position
change or transposition is what is usually done in practice.

The two images in correct relationship C are now viewed by
the two eyes, w r hich instead of being fixed upon a single, common
original, have an original image for each eye. Thence the process
duplicates the steps D and E as first described for direct vision.

Referring again to Figs. 5-2 and 5-3, note that the infinity image
is above center in the visual picture plane, but below center in the
negative. Images A and C in the right picture plane occupy the
lower left quadrant, while B and D are in the upper right. In the
negative these positions are just reversed. Thus we find that the
right negative is simply the right picture plane reversed top and
bottom, right and left. The same is true of the left negative and
left picture plane.

Thus although the relative positions of all objects within the
field are inverted, the field at the left is the true left field and that
at the right is the true right field.

This brings us to a basic factor in stereoscopy. If the photograph
is identical with the image of the picture plane, the photographic
images of each object will occupy the position in the photograph
which the visual image occupied in the picture plane.

The two eyes do not converge upon a single image to look at
it. On the contrary, the stereogram provides a separate image for
each eye, and we look through the image, in a sense. The two
images of any given object are so spaced in the stereogram that
the visual paths from each eye to its respective image, will, if con-

MOUNTING, TRANSPOSITION AND SPACING 79

tinned, cross at a distance equal to the true distance of the original
object.

Thus the stereogram is not a physical object to be visually ex-
amined. Rather it is simply a guide or stencil which when visually
projected causes the two separate images to fuse at some point
out in space equal to the original distance of the object. This
visual projection, as can be seen from the diagram, is a very real
thing and not at all fanciful theory.

As we have seen, direct stereoscopic vision is possible only by-
virtue of a psychological visual projection, so this geometric visual
projection is quite simple, yet it is the keystone of all stereoscopy.

Thus it will be seen that the photographic image must be in-
verted to cause it to correspond to the original picture plane im-
age. However, in actual practice it is not usual to revolve the
images after they are cut apart.

METHODS OF TRANSPOSITION. Glass-based images. When glass
plates are used in the camera, and when the positives are printed
upon glass, the transposing printing frame is used. This is a frame
which is longer than the negative, sometimes half as long again,
sometimes twice as long. The frame which is as long as three
single images is the most common. The negative is placed at the
left side and the positive at the right. The two overlap in the
center where the printing opening of the frame is located. The
exposure is made. The frame is opened and the negative is pushed
to the right while the positive is pushed to the left, and again
the exposure is made. Thus the left negative im^ge is printed at
the right of the positive and vice versa.

It can be seen that this principle presupposes the *ise of rigid
materials which will conform to the limits of the frame used, but
glass is not the only material that can be used. If the positives
are made upon stiff sheet film, the transposing frame may be
used, and even when both negative and positwe are of roll film
it may still be used if care is observed,

Film negatives. It is customary when the negatives are made
upon film, to cut the two images apart, reverse the positions and
re-attach the two images with a strip of adhesive tape. Quarter-
inch cellulose tape is excellent for the purpose, although if the
negatives are not kept in individual envelopes, they will stick to-
gether in time.

80

THREE-DIMENSIONAL PHOTOGRAPHY

Many stereographers use a diamond point and cut glass nega-
tives in a similar way. Obviously when this is done the printing
requires only a single exposure as when printing an ordinary nega-
tive. This is a distinct advantage when printing upon paper, as
paper has not sufficient rigidity to make its use in the transposing
frame practical.

Print transposition. Some stereographers make their dual prints
from the untransposed negatives at a single exposure, then cut the
prints apart and mount them singly after transposing. This is the
most difficult method as it means that every print must be indi-
vidually transposed and that print alignment must be preserved
during the process of mounting. Both rubber cement and thermo-
plastic mountants are difficult, because the prints are inclined to

Fig. 5-tf. Stereo Realist mounting kit. The case
includes at upper left, the heater; center, film
forceps and sealing iron; right, film cutter. Low-
er left, the sorting box for films and lower right,
the mounting jig.

MOUNTING, TRANSPOSITION AND SPACING 81

shift slightly during the mounting, and an error of half a milli-
meter is sufficient to produce a poor print.

5jmm Stereo grams. Modern stereograms made on 35mm film
are the most easily mounted of all. The pictures are simply cut
apart, the film edge providing a positive horizontal alignment
base, or in the case of the Personal, a dual punch makes it possi-
ble to punch out both frames at once thus providing positive sep-
aration and alignment.

Realist films may be jig mounted. A mounting guide or jig is
provided for use with heat-adhesive masks. The mask is placed in
the jig, the cut films are placed in corresponding wells which
align the film, and the films are sealed into the adhesive mask
with a heated sealing iron. The mask is then mounted between
glass slips and sealed with tape. A special cutter is provided to
insure cutting the film perpendicular to the edges.

An alternative is a cardboard mount which is almost as rigid as
the glass slide. It has accurately diecut pre-formed pockets into
which the film is slipped. As the pockets are already spaced, it is
only necessary to keep the left and right films in correct order,
slip each into its own pocket and the slide is ready for use. This
method of mounting is proving popular for both hand viewing
and projection.

Another type of mount is the Permamount. This is a plastic
base into which three channels are molded in the form o a very
squat "H." A glass slip is dropped into the cross channel, then
the films are laid in the two end channels and another glass on
top. The whole is sealed and label provided by a die-cut, self
adhesive panel which seals the assembly into a permanent unit.

These are provided with three separations which compensate
for projection base and permit all types of subject from close-up
to distance to be projected in any sequence without adjusting the
projector control,

The Personal films are mounted in a similar way. Blank "reels"
familiar to those who have tised the ViewBaaSfer are provided,
These have accurately formed pockets into which tee film is
slipped from the edge inward. Because of the Sblanfc tab at the
end of the film this may be done ^Mhoat toncbiiag the film* The
accuracy of alignment is extremely satisfactory^ and the system so
easy that a child can learn to <lo it in a few u&lwttes. These reels

82 THREE-DIMENSIONAL PHOTOGRAPHY

are stamped with the consecutive number of the films as well as
the semi-circular and square notch symbols so that there is no
danger of confusing the right and left images. They are marked
in the camera and cannot be confused.

Because so many 35mm stereograms are destined for projection,
and because projection slides should be more precisely aligned
than others, what are the limits?

This all depends upon the individual spectator, so all that can
be said is to give a low average for the limits which apply to the
alignments discussed below.

The two images should not have more than one-half prism
diopter vertical deviation, that is a deviation which a one-half
diopter prism would compensate. This is equal to one unit de-
parture from identical positions for each 200 units of screen dis-
tance. So, if the screen is located four meters from the spectator,
the greatest advisable vertical difference would be two centi-
meters, or not quite one inch. Most stereographers prefer half
that as a working limit. And remember the angular displacement
of the visual axis; the amount of movement demanded of the eyes
depends upon the distance from screen to spectator, not screen
to projector.

Laterally the deviation should be less than three prism diopters
or about 12cm, say five inches, for the most comfortable viewing,
but usually the spectators can tolerate four and sometimes five
prism diopters provided the relative positions are such that this
means convergence, never divergence. Most people find a half
prism diopter divergence uncomfortable, while one prism diopter
divergence is impossible for many people. Exact limits of all
vergences depend, as has been repeated so often, upon the indi-
vidual. (Distances given assume 400 cm screen distance.)

Figure 5-7 shows the most frequent mounting errors in exag-
gerated and diagrammatic style. A is the normal mount with the
two images correctly aligned.

B shoivs the result of having the camera tipped endways. Such
a pair must be mounted as shown by the dotted lines in the right
figure so as to preserve the original horizontal alignment, but if it
is more than a degree or so off vertical, no correction should be
attempted because painful eye "twisting" will result.

MOUNTING. TRANSPOSITION AND SPACING

Fig, 5_7. Mounting errors.

C shows the result of cutting the films obliquely along the
sides. Only one image may thus be trimmed and the direction
may be either right or left, but in any event misalignment results.

D is found when one lens has a greater focal length than the
other resulting in one image larger than the other. This may not
be anticipated when using stereo cameras of reputable make.

84 THREE-DIMENSIONAL PHOTOGRAPHY

E is the result of perpendicular alignment but with one image
lower than the other causing a misalignment of horizontal homo-
logues.

ALIGNMENT. Lateral spacing. The lateral spacing is not critical
as far as relief alone is concerned, but if the stereogram is to ex-
hibit the same amount of relief as was seen in the original scene,
it is necessary that the prints be carefully spaced. This is done by
measuring the distance between the images of the same object at
infinity, such as the tip of a tree or a distant steeple. This distance
should be equal to the distance between the centers of the lenses
of the stereoscope. However, the error resulting from a slight
change in this measurement is not obvious and in many cases does
not injure the stereogram in any way.

Often 35rnm stereograms intended for projection have their
separations varied deliberately so that the separation control of
the projector need not be used during an exhibition. As will be
explained this prevents serious physical discomfort for many of
the spectators.

The separation of the unit images Is then, not at all a critical
factor, its importance being largely theoretical.

Convergence range. Lateral spacing affects the two images as
units, but in each stereogram, if there is relief, there is also a
variable convergence. If infinity and distant objects are alone in
the picture, the convergence from one to the other is slight. If all
objects in the stereogram are closely grouped, the change in con-
vergence is small, even if the group lies near the camera, but if
the stereogram includes both very near objects and distant ones,
then the convergence range or change in looking from near to
distant objects is large. This range is fixed at the time of taking
and the "remedy" must be applied before the exposure is made.
It is advisable to keep such range less than three prism diopters
when practicable.

Vertical alignment. While the eyes have a wide latitude in
fusing images of various degrees of lateral spacing, a small differ-
ence in vertical position causes an appreciable eye strain, and if
the discrepancy is more than imm, many people cannot fuse the
images at all. In good stereo this alignment should be kept within
an error of .25mm or less, and in the 35mm stereogram, the limit
is about i/6mni (a. 1 75mm = i/ 2 A)

MOUNTING, TRANSPOSITION AND SPACING 85

Rotary alignment. The worst fault of all is to have the true per-
pendiculars leaning at one side or the other, If the true perpen-
diculars in the two images are parallel and both slightly inclined,
the only immediate effect is a sense of visual discomfort. If the
true perpendiculars to the camera base line of the two images in-
cline in different directions the effect is at once noticeable. If the
inclination is slight, a definitely painful sensation is experienced
when viewing, but if the difference is great the eyes see two dif-
ferent images (diplopia). The term "perpendicular" always refers
to those lines perpendicular to the camera base, common to both
images. If the negative was made with a laterally tipped camera,
the images must be mounted with this degree of tilt. Thus in
stereo mounting, camera side tilt should never be rectified by
trimming as is the case wth the planar print.
The negatives, whether made in a stereo camera or in a single
camera with sliding base or stereo reflector, have a common base
line. This is a good guide for correct alignment. If the print bases
are also kept uniform, there is little danger of rotary misalign-
ment. However, if the two images have been made by a single-
lens camera held freehand, then it is necessary to determine the
perpendiculars by visual proof. The two prints are placed in the
stereoscope and turned about their centers until the comfort-
able matching position is found. This is marked and the prints
trimmed to that perpendicular. This operation requires some ex-
perience, and for that reason freehand exposures are best left until
some facility in stereo technique has been acquired.

** * * * * *

B <r> A C B A CD D C

Fig.

CADooB CD*o A B

Fig. 5-$&,

When the distance between infinity images is correct, aiad l>otfa
horizontals and verticals are in correct aligewent;, tibere remain
only those factofs of BK>mitiag wbfcb are more or less a janatter of
taste.

WINDOW TRIMMING. Figure g-8a and b show tt*e relative boo-

8f>

THREE-DIMENSIONAL PHOTOGRAPHY

zonta! distances separating the two images of the various objects
of Fig. 5-4 (homologous distances). Figure 5-86 shows the disposi-
tion of these homologous points in the un transposed stereogram,
while Fig. 5-8*7 shows the homologous distances of the transposed
image. Notice that the distance between points A-A is the least of
all and that as the original distances of the objects increase, the
separation also increases.

In the finished stereogram the distance between homologous
points is directly related to the distances of the original objects

Fig. 5-9. TJie Window (A). This is the conventional window, placed

nearer the eyes than the flower. The flower spacing is constant in all

examples, so the spacing of the mask shows the variation which places

the window at different distances.

Fig. 5-10. The Window (B). Here the window is in the plane of the
tip of the stamen of the flower.

MOUNTING, 'TRANSPOSITION AND SPACING S7

from the camera. The less the homologous separation, the nearer
the object.

If a mask be prepared with two openings whose corresponding
sides (that is, the two left sides, or the two right sides) are sepa-
rated by a distance of, let us say, imm less than the homologous
distance of the nearest object in the foreground, and if this mask
is laid over the stereogram, the stereoscopic appearance of the
mask will be that of an opening or "window" nearer the observer
than the nearest object. In short, the appearance will be that of
looking out through a window at the scene.

Obviously if the prints are trimmed exactly to match this mask
and are mounted upon a black or dark card, the general effect
will be the same. The mount will form a window nearer the eyes
than the scene itself. This is known as "window trimming." Many
stereographers consider it essential. However, the mere fact that

Fig. 5-1 L The Window (C). The window is in the plane of the petals

of the. flower.-

striking effects are often obtained by deliberately throwing the
window back into the mid-distance is sufficient proof that conven-
tional window trimming is nothing more nor less than a personal
preference of the individual. (Figs. 5-9, 5-10, 5-11, 5-12, 5-13, 5-14*
and 5-15.) (Note that the separations of the flower images are the
same in all seven slides.)

To produce the window effect it is necessary that the images be
so trimmed that they appear to be farther away than the frame
produced by the edges of the prints. The "frame" must conform

88 THREE-DIMENSIONAL PHOTOGRAPHY

to stereoscopic principles, both must be the same size, the hori-
zontal edges must exactly coincide, and the corresponding edges
must be regarded as homologous points in the stereogram. It will
be found that when the trimming has been completed, the left
unit picture contains more of the right side of the subject than
does the right unit picture.

It must also be remembered that good stereoscopes are so made
made that the field is fully masked, so the effect of trimming for
the window effect is lost anyway. The same thing applies to the
dictum that only black card mounts shall be used. In a good

,, ' '"*,?' , ": i3fii^^mmmM
%# ^^ WB^^S

* ^-iitAil

9*&&'$

Fig. 5-12. The Window (D). The window is just back of the flower.,

but in front of the background. This is perhaps the most effective

position for this particular subject.

l^^f

Fig. 5-13. The Window (E). The window is in the foreplane of the

background. This is also very effective and might be preferred to D

by some technicians.

MOUNTING, TRANSPOSITION AND SPACING

Fig. 5-14. The Window (F). The window is in the most distant plane
of background, but it obscures details in the near background, thus

causing confusion. This is the least desirable of the group, with A the

next least desirable, although A is the position which is supposed to

be essential. (Hibiscus used as subject.)

Fig. 5-75. Angular window. The plane of the window is not neces-
sarily perpendicular to the visual axis. It may lie at an angle as shown
here, although the position is not recommended for any but demon-
stration purposes.

stereoscope the mount is not visible, so what difference can it
make? Many stereographers use various colors of mounts to dis-
tinguish among various classes of subject matter. This applies to
paper prints and viewers intended for them.

In modern stereography, the films are mounted in ready pre-
pared mounts. These are occasionally supplied with a special

90 THREE-DIMENSION \L PHOTOGRAPHY

window for dose-ups, but this is obtained only by the sacrifice oE
portions of the film area, making the picture field considerably
narrower than normal. The value of such "special window"
mounts is open to serious question. Most 35mm workers use the
"standard" mount for all types of subjects.

It is essential that the two images be vertically aligned with
great care, and that their true perpendiculars be both perpendicu-
lar and parallel. It is advisable to have the spacing of homologous
objects at infinity equal to the spacing of the stereoscope, but fur-
ther than that the arrangement is purely a matter of personal
preference with no valid arguments for or against any particular
stvle.

CHAPTER 6
STEREOGRAPHIC TECHNIQUE

ONE HUNDRED YEARS AGO the photographer, professional or
amateur, was assumed to be a stereographer as a matter of
course. All photographic shops carried full lines of stereo equip-
ment and stereograms by a number of nationally known stereog-
raphers were readily available.

Later, specialization set in and stereo become a special field and
one in which the amateur took the lead. In that fact we have the
key to the loss of stereo popularity. At the end of the last century
it seems that in any activity human free will was abhorred as posi-
tively obscene. Everything from dawn to midnight w r as regulated.
In short the period w T as precious in the extreme. Stereo seemed to
fall into the hands of a group which was unusually rule-minded.
The result was to surround stereo with a wall of rules and regula-
tions which left only the bare possibility of making an occasional
exposure. It seems unlikely that any stereographer could observe
all of the rules and make more than a half-dozen stereograms in
a year!

Today stereo is popular, and fortunately our mentality is such
that when w r e hear of a rule we immediately want to see if it can
be successfully broken in stereo it usually can. Today the stere-
ographer considers an afternoon a loss if he fails to expose a half
magazine, that is, some 15 stereo exposures; and 50 or 100 is a
normal week-end production. Obviously these are not master-
pieces of art, nor are they intended to be such. The modern ama-
teur has learned that stereo reproduces what he originally saw,
and the stereo camera preserves the highlights of pleasant week-
ends or vacations. Many of the stereograms are beautiful, some of
them exquisitely so. But the point is that the stereogram of today
is made, not because the stereographer believes himself to be a
great artist, not to impress a group of friends or club members, not
as an excuse to strut pompously before a group, but purely be-
cause of the pleasure it affords the maker and his friends. This
fortunate and wholesome attitude has revived stereo and will
keep it alive.

Therefore, in this discussion of stereo technique, we shall de-

92 THREE-DIMENSIONAL PHOTOGRAPHY

liberately cast off the old burden of ridiculous rules and discuss
those common elements of techniques which affect the result.
You will not find a prohibition in the chapter, only a few sug-
gestions with an explanation of probable results if you do
so-and-so.

The camera is yours. You bought the film. If you wash to lay
the camera on its back and shoot the whole roll making empty
pictures of the room ceiling, that is your privilege. In fact if you
do the same thing outdoors you might get some interesting cloud
effects! You do whatever you wish with your camera. If you find
some of the disappointments foretold in this chapter, you will not
have wasted time and film because to the hearsay of this discussion
you will have added direct personal experience for which nothing
else can adequately be substituted.

By all means make experiments. Most useful additions to hu-
man knowledge have been made by skeptics who wished to prove
(or disprove) some current belief. There have been some highly
successful stereograms (and photographs as well) made by a de-
liberate violation of some widely recognized rule.

In any such discussion as this it is necessary to make decisions
as to what subjects are to be included, what to be ignored. Stereog-
raphy overlaps planar photography, and most of the purely tech-
nical aspects of both fields are identical. There is no room to
include all conventional photographic technique, and no reason
for so doing as it is a subject which has been repeatedly discussed
by a number of writers. At the same time there are some aspects
of conventional photography which are of such particular interest
to the stereographer that they must be discussed at length, as in
the two chapters which follow this. Therefore the points discussed
will, in many instances, be familiar ones; some will be new.

This discussion will, generally, assume that color film is being
used because color is one of the important elements of stereo.
It may be said, however, for the benefit of those who use mono-
chrome processes, that the negatives made should be rather soft,
that is fully exposed and not overdeveloped.

Color exposure, of course, permits little control through ex-
posure variation, but the effects of under and overexposure affect
stereo more adversely than in the case of planar photography.

STEREOGRAPHIC TECHNIQUE 93

Filters are quite as important as in planar-monochrome work,
but this subject is discussed in detail in a later chapter.

Choice and arrangement of the subject is a matter of pictorial
treatment and will be discussed later.

The stereo camera differs from the conventional in being two
cameras joined side-by-side, and the actual manipulation of the
camera must take this into consideration. Because the camera
lenses have the relationship which characterizes the two eyes, it is
desirable that this relationship be maintained. Thus if you stand
upon a hillside, your feet, ankles and legs are so positioned that
the body is maintained in a vertical position, and your eyes are
maintained in a common, horizontal axis. When you use the
stereo camera, you should take great care to see that the camera is
not tipped up at either end.

It is often said that the stereo camera must be kept perfectly
level. This is far from the truth. You can tip it up or point it
down with far greater freedom than the planar camera because
the stereo camera realistically retains the vertical perspective of
nature. In the field of drawing, we have only that perspective
which applies to the horizontal. Our world is relatively flat, so the
need of vertical perspective has not been felt until recently. As
a result, a picture of a tall building, made with a conventional
camera, violates our training (although it is perfectly normal, be-
cause perspective does act vertically as well as horizontally). We
say the building appears to be "falling over backward/' In the
stereogram the building has a normal appearance, and if the head
is tipped upward in viewing such a stereogram, the realism be-
comes astonishing.

It would be well, perhaps, to digress a moment and explain that
the cooperation of the spectator can add tremendously to the
effectiveness of stereo. If the spectator does not wish to co-operate,
he is not interested and the loss is his own. This is true of many
phases of stereo, and forms a significant part of technique. Do not
waste time trying to convince anyone who is not interested or who
wishes to argue the question. Stereo is for enjoyment, and if any-
one prefers to indulge in other forms of pleasure, surely the choice
is his. But those who really do enjoy stereo find they can increase
viewing pleasure by just such ''orientation" aids as bending for-
ward to view stereograms made with the downward pointing camr

94 THREE-DIMENSIONAL PHOTOGRAPHY

era and vice versa. The reason is that the sense of balance changes
as the head is moved, and when the balance of looking upward
is present, the appearance is rendered even more realistic.

But to get back to the level camera. As a rule the camera should
be level from side-to-side, that is, one lens should not be higher
than another. The reason is perfectly simple. The stereo picture
cannot be twisted and corrected as could a similar planar photo-
graph, because the common central axis must be retained if un-
comfortable vision is not to be encountered. If the camera is
tilted, the pictures are mounted w r ith the scene at a slant. If this is
done, the stereogram is visually acceptable, but hardly satisfactory
esthetically.

As far as the rule is concerned it might be stated thus, "If you
want your pictures to be straight, keep the camera level from side-
to-side.'* As a matter of fact, the camera can be side tilted to pro-
duce some highly fantastic effects, provided the conditions are
right.

Another factor is that of the tripod. Color film is still relatively
slow, and there are very few people who can make a needle sharp
film with a hand held camera operated at a longer exposure than
1/50. Most of you w r ill ignore the advice, of course, but those of
you readers who do make use of a tripod will be more than repaid
by the great improvement in the quality of most of your shots, and
you will be far less handicapped in losing those subjects which
require 1/10 second or more exposure.

The stereogram should be sharp throughout if the full realism
is to be retained. However this "rule" should not be given too
much consideration when the loss of a picture is involved. If you
can obtain a satisfactory stereogram which includes some blur,
and cannot make an exposure at all with the aperture necessary
for all-over sharpness, by all means do the best you can. A stereo-
gram of an interesting subject, even if subject to some technical
criticism, is far better than no stereogram at all.

The "rule" actually originated in the days when extreme soft
focus was a fashion and every photograph had a series of abruptly
changing planes. This was said to imitate the effect of human
vision, although why such an absurd statement should have been
made by men and women whose own eyes constantly wander from
point to point, cannot be understood. Many stereographers went

STEREOGRAPHIC TECHNIQUE 95

heavily into soft-focus stereogram making, with results which are
better imagined than seen.

But, all in all, the greater the sharpness of the stereogram in all
planes, the more satisfactory it will be.

Of course to produce this effect, inasmuch as the best stereo-
grams have some relatively nearby object and many of them in-
clude backgrounds at infinity, a small lens aperture must be used.
When possible, apertures of f.'ii and f/i6 are advisable, but ex-
cept under the most intense light, this means exposures of such
length that the tripod is necessary.

There are many times when it is necessary to use larger aper-
tures, even to the extreme of .'3.5. Then, too, the shorter the
focal length, the greater the aperture which can be used without
loss of too much depth of field. In short, the matter is one which
you should decide for yourself, but make a practice of referring
to depth of field tables (or to the depth scale on your camera if
it has one). Then use the smallest aperture which includes the
desired field. If you can cover the field at f ,/8, there is little point
in stopping down to f 16, and the quadruple exposure gained
aids a lot in balancing exposure against necessary shutter speed.

That matter of having an object relatively near is not a rule at
all, it is simply a reflection of normal conditions. It is not often
that you find yourself in a position where there is nothing before
your eyes for a distance of 100 feet or so. Upon a cliff edge or at
a window in a tall building perhaps, but such conditions are not
common.

When you take a walk in the country, or in a large park, you can
see quite easily that details in the distance, say at 100 yards or less,
are not so sharply differentiated in spacing and relief as are objects
nearer at hand. Of course the size of the object enters into this,
but for ordinary objects, people for example, a distance of 15 or
20 feet reveals more contour than can be seen at a distance of 100
feet. It is just this fact that makes the 15 foot stereogram more
definite in relief than the one made at 100 feet.

This of course leads straight into the question of the distance
limits, but this has been discussed elsewhere in this book. It is
enough to say here that thousands of amateurs who never heard
of the limits "have made stereograms at 30 inches and have been

96 THREE-DIMENSIONAL PHOTOGRAPHY

satisfied with the results. Nothing more than that need be said
here.

The subject of close-ups, however, does bring up a most im-
portant question of stereomatics. Those who are disappointed
with their stereo souvenirs of a vacation are usually so because
the stereograms fail to reveal certain things which are affection-
ately remembered.

Let us borrow from the motion picture. The successful motion
picture has two elements of the utmost importance. First it has
continuity. The amateur film should be so made that it tells its
story so well that most of it could be grasped without the use of
titles. (We refer of course to the usual silent amateur film.) The
second element is that of visual transport. The audience sees a
scene as a whole. A distance change gives a more detailed aspect.
Another change and the semi-close-up concentrates attention upon
a relatively small area and picks up details and finally the close-up
pins attention to one particular object or character, and detail is
paramount.

Let us take a simple example. You take a walk in the park and
see a tree in bloom. You want to preserve this scene, so you make
a shot of the tree. When the stereogram is done, you show it to
your friends and find that to tell the story you have to talk on
and on. You tell of the day, where you found the tree and the
exact appearance of one of the flowers. The slide may be beauti-
ful but the narrative easily becomes boring. Now let us see how
this story can be told stereomatically.

1. Entrance to the park, groups of people visible within the
park.

2. A portion of the park. The tree in question not particularly
noticeable among many other trees.

3. The selected tree fills the frame (with due allowance for
spacing of course).

4. Semi-close-up shows one or two branches. The individual
flowers are visible, but structure may not be.

5. Close-up. The structure of the individual flowers can be
seen easily.

Certainly it takes five exposures instead of one. But when it is
all done you have a complete narrative of your experience. In-

STEREOGRAPHIC TECHNIQUE 97

stead of depending upon your (probably inadequate) power o
description, the spectator can enjoy the same sequence of emotion
which you felt when making the slides. You do not have to em-
bark upon a long and boring dissertation. The slides tell more
than you could relate in a half hour. Try it and you will be con-
vinced.

A vacation, for example, could start out with the preliminary
preparations. It would include: (a) start, (b) the road back home
as you leave town, (c) a few sequence shots along the road, in-
cluding road signs to provide location, (d) camp entrance, (e) un-
packing, (f) the first night, and then start the actual record of
camp or hotel or any other vacation spot.

Many make a good start and then let the sequence lag, with the
result that the story is spoiled. Do not forget the departure, the
trip home, the arrival and the first evening at home.

In short, if you want to make the most of your stereo record
without it becoming burdensome, prepare a tentative scenario
which of course is always subject to alteration and addition at
any time.

In closing it is well to point out that the slide does not show
what you actually saw at the time of making the exposure, it
shows what you should have seen at that time had you used your
eyes! In short, do not become so preoccupied with your principal
subject that you overlook undesirable elements in the background
and surroundings.

These are the principal points of technique other than those
discussed in specific chapters in other parts of this book. You
should feel as free with the stereo camera as with the planar type,
and as long as you maintain the side-to-side level position, try to
get sharp definition and observe the normal technique for con-
ventional cameras, you can do about as you please and obtain
good results. Yes, you can even use unconventional angle shots if
you watch the background.

NON-WINDOW MASKING. There are masks used occasionally in
stereo which are not used for the purpose of producing a specific
window effect; or perhaps, because the window is produced by
them, it might be more truthful to say that the window is not the
primary object of such masking.

Of course there is the fundamental mask which is used to elimi-

98 THREE-DIMENSIONAL PHOTOGRAPHY

nate portions of the camera field which contain distracting ob-
jects; to limit the field to the object of primary interest. This is
usually more of a commercial or record type of masking than
pictorial. But this leads into the general field of masking.

Planar pictures are usually masked to a rectangle, less often
to round or oval and at times "fancy" masks are used such as a
star, diamond, leaf, fish or the like. We are not concerned with
any question of taste or advisability; only with the fact that these
various masks can be used with stereo if desired. As a matter of
interest it may be added that the keyhole mask often used in
planar photography is more effective in stereo because the win-
dow placement plus the shape gives the effect of really looking
through a keyhole.

The actual masking is done just like the window masking. The
two openings must be of the same size and shape, and the two
must be carefully aligned. This is usually accomplished by care-
fully drawing the outline upon a piece of tracing paper which is
then used as a printing mask.

Masks are positive or negative. The negative mask is the most
popular because without further mounting it provides a black
frame for the stereogram.

The negative mask is made by cutting out from opaque paper
two shapes corresponding to the mask openings. These are aligned
and attached to a piece of tracing paper. Guide lines are marked
upon the paper.

This negative mask is placed in the printer and a sheet of paper
exposed behind it. On the back of the sensitive paper, the out-
line of the mask is traced. This paper is then exposed to the trans-
posed stereo negative, the pencilled lines serving as a guide to
locate the centers of the object symmetrically within the mask
openings.

The choice of mask is largely determined by the characteristics
of the negative. If the negative is thin and open, so that it gives a
dark to black background, a positive mask is used (Fig. 6-1), one
with the mask openings cut from opaque paper. If the back-
ground is medium to dark in the negative (medium to light
printer), the negative mask is used. Thus through tonal contrast
the shape of the mask is visible.

STEREOGRAPHIC TECHNIQUE !K>

It is not advisable to carry the use of these masks to the extreme.
A set of four or five negative window masks may be used nor-
mally, but the fancy shapes should be reserved as a kind of spice

Fig. 6-1. Masks (A). There is little value in a mask when the hack-
ground of the picture and the mask have similar tonal values. The

shape of the mask is lost. (The flower is a night-blooming cereus.)

Fig. 6-2. Masks (B). Fancy shapes may be used, but they are rarely

successful unless given a conspicuous position. This one places the

window in the plane of the stamen.

to the collection, and used sparingly to preserve the novelty ap-
peal (Figs. 6-2, 6-3 and 6-4).

FILING AND STORAGE. Any collection of stereograms rapidly
grows to such an extent that no hit-or-miss filing system (or lack

100

THREE-DIMENSIONAL PHOTOGRAPHY

of such system) will serve. Therefore the stereographer should
start some kind of filing system with his first roll of pictures.
There are special cabinets of various types on the market, and

W^>^18r''' : -

Fig. 6-3. Masks (C). The diamond shape is acceptable if novelty
masking is essential, but plain masks, either circular or rectangular,

are best.

Fig. 6-4. Masks (D). Of all the non-rectangular shapes, the circular
(and oval) is perhaps the most generally acceptable.

the choice will depend largely upon the type of mounting you
use.

Glass and other thick slides are best filed in a drawer cabinet
equipped with guide slots to keep the slides separated. These

STEREOGRAPHIC TECHNIQUE

101

slots are usually Indexed and blank cards supplied. When card-
board mounts are used, two slides may be filed in each division
thus doubling the capacity of the drawer.

However, when rigid card mounts are used, the individual di-
visions may be omitted, but it is advisable to have the drawer
divided by a partition about every two inches more or less to pre-
vent a small number of slides from becoming mixed in a long
drawer.

It is not necessary to buy a commercial cabinet. In fact, if the
collection promises to grow to any great extent it would be ad-
visable to make or have made a cabinet with a capacity of not
less than 5000 slides. You should allow about 18 inches of drawer
space for each 100 glass slides or about eight inches for each 100
rigid card slides.

INDEXING. The collection does not have to be very large before
indexing is necessary and this necessitates adding a number to

1 '

i

1 I '

1 I '

I

1 I '

1 1

1

1 1 !

1

'. I

1 I :

1

1 '

I

1 I

to .2

B C

2

Fig. 6-5. Sli de Indexing.

102 THREE-DIMENSIONAL PHOTOGRAPHY

each slide. To make the collection readily accessible, this number
should be placed upon the edge of the slide so that it may be read
without removing the slide from the drawer.

The glass (or plastic) slide may have a small number, written
upon paper and sealed over the edge by a strip of transparent,
cellophane tape.

It is easy to work out a complete indexing system for rigid card-
board mounts. These mounts are stacked so that each slide over-
hangs the one below it by i/ s inch as shown in Fig. 6-5 (i). In
this position a color stripe is painted over the whole pile of ten in
one straight line as shown (B). When the pile is straightened up
as in (2) of the same figure, the colored portions assume a stag-
gered sequence. If any slide is removed its position is easily ascer-
tained by the break in the regular sequence. This provides the
basic group indexing for ten slides.

The second ten (11 to 20) are indexed by painting the stripe
at D of Fig. 6-5 (i). This will place the staggered indicia at the
right of center.

The actual numbers are color coded in spectral sequence, i.e.,
1-20 =z violet; 21-40 = blue; 41-60 = green; 61-80 orange; 81-
100 red. This provides for a sequence of numbers up to 100.
For multiples of 100 the center stripe (C) is color coded.

1-100 = plain 501-600 = yellow r

101-200 = gray 601-700 = orange

201-300 = violet 701-800 = scarlet

301-400 blue 801-900 = brown

401-500 = green 901-1000 = black

For the second thousand a second stripe is added to (C) so there
are TWO narrow stripes down the center. The marking is car-
ried on to any desired extent.

Classification is also desirable, and for this purpose the ends
(A) and (B) are color coded according to any color code you may
wish to use. For example you might have for (A) the following:
violet = scenic yellow = water

blue i= flashlight orange = action

green = figure red = animal

For (B) you will have a code for each of the "A" classes. For ex-
ample under Green A (figure) you might have:

STEREOGRAPHIC TECHNIQUE 103

violet == portrait yellow = group

blue = draped orange = genre

green = nude red = action

If you wish to carry this classification to the extreme you might
consider such a subject as a nude dancer photographed with strobe
light, then at "A" there would be a combined blue and green tab

Fig. 66. Realist filing cabinet. The three
drawers hold 48 glass or 96 cardboard mounts.
The lower drawer takes the viewer. This com-
pact cabinet affords protection for slides and
viewer in easily portable form.

while at **B' 7 there would be a combined red-green tab. Thus you
can provide visual indica for a whole collection which will enable
you to select any individual slide by reference to an index or to
select any type or group without referring to the index.

Some such classification is essential if the collection is not to
lose its value, because hunting through even two or three hun-
dred slides is such a tedious task that too often the effort will not
be made.

It is difficult to overemphasize the importance of not allowing
your collection to become unwieldy through absence of adequate
indexing.

CHAPTER 7
FLASH IN STEREO

MOST MODERN STEREO CAMERAS are equipped with internal
synchronization for flash, but too many stereographers
neglect this technique and so lose a great deal of pleasure which
would otherwise be theirs.

For some reason the idea persists that flash is restricted to mak-
ing photographs at night, and a great number of stereographers
still lower or extinguish the room lights wiien preparing to make
a flash exposure. This is all wrong. Flash is intended for use
under normal conditions of artificial lighting or in full daylight.
That is the reason for the synchronized flash instead of the "free"
or "wild" type.

Flash is a technique whereby the photographer can produce a
light of sufficient intensity for making the exposure in such a
short time that the great intensity is not painful and has no
significant heating effect. To produce this light at just the right
instant, synchronization is used.

For older types of shutters, there are various types of mechani-
cal synchronizers which ignite the bulb and open the shutter at
the same instant. The whole action may be mechanical, or the
motion of the shutter may close a contact to ignite the bulb, or
pressure upon a button may ignite the bulb and activate a sole-
noid which operates the shutter. But in modern cameras, the con-
tacts are usually built into the shutter itself so all that is necessary
is to provide a socket for the bulb with associated reflector and
the battery case.

Synchronization may be either for a 5-millisecond delay or for
a somillisecond delay type of bulb.

The F (SF or SM) bulb has a duration of about 1/100 second
so it may be used with slow shutters in inexpensive cameras for
relatively high speed shots. The M (5 or 25) bulbs are wire-filled
and have a broader peak, so they are best used with a shutter
which can be set for the desired speed. Synchronization set for
one of these types will not necessarily serve for the other, so be
sure which type your camera or gun is set for. The midget M bulb
(5 or 25) is wire-filled, the F bulb (SF or SM) has no filling. The

104

FLASH IN STEREO 105

M has about 50 percent greater brilliance, but on short exposures
some of this is lost by clipping out the peak brilliance with short
exposure.

Although these designations were received from an authori-
tative source, there seems to be some confusion about terms. As
used here ct F" simply means the "empty" bulb and the tk M" a
wire-filled bulb.

For general use with Kodachrome, the F bulb is probably the
best. Choice often depends upon various circumstances of course,
but without good reason for using another type, we recommend
the F bulb, unless your camera shutter is synchronized for other
bulbs. These bulbs provide an excellent color match for arti-
ficial light film without a filter.

We shall not give details of the equipment and its attachment
to the camera as this information is given with each flashgun as
it applies specifically. With the Realist, for example, the flash re-
flector slides into the "shoe" on top of the camera, the shutter set
at 1/25 or 1/50, a bulb inserted in the socket and the exposure
made as usual. The Personal and Stereo Vivid have synchronizing
contacts of somewhat different design. More elaborate externally
synchronized guns are attached according to the manufacturer's
instructions. The Verascope uses a Busch gun of condenser type.

ExposuRE.The question constantly arises, "What exposure
shall I give with flash?"

Although the flash has high intensity, it is used so close to the
subject that its effect varies sharply with distance. According to
the inverse square law, the exposure at eight feet must be four
times that at four feet. For this reason, the flash exposure is based
upon "guide numbers." These guide numbers take into considera-
tion the brilliance of the light, the sensitivity of the film, and the
distance of the subject. Thus the guide numbers change as the
film sensitivity changes.

Let us take the F bulb, used with film of 10-16 ASA sensitivity.
Reference to the exposure guide shows that up to 1/100 second
the guide number is 44. We now turn our attention to the sub-
ject. The focusing scale tells us that it is eight feet distant. We
therefore divide 44 by 8, which gives us 51^. We therefore set the
diaphragm at f /5-6 which is close enough to the 5.5 as determined
by computation.

HM5 THREE-DIMENSIONAL PHOTOGRAPHY

You may avoid complex mental computation by using a guide
number close to that given. Thus, for a nine foot distance, you
may use 45 instead of 44, and the error will not be at all serious.
In fact surroundings often alter the effective light intensity to a
considerable degree, yet there is enough tolerance to give, usually.,
good results from an approximate guide number.

There are conditions which call for some departure from the
computed exposure. For objects which are very dark, use one-half
stop larger aperture, for those which are light, one-half stop small-
er. Note that this also applies to rooms. In a small room with
white walls a reduction is required, but in a large room particu-
larly with dark walls, use the increase in exposure. For outdoor
work, use a full stop increase. To do this compute the exposure
first and then add one stop, do not use one-half the original guide
number.

It may be that you will find it necessary to alter the guide num-
bers after some experience. Different types of reflector make a
difference in the intensity of light reflected from the subject, and
that is the factor of importance. If the light is concentrated it will
be brighter, but if a wide beam reflector is used, it will be less
brilliant.

FLASH POSITION. There is a distinct relationship between the
effect produced and the position of the flash bulb in relation to
the lens. If the light is close to and directly above the lenses, the
lighting will be very flat, and projections along the camera axis,
or in the direction of the camera, will be minimized. For example
if a model is facing the camera directly, the nose will be short-
ened, the facial expression altered, the eyes set flush in the face
instead of being in their sockets. In short the face is flattened and
robbed of expression.

If the flash is supported at the side of the camera, and lies at the
same side of both lenses, there will be enough cast shadow to
remedy, to some extent the fault just noted. This is the position
used in most press photography, and for purely record work it is
acceptable, but it is still far from good.

The best position for the flash is above and at one side of the
camera, the distance from the lens at least 30 inches, and more
if the subject is more than four feet from the camera.

The illustrations which illustrate these points were made with

FLASH IN STEREO

107

a display mannikin as the model so that any difference must neces-
sarily be a result of the technique, not of any change in the model.

Fig. 7-1 .

Figure 7-1. Note that there is a slight halo of shadow all around
the head. You will also see that this halo at the right is just a bit
more broad in the right image than in the left one. Examine the
nose with a stereoscope, and you will see that its projection is
somewhat vague. In fact you are inclined to overlook, the fact that
there is a nose there.

Fig. 7-3.

Figure 7-2. This is made with the flash mounted at the side of
the camera upon the conventional press-camera bracket. Note that
the shadow at the left is somewhat wider than it isi in the first
illustration, but of much more importance, note that the nose

108 THREE-DIMENSIONAL PHOTOGRAPHY

now lias a far more definite modelling than in the first. However,
the pointed shadow beneath the chin is about the same as in the
first and is not at all satisfactory. In short this is better than the
first, but not much.

Fig. 7-5.

Figure 7-3. The light was held, freehand, about 30 inches from
the camera, above and at the right. It was fired by an extension
cord and is typical of * 'extension flash/' Note that while the left
ear and cheek are in deep shadow, the face modelling is far better
than in either of the others, and that even in the flat, the face
appears to have more roundness than in the first two. Compare
the lips with those of the other two. The modelling is far better.
In short, although the total relief of the head is not greatly altered,
the modelling of detail grows as the lighting becomes more angu-
lar. This is in direct accord with the visual effect, as all portrait
photographers and all artists know. Light affects stereo quality
simply because it affects visual quality.

These exposures were made under average amateur conditions
by an amateur stereoscopist of average experience to avoid any in-
fluence which might result from professional technique. They
were made with the Rolleidoscope with the Kalart Master gun
and extension, using F bulbs and Plus X film.

Pink y<?. When the flash is almost in line with the camera lens
a peculiar effect results, one in which the pupils of the eyes have
an unusual pink coloration. Two explanations have been offered,
both similar. One is that the light reflects at the retina and fills
the eye with a reddish glow, the other is that a fluorescence is ex-

FLASH IN STEREO 109

cited similar to that observed in animals' eyes when illuminated
by a light almost in line with the eyes of the observer. One thing
is certain, the effect does come as a result of too close alignment
of flash and lens and it may be cured by moving the flash to one
side. Occasionally there is a pink eye found in color shots made
with the side mounted flash, if the bracket is close to the camera,
but you will not discover it in extension flash shots.

As the effect amounts to a grave distortion of color value, it
naturally spoils the picture. For that reason alone the use of ex-
tension flash is to be recommended.

Kalart has introduced a gun which is a compromise between
the "anti-pink-eye" and the bulky extension. It is a gun of con-
ventional appearance, but the bottom of the battery case is fitted
with a shoe which slides into the Realist stirrup. All you do is
to slip the shoe into the stirrup and the gun is ready for use. The
bulb is raised several inches above the usual position and cuts
down the number of pink-eyes.

Staring Eyes. You have all seen the flash shots in which every-
one seems to have his eyes popping out of his head. You are also
no doubt aware of the explanation that the bright flare of the
flash causes it. However you might stop to consider that the flash
is over and a thing of the past before any muscular reaction can
take place in the human body. No, this pop-eyed expression is one
normal to everyone when in the dark! The flash is so quick that it
only catches the expressions which are normal to darkness. Moral:
Do not turn out the room lights before making a flash shot.

THE COLOR FACTOR. Now we come to one of the most hotly
discussed factors in all flash work. This has to do with the color
effect of the flash. Some use daylight film with no filter, some use
daylight film with the incandescent filter, some use A film with
the daylight compensator, some use A film without filter. What is
correct?

The correct procedure depends upon circumstances. First of all
there is the color temperature of the flash. This varies with differ-
ent types of bulbs, and of course electronic flash is still different.
The F bulb is about ggooK (Photoflood = 3400; HP tungsten =
3200). The M bulbs run about gSooK. On the contrary* the elec-
tronic flash is more blue than "normal" daylight and from our
tests seems to run close to 6800 to 7oooK.

110 THREE-DIMENSIONAL PHO TOGRAPHY

Then too, room walls affect the color, and as these are usually
a warm tint such as cream, ivory or tan, they tend to lower the
over-all color temperature. As this is so variable, we shall limit the
discussion to the actual flash color.

Standard Balance. Daylight film is usually balanced for 5gooK ?
the tungsten films are corrected for projection bulbs at about
3200 and photoflood are corrected for 3400. But in actual use
both the projection and flood bulbs tend to fall below this color,
photoflood bulbs particularly fall off rapidly with use.

Thus, if we assume the color temperatures of flashbulbs to be
accurate (and they are more accurate than the average tungsten
or photoflood), the corrections may be ascertained.

When using the F type bulb, use no filter with either type A or
tungsten films. The light is very slightly warm for A and cool for
tungsten, but the variation is not more than other variations to be
expected in commercial products, so with the F type, rapid bulb,
use no correcting filter when using either of the artificial light
film types.

For use with daylight film, use the usual filter which would be
used with this film for artificial light use.

The M type bulb, at gSooK, is the standard flash color for other
bulbs except the blue series. For use with photoflood film (type
A), the H&H C l/ 4 is used and with tungsten film the C i/ 2 is indi-
cated. With daylight film a Bs filter would be indicated rather
than the 64 for F bulbs. Corresponding filters of other series will
provide similar corrections.

If the electronic flash is used, the color temperature may be as-
sumed to be 7oooK more or less or one correction step more blue
than the light for which daylight film is corrected. Daylight film
would therefore call for a C i/ 4 filter, photoflood would need the
GS and tungsten film the C6 for correction. However, it is advisa-
ble to use an artificial light film type for flash work, rather than
the daylight type, unless electronic flash is used, when daylight
film is preferable.

Blue Bulbs. There are times when it is necessary to use day-
light film, such as when balancing flash and daylight. In this case
it is better to filter the light at its source than to use a filter over
the camera lens.

The easiest way to do this is to use blue bulbs, but they are

FLVSH IN SIEREO 111

more costly than the clear type and have a decidedly lower output.
For example the Sylvania SF (F) has a peak lumen output of
800,000, the Press 25 (M) has 1,250,000 while the 256 (M-blue)
has only 500,000; this is actually less than the output of the F
type of clear bulb.

It is economical to dye your own bulbs. This may be done by
dipping them into a solution such as Jen-Dip provided you dye
two or three dozen bulbs at once. It is not economical otherwise
as the dye is highly volatile and often when you return to dye
another small lot you find the bottle empty!

Dipping is also advantageous in that the dye strength depends
upon the time of immersion and by care you may make up your
own correction to give just the effect you want.

Another method is to use a filter over the reflector. This is least
costly, and It works all right. The filter used is ordinarily incor-
porated with a protective shield which prevents particles from
flying if a bulb should burst, as occasionally does happen. We
have found the combination entirely satisfactory. However it must
be remembered that the low efficiency of the blue bulb is due to
the absorption of the more plentiful red component, and any filter
w T ill have a similar effect, so whether you use blue bulbs, dipped
bulbs or filter, use the guide number for the blue bulb which
corresponds in size to the bulb you use.

MULTIPLE FLASH. The usual flashgun will ordinarily handle
two bulbs, in extension work, without difficulty. If more than two
bulbs are to be ignited at once, it is advisable to use some ignition
system designed for this heavier load to insure prompt and sure
fire. For elaborate layouts a special powder pack is used, but for
intermediate use, such as from three to six or eight bulbs, the
simplest solution is the use of a gun which incorporates a con-
denser. This stores up a charge and w r hen released hits the bulbs
with a heavy "kick."

As a rule it is advisable, when using extension flash, to have no
bulb in the main gun, as this would tend toward the flattening
effect typical of any gun attached to the camera. It is necessary to
have a shorting plug in the main gun w T hen bulb filaments are in
series. Usually one is above the camera, at one side and nearer
the subject than the lens; the other is at the opposite side, lower

112 THREE-DIMENSIONAL PHOTOGRAPHY

than the lens and about twice as far from the subject as the first
one.

DAYLIGHT FLASH. To most amateurs flash means night, but
there are many times when the value of flash in full daylight far
outweighs its value as a night illuminant.

First of all let us catalog some of the ways in which flash may
be used in daylight.

Shooting against the sun; shooting where the light is spotty;
where there is a big differential between foreground and back-
ground illumination; exposures on the beach and other locations
of intense light; making shots in mist or fog; making shots in the
rain; stopping rapidly moving objects; groups under overhead
shade; modelling groups in cast shadow; adding intensity for
morning and evening shots; adding accents to otherwise flat illu-
mination . . . and more may be added.

In short, the daylight flash adds to the picture in intense sun,
in bright sun, in hazy light, under overcast skies, in very low light
and in mist, fog or haze. Which may all be summed up by saying
that flash can add to the picture quality in any kind of daylight. It
would be going a bit far to say that any outdoor picture can be
improved by flash, but for any type of subject in which the center
of interest is confined to the foreground, the flash will add to the
picture in about 75 percent to 80 percent of all exposures.

At first it seems foolish to make use of such an insignificant
light source as a flash when we have full sunlight, but after all
let us consider that even with a midget 25 bulb ten feet from the
subject the exposure is 1/50 at f/g. This is just about the exposure
for intense sunlight, assuming an index of 8 for both. Thus we
may consider that at about ten feet, more or less, the intensity of
the flash and of the sun are about equal.

But note this difference, at 20 feet the sunlight has not altered,
but the flash is i/ 4 sun intensity. While the sunlight remains un-
changed, the flash dies at this rate: 30 feet = 1/9; 50 feet -=2 1/25
and so on. For all practical purposes, and under the conditions
stated, the flash effect will die out at not more than 30 feet. If
the sunlight drops to half normal, it will balance at between 14
and 15 feet and so on. So, in the flash we do have a light source
which is comparable in intensity to sunlight, provided the sub-
ject is located at a relatively close distance.

FLASH IN STEREO 113

Flash is and can be combined with daylight to very good ad-
vantage, and while some experience is necessary to obtain the
the best results, the beginner with daylight flash will obtain bet-
ter shots with flash than he would without it.

FLASH COLOR. One problem which may or may not give you
trouble is that of the color of the light. Flash usually has a color
which is definitely red compared to daylight. If this color is
used as an auxiliary to daylight the effect is not unlike that of
using a red or pink spot for accent in a white flood light. This
point has been the subject of much bitter argument, but it is
a fact that the warm flash can be mixed with the cold daylight
in many instances, but for this discussion we shall assume a match
is to be made.

If type A film is used without a filter, the effect will be very
blue, contrary to the red of the flash, the latter being almost
normal for the film. Therefore it is possible to use a "red" filter
to balance the two. Inasmuch as a red 3 is required to balance
the film to daylight and a Ci/ 4 can be used with the flash, the
midpoint would be about a Ci.

This is a false balance! The balance which does occur is that
which depends upon simultaneous color contrast rather than an
orthochromatic reproduction. That is, where the daylight is
predominant, the colors will be too blue. Where the flash is
stronger, red will be strongest, and true color will occur only
in those places where the lights fall upon a surface in equal in-
tensity. Thus this balance is one which can be used to the best
advantage only after a considerable experience, but then it can
often be used for some very spectacular effects.

The true balance lies in working under daylight conditions,
that is either with daylight film or type A with the full strength
compensator. The flash used is blue, either a blue flash filter or
blue bulbs. Here again the reflector filter forms the most satis-
factory practical compromise.

It may be added that if an electronic flash of sufficient power
(300 to 500 w/s) is available, it may be used with the daylight
combination but it is advisable to use the type A film with a
corrector one degree lighter than usual, because the electronic
flash is as blue as daylight and usually somewhat more, ranging
often over 7000 Kelvin.

114 THREE-DIMENSIONAL PHOTOGRAPHY

FLASH TECHNIQUE. Under the usual conditions, the flash is the
specific illuminator of the principal subject. The exposure is de-
termined by using the usual guide number for distances of ten feet
and more. For nearer distances the aperture is a half stop smaller
than that called for. It has often been asked why you can double
the amount of light without overexposing; that is, when you
have daylight and then add the flash and give an exposure which
either one would require, why does not overexposure result?
There are several reasons. First, you are working outdoors, where
at night you would use a stop larger than the guide calls for, so
you are using the flash at what is really half strength. Second,
you use a flash because the subject or some part of it is shaded
and you wish to subdue the heavy contrast. Third, this light
intensity is present only near the camera and the greater portion
of the film area is lighted by daylight alone. Fourth, the presence
of a 100 percent increase in exposure is visible, but it does not
injure the film unless a base exposure much greater than normal
is used. In short, film latitude takes care of a lot of it. With all
of those factors involved, it is easy to see why the use of "normal"
flash aperture and time is not excessive.

The flash should, when possible, be extension flash, because the
centered flash can and will produce all of the undesirable results
which characterize it in night work; flat relief, pink eyes and all.
In outdoor flash, you are making a flash exposure with the day-
light acting as a base level of illumination.

SPECIFIC SUBJECTS. -Now let us consider some of the types of
subject which were first mentioned.

Against the sun. Flash is not used for contre jour effects, as
that requires a low level in the main figure to be contrasted with
the brilliant background. However, there are many occasions
when the desired angle is not such that the normal light is good;
often this angle is more or less against the sun, as when shooting
generally south at any hour. Flash will permit a fully exposed
subject under such conditions.

Spotty light. It your model is posed in the shade of a tree, it is
probable that the light passing through the leaves will fall upon
the model as a spotty pattern of light and shade. This effectively
kills modelling and gives an appearance of exaggerated freckles.

FLASH IN STEREO 115

Unless desired as a form of bizarre pattern, such a light pattern is
highly irritating.

Flash should be used at a level slightly higher than the sunlight.
This will effectively wash out the spots.

Local co/o)-. This too is encountered among many other con-
ditions, beneath trees especially in the spring when the leaves are
light in color and less dense than later in the season. The leaves
cast a very unpleasant green tint over the flesh tones, which
effectively spoils the picture. Here it is often possible to make
use of the mixed balance, that is a clear bulb, and with type A
a compromise filter. Another common source of local color is
a wall or other large reflecting surface.

Massed contrast. This situation often occurs in parks. The
model is placed beneath a group of trees, but the open back-
ground is fully illuminated. Under this light, modelling tends
to become flat, and if an adequate exposure is used the back-
ground is washed out by overexposure.

Flash will illuminate and accent the model, but it will die away
so that a group of trees of any size will retain their shaded appear-
ance. At the same time the background will receive only normal
exposure. Thus the best features are preserved, with full, modelled
exposure of the figure.

Detail contrast. A. model on a beach or in some other exposed
location will ordinarily wear a hat or have some other protection
from the sun. If this is discarded a squint results. If it is not
discarded, the cast shadow appears in the picture as heavy, often
blank shadow.

Flash used at balanced intensity will illuminate the face in the
shadow, but will not entirely erase the shadow so that normal
appearance is preserved. The shadow is so light and delicate that
it appears quite normal to the subject and not as a defect.

Mist and /og. Pictures made in the midst of fog are rarely sat-
isfactory because the fog seems to be more solid and opaque than
it really is. The use of flash illuminates the fog itself as well as the
subject, and this luminous fog has the tenuous appearance of the
real thing. Such technique is almost essential to such subjects.

Rain. Usually when a shot is made in the rain, it is better to
have the rain unmistakable, otherwise it may look like a poor
exposure. If flash is used, not only will it overcome the flat light-

116 THREE-DIMENSIONAL PHOTOGRAPHY

ing incident to the cloudy light, but it will highlight the rain it-
self and by the sparkling reflections, add much to the brilliance
of an otherwise dull lighting.

Stopping motion. Often in dull or hazy light, the problem of
stopping motion becomes an almost impossible one with which
to cope. Here the flash again saves the day, but naturally the flash
intensity is gaged for the shutter time demanded, which means
that it will far override the normal light with the consequence
that the portion of the scene outside the flash area will be
very dark. In short, the daylight is dull and of low intensity, so
outside the flash area, this appearance will persist.

OTHER SUBJECTS. It would be impossible to catalog all of the
possible uses of daylight flash, because almost every subject can
be improved by flash, but each presents its own problem as to
just how the flash is to be used. There are no hard and fast rules,
and each subject must be treated according to its own character-
istics. This is difficult for the beginner. That is why we have
listed the foregoing common types of subject. Using them as a
rough guide, it is easy to make experiments, so that experience
may be gained. Once you have experience you will find not only
that you have confidence in the use of the flash, but you will find
it hard to work without it.

The technique of daylight flash depends largely upon two fac-
tors, position and intensity.

Position. As stated in the first part of this chapter, the position
of the flash will give you any type of modelling from flat to full
relief, depending upon the distance between the flash and the
camera lenses. It may be added that even the "pink eye" condi-
tion can be used to advantage. If you ever wish to have direct
reflections recorded, then the nearer the flash to the lens the
better. This is of value at times for special effects.

On the contrary, the extension flash will prove just as effective
in modelling in daylight as it is in night work.

Intensity. In daylight flash, you have two lights which can be
balanced in various ways. You have the normal daylight, which
establishes a basic level of illumination over most of the scene.

You also have the flash whose intensity is within your control.

FLASH IN STEREO 117

Thus you can have the flash intensity, at the principal object,
below daylight, equal to it or greater than it is.

You always have a certain daylight level upon the principal
subject, so the flash is never the sole illumination. The subject
illumination, therefore, is the daylight plus the flash. Because
the subject is often lighted at a lower level than the general
scene, that is when there are cast shadows, the flash usually (but
not always) is designed to raise the level to that point where it
will equal or fall slightly short of the general daylight level.

The principal object will have an appearance which is deter-
mined by flash plus daylight, while the general scene will have
the appearance due to daylight alone. The two lights are self-
blending as to intensity, and no sharp line of demarcation will
be seen, regardless of the flash intensity.

(A) Flash below daylight level. This is the normal procedure
when the normal guide number is used to select the flash ex-
posure. (B) Flash equal to daylight. This tends to eliminate cast
shadows completely, and to substitute flash modelling for natural
modelling. (C) Flash brighter than daylight. This is used for
stopping motion and also for pictorial effects because it causes
the general scene to be darkened to a degree depending upon the
relative intensities of the two lights, providing the principal sub-
ject is correctly exposed.

In all of these conditions it is assumed that there is some prin-
cipal object, a figure or a group of comparatively limited size.
Flash will not ordinarily be satisfactory if the distance of such
object from the camera varies through a distance equal to more
than one-third the distance of the nearest part of the subject from
the camera. For example, if a group of people is so placed that the
farthest member is three feet away from the nearest, then the flash
should be at least nine feet from the nearest member.

The relative intensity of light upon the nearest member will
be 9X9 or 81 units, that of the member three feet farther will be
i2X !2 or 144. Taken in inverse order the light intensities are on
the order of 1/81 to 1/144, so the farthest object receives a little
more than half the illumination of the nearest.

As we have seen, the number 25 bulb will approximately aver-
age daylight intensity at ten feet. This of necessity is a very vague
statement, as ''average" daylight may be twice as bright as the

118 THREE-DIMENSIONAL PHOTOGRAPHY

flash or even more, or it may be one-half or one-third. There is
no definite intensity level for daylight which may accurately be
called "normal".

A number 3 bulb has twice that guide number at 1/25, which
means four times the intensity. A number 2 has about twice the
intensity. This simply means that a number 3 bulb at 1/25 has
the effect at 20 feet which the 25 has at 10! Not much gain is
there? The inverse square law places a most stringent limitation
upon the effectiveness of artificial light.

To make matters worse, if you wish to extend the range to 40
feet, you will need four number 3 bulbs at 1/25! So, the practical
limit must be taken as something less than 25 feet, and for the
small bulbs which we ordinarily use, that limit is about ten
feet. Of course when we make use of a blue filter or blue bulbs
we draw the limit even nearer the camera.

This does not mean that the flash is ineffective. Far from it.
Once you use dayflash you will always use it, but you will soon
learn that it is primarily used for nearby objects and not for the
illumination of large areas. You will learn that it is used for light
correction rather than for primary illumination.

Once you learn the true usefulness of flash, its great advan-
tages and its limitations, you will be able to make use of it intelli-
gently. When you can do that, you will find that its value is hard
to overestimate.

Yes, it does add to the initial cost, but consider it this way. You
expose 16 pictures at a film cost of a bit less than $4. You pay the
costs of a day's outing, perhaps $10. You encounter some subjects
which cannot be duplicated, with the value of $*.**. You find
that the effectiveness of the shots is gone, ruined by poor lighting.
$14 in cash and a day full of imponderable values gone! You add
about $2 or say 15 percent and you get 16 superb shots . . . does
that make flash an extravagance or an investment? Better four
good shots than 16 poor to mediocre ones. It is not a picture
which has value to you, but a good picture.

CHAPTER 8
COLOR IX STEREO

MODERN STEREO is COLOR STEREO. Not only is color predomi-
nant but it adds specifically to the realism which is the great
characteristic of stereo. No black-and-white stereo can approach
the realism of the color stereogram, and this is true of a white
statue as of any other subject. There is no monochrome in nature
and those objects and scenes which w r e call monochromatic are
in reality colored. We have, in another portion of this volume,
discussed the importance of color perspective in regard to stereo,
and have mentioned the fact that color accuracy is of far greater
importance in stereo than in planar work. In the planar, color
does no mc^re than add to the attractiveness of the subject, in
stereo it is a definite factor in the establishment of spatial relation-
ship, the very core of stereo.

COLOR ERROR. We may as well face the fact, we all know it.
Average results with color film are not satisfactory to those who
have a discriminating color appreciation. More than that, many
such color shots are not satisfactory to anyone at all. This condi-
tion is an intolerable one as far as stereo is concerned, but what
can be done about it? Where does the fault lie? Can it be
overcome?

Quite naturally, you think the fault lies in the film; and in so
thinking you are making a grave error. The film is not perfect,
of course, but it is quite good enough to satisfy a competent col-
orist when given a chance to do its work. No, dear reader, the
trouble lies with you, or possibly with the dealer who did not store
the film properly, but in all likelihood the fault is yours and
yours alone.

It is all very well to sit back and let modern technology carry
your burdens, but there is a limit. You must do a few things for
yourself. Consider the days before color film. What was done
then? The color photographer laboriously made three exposures,
and without really satisfactory red-sensitive plates. He developed
them giving each one just the development it needed. He prob-
ably resorted to intensification and reduction. He petted and

119

120 THREE-DIMENSIONAL PHOTOGRAPHY

babied those negatives through processing. When they were done
he made three positives, maybe dye masters for imbibition,
maybe carbons for carbro, perhaps in some other medium, but
he made them. Then he superimposed them and if anything had
gone wrong the whole thing was ruined.

Can you, without being ashamed, refuse to spend five short
minutes, perhaps less, to improve that color film you are making?
Think of it! There was a time when you would work, work hard
for two or three days or more to make just one color print. Now,
by adding four minutes to the time spent, you can bring your
color films up to a quality level you never dreamed of. Isn't it
worth it? Particularly when you think of the great influence of
color upon the stereogram, isn't it worth it?

Color films are balanced for a certain kind of light, a light
w r hich has a certain color balance. Daylight films are balanced
for "normal daylight," but that is a kind of daylight which we
rarely if ever see. It is just another non-existent average condi-
tion, or almost non-existent, Fortunately color film or our color
perception or both have such great latitude that the actual light
used may be considerably off-standard and still give color results
which are satisfactory. But there is a limit, and much of our day-
light I should say offhand about 50 percent of it is beyond the
limit. So, a lot of amateur color exposures are not satisfactory.

The truth is that daylight film does reproduce the color which
existed at the time, but we find it unacceptable because we did
not see it that way. In other words, with the exception of lights
which are deeply and brilliantly colored, our color perception
rapidly adapts itself to any color and when the adaptation is com-
plete, that predominating color is "white" to us.

The purpose in having two types of film is based upon this
fact. Daylight film reproduces color naturally under normal or
average daylight, but if we use that film for artificial light, even
the brilliantly "white" photoflood, the film comes out definitely
red. The actual appearance of the scene, to eyes adjusted to day-
light, is that shown by the film; but because, as soon as the lights
are turned on our color standard shifts to make the red-yellow
artificial light "white," the daylight standard is not acceptable.

You are familiar with the deep yellow color of any light bulb
which burns in daylight. It does not change color at night, only

COLOR IN STEREO Ii>l

our color perception changes to transform this golden yellow
into "white."

"White" might be defined as the color of the predominating
illumination.

Now with such a change as that, is it any wonder that the
color value of daylight can change without our being aware of
it? In fact even training cannot teach us to perceive such changes
by our unaided visual sense, unless the change is relatively great.
But any change perceptible to the most delicate color perception,
is many times as great as the amount of change necessary to throxv
the color film off balance.

So, if you want to make good color films, the best %vay is to do
what you can to assist the film in its job.

COLOR TEMPERATURE. Under certain laboratory conditions,
there is a very close relationship between the temperature of an
object and the exact color of the light which it emits. This re-
lationship has been studied and the color-temperature scale
worked out. The color is designated in degrees of the Kelvin
scale and is usually written "K." We are not going into such fac-
tors as a "black body" nor "absolute temperature." It is enough
to know there is a scale of color temperature which we can use,
and which is reliable.

In this scale we have a constant change characteristic of a hot
body growing steadily hotter. The first appearance is red, as the
temperature rises, so does the predominant hue, and with green
added to red we have the yellow of incandescence. Later the
growth of blue produces a true "white" and as this is raised, the
red decreases and the color becomes definitely blue-white, then
blue. Note that in all of these changes, we do not have a hap-
hazard scramble of colors, but simply a changing balance of red
and blue, the colors which represent the lowest and the highest
heat of incandescence.

Because this is true, all we need is a set of filters divided into
two categories; one set to absorb blue the other to absorb red.
If these are graduated in strength, you can convert the photo-
graphic color balance of any common light to that of any other.
You can run the scale with Kelvin degrees. You can use any type
or make of color film with any ordinary kind or color of light (of

122 THREE-DIMENSIONAL PHOTOGRAPHY

sufficient photographic intensity), and obtain a color picture of
better than average color balance. It is as easy as that.

You can do this simply by placing a filter over each camera
lens, but to select the correct filter you must know the color
composition of the existing light and how far it is out of normal
balance. In short you just know the color temperature of the
existing light and the color temperature for which the film is
balanced. By filter compensation you can then easily coordinate
film and light balance.

Practically, this is easy and pays excellent dividends in increased
color quality. There is just one thing against it. It is just too much
trouble to carry around the necessary filters and to make a CT
reading of the light. Too-muchtrouble! The same old thing
which is responsible for such a tremendous amount of photo-
graphic inferiority among the work of our amateurs.

So much for "Why," now for the "What" and "How."

Consider these color temperatures:

Dull red-hot iron ... . . . 8ooK

Candle - i85oK

Ordinary electric bulb . ... agooK

Projection lamp, special . . . gsooK

TUNGSTEN COLOR FILM . ... gsooK

SF/SM flashbulbs . 33ooK

Photoflood lamp 34ooK

TYPE A FILM 34ooK

Wire filled flashbulb 38ooK

Blue photofiood 5OooK

Noon sunlight 54OoK

Sun in blue sky 59ooK

Electronic flash about 7oooK

Blue sky 9000 to 3o,oooK

Suppose you have Type A film in your camera, and you wish to
make a sunlight exposure. What should you do? The film is bal-
anced for light of 3400 K, the light you will use has a CT of
54<x>K } a difference of 2oooK. The sunlight is higher in tempera-
ture than the film balance which means it is more blue. You must
use a filter which will absorb a sufficient amount of the blue of
sunlight to reduce its effective CT to 34OoK. This is exactly what
the familiar conversion filter does.

If you use daylight film, balanced for sgooK and you are using
photoflood light at 34ooK, you have a film balanced for blue and

COLOR IN STEREO I1Z.S

a light which has too little blue. How can you make up the de-
ficiency of blue in the photoflood? You cannot build uf> the blue
content of the light, but you can remove enough of the red to
leave the blue in proper balance. So you use a red-absorbing filter
(bluish) of the correct density.

But what if you are using daylight film and an SF flashbulb,
5gooK film and sgooK light?

The light is ggooK, Type A film is balanced for 34ooK. The
difference is insignificant, so you can safely use the same filter
you use for balancing daylight film to photoflood light, and by
the same reasoning you need no filter to balance Type A film to
SF or SM flashbulbs.

Thus adequate color control depends upon just two things,
knowing how to determine the color of the light used and how to
determine the filter necessary to balance this light to the film you
are using.

Fortunately artificial light sources may be assumed to conform
to published ratings which means the difference between actual
and nominal rating is not enough to spoil the color effect. In fact
most light sources conform to average ratings. Therefore your
light measuring problem is limited largely to daylight conditions.
The measurement is done by a color temperature meter. This
may either be a visual meter, an electric CT meter or an attach-
ment for the G.E. exposure meter.

If you can distinguish between purple and blue, that is, whether
a blue color has any tinge of red in it or not, you can use the
visual meter successfully. You simply find three visual filters one
of which is purple, one of which is pure blue and one of which
is neutral or in-between.

If you doubt your color sensitivity, then use the attachment for
the exposure meter which gives you a definite reading unaffected
by personal visual error, or an electric CT meter. .

As to the filters necessary, you can obtain filters from Eastman
and from Ansco for certain corrections of their films, but the
writer prefers the Harrison & Harrison correction set which has
11 blues, ii reds or a total of 22 filters for each lens, a grand
total of 44 filters for the stereo camera. These filters make it possi-
ble to correct the balance of any light from 2&ooK to Q5ooK for

124 THREE-DIMENSIONAL PHOTOGRAPHY

use with any film, and when using daylight film this is extended
to 3o,oooK.

These filters are arranged in graded steps, the temperatures
corresponding to the steps are: 2800, 2900, 3000, 3200, 3400, 3600,
3800, 4300, 4800, 5400, 5900, 6500, 7100, 8000, 9500, 11,000,
13,000, 16,000, 20,000, 25,000, 30,000. The blue filters (red ab-
sorbing) are indicated by B, the blue absorbing (pink) filters are
"C." The depths of color range by numerical values, i/, 1/4, y^>
i, 2, 3, 4, 5, 6, 7 and 8.

Thus to correct daylight film for wire filled flash, you find that
the film is balanced for sgooK, the flash has a color of 38ooK and
to compensate it is necessary to use a 62 filter.

Let us assume that, like many others, you make a practice of
using only Type A film with the corrective filter. This is your
scheme of conversion for several different lights:

Ordinary house lighting electric bulbs Bi to BI/

Projection bulbs BI/

White fluorescent None

Type SF or SM flash . . None

' Wire filled flash Ci/ 4

Morning or evening light . . Ci/% to Ci

Noon sun Cs

Bright sunlighted day C$

Average daylight 04

Daylight fluorescent 4

Open bright sky, no direct sun. .05 to C8 (Max. n,oooK)
(The "B" filters are blue in color, the "C" type is pink.)

But this is not all. When you gain experience in the use of
correction filters you can use them to correct "local" color such
as a reflection from a red brick wall. The color meter will give
you the analysis if you use it to measure the actual light which
falls upon the subject.

You can also make a color reading, and then deliberately over
or under correct, to produce an effect which is warmer or cooler
than normal, and for certain experimental and "effect" purposes
you can swing very wide of standard and produce deliberately
falsified effects.

In short, color is a most flexible medium, and once you gain
control of it, you will find that instead of being one of the un-
predictable factors, it is of very great service. For example, one
photographer produced beautiful, blue-haze effects when making

COLOR IN STEREO litf

stereograms of mountain scenery. Others obtained a suggestion of
the effect, but the blue was dull and inclined to be leaden while
this man's work was practically of Maxfield Parrish quality in the
blues. How? After considerable experimentation he had found
that exact strength of blue correction filter which would clear up
the blues without too greatly falsifying other colors. His greens
were slightly blue-green, the edge taken from pure red but these
effects were not at all obvious and passed unnoticed unless they
were sought deliberately.

This did not constitute a correction to normal, but an over-
correction designed for one specific purpose.

To return to the more commonplace, you will usually find that
the average, normal light in your part of the country results in
color which is consistently red or consistently blue. True, many
people do not recognize these effects, but merely think of the pic-
ture as being "a little dull" or "a little muddy/' Look at the
whites and at the strong colors. Study them. You will almost al-
ways find that a flaming red, for example, has a deadened appear-
ance, just a hint of maroon which is really a hint of wine, the
tinge of blue. Or it may be that your blues are just a little dull
and lifeless and all of them carry a suggestion of purple or ma-
genta, hardly perceptible. You are overrunning in red. Often the
slide will be distinctly brown-purple-dim-muddy which indicates
a heavy overrun of red. But color correction can give you color
shots in which 80 percent at least will be above criticism other
than the most searching.

Certainly color correction filters are far more necessary in color
photography than ordinary filters are in black and white work.
For years, amateur photographers did without filters, and when
they were introduced they were simply "cloud filters*' one degree
of yellow. It took more years for the amateur to become "filter
conscious/*

We first took our color film as it came and made the best of it.
Then we adopted the conversion filters and the flash filters, and
were happier. Finally the control filters were introduced, but they
failed to win popularity because ot the high cost of color meters
and the difficulty of using them. Now that we have color meters
which are reasonably priced and not difficult to use, we are enter-
ing the color filter stage comparable to the third stage of black-

126 THREE-DIMENSIONAL PHOTOGRAPHY

and-white. In short we are slowly becoming filter conscious in
color work.

Yes, it does involve trouble. You must carry a certain number
of filters, you must make the color reading, you must apply the
correct filter. All that takes time, as much as two or three minutes,
but when the stereogram is complete, it w r ill be with you for years.
Is your time so valuable that you cannot devote two or three
minutes to producing a good slide when otherwise you would
only have a poor one?

Or look at it this w r ay. You will spend a whole day, and use a
magazine of film in making 20 exposures. Careless color technique
means you will probably get not more than five good slides. But
one hour divided among the 20 exposures, at the most, will in-
sure at least 15 good color slides. If you can devote eight hours
to five slides or more than 1.5 hours each, you can surely devote
nine hours to 15 slides, which is considerably less than a half hour
each! And we do not even mention the money invested in the
lost film.

If color is worth doing, it is worth doing well. Take your time.
It is better to make a half dozen good slides in one day than to
produce a hundred mediocre ones which you will not value.

Most stereographers prefer to use one type of color film for all
purposes, thus avoiding the necessity for two cameras or for
wasting film. Therefore it will not be necessary to have the full
stereo set of 44 filters. The table below indicates the Harrison and
Harrison color correcting filters to adjust any film to any ordinary
light.

Film

Light Daylight Type A Mazda

"3200" tungsten 65 Bi/fc None

Photoflood 64 None Cy 8

Wire filled flash Ba Ci/4 Ci/ 2

Late daylight Bi/ 2 * Ci* 2*

Noon sunlight B*4 2 03

Sun and sky Haze 3 Q.

Average daylight d/ s 4. 5

Shade Strobe-light Cy 4 5 C6

Sky light Ci/2** C6** 7**

*May vary to call for Bi, Cy 2 and Ci.

**May vary to call for as much as C2, 7 and C8 respectively, or
even enough to necessitate up to 7 with daylight film.

COLOR IN STEREO 127

Thus 7, 9 or 11 pairs will meet most requirements, while the
addition of B8, Bi and Bi/o, respectively, adapts the three films to
ordinary home lighting. I would suggest the extra pair for late
daylight and an extra one for blue sky light, plus the ma/da home
light, making a total of ten pairs of filters, or less than half the
full set. Such a series of ten pairs should meet 98 percent of the
requirements as long as you stick to one film type.

Incidentally never expose any film out of doors without a filter
because of the ultra-violet exposure which will result. When day-
light film calls for "O" filter, use a haze filter.

Color Temperature (CT) M^fcr. There have been a number
of CT meters offered from time to time. One of the earliest was
the Eastman which we are told is no longer available. Of the
more recent models, the first was the visual meter of H&H. This
is a disc-shaped black plastic case with a graded series of visual
filters around the edges. The filters range in color from" definite
blue to an equally pronounced purple. A shield over the side of
the meter obscures all but three of these filters, although by ro-
tating the body of the meter, the full sequence passes behind this
triple mask.

A white reflecting sheet of controlled color is used as a stand-
ard. It seems strange to speak of the color of white, but there
are a great many colors ranging from pale cream to very pale
blue which pass ordinarily as "white." This reflecting sheet is
of a white color which reflects red and blue equally.

The reflector is held so it is illuminated by the light which
falls upon the object. The disc is then rotated while looking
through the filters at the reflector. There will be one set of three
adjacent filters; one appears blue, one seems to be a blue-gray and
the third shows a distinct tinge of purple. This is the end point,
and the reading of the scale at that point indicates the color tem-
perature of the light.

This meter is good, as far as the instrument Is concerned, but
most people do not have the sensitivity to color which makes it
possible to arrive at a positive end point, and even those who do
have sensitive color perception, are often unsure and so may be
off one full point of the scale.

To overcome this difficulty H&H brought out their color at-
tachment. This is a square case which contains a slit diaphragm,

log THREE-DIMENSIONAL PHOTOGRAPHY

adjustable in width, and a pair of filters, one red and one blue.
A plunger in the bottom serves to change the filters in the aper-
ture.

The attachment has a rectangular window which fits the G.E.
exposure meter when the hood is removed. To use it, the attach-
ment is placed upon the meter and the incident light allowed to
fall upon the filter. The diaphragm is closed until the meter reads
either "10" or "20". Then the plunger is released to change filters
and the meter reading taken. A scale attached to the filter case
then indicates the color temperature which corresponds to the
second meter reading.

As meter cells may vary a little in color sensitivity, the filters
of the attachment are graduated in density and may be balanced
to correspond to the meter sensitivity by moving one or both
sidewise^ The meter is quite satisfactory, and appeals to many
because it is inexpensive and because it permits the use of only
one meter unit for both exposure and CT reading.

Another type of meter is one designed and built for just one
purpose, that of reading color temperature. The Guild labora-
tory uses the Rebikoff, a Swiss meter of better than 2 percent
accuracy. This meter, instead of the usual scale, has an arc made
up of a blue sector, a red sector and a heart-shaped black zero
point set in a transparent housing so the needle position may be
seen from either side.

A dial is set upon one side of the meter in a position similar
to that occupied by the dial of most exposure meters. However
this dial is transparent, and beneath it are located the two cells
with their filters. The dial itself carries a semicircular black mask
so that, as the dial is turned, the areas of red and blue exposed
to the light, constantly vary. Because this dial must be exposed
to the incident light it is convenient at times to hold the meter
in the palm and look down upon it and at other times to hold it
up before the face, directed away from you. The transparent case
makes the needle position easily visible in either position.

Th dial carries an index point which directly indicates the
color temperature of the light. It also carries a symbol of the sun
and of an incandescent bulb. When either of the two latter sym-
bols lie opposite a fixed index mark on the case, no filter is neces-

COLOR IX STEREO 129

sary for the film indicated, i.e., daylight film if the sun is in-
dicated and an artificial light film if the light bulb is indicated.

So far we have had only one trouble with this meter. It is so
sensitive that the user tends to make corrections where none are
necessary. In a series of tests made under conditions where the
light varied only 300 or 400 degrees from normal, a degree of
change ordinarily considered insignificant, the slight correction
indicated has resulted in a distinctly improved color balance.

As to the value of a CT meter, if we had to give up one meter
we should give up the exposure meter and keep the CT meter.
Exposure CAN be estimated with reasonable success especially
when the exposure tables are memorized, but CT changes can-
not be recognized without instrumental aid. Our color has im-
proved so markedly since the CT control was instituted that
we regard it as invaluable.

In closing this discussion one thing should be made clear. You
need know nothing whatsoever about color temperature theory.
Your exposure meter is calibrated in numbers. Those numbers
have definite meaning, but most people know nothing except
that they are the "meter reading" and these have just as much suc-
cess with exposure meters as do those who know the specific na-
ture of the units indicated upon the meter scale. Just so in the
CT meter, you need not know what Kelvin degrees are, you
need not know what color temperature means. All you need to
know to use the CT meter with every success is that a certain num-
ber on the meter indicates a certain filter with the film you are
using.

For example, suppose your meter reads 4800 some afternoon.
You refer to the filter table and you find that if you are using
daylight film you will need a Bi/ filter; or if you use Type A
Kodachrome, you will use a Ci filter; or if you use Ansco film or
Type B Kodachrome you will use a Cs filter. If you use Eastman
or Ansco filters, the correct ones will be indicated.

The meter is plainly marked, the filters are marked and all
you have to do is to match the two figures without knowing any-
thing about what the numbers actually refer to. Moreover, if
you use a Rebikoff meter, you get a plain signal if the light is
within satisfactory range of either daylight or artificial light film,

130 THREE-DIMENSIONAL PHOTOGRAPHY

and if it is not, all you have to do is to swing the symbol cor-
responding to your film (sun for daylight film, bulb for artificial
light film) under the index and the meter needle will tell you if
the light is redder or bluer than it should be and give some
approximation as to the amount of unbalance. Then you can,
with tolerable success, select an unidentified filter for partial cor-
rection.

Color correction is not new, but it has not yet become popular.
There is little doubt that when amateurs have had an opportunity
to actually see for themselves the tremendous improvement which
it makes possible, it will take its deserved place as being just as
important as exposure control.

CHAPTER 9
PICTORIAL STEREOGRAPHY

ANY STEREOGRAPHIC EXPOSURE must include such technical con-
siderations as exposure, filters and aperture, and it must
also include certain fundamental esthetic considerations which
may roughly be divided into txvo groups. The first group includes
those factors which, while not strictly photographic, have much
to do with the stereo perception, including the type of lighting to
be favored. The second group includes the elusive considerations
of pictorial composition as applied specifically to stereography. As
no stereographic reproduction can be made without giving more
or less consideration to these factors, we shall discuss them to-
gether.

The whole subject of three-dimensional photography has often
been discussed so irrationally that we must pause for some clarifi-
cation. In the first place, three-dimensional photography is not
necessarily identical with stereoscopic photography; and even
more decidedly, binocular vision (or photography) must not be
regarded as identical with stereoscopic vision (or photography).

Any photograph which exhibits a definite appearance of depth
may correctly be called three-dimensional. Any good photograph
examined in a concave mirror exhibits a three-dimensional quali-
ty, but it is not stereoscopic.

PSYCHOLOGICAL DEPTH PERCEPTION. It is evident that we must
consider the true nature of visual depth perception before we can
arrive at any conclusion about the nature of stereoscopic vision.
The several factors which combine to provide the visual percep-
tion of depth or distance are:

i. Geometric Perspective. This is inherent in every planar
photograph and in good drawings (ancient and oriental art ex-
cluded). It is the arrangement upon a single plane of those aspects
of an object which in reality occupy three dimensions. For ex-
ample, a side of a cube is not shown as a square but as an irregu-
larly quadrangular shape with the far side much shorter than the
near one.

In actual fact perspective relates to the size-in-width element of
the depth factor, and is always limited to those planes which are

131

132 THREE-DIMENSIONAL PHOTOGRAPHY

parallel to the film plane. It is the most potent device yet found
for lending to a picture upon a single plane, a suggestion of depth.

2. Contour Interference. It a tree is shown in front of a build-
ing, the side of the building which lies behind the tree cannot be
seen. Hence we know at once that the tree is in front of the build-
ing. If the lower part of the tree trunk were hidden by the
building, we would know that the tree was standing behind the
building. This is the principle of interfering contours, the hiding
of objects or portions of objects by other objects which are nearer
the observer. Alone, it w r ill give a fairly accurate index of relative
depth.

3. Size Diminution. This partakes of both the preceding fac-
tors, and is the well-known effect of apparent decrease in size as
an object becomes more distant. It is usually assumed to be an
inherent phase of perspective, and is usually included as such
when the principles of perspective are explained. However, it
should really be considered apart from, or at any rate as a distinct
subdivision of, perspective, because of its great influence upon
the appearance of non-stereoscopic depth.

4. Light and Shade.^The forms and intensities of contour
shadows are so important that they, together with the careful ap-
plication of perspective, form almost the sole means for imparting
to a plane image a simulation of depth. Just as an angular or
broad light gives to the plane image a false appearance of depth,
so in stereoscopy does the same light enhance the true stereo effect.
(Compare Contour Gradation.)

5. Haze. The effect of haze, sometimes called aerial perspec-
tive, aids to some extent in the perception of depth. It is common-
ly used in planar photography to afford separation of planes (i.e.,
stereo relief), but when applied to stereography the result is usu-
ally an unfortunate appearance of a weak, washed-out print.
(Compare Color Perspective.)

These five factors apply equally to planar reproduction, and in
fact they constitute the principal devices which are used to suggest
depth in the planar photograph. There are two additional factors
which are purely stereoscopic, and which must be given careful
attention by the stereographer who wishes to attain success in
stereo pictoriallsm.

6. Contour Gradation .Contour gradation is somewhat anal-

PICTORIAL STEREOGRAPHY 133

ogous to the effect of light and shade. In that, we consider the
effect of shadow gradation, and one important purpose of shadow
gradation is to suggest, upon a plane surface, the contour of an
object. In stereo we have the actual contour shown in relief, so
the shadow plays a minor role in that respect. Now the gradation
of both shadow and contour play a purely pictorial role.

If the scene shows scattered rocks, the contour gradation is a
series of isolated shocks, a jump from one to the next, but if the
object shown is the curving wall of a pool, there is a continuous
recession of contour w r hich the eye follows easily and smoothly.

Interrupted relief which might not be observed in a planar
composition, which might indeed contribute to the continuation
of shadow gradation, will in the stereogram be rough and broken.
On the contrary, a continuous contour may change in color and
be spotted with abrupt patches of shadow and highlight alternat-
ing, yet the smoothness of flow is not interrupted. This complete
antagonism between planar and stereo pictorial character must be
observed, because the planar system alone cannot provide the flow
character desired. It is completely overshadowed by the new ele-
ment of contour gradation. The relation between contour grada-
tion and shadow gradation is that existing between substance and
shadow.

7. Color Perspective. Atmospheric perspective has traditional-
ly been a potent device in the hands of the planar pictorialist, but
unfortunately many of the locations which provide the greatest
amount of pictorial material have such clear air that the haze ele-
ment is invisible. Here again the stereographer has the advantage
because he can always make use of color perspective. In fact it is
one factor which tends to lower planar quality while it enhances
stereo quality.

Color, like shadow, has an intensity which in part depends upon
the size of the area and the distance of the color. Given a certain
patch of color, it loses intensity as well as size as we recede from it,
and this characteristic persists even when there is no observable
haze. It is also perceptible over the relatively short distances with-
in which stereo is most effective, something which is not at all
true of haze unless the haze is a definite mist or fog.

In the stereo color slide, this color recession is so characteristic
toward a common neutral tint, that it has been the cause of con-

134 THREE-DIMENSIONAL PHOTOGRAPHY

siderable complaint about color shots. The charge Is made that
the color film is defective. Not at all; the film simply records
the normal loss of color intensity which we see but do not ordi-
narily perceive. It is just another example of the photograph show-
ing details of a familiar object which escape direct vision.

In the stereo color slide, this color recession is so characteristic
of distance that it actually helps in making distance specific. Its
importance is hardly surpassed by any of the other extrinsic fac-
tors. True, it is not as obvious as size diminution or overlapping
contours, but realism is a subtle characteristic and color perspec-
tive or stereo-chromatism does, perhaps, even more than these to
increase the essential realism of the stereogram.

It will be noted that these factors are all purely psychological.
None of them is definite, none (except perspective) is subject to
measurement, none of them is inherent in the object or scene, and
all of them are subject to different interpretations by different
people.

The only factor of depth perception which is inherent in the
physical conditions of the view and the viewer is that of differen-
tial parallax, that precise difference between the two members of
all pairs of details afforded by the two points of view. This, and
this alone, is the stereoscopic factor. The other factors simply
afford a greater or lesser intensity to the general appearance of
depth perception.

The inherent factor of differential parallax is essential to stereo-
scopic perception. If it is absent there is no stereo; if it is present
there is stereo. This is not true of any of the psychological factors.
They may enhance the stereo effect, but they can neither produce
it nor destroy it by their presence or absence.

Therefore, when making a stereogram, it is essential that differ-
ential parallax be present. This is automatically assured by mak-
ing the two exposures from different points of view. If it is con-
venient to include any of the other factors, they will add to the
effectiveness of the result in some measure; but if it is not practical
to include them, the fundamental stereo effect will be present
even if not one of the psychological factors is included. Perspec-
tive, of course, cannot be eliminated.

In making the stereogram, it must always be remembered that
most of the extrinsic factors mentioned are not peculiar to stereo.

PICTORIAL STEREOGRAPHY 135

They are factors present in everyday direct vision, but of which
we are not ordinarily conscious. We do not realize that the degree
of stereo relief which we see in everyday life varies with the light-
ing, and with other extrinsic factors. We do not realize that the
farther away an object is the flatter it really appears.

We have had a lifetime in which to balance knowledge against
appearance. Hence a tree trunk five hundred feet distant, looks
as if it w r ere just as round as one ten feet away, yet we know that
to our vision it really presents a shape of a flattened oval. In fact
xvhen we know this and train our vision we can actually perceive
the diminishing relief in the distance, but commonly our mental
police step in and say, "Now, you know that tree trunk is round,
so see it that way" and we obey. In fact we do not think about it.
We see the tree, w r e know r trees are round, the tree stands in re-
lief so we simply accept the relief as full relief with never a ques-
tion as to whether it is round or flat.

This visual trick is common in almost every phase of vision.
We see the invisible because w T e know from experience what it
looks like, but if you want to know the truth about visual per-
ception, show a strange object to a group of five people, let them
look at it ten seconds. Remove it and ask for a description. You
will get five wholly different descriptions. It has been tried too
many times for there to be any question about it.

So we are not introducing technical tricks, merely using in
normal stereo the characteristics of normal vision. For example,
the practical application of some of these extrinsic factors are:

GEOMETRIC PERSPECTIVE. This is the factor which is ideally
illustrated by a pair of railway tracks or even a roadway extending
in a straight line directly away into the distance. Such a strong
perspective is rarely acceptable pictorially, but the roadway wind-
ing back into the picture is a device commonly used by planar
pictorialists.

In stereo the principle is applied by selecting a point of view
from which several objects are to be seen in relatively receding
planes. It is always advisable to have some object in the immediate
foreground to establish the strongly stereoscopic reference. Then
a succession of objects farther and farther away leads the eye into
the distance and enhances the effectiveness of relief.

It must be borne in mind that this and other devices do not

136 THREE-DIMENSIONAL PHOTOGRAPHY

actually increase the stereo relief. They only make more apparent
the relief which exists, just as when we look down a wooded glade
we obtain a sense of greater distance than when we look out over
the ocean or across a vast plain. The enhancement of effect is no
more and no less than is given to actual direct vision under similar
conditions. This is true of all the psychological devices.

CONTOUR INTERFERENCE. In selecting the point of view, any
objects which are located partly in front of other objects more
distant will enhance the stereo depth. The comparison of the
glade and the open plain applies here as well. The effectiveness is
increased just as it is for direct vision.

One need not fear for the confusion which exists in the planar
photograph those accidental mergers which tie up objects in the
foreground to those in the background. The classic example of
having a tree growing out of the head of the subject need not be
feared in stereo, for the tree will appear in its correct position far
behind the subject, and the two will interfere no more than if a
similar distance divided them laterally.

SIZE DIMINUTION. This will be seen in any composition in
which the two preceding factors have been given place. It is as
automatic as perspective, provided there are objects included
which serve to bring out the factor.

In this connection it may be remarked that the surprising and
frequently ridiculous size relationships seen so often in planar
photography are never encountered in stereo. For example, take
the photograph so often made for reasons of comedy in which a
man is stretched out upon a park bench or the like, and photo-
graphed with his feet a yard or so from the lens and his head eight
or nine feet away. The feet appear to be about two feet long.
When the same thing is made in $tereo, the man appears to be
proportioned with complete normality and appears in the posi-
tion he actually occupied.

The same thing is true of the perspective of tall buildings. The
stereo camera may safely be tipped upward to include the top of
any skyscraper. When the picture is viewed, all sense of falling
backward is lost and you seem to be looking upward in the most
natural manner. In viewing such stereograms, the realism is en-
hanced by holding the stereoscope so the head will be tipped at
an angle similar to that necessary to view the original scene.

PICTORIAL STEREOGRAPHY 137

All of the so-called distortions of violent perspective are lost in
stereo, and the stereogram has a wholly normal appearance. Per-
haps there is no stronger evidence than this that the conventional
photograph is highly artificial, while the stereogram is wholly
realistic and does actually repi~oduce direct vision to the last detail

LIGHT AND SHADE. If you look at a large white ball with the
light coming from directly behind you, the shape of the ball is not
definite. It may be a flattened or an elongated ball, but you cannot
be sure. If a planar photograph is made, the ball has the appear-
ance of a disc, but in a stereogram exactly the same vagueness of
form will be noted which is characteristic of the direct vision.

If the ball is lighted from above and at one side, the contour
shadows make its shape unmistakable to direct vision. To the
planar photograph these shadows impart a character which en-
ables the shape to be "read.'* That is, we do not actually see it
standing out in relief, but we can see from the shadows that it
does have a round shape. The stereogram gives us the same defi-
nite perception that is seen in direct vision.

If a disc is skillfully painted with imitation contour shadows, it
does not readily fool direct vision, nor is the deception effective in
the stereogram. Both show it for what it is, while in the planar
photograph the painted disc and the true ball are hardly dis-
tinguishable, if at all. Thus it will be found that stereograms are
more decisively visible if a quartering light is used. This applies
to portraits and to landscapes, to large objects and to small.

Dramatic lightings may be freely used in stereo, because the
absence of the modelling light does not eliminate stereo relief. But
empty shadows should not be used, nor should light be used as a
positive element in the composition, for in the stereogram the
light will follow the contours of the surfaces upon which it falls,
and these will not have the same pictorial weight as in the planar
photograph. In a planar photograph a shadow or a highlight is in
essence a solid. In stereo a shadow is simply a shadow without
form other than the form imparted by the surface upon which it
falls, while light is simply light. These values are those of real
life which is considerably different from the "solid" values they
have in planar composition.

One exception to this is the definite beam, as from a spotlight
or a sunbeam from a window. If made visible by dust or smoke,

138 THREE-DIMENSIONAL PHOTOGRAPHY

this beam will have the same position in space it exhibited to
direct vision and so becomes far more effective than when used
in the planar photograph. Anything which has definite visible
form, whether tangible or not, will have that form reproduced.

COMPOSITION. Among the usual elements of composition, the
value of unity, harmony, and balance are as important as in planar
photography, but it must be remembered that the pictorial "steel-
yard" must now be applied from side to side as usual and also
from near foreground to far distance as well, something entirely
new in pictorialism. The patch of sky or the dark cloud no longer
has its power of balance, because it is in the far distance and its
"balancing .weight" is much less than in the planar picture.

Composition in stereo is more difficult and more complicated
than in planar pictorialism, but for that very reason it has poten-
tialities which the planar form could never exhibit. To make use
of a mathematical analogy, one might say that the planar pictori-
alist is limited to addition and subtraction, while the stereogra-
pher has the possibilities of division and multiplication as well.

Stereo composition is at present largely unformulated, which
perhaps is just as well, for too much formalization of composition
creates a stiffness and artificiality which has very little in common
with esthetics. However, enough has been said to point out the
general path to be followed.

The point o great importance in relation to stereo composition
is that such a thing actually exists and must be taken into con-
sideration. Once aware of the necessity for three-dimensional com-
position, the stereo pictorialist will lose no time in attacking the
problem.

Strangely enough, among present-day stereo enthusiasts the
shibboleths of the old pictorialism are still regarded as being po-
tent. The result is highly amusing, because more often than not
the attempt to compose according to the planar standard simply
results in a stereogram which actually is inferior to one in which
no attempt at composition has been made. For example, haze or
atmospheric perspective included in a stereo no longer suggests a
soft and mysterious distance. The distance is there in all reality,
but the haze often appears to be simply the washed-out tone of a
poor print.

PICTORIAL STEREOGRAPHY 131*

NATURAL AND PICTORIAL BEAUTY. The beginner in photogra-
phy who has ambitions toward the pictorial, makes endless nega-
tives of "perfectly beautiful" scenes, and almost without exception
the prints bring only disappointment. He then starts to study the
rules of pictorial composition, and slowly he learns that very
rarely does the beautiful original produce a beautiful picture. On
the contrary, the prints he eventually exhibits will be made from
subject matter which would be passed by the casual observer with-
out the slightest suspicion that it had pictorial possibilities.

Pictorialism is based upon a set of wholly artificial standards.
No one can ever claim that there is any beauty in a water faucet
or a couple of eggs upon a kitchen drain board, yet pictures of
such subjects have been made; pictures which no discerning
critic can say are not beautiful. Thus the artificiality is not the re-
sult of preciousness, but has a very solid foundation. We are not
dealing with certain schools of thought or faddism. It is a solid,
inescapable fact that many of our most beautiful scenes sink into
insignificance when photographed, and that many of our most
beautiful outdoor photographs have been produced only because
the photographer has developed his power of camera discernment.
This is true because the values of an object in space are not its
values when reduced to a plane. In the planar picture mutual
effect of two objects distantly separated in space, but adjacent in
the plane, must be carefully considered. Hence it is inevitable that
in any planar picture the values must be decidedly different from
those of the original.

In stereo this is not true. The values of the original are repro-
duced in the stereogram, hence the stereo slogan which has done
so much to make it popular, "In stereo you get exactly what you
saw" and that is true. The stereogram is a truthful reproduction
of the original, which is never true of the planar reproduction.

The beautiful scene almost always yields a beautiful stereogram.
However, this is apparent to the beginner more than to the ex-
perienced photographer. The writer has been deeply amused by
the criticism of stereograms offered by pictorialists. It is quite
obvious that such criticism has been made while studying the print
directly, not the stereogram in a viewer. Such criticism is not only
unjust, but its absurdity is manifest to anyone, skilled or layman,
who hears the criticism while studying the stereogram in a viewer.

HO THREE-DIMENSIONAL PHOTOGRAPHY

His comment is most often "Why, that is not true! That is not
visible here. No such condition exists/'

The stereogram cannot be criticized except when correctly
viewed, for the subject of the criticism is not a photographic re-
production, but a synthetic image which is actually created within
the brain and which cannot be "seen" without the simultaneous
viewing of the two units. Moreover, as much as we like projection,
it is an error to undertake serious criticism of a stereogram as
viewed by projection. This should only be done while the stereo-
gram is viewed in an accurately orthostereo viewer.

Let us take as examples two very commonplace errors made by
the beginner. First we have a small lake, perhaps a mile across.
The near shore is nicely curved and presents the requisite trees
and shrubs. The whole is idyllic. But the sweep of water is the
thing which gives the scene its beauty. The poor beginner makes
his print and is dismayed to find his beautiful lake a mere smear a
quarter-inch wide across the print, with the trees at one side rising
abruptly to cut the composition in two.

The stereographer makes the same shot from the same position,
but he has no disappointment because his stereogram reproduces
the broad expanse of the water, giving it the weight it has in the
original. The water is not a quarter-inch strip, but a surface ex-
tending a mile back into the distance. The relative pictorial values
are just about upon the scale of one quarter-inch to a mile!

The next error is that typified by the profusion of growth in a
swamp. Most people think of a swamp as an expanse of soggy
grass and mud. On the contrary it is a place of luxurious growth,
tall plants and shrubs, gigantic trees, lush flowers, and mirror-like
pools. It is a subject which no amateur can resist, yet again the
planar print is a disappointment. The growth is so luxuriant that
the print presents a confused tangle of detail, all more or less in a
monotone of gray. One can point out the various features, but the
spirit of the swamp is missing, and the print is simply another
failure.

In the stereogram, this detail is resolved into its original three
dimensions. We can peep through the tall leaves of a lily to see
the sparkle of the swampland pool. The rising columns of the
trees with their patterns of cast shadows lend emphasis instead of
confusion. We look deeper and deeper into the depths of this un-

PICTORIAL STEREOGRAPHY 141

known terrain, and the stereogram preserves the spirit of the
mysterious which the sight of the real thing brings.

The plane picture depends upon emphasis of some object or
group of objects. These must stand out from the rest of the com-
position by reason of tone, contour, or psychological emphasis.
In the stereogram the almost perfect monotone of the swamp pic-
ture attains far greater emphasis in the viewer than the most care-
fully composed plane picture. The emphasis is that of nature, not
the artificial emphasis of the pictorial ist.

Beauty in the stereogram depends upon the taste of the stereog-
rapher. If he selects a truly beautiful subject, the stereogram will
automatically preserve all of that original beauty.

The pictorialist who turns stereographer must also use care
in selection. If he follows his usual method and seeks to make a
picture of some very commonplace subject simply by careful
treatment, he will meet only with failure. This is the reason xvhy
pictorialists rarely become enthusiastic stereographers and why
they usually condemn stereography calling it a purely record
medium. The stereogram will reveal the commonplace subject
as commonplace. The magic of pictorial composition is gone.
The stereogram is the only reproductive medium known which
will reproduce an original in its exact original appearance.

The stereo-pictorialist must rely upon nature for his beauty
and be content to record it as it is. On the other hand, he has un-
limited choice of point of view, lighting, selective arrangement,
and so forth. The only thing that he must remember is that his
choice depends, not upon the tricks which can be used in plane
photography, but upon the effect of each choice upon the actual
appearance of the subject which is to be stereographed.

There is sufficient beauty in nature to satisfy the stereographer,
certainly, so instead of trying to create beauty from ugly raw ma-
terial, let him be content to present existing beauty in its most
appealing aspect. Such is the basis of stereo composition.

There are no limitations as to subject matter. Nothing is un-
suitable except a subject which is already planar. There would
be little point in making a stereogram of a fiat page of a news-
paper. Anything which has solidity, which lies in more than one
plane, is a subject which will provide a better stereogram than
photograph.

142 THREE-DIMENSIONAL PHOTOGRAPHY

The idea that stereo is not a medium for esthetic expression
and is suitable only for purely record work, is wholly wrong.
Stereo is capable of producing results which are as beautiful as
those of any other graphic art, if not more so. The fact that we
have not perfected a method of attack, that our taste has not be-
come trained to the third dimension, argues only our own short-
comings, not those of the medium.

Generally speaking, the attack is one which must be worked
out individually. We can observe the rules in so far as we know
them, but the ultimate judgment must be based upon the effect
obtained, just as it is in any pictorial work.

There are at present several stereographers who are already
making progress in the field of stereo pictorialism. Many of
these are Stereo Guild members, none of them from the ranks of
the planar pictorialists. It grows more and more evident that the
stereo pictorialist is a genius apart and the ranks must be filled
from those heretofore unknown. The planar pictorialist, through
long years of habit insists upon bending his stereo composition to
fit the rules and limitations of planar composition which are in-
tolerable and fatal to stereo.

We shall have cause to mention a few of these. This does not
mean that those mentioned are the only ones. We should like to
give credit to the many stereographers who have done outstand-
ing work, but because of space limitations we shall mention only
a few of those whose work is typical of some specific form of
stereography. It is so much easier to discuss an actual stereogram
than some synthetic ideal which does not exist.

The new stereo pictorialists have already shown a surprising
variety in treatment as well as a high and admirable degree of
individuality. Much of the work is excellent, most of it worth-
while. Any subdivision such as the writer may attempt comes only
from the necessity of having some kind of classification which
will make discussion possible.

SCENICS. One stereographer bemoaned the fact that he could
not get to the West where "You can stumble, fall down and get
a grand picture from the exposure you accidentally made while
falling." That belief is widespread and every stereo exhibition
has scores of scenes made in our National Parks. They are defin-

PICTORIAL STEREOGRAPHY 143

itely things of beauty and no one would call them stereographic
failures. But such scenes are not the only ones worthy of repro-
duction. In fact such scenes depend upon the majesty of the
original and demand only technical skill on the part of the
photographer.

On the .contrary there are many scenic subjects whose beauty
is revealed only to the observant, and when such a scene is used
to make a beautiful stereogram, the stereographer deserves more
credit than when his subject is such that a deliberate effort would
be required to produce a poor picture.

For example there was such a subject reproduced by Faulk-
coner; the locale, the Indiana woodland; the time, early spring;
the central subject a modest little waterfall. At first glance the
slide seems to be in monochrome, similar to a brown-black toned
slide. Nor is there anything in particular to recommend the com-
position when viewed as a planar picture, but when the stereo-
gram is viewed as such, the whole thing changes. The mono-
chrome resolves itself into tones of gray, green and brown, ac-
cented by a small tree filled with brilliant red and a smaller
shrub sprinkled with white. The whole thing is a subtle compo-
sition, the first glance does not reveal the scene. It grows and
continues to grow upon one, as element after element impresses
itself upon one. Finally when it has been absorbed the spec-
tator realizes that the scene is one of breath taking beauty and
that it involves a subdued but highly intricate color pattern.

This scene is a highly successful scenic stereogram but one for
the connoisseur.

Another subject is the pictorial representation of historic sites,
a field in which Short is doing excellent work, although his spe-
cialty is the general scenic beauties of the Atlantic Coast from
Florida to Maine. Autumn foliage is a subject which attracts
many workers, but there are few who concentrate upon the color
pattern regardless of the spatial pattern.

Figure work is a fertile field, but the usual pictorial euphe-
misms of planar figure study are not appropriate to stereo. There
is a straightforward realism in stereo which is easily adapted to
fantasy but which is incompatible with the little hypocrisies dear
to the heart of the portrait and figure photographer. Dramatic

144 THREE-DIMENSIONAL PHOTOGRAPHY

lighting is more useful in stereo than in planar work, but if light
is to be used as a spatial solid the beam must be made visible
by suspended particles such as smoke or powder, if it is to be
used as light it must not be given the pictorial weight of a solid.

Smoke, soap bubbles, tenuous veils, in short all objects whose
physical mass is subtle are reproduced in stereo without taking
on the appearance of being carved from cardboard as in planar
work. These accessories to the figure have been used to good ad-
vantage, and occasionally as primary subjects rather than as ac-
cessories. Dunnigan has done excellent work in this field.

Stereo nudes are plentiful, good stereo nudes among the rarest
of all stereograms. Among those whose work has come to our at-
tention, we admire that of Hilborn. His nudes are living models,
in unconsciously natural poses of everyday life. They are for
the most part girls of natural grace who are not posed in stiff,
formal, ridiculous poses, but who are living life naturally and
easily, without clothing. The whole spirit of most of the Hilborn
studies is that the models are simply not conscious that such a
thing as clothes exists, and that there is nothing at all strange or
unfamiliar with the condition of nudity.

But we cannot catalog every possible subject for the stereo pic-
torialist. We must proceed to generalizations if this chapter is
to be kept within reasonable limits of space.

MODERN TREATMENT. The planar pictorialist argues that he
is free from the bonds of realism, that he can embrace the whole
universe of impressionism. That might evoke a number of state-
ments which perhaps should be left unwritten. The writer has
a deep admiration for some of the truly great impressionistic
painters, but he has never yet seen a photograph presumably made
after the impressionistic, modernistic or surrealistic schools which
appeared to be more than a meaningless conglomeration! Perhaps
the physical limitations of the medium prevent such a purely
spiritual expression; perhaps the limitations lay with the pho-
tographers. At any rate the planar photographer has a real prob-
lem before he can bend the characteristics of the planar photo-
graph to a genuinely impressionistic result but the stereogra-
pher has no such limitation. His medium presents him with the
possibility of retaining all the subtle elements of realism, all of

PICTORIAL STEREOGRAPHY 145

the inimitable coloring of nature combined with a presentation
which is more unreal, more fantastic and at the same time more
harmonious than even the manual artists have achieved! Truly,
stereo is the photographic medium par excellence for genuinely
abstract presentations.

The stereo abstraction may be a wholly normal subject, given
a definite atmosphere through considered treatment. Another,
although using a normal original, attains a fantastic appearance
through pseudoscopy. Then too, there is a third type which is
purely abstract and is achieved by making a stereogram of an
original which never existed in the form reproduced. For example
a moving point of light is registered as a continuous line. But let
us consider a typical example of the first type.

The subject, a dead tree, fallen, with white limbs reaching
toward the sky from the surface of a river. Not so much of a
subject perhaps, but treatment made the picture. The stereo-
gram was made at night with flash. There is no light in the pic-
ture, no reflections from water or background, only those gaunt,
ghostly limbs stretching upward. Nor was there the safe haven of
the artificiality of paper, the thing was real, with all of its atmos-
phere of fantasy, of obscure premonition, of the atmosphere of
death. It was only a few dead branches, inanimate, yet that picture
carries more suggestion of terror than any I ever saw upon a plane
surface. It is simply a stereogram of intangible atmosphere. We
have seen three such stereograms by three different pictorialists.
All were highly effective.

But now to the second type, what we might call, and with very
good reason, the stereo surrealist school. The technique is ab-
surdly simple, the absence of transposition, pseudoscopic view-
ing. It is a trick often used to make "puzzle pictures," it has been
used repeatedly to emphasize the form of certain objects, but
when the method is combined with a nice discrimination of form
and color, the inversion of relief causes such a change that the
subject is not only unrecognizable, it can often not be recognized
when its identity is known. Yes, it is the simple shifting of eleva-
tion and depression into their reverse phases. The result is in
effect a pure abstraction. There is not only color without apparent
cause, just color valuable for its effect alone, but there is that

146 THREE-DIMENSIONAL PHOTOGRAPHY

which no painter of the abstract has even dreamed of achieving,
a definite pattern of relief which is as detached from reality as is
the color. This, too, has been used by many stereographers with
a wide variety of subjects. (See Chapter 16.)

The disinterested person will very probably find the stereo ab-
straction more attractive than the painted one. Perhaps it is be-
cause the stereographer has the aid of nature to preserve propor-
tions while form disappears as such, perhaps it is the subtlety of
tone, color and contour gradation in which photography (plus
stereo) is unexcelled, but whatever the reason, the effect is an utter
sublimation of realism into pure impression.

There is another type of abstraction which has been given a
great deal of attention. Perhaps many of you recall the Picasso
story in Life, telling of the efforts of the artist to paint pictures
with light, that is by waving a flashlight around before the camera.
If so you may be surprised to know that for months before this
article appeared stereographers were playing with just such "light
pictures." But they were all in three dimensions and in multi-
color! Dahl produced some excellent examples early in the game,
and many have followed. We saw one in which a fiery female
figure stood life size in a dimly seen garden, with multi-colored
veils streaming behind her, all "painted" with flashlights and col-
ored filters.

Line was abandoned as the sole medium early in this move-
ment, now the light source varies from a pinhole to a beam four
or five inches wide, so that splashes and waterfalls of fire may be
produced.

This may not be art, but it is highly amusing. One amateur
said he had to quit it because he used up twenty magazines of
color film in one week playing with what he called "pyrographs"
or fire pictures, not an inappropriate name.

FANTASTIC REALISM. Directly opposed to this non-realism are
the super-realistic stereograms made with electronic flash, and
those made with stroboscopic lights (rapid succession of exposures
upon one film). Not only do we see leaping figures frozen in mid-
air and such subjects which are more or less familiar to us, but
we have a group of such subjects which reveal exquisite beauties
never before seen by man.

PICTORIAL STEREOGRAPHY 147

Perhaps the most outstanding work of this kind is done by the
Kents. The subject in general, birds, the one specific stereogram
in mind at present, a cardinal (red bird), in the air, braking de-
scent preparatory to alighting, but a good two feet away from the
landing place. The slide is perfect. Every feather visible, the scales
on the leg skin clearly visible, in short all detail as clear as if the
bird were stuffed, yet the taxidermist does not live who could im-
part that air of tenseness as the widespread wings are controlled
with exactitude, and the whole body just ready to land. Nor is the
color inferior to the other qualities of the slide. In short it shows
what has happened before our eyes time after time yet which we
have never been able to see because of the inefficiency of our
vision.

Certainly that is a record shot, it has great value of scientific
nature, but despite that, there have been few photographic pic-
tures which could rival it in pure pictorial value! In fact, recalling
the statement of the conventional pictorialist "Stereo is fit only
for record work" one is tempted to reply, "Perhaps, but in stereo
even the records can be superlative pictures."

But life is too short for recrimination and quarrels. The fact is
indisputable that stereo offers quite as many pictorial opportuni-
ties as does planar photography; stereo offers several pictorial op-
portunities which are impossible to the planar photographer by
reason of the inherent limitations of his medium; stereo pictorial-
ism presents a challenge because its potentialities have not yet
begun to be explored; and because the stereo picture is appealing
to everyone inasmuch as it represents the realism of nature ex-
pressed by the individual stereographer instead of being a cold
artificiality appreciable only by those who have been schooled in
the complex theory of pictorial composition of a rapidly pass-
ing era.

DISTANCE LIMITATIONS. The question of distance limitation
involves only personal preference and effectiveness, not any fixed
rules. It is true that the older stereo technique did try to impose
such limitations and there are even formulae which prove mathe-
matically the existence of limitations imposed by natural law. The
trouble with these "laws" is that they are transgressed daily by the
new stereographers who never heard of them, and the results are

148 THREE-DIMENSIONAL PHOTOGRAPHY

quite satisfactory. A "law" which can be ignored without percepti-
ble result certainly has little authority!

We have a formula to tell the amount of separation between
objects which can be seen at various distances, and the formula
would be dependable if only human vision were uniform; but
the sad fact remains that no individual can predict what another
can and will see. Therefore the formula can only give values
which apply to those individuals whose vision is accurately in
accord with the laboratory standards, perhaps one person in a
thousand. We have another formula which proves conclusively
that normal base stereo cannot be used at a distance nearer than
well ten, eight, five feet, take your choice! One marvels indeed
at the power of mathematics or should one say the dexterity of
the mathematician? There is a good, sound formula to prove just
about anything that anyone wishes to prove. Again the trouble
with the formula is that it is meaningless because it does not even
approximately conform to demonstrable fact.

What you can do with stereo depends upon your stereo sensi-
tivity, and with all due regard to the theorists, stereo sensitivity
involves more than stereopsis, at least more than the term stereop-
sis is assumed to include.

For example, there are two objects, both at a considerable dis-
tance; one is slightly nearer than the other. According to the
formula, the distance separating those two objects must be at least
such-and-such to enable you to see that one is nearer than the
other. When you actually look at such objects you (as a rule) can
easily see the separation and by careful observation you gain some
idea of the amount of distance between them. Here stereopsis is
superior to the formula. If they are moved closer together by a
certain amount, conditions change. When you first look at them
you can tell which is nearer, but if you study them you lose assur-
ance and cannot tell which is nearer. Here the tenuous stereo
sense has operated but stereopsis is now undependable. This
subtle sense is difficult to identify but the writer feels sure that it
is only the normal operation of dynamic parallax. As long as the
eyes move from object to object, the stereo sense is most acute,
and as soon as the eyes become fixed, the acuity drops. At least
this is true for general conditions and there is no reason why it

PICTORIAL STEREOGRAPHY 149

should not also explain the discrepancy between the delicacy of
the stereo sense, and that of stereopsis as usually defined by meas-
urement.

One thing is undoubtedly true, the conventional stereo limita-
tions are certainly far more restricted than is warranted by aver-
age normal vision. But we must remember that these limitations
were worked out by men leading a largely sedentary life, by stu-
dents, by city dwellers, by those in whom we should expect to find
the lowest degree of stereopsis. When the average of outdoor
dwellers, foresters, hunters, seamen, farmers, small town and vil-
lage dwellers, and particularly those who habitually make use of
stereograrns are examined we find the actual value of stereo in-
finity jumping to from two to three times the stated limit!

As for near distance, there seems to be even less logical reason
for its imposition. There is no near limit for stereo. The nearest
distance the writer has employed is (theoretically) i.5mm, using
a micro objective of that focal length. However, not to quibble,
the nearest distance at which stereo may be used with a normal
lens separation is the nearest distance at which the homologous
images fall upon the film area provided, a matter of inches, not
of feet. Of course, as explained in the chapter on closeup work,
this involves a discrepancy in field and loss of picture area; the
tall, narrow panel is not esthetically attractive, but as far as stereo
is concerned, the image is reproduced and it is reproduced with
that degree of relief which would be seen by the eyes at the same
distance.

The trouble lies in the "mental policing" which has been men-
tioned before. We are not consciously aware of the flattening of
round objects by distance and similarly we are not aware of the
exaggerated relief which we actually see when an object is held
close to the eyes, but stereography is certainly not responsible for
the vagaries of human vision. The stereogram represents just that
degree of relief which would actually be seen by direct vision,
always provided of course that the lens separation is equal to the
interpupillary. The mechanism of parallax is as automatic as that
of perspective. There is one factor, however, which does exist in
the extreme close-up: those made at a distance of about one foot
or less. The dissociation of accommodation makes the extreme

150 THREE-DIMENSIONAL PHOTOGRAPHY

convergence noticeable, and the actual muscular tension, as it
changes to follow the extreme depth changes, can be felt but
not painfully. The effect is rarely noticed unless the convergence
depth is considerable; and it is rarely experienced when the
original distance exceeds 18 inches.

True, many stereographers, particularly those of the old school
who pride themselves upon their orthodoxy, will refuse to look at
a stereogram made at two or three feet, exclaiming that it is "a
caricature," "intolerable," "a travesty." I often wonder if they
can really be sincere, if anyone is really capable of such self-deceit,
if anyone can refuse the actual evidence of his own eyes to such
an extent.

There is always the proof of the pudding. The chances are that
you have a camera which focuses down to 2 1/% feet. Make a picture
of a rose, for example, at this distance. Compare the stereogram
with the original, looking at the latter from the same distance of
go inches. The stereogram is no more difficult to view than is the
real rose! The stereogram shows no more exaggerated relief than
does the actual rose! And moreover nothing but the stereogram
made at 30 inches can show the true appearance of a rose seen
at 30 inches.

Short has done remarkable work with outdoor subjects. One
is a group of small salamanders in a setting of moss and violets
which is worthy of acclaim. Although made at about 18 inches,
there is nothing in the slide to indicate the fact. Only the size
of the known objects provides a hint, but nowhere is there any
indication of depth distortion. Brooks also does superb floral
work at very close distances as well as definite macro. All of his
slides can be viewed with complete comfort and exhibit wholly
normal stereo relief.

Our frank advice to you is wholly to ignore all such "laws"
which can so easily be demonstrated to be meaningless. We do
have stereo laws, real laws, laws which can be demonstrated and
which are always in effect. Those we respect; but we cannot re-
spect "laws" which have been promulgated simply because they
substantiate the defective stereoscopic vision of some one indi-
vidual or group of individuals.

MATHEMATICS VERSUS SUGGESTION. Just a word about mathe-

PICTORIAL STEREOGRAPHY 151

matics. This is an exact science, the nearest approach to the abso-
lute man has made. It is invaluable in most human activity. To
question either the exactness or value of mathematics would be
absurd. The trouble is that in stereo, theorists have used perfect-
ly sound mathematics to "prove" their inoperable theories, and
because the mathematics are sound, the reader accepts the proof.
The fault usually lies in the assumption of a false premise, which
is, of course, not evident in the mathematical presentation.

But when mathematics explains (not proves) phenomena
which have been substantiated by the only tenable form of proof,
actual demonstration, we discover the full value of the science.

Still, there are discrepancies which cannot be explained clearly
by anything but analogy to other sensual experience.

There certainly is a limiting distance beyond which the eye
does not distinguish parallactic difference. There is beyond ques-
tion a true stereo infinity. Moreover that limit can be precisely
evaluated provided we can devise some accurate measurement of
the fundamental sensitivity of the particular individual involved.
Once the basic evaluation has been made, the physical stereopsis
of that individual can accurately be predicted. All this is of the
utmost value in certain stereoscopic studies.

While officially stereo infinity lies at 670 meters, somewhat less
than a half mile, among outdoor people, experience and observa-
tion indicate that the limit actually exceeds a mile. It is doubt-
ful that anyone has a stereo infinity at a distance greater than
21/2 miles. But, if you make a stereogram of a scene which em-
braces a depth of five miles or more, you can actually see in the
stereogram, the natural visual relief extending to the full limit of
the scene. You can even see this in a stereogram in which the
depth is 20 or 30 miles, as in some scenes which include distant
mountain ranges.

Why can you see relief through a depth several times as great
as the value of stereo infinity?

The explanation is the familiar one. The stereogram shows you
just what you would see if you were looking at the real object.
When you see such a scene, you are not conscious that your per-
ception of stereo relief ceases at about a mile, more or less. You
"see" the relief right back to the limit of vision. Why?

152 THREE-DIMENSIONAL PHOTOGRAPHY

In this chapter we have already discussed several extrinsic fac-
tors which enhance the appearance of relief although they do not
affect the true relief. As true stereo relief becomes less and less
evident, these extrinsic factors take over and there is a gradual
replacement until, at stereo infinity, they have taken on the whole
load of relief.

Stereo perception does not mean the ability to see that depth is
present, it does not mean that contours can be interpreted. That
can be seen in a planar photograph, and a person who has unfortu-
nately lost one eye often has a very good judgment of distance.
The perception of stereo relief is that sensation which enables us
to see (not judge or interpret) the most minute details of differ-
ences of depth. It is something impossible to describe to those who
have not experienced it, something which can with difficulty be
distinguished from the extrinsic evaluations when it is described
to those who have stereo vision, but which is instantly and obvi-
ously apparent as soon as a stereogram is viewed.

The difference is overwhelmingly great, yet it is subtle. For that
reason, even when one is aware of the problem involved and
makes every effort to determine the boundary between stereo and
extrinsic depth evaluation, the limit can be only approximately
defined.

Because the stereogram reproduces all of the effects of the ex-
trinsic factors as well as those of the intrinsic parallactic factor,
the stereogram shows you what the eye would have seen. True,
those who have studied the subject for years can see the flatness of
distance either under direct vision or when viewing the stereo-
gram.

But after all, this is merely the splitting of technical hairs. You
are not interested in the psychology or the physics or the mathe-
matics of stereo. You want to make attractive stereograms. That is
no doubt the full extent of your goal, even as it represents the
ideal ultimate of stereoscopic photography.

Then, the one thing you want to know is this: The limitations
of the normal stereo camera are the limitations of a normal pair
of eyes, because "What you see, you get."

At the same time, objects relatively near, say at distances of
three to 20 feet, are the most clearly delineated to vision because

PICTORIAL STEREOGRAPHY 153

truly familiar objects are usually seen at such distances. You will
find that stereograms made within these limits will usually afford
you the greatest degree of satisfaction. This is not true because
of any limitations imposed, but because it represents that visual
zone which is the most satisfactory in direct vision; and any ex-
ceptions which apply to vision apply to stereo.

CHAPTER 10
THE NUDE IN STEREO

PROBABLY MANY OF YOU who read this have no experience in
pictorial work, either in photography or any of the arts. If so
you probably wonder why the importance of the nude is stressed
to such an extent. In short, why photograph the nude?

There are many reasons. The nude is the human being; the
clothed figure is hardly more than an advertisement for the cloth-
ing industry. You cannot effectually pose the draped figure until
you know, thoroughly, the form and function of the body which
supports the clothes. Then too, the human figure, particularly the
female, is the most beautiful object we know. This is not a matter
of the forbidden or of the unknown. It is a simple matter of color
and texture, of contour and shadow gradation, of line and inher-
ent grace of movement. If a clay figure could be made which ex-
actly duplicates the body, it would serve almost as well as the
living human. However such imitation is impossible, for at each
movement there are subtle changes of the underlying skeleton
and muscular structure which are reflected upon the surface. One
of the great attractions of the nude lies in its infinite variety.

The nude figure comprises the sum of all the important prob-
lems which the photographer has to face. If you can make a good
nude, you can make a good photograph of anything. Nothing
else can give the training in lighting and posing which this does,
and nothing can lead to such a long series of disappointments;
that is, if you take real pride in your work and are not satisfied
with a mediocre result.

But all of this has to do with technique. There is to this prob-
lem another aspect, which is so important and so wholly absurd
at the same time, that it must be straightened out. This is the
layman's conception that the nude is offensive. For the most part
we have two great groups. One group states that the nude is ob-
scene per se. The other group simply replies that the artist should
be above such considerations and be permitted special privilege
in the matter. Both points of view belong to an age which could
produce the inquisition and the witch hunt. Both are so utterly

154

THE NUDE IN STEREOGRAPHY 155

absurd that they should be ignored were it not for the fact that
these two groups are numerically so great.

The first is perhaps the more excusable. True, it is difficult to
conceive of the unwholesomeness of the mind which can possibly
perceive anything offensive in the human figure; but that is still
more easily understood than the mind which believes that while
the nude is offensive, it must be made use of by a restricted ele-
ment of society for artistic purposes. The latter group prates of
idealism, and neutral attitudes, which is sheer nonsense. The very
ones whb use the statements disbelieve them. They are excuses,
not reasons.

The whole situation arises from the fact that as a nation we
are incredibly adolescent. The question of sex is involved, of
course. But the whole trouble lies in the general unwholesome
attitude toward sex. After all what is it really? It is the sole force
which is responsible, of course, for the population of the world;
but more than that it is the sole force which has resulted in our
progress, which is directly responsible for every great accomplish-
ment of man, of all great art, music and literature, of all inven-
tion and discovery. It is a scientific fact that it is the factor which
prevents man from stagnating, from becoming dull, stupid, mean,
treacherous and generally repulsive.

The origin of our national fear of sex is difficult to discover.
Of course it all stems from the older eras when a woman was a
chattel, a thing purchased and owned purely as a plaything for
man. But while we have progressed in every other way, we still
make the degradation of sex a principle of social progress! We
permit the most disgusting abuses of eating and drinking, a man
may take to drugs or dissipate his whole substance in gambling
and we are indulgent, but if he is "guilty" of the most natural sex
activity he is condemned. It is simply beyond any intelligent logic.

Yet those who are most bitter against any idea of sex refer to it
constantly throughout their lives. Such words as "he" and "she,"
"man" and "woman," even "boy" and "girl," or "king" and
"queen," "horse" and "mare" and so on through hundreds, yes
thousands of words, are differenced only by sex. These very people
are the ones who would rebel most strongly against becoming
asexual. In fact, the psychologists inform us that those who are

156 THREE-DIMENSIONAL PHOTOGRAPHY

most rabid in anti-sex crusades are those whose minds are pre-
occupied by the subject.

Sex next to life itself is our most vital driving force. It is the
most precious gift we possess. It alone gives us a touch of divinity
and makes of every one a creator. It is not only pure and whole-
some, it is the most sacred thing known to man. If man cares to
deliberately defile it, that is his misfortune, but it does not affect
the status of sex itself.

The wholly innocent person is unaware of sex, and the nude is
the very mirror of innocence. But, and this is important, when a
bathing suit is donned, does it add to modesty? Again our psy-
chologists tell us that the brief covering only serves to make con-
spicuous the areas covered, where before they were but com-
ponents of the whole without emphasis. More practically it is
known that pictures made with base intent always make use of a
certain drapery to enhance the suggestive effect. In short the pro-
ducer of objectionable pictures and the anti-sex crusaders favor
the same degree of obscurity. We should hardly care to point out
that similar minds think along similar lines, but there it is.

On the other hand, there is no new situation which loses its
novelty quite as quickly as being in the presence of the nude. For
example, I recall one picture which was fairly elaborate. There
were some six or eight nude models in the composition, and work
had been proceeding for several hours. Finally the chief was satis-
fied and signalled for lights and the exposure. Nothing happened.
The cameraman and the electrician had become bored and had
started a sharp argument. As I was near I had heard the whole
thing. The subject of the argument was, you will be surprised to
learn, the abilities of two rival baseball pitchers of the season.
The presence of a half-dozen very beautiful nude models meant
no more to these men than the painted background. Later I
asked the electrician about it. His reply tells the whole story.
"Aw! You get fed up with naked dames in ten minutes. If they'd
put some tights and ruffles on 'em and let 'em dance, you'd have
something!"

To sum up the situation, there is nothing more inherently ob-
jectionable in the nude than there is in a new automobile of
which you might wish to make a picture.

THE NUDE IN STEREOGRAPHY 157

It is true, of course, that highly objectionable photographs
are made and circulated, but these offend good taste far more than
they offend morals. However, because so many of our fellow citi-
zens limit the term "moral" strictly to sexual morality the law
has been forced to take cognizance of the fact.

It became necessary for the law to create a line of demarcation
upon one side of which poses would be acceptable, on the other
side objectionable. The result was hilariously funny, but after all
it was as well done as could be expected when a physical limita-
tion is applied to subjective reactions tinctured by idiosyncrasy.
The law simply says that if the pubic hair shows in the picture, it
is necessarily objectionable, but if it does not show, the picture
is acceptable! The processing laboratories accept that legal dif-
ferentiation, so if you want your color films returned, be sure no
pubic hair is shown in the pose!

Of course it is absurd! It is perfectly easy to make a photograph
with the pubic hair fully displayed yet to have that the most inno-
cent of poses. At the same time it is even easier to make a photo-
graph in which the pubic hair, the breasts and the face are all
concealed and to have that a pose of the highest suggest iveness.
For example see some of Boucher's paintings, the prized posses-
sions of some of the great museums. But as long as the difference
lies in the mental attitude of the model, (and the physical ex-
pression lies in nuances of muscular tension too vague to be ver-
bally described, but instantly apparent to vision), what can you do
about a legal differentiation?

Too, when you have certain people, many of them active in
anti-nude crusading, who find unlimited suggestiveness in almost
any photograph of an attractive woman, and when at the same
time there are thousands of people who have no reaction whatso-
ever to a nude study other than to admire its beauty, how can
you apply a differentiation? If we should allow laws to be based
upon the mental wholesomeness of the individual we should be
lost.

No, ridiculous as it is, undignified and unseemly as it is, the law
is perhaps as good as can be expected as long as some kind of
limitation must be imposed. Of course, the only sensible position
for an intelligent society is to have no such limitations at all. Any

158 THREE-DIMENSIONAL PHOTOGRAPHY

kind of censorship is but an evidence of the desire of one group to
impose its ideology upon another, the germ of dictatorship. It
may be assumed that most adults are capable of guiding their own
lives, particularly as long as they happen to be so fortunate as to
possess that one characteristic whose absence seems to be typical
of the usual "sex" crusader, namely, good taste.

In time this will come. In time, and I expect to live to see the
day, all sex taboos will be things of the past, and then we can
work freely with the nude just as we do with a rose or any other
object of superlative beauty.

There has been a great deal written about the pose. One school
"idealizes" the model by obscuring shadows or by the use of some
absurd drapery which only accentuates the features which the
photographer thinks he obscures. He should learn the true nature
of the nude. Another school assures us that only those poses are
permissible which shriek to high heaven of artificiality. Their
argument is that the offense lies in naturalistic poses. They too
have been contaminated by the universal poison.

Consider this. Except for reasons of physical comfort, there is
no reason why you should not converse, play bridge or eat lunch
with one or more nude people. Of course, the average figure being
what it is, this might not contribute much to esthetics, but much
of our lives is subjected to unesthetic influence. Consider also
that clothing is merely a collection of rags hung upon the frame-
work of the body. The original purpose was protection. Today de-
signers o clothing are quite frank in saying they are guided by
the importance of making women more attractive to men and to
hide the deficiencies of the natural figure.

You think this extreme? If you ever use as a model a girl who
has a "fashionable" figure you will never again doubt it. The
emaciated, flat-breasted, masculine figure which is "fashionable"
is utterly hopeless for figure work. Rather select a model who is
somewhat "dumpy," not extremely so, but too much so to be
fashionable. You may probably find a figure which approaches
perfection. Women do not like this statement and argue against
it, but their mirrors tell the true story. The feminine figure has
breasts, a waist of certain constriction and widened, oval hips.

But most of all, clothing hides deficiencies, in some instances
almost deformities, which are so common. It might be said in

THE NUDE IN .STEREOGRAPHY 159

passing that the most vigorous enemies of the nude are the un-
fortunate women who need such camouflage.

Therefore, the presence or absence of clothing should not be
regarded as of more than the most insignificant importance. And
this in turn leads us back to the original question of the pose.

The pose of the nude is best when the action expressed is wholly
normal. Lounging in a chair, lying on a day bed, playing with a
kitten, picking flowers, eating lunch, reading, swimming, or any-
thing else which is normal and which gives rise to normal physical
motions. The only requisite is that the model be wholly uncon-
scious that the state of nudity is in any way unusual.

One writer has cautioned against posing the nude out doors
without a blanket or some such protection. Such consideration
is admirable of course, but the blanket introduces a false note,
and I must add that this man's models must have been far more
delicate than those with whom I have worked. They seem to care
no more about the presence of rocks, sticks and stones than would
a man. And certainly when you take away the model's clothing
for a pose in natural surroundings, the blanket is definitely out
of harmony.

The selection of a model is often difficult. It is not so difficult
to find a model, but it is difficult to find one who enters into the
spirit of the work, who will try to work with you in your prob-
lems of composition, one who will learn and advance in ability
as you do. Of course you can always hire a professional, but this
is inadvisable. First the rates are extremely high. Few of us can
afford $50 an hour. In the second place, the professional is ac-
customed to professional photographers, whose work is purely
commercial and whose work is carried on in an atmosphere of
blase boredom. She will be contemptuous of your first stumbling
efforts, and will never be in sympathy with your work. The idea
of trying to create a picture for the simple joy of doing it lies
beyond the comprehension of the commercial world. No, by all
means find a sympathetic amateur. Your progress may not be as
fast, but it will be much better. Moreover you will have a chance
to develop your own pictorial individuality without the influence
of the stereotyped commercial world.

Before starting the actual posing, you should give some time

160 THREE-DIMENSIONAL PHOTOGRAPHY

to a study of art anatomy. Any pose which can exist is one which
superficially interprets the function of bones and muscles beneath
the skin. Become at least somewhat familiar with this underlying
structure if you wish to avoid poses which are unnatural and
awkward. A whole volume could easily be devoted to this one
subject, so instead of trying to present a too brief resume, we
shall only advise you to obtain a dependable reference text, (your
local library can possibly supply it) and go through it conscien-
tiously.

When starting the first session with your model, you will both
be somewhat self-conscious. There is no use trying to do any
camera work as long as any trace of this feeling exists. The model
will probably lose it before you do. At any rate, have another
woman present, preferably the mother or a sister of the model.
Put in the first study period by having her assume certain poses
illustrated in the art anatomy. See for yourself the skeleton and
muscles as they function. Once you have the key it is surprising
how clearly you can really see these submerged motions. Have
your model leave off her robe even during rest periods at first.
Cultivate the complete and wholesome normality of atmosphere
in the studio. A single 30 minute period at most should suffice.

The next session can be devoted to elementary figure work. If
you can possibly find a copy of Fred Peel's Shadowless Figure Por-
traiture study the poses and especially the fragments. Use one
period and at least 15 exposures of the elbow alone. Do the same
with the knees, then to the ankles with feet, the shoulders, the
head and neck from the rear so as to exclude the distracting ele-
ment of the face. Give very careful attention to the hands and
wrists. Of course the model need not be undraped for this type of
study, but it is highly advisable that she be so. Just so long as you
continue to think there is anything in the least abnormal in the
state of nudity you will fail to produce really good work.

The torso from the rear is worthy of several sessions. With each
change of pose there is a shift of the line of the spine. This line
has naturally the recurved form which is characteristic of the
Hogarthian Line of Beauty. Other lines branching to one or both
arms illustrate harmony of line. Alteration of masses by shifting
lights gives opportunity for endless studies.

THE NUDE IN STEREOGRAPHY 161

The torso front introduces more problems. The simplicity of
the back is gone. The breasts make up one distinct area; the abdo-
men another and they are separated by the waistline. This divi-
sion must be handled to eliminate any division of mass, and at
the same time you have a complicated problem of contour. When
you have mastered the elements of a static pose, go ahead to others
and note the change of both breast and abdominal lines as the
model leans forward, or raises her arms, or when she lies supine.

When you have made at least 100 fragmentary studies you are
ready to start serious figure work. Again simple, natural poses are
the best. From your study of anatomy and the fragments you have
made you should be able to see grace or awkwardness of line in
any part of the figure. If the line is right, note the masses, study
contours and the effect of lighting upon them.

If space were available it would be possible to give a num-
ber of possible poses, and to give several warnings as to what to
avoid, but if you find you need them, there are several such texts
available on the market. The better figure photographers however
have developed their own taste in the matter. It is advisable how-
ever to present a few "don'ts":

Avoid dramatic poses, theatrical ones. Despair, worship, fear,
adoration, any strong emotion should be avoided. Your model
might be able to give a fair facial expression to such emotion, but
not to her body and always remember the body tells far more of
the mind than does the facel The same thing applies to the pre-
cious and artificial poses so popular a half century ago. Avoid the
allegorical pose as you would the plague. Stick to the normal,
everyday pose. You have a model, for the moment the archetype
of man's indispensable and adorable helpmeet. Photograph her in
character. Do not ask your model to be an accomplished actress
as well as a model.

So much for the question in general. Now what about the nude
as it appears specifically in stereo? Stereo does impose an addi-
tional load upon the photographer of the nude, just as it does in
all pictorial work. The planar photographers have repeatedly said
that in stereo the nude ceases to be a nude and is simply a naked
woman. Surely that is what any nude should be!

In all stereo work we are faced with the task of making reality

162 THREE-DIMENSIONAL PHOTOGRAPHY

beautiful. The artist simply omits objectionable features; the
planar pictorialist hides them in some "dramatic" shadow, the
stereographer either removes the element or works it into the
harmonious whole of the composition!

The stereographer is forced by his medium to abandon the de-
cadence of an effete civilization and to return to the unsullied
wholesomeness of nature. More than that he must extend that
same attitude to every minute feature of his scene. That paper
plate in the background, a souvenir of some picnic party, will not
be a mere white patch easily removed by retouching. It is a paper
plate, round and light in color and foreign to the setting. See it
and remove it!

Nor can you indoors depend upon tonal mergers to effect your
composition. The dark wall behind the model's head will be far
behind her and you cannot merge the two. You can kill all detail,
but the relief of the lighted part of the body will establish it in a
plane in front of the wall.

In the pose itself, you cannot disguise an awkward elbow angle
by swinging the arm back to make the angle more obtuse. As long
as that angle persists in any direction it preserves its awkward ap-
pearance. Forget foreshortening, it doesn't exist. That is a feature
of planar reproduction.

Do not study the pose in the finder, you will be deceived by the
planar effect. Do not study it from a position beside the camera
at first, you will see a different aspect. Study the pose by looking
just over the camera. Base your judgment upon the appearance of
the model as you see her. The stereo camera will reproduce what
you see.

Then of course, learn to use your eyes. Your first attempts will
be disappointing, and you will swear that the picture does not
resemble the original pose. Oh, but it does! Precisely! The trouble
is that you have not yet learned to see. Most people do not actually
see one-tenth of the things upon which their eyes fall. You will
be disgusted with yourself at first because of the many disturbing
elements which you overlook. Take plenty of time. Study the
pose. Have your model shift the pose slightly, or better yet, make
several exposures in each of which there is but a minor change.

THE NUDE IN STEREOGRAPHY 163

Study all of them. Select those which you like best and then con-
tinue the study until you decide why you like them.

Show all the poses to the model. Have her select her favorites.
See if they are the ones you chose. (They will not be.) Find out
why she prefers them. Explain your choice to her. The discussion
will help both of you tremendously.

If you have a projector, project the slides. Block off one lens of
the projector and study the planar picture. Then change to stereo.
Note the great change. This will do more to teach you the differ-
ence between stereo and planar than could be done by a book
twice the size of this one.

When you have had a little experience, start studying the model
stereoscopically. Study the pose over the camera. Then move to
the left, then to the right. Study the pose from several angles. Try
to obtain a pose which is satisfactory from every angle.

Next obtain a pose which is fully satisfactory from the camera
position, but which is decidedly unsatisfactory from the side
angles. Make this exposure also. Study the two and compare them.
You will find that the pose which was satisfactory from all angles is
decidedly superior to that which was satisfactory only from the
camera position.

This is true because when we look at an object, we make use of
stereoscopic vision subconsciously. It operates all right but we are
not specifically conscious of it. Therefore we concentrate upon the
planar effect. But with the stereogram we remain clearly conscious
of the stereo relief. Mentally we shift our position to look around
corners. We try to estimate the appearance of the subject from
other angles. This seems to be carrying an argument to the hair-
splitting stage, but one trial will convince you of the undeniable
reality of the fact.

In planar work a model is often moved a few inches out of
natural position, or she is instructed to look slightly more toward
the camera to give a better view of the face even though she is
looking off into space. This is known as "faking." Do not try it
with stereo. The fake is obvious in the stereogram.

Then too there is the oft-repeated statement that women in a
photograph appear shorter and stouter than in real life. Frankly
I do not see it that way, but perhaps those who are not trained to

164 THREE-DIMENSIONAL PHOTOGRAPHY

analytic vision are so deceived. At any rate I have had others speak
repeatedly of the fact that stereo does not have this effect. I have
had several women say about a stereogram, "Oh! I like that. It
doesn't make me look fat as I usually do in a photograph 1"

Another point. Often a model will wear her shoes in the studio
only to kick them off at the last moment. Avoid this. The shoe
marks the foot, and often the dye rubs off upon the toes leaving
a decidedly unwashed appearance, which is far more obvious in
the stereogram than in a planar photograph.

Very often too, the model will have a scratch or bruise upon her
ankle or calf. Cover the mark with makeup before the picture and
blend the edges carefully. Slap-dash makeup is as apparent in
stereo as in real life.

Another error which is made is to leave some jewelry upon the
model. Do not do it. Jewelry is clothing which turns the nude
into a semi-drape and ruins the whole atmosphere. Rings are easily
forgotten, but I could never understand the reason for one model
posing with an elaborate wrist watch as her only article of apparel.

Facial makeup should be kept moderate. In fact, heavy makeup
in any stereo is a mistake. If the fingernails are enameled the toe-
nails should be covered to match, but as a rule the better result
follows when all nails are natural. The fingernail enamel is the
stepping stone between nothing and jewelry just as that lies be-
tween nothing and clothing. It is the touch of the artificial which
can adversely affect the result. The same is not true of facial make-
up as long as it is used to accent natural features. The unpleasant
habit of many girls of repainting the shape of the lip is as ludi-
crous in stereo as it is in real life, and not a bit more deceiving.

The hair is a matter of argument. Of course any elaborate hair
dress is out of place. But as few models today have long hair, this
is not a problem. Any simple style of hair dress will usually serve.
Here too remember the three dimensional relief of stereo. A hair
dress which might appear simple in planar might reveal too clear-
ly the artifice employed, when the stereogram is viewed.

When you can make really good stereo nudes, start dressing
your model. First go to the bathing suit. Note the difference. First
of all there is the color disturbance. No longer is there the flow
of natural color. Even a "flesh" colored bathing suit does this.

THE NUDE IN STEREOGRAPHY 165

Next note that the beautiful breast line is ruined, and has become
a shapeless bulge. A torso study which is beautiful in the nude
now becomes ridiculous. There is an ugly accentuation of the hip
line. In short all of the esthetic attraction has gone.

Next add hose and shoes to the bathing suit. What is the result?
For the first time you do have a definite suggestiveness. You will
not like it.

Next have the model dress in all of her clothing, less the outer
gown. This may be stepins and brassiere, or it may be a slip or
whatever she happens to use. Here you have the secondary foun-
dation upon which the dress rests. The curve of the lower spine
is gone. The recurve beneath the buttocks is concealed. The femi-
nine contours have been transformed into a set of curious curves
ending in almost geometrically flat planes. Without your prior
anatomical knowledge you would have no conception of the true
framework beneath.

The characteristic shape of breasts and hips disappears. Both
are compressed into shapelessness. By contrast the waist seems to
bulge. The whole figure becomes amorphous.

Finally the outer gown is applied. Now what have you? The
foundation used has ruined the figure, and now the gown can only
decorate the shapeless figure beneath. So the gown itself is decor-
ated and made attractive. The natural beauty of the feminine
figure is replaced by an example of the dressmaker's art, and what
we admire is not the disguised figure but the dress!

But no matter. You now have a draped figure with which to
work and with which to do your best.

Now I want you to imagine two stereographers, each with his
own draped model, both trying the same pose. A has been through
the training with the nude, B has never worked with a nude
model.

B will be worried. He will say, "Well, I don't know just what
it is, but you don't look right. Try walking away and coining back
to the pose." The model does so, and there is no improvement.
B suggests changed hand positions, turning the body, bending
forward and bending back until the poor model has become as
stiff as a wooden figure.

A looks at his model and says, "Just a bit more weight on the

166 THREE-DIMENSIONAL PHOTOGRAPHY

right leg. Drop the right pelvis line a quarter inch, rotate the left
thigh a hair." He studies it a moment. "Center of gravity back
an inch and drop the right shoulder just a bit." He looks again,
makes the exposure and gets just what he wants.

B works all day and gets nothing because he has not the slightest
familiarity with the object of which he is making photographs, he
directs in terms of dress instead of terms of bones and muscles!

Even the actual folds of drapery usually respond to a good bone
position, more easily than otherwise, at any rate. So if you know
the figure as a result of personal experience, if you can visualize
the figure beneath the gown, and the true figure beneath all the
restraining garments, the muscles beneath the skin and the bones
beneath the muscles you will be a long, long way down the road
to successful stereo figure work.

CHAPTER 11
ANAGLYPHS

SINCE THE DAWN of stereo there has been an endless search for
some method whereby stereograms of unlimited size could
be seen. The first solution was the original anaglyph. Today this
is known as the chromatic anaglyph, because we have a superior
new type, one which approaches the ideal. This is the polarization
anaglyph, which is more commonly known as the Vectograph.

It is known that the stereo impulse depends upon the simul-
taneous stimulation of the two eyes by two different stimuli. It
is also desirable that these two stimuli should be placed so that
no excessively abnormal vergence is needed to view them.

It is also known that in additive color mixing, any pair of com-
plementary colors will produce white, and that this theory is borne
out in practice to a considerable degree. It is further known that
in subtractive mixing, two complementaries will produce black.

It would seem, therefore, that if a mixture, for example, of
orange-red and blue-green will produce white, the separate stimu-
lation of the two eyes by red and green simultaneously should
produce a sensation of white through stereo fusion. If this is true,
then a red picture and a green one presented separately to the
two eyes should produce an achromatic image (neutral or white).
If the red image is blanked out by a red filter and the green by a
green filter it should produce the necessary stereo differentiation.

Thus, the first anaglyph consisted o a stereogram made by
printing a picture in red, and another picture in green was print-
ed directly over it. This was viewed through a pair of "spectacles"
containing one red and one green lens.

The result was mixed. To some people there was an excellent
stereo effect; to others there was a stereo relief visible but upon a
background of alternating flashes of red and green. This latter is
said to be a visual-pathologic condition, but if so it is widespread.
The anaglyphie motion picture was first presented publicly three
or four decades ago and was withdrawn because of the complaint
from those who suffered headaches resulting from the color alter-
nation. In fact, this phenomenon was so prevalent that its name,

167

168 THREE-DIMENSIONAL PHOTOGRAPHY

"color bombardment" was a commonplace in the photographic
and motion picture industries some years agol

The writer experienced definite color bombardment, even from
still anaglyphs when he first saw them. The discomfort rapidly
lessened, and today the anaglyph, either moving or still, provides
perfectly neutral fusion and comfortable stereo. However, the un-
comfortable color bombardment effect is constantly met with
among many people. It may be that this is something which is
subject to a developed visual skill. We have not performed nor
are we aware of any such experiments as might prove or disprove
this theory. It is simply offered as a question as yet unsettled. The
evidence might seem to suggest it, but the evidence is not suffi-
cient in quantity nor has it been subjected to a sufficiently search-
ing examination to warrant a definite statement.

The fact remains that some people view the chromatic anaglyph
comfortably, others do not.

The method described above is widely used in Europe for
periodical illustration and for the preparation of mathematical
drawings, particularly in geometry and trigonometry. A German
army periodical was printed with almost 100 percent anaglyphic
illustration. Since World War II, in Russian Berlin, a whole
series of mathematical texts has been printed with anaglyphic
illustrations.

However, strangely enough, the continental anaglyphs are poor-
ly done, the ink colors have not been properly balanced to the
filters used, and there is a strong residual image where there should
be none at all. In this country we prefer to place the red and blue
inks far within the spectral cutoff of the filters, and we achieve an
almost total blanking out of the unwanted image. For this reason,
we usually print in a blue rather than green for the one color
and a fire red instead of vermillion for the other.

Amateur anaglyphs are often made. The images are either
toned with blue and pink toners (as used in the once popular
Chromatone color process), or they are printed from dye masters
in the popular dye-transfer method. In halftone printing, two en-
gravings are made with about a 30 difference in angle. These
are then printed in blue and red. For example, a peacock blue
and fire red have been used in the Guild laboratories.

ANAGLYPHS 169

Reproduction of Anaglyphs. The question of restoring ana-
glyphs to their normal stereo form has often been raised. This is
very simple. The anaglyph is copied upon panchromatic film
using photographic filters which will separate the two ink colors.

When the same copy is used with both a tricolor A and a tri-
color C filter, the A filter will record the blue printed image and
the C filter will record the red printed one.

If the anaglyphs are of the polarization type, the camera filter is
a simple polarizing filter. One copy is made with the filter blank-
ing one image and the next exposure is made with the filter rotat-
ed through 90. Panchromatic film is not necessary in making
polarization anaglyphic restorations.

Polarization Anaglyphs. This is the process which is known as
the Vectograph and for some reason it has never been freely of-
fered to the amateur. The Vectograph process was developed by
the Polaroid Corporation and was widely used in World War II
with the utmost success. It was later adapted to a unit for home use
for ophthalmic therapy by Bausch & Lomb, and has had limited
use among amateurs. It is highly satisfactory in every respect and
solves many problems as old as stereo. The pictures are in full
tone, monochromatic (color is being developed) and have all of
the quality of a normal print. They exhibit the unmistakable
relief found in normal stereograms, and, in fact, offer a solution
to the problem of an album full of stereograms which are to be
looked at in exactly the same manner as any collection of snap-
shots. They may also be projected in any standard single frame
projector, and if made as motion picture film they may be pro-
jected in full relief by any standard home movie projector.

Polarization anaglyphs are made from normal stereo negatives,
but instead of being printed' in complementary dyes, they are
printed by the dye transfer process in a dye which crystallizes in
dichroic form. There are many substances which will react cor-
rectly, but the formula developed by the manufacturers is ex-
cellent.

The secret lies in the base used. This is an optically active
plastic (transparent) base. By printing the image on the two sides
of this active base in the correct type of crystalline image, the two
images are polarized at 90 to each other. The examination of

170 THREE-DIMENSIONAL PHOTOGRAPHY

such a dual print through the usual gD goggles used for common
stereo projection produces the full stereo effect.

Neither the base nor the image is effective alone. Only the com-
bination of the image made up of crystals of the proper type with
the optically active base will produce the effect. It need hardly be
added that the crystalline structure is not visible to the eye. The
images have the full quality of any fine dye transfer print.

The normal Vectograph is a transparency. If it is desired to be
used as a print, the back is coated with a solution of very fine
aluminum powder in a suitable vehicle. The aluminum serves the
same purpose as the white paper in an ordinary print, but to pre-
serve the essential polarization it is necessary to have this reflect-
ing layer of metallic material. Hence the aluminum.

However, the Vectograph is at its best as a transparency. In
relative large sizes, 8x10, 11x14, or larger, and placed in an il-
luminated frame, Vectographs form one of the most striking of
stereo exhibits.

The Vectograph would be ideal for home movies if it were not
for the complex mechanical installation which would be neces-
sary. Also the quintuple film footage would boost costs, one
double length negative, two positives and a transfer base that is,
assuming that the original negative is exposed with alternate right
and left frames. The only practical solution would be through
central laboratories such as those used now for normal film
processing.

There is no such disadvantage with the still picture. Any com-
petent amateur can quickly learn to make good prints, and if he
happens to be versed in dye transfer already, he has very little
to learn.

The negative is made with an ordinary stereo camera. Enlarge-
ments are made on washoff film and developed in warm water. In
this step, one negative is reversed so that the prints are mirror
images of each other; placed face to face, they correspond. The
two images are carefully aligned vertically. Lateral alignment is
based upon normal stereo separation. The films are hinged to-
gether book-form with adhesive tape, then immersed in the spe-
cial dye bath. The two films are removed from the dye bath,
rinsed and the special base sheet inserted between the two, and

ANAGLYPHS 171

the whole run through a photo wringer. The base sheet is "stripped
off, with one stereo image on the back and one on the front.
Viewed with gD goggles, the image is seen in full stereo relief.

Unfortunately, at the time this is written the Vectograph chem-
icals are not available, but it is to be hoped that they will again
be offered in the near future. If for any reason that is not done, we
hope the formulae will be made public. This process is far too
valuable to be lost to the stereo world. We have been informed by
Mr. Joseph Mahler, the originator of the process that some de-
cided improvements have been made in the process, and that these
will make i6mm motion picture film production wholly practical.
It will also have added possibilities for the makers of trans-
parencies.

Anaglyph ic Spacing. In printing the anaglyph it must be re-
membered that this is a true print, held in the hand. We are
fully aware of that fact, and the eyes naturally converge upon the
physical sheet which bears the image. For this reason it is usually
advisable to print the anaglyph so that objects in the plane of
primary interest shall be registered as nearly as parallax will per-
mit, leaving greater differences to be exhibited by images of less
important objects.

See Chapter 14 for general discussion of polarized light in
stereoscopy, and additional material about the Vectograph.

Integrated Analaglyphs.A third type of stereogram which
properly falls into the anaglyphic classification is one in which
the two images are divided into narrow vertical strips in which
right and left elements alternate. These stereograms are made to
be viewed without any filters, viewers or other accessories.

Such stereograms have been made in several types. The original
consists of alternate strips set behind a grid of alternating opaque
and transparent bars. In theory the right eye sees past a bar
obliquely to perceive a "right" element, while the left sees past the
bar from the other side to see a "left" element.

In the resulting stereogram, a certain degree of stereo relief is
apparent when the eyes are within a certain restricted area. Mov-
ing the head to one side or the other introduces either neutral re-
lief or definite pseudoscopy.

To overcome inherent defects, the black bar grid was replaced

172 THREE-DIMENSIONAL PHOTOGRAPHY

by lenticular elements so the alternating ray paths could be pro-
duced without wasting half the brilliance upon opaque bars and
to remove the unsightly grid.

The lenticular film considerably improved the unsightly "pris-
on bar" effect, but it did not make the image three dimensional
from every convenient point of view. This lenticular idea was
applied to a movie screen after the expenditure of a fortune. But
as could have been predicted, it worked only when the eyes were
correctly aligned with the screen. For that reason it was dropped
only to be re-invented (?) by the Russians after the patent reprints
had been made public. This is the much vaunted Russian stereo
movie. Of course, in Russia it is possible to maintain audience
rigidity by edict, but here the audience pronounces the edict
against the exhibitor.

Now let us consider a few elementary facts:

If stereo is to be seen without any type of viewer, it is clear
that the rays from the left image must reach the left eye, with no
more than a small residue visible to the right eye, and vice versa.
If an integrated stereogram displays stereo relief, this condition
prevails, and you can actually erect a screen into which holes are
cut 21/2 inches apart. If you apply your eyes to these holes, you
will see stereo. Now imagine such a screen, behind which a num-
ber of people are placed.

1234567
LRLRLRLRLRLRLR
oooooooooooooo

Here we have the 7 pairs of eyeholes with 2i/ inches between
every adjacent pair. If you are looking through pair 4, you will
see stereo, but if you move to the left and look through holes
which include the R of 3 and the L of 4, you will be looking
through a R hole with your left eye and through an L hole with
your right eyeand you will see pseudoscopic relief.

But suppose that all of the even numbered pairs are removed,
and you are looking through pair 3. You move to the right until
your left eye is looking through R of 3; then your right eye is
obscured by the blanked out 4!,. Seeing but one image you see
no stereo relief.

ANAGLYPHS 173

Now suppose the screen is removed. You will still have to as-
sume the same positions to see stereo relief, and if you move to any
extent you will either see pseudoscopic relief or neutral (planar)
relief. In no position will the relief be of the clear cut, unmistak-
able type which is common to the normal stereogram and which
is exhibited to a high degree by the Vectograph.

There is a definite field for the integrated stereogram. It has
been used for years for display in store windows, an excellent
adaptation of the method because the alternating stereo and pla-
nar aspects cause the image actually to move and to draw atten-
tion. If it can be published it will be of greater value. But for
stereo quality, no type of integrated stereogram has yet exhibited
a degree of sufficient quality to be acceptable to the stereographer,
let alone to win him over.

Just what is the situation?

In movies, you must seek the proper head position and you
must maintain that position rigidly. There is nothing you can do
which will so quickly produce an agonizing headache as holding
the head and eyes in one fixed position. If you lean over to whis-
per to your neighbor, you lose the stereo effect.

On the contrary, if the movies are polarized you wear goggles
which are to all intents and purposes identical with the Polaroid
sunglasses you habitually wear out of doors, even to the shape of
the frames. The only difference is an invisible one, the direction
of the axes of polarization. You sit normally, you move about,
you speak with your neighbors, and unless you tip your head to a
decided oblique angle, you continue to see the full stereo relief
upon the screen.

Then what does the integrated stereo have to offer? Inferior
relief and stringent limitations in exchange for the privilege of
taking off your sunglasses!

In still work it is much the same. Relief which satisfies only
those unacquainted with true stereo, changing relief when the
head is moved, necessity for remaining in one of several fixed
positions. You must make an event of looking at the picture. On
the contrary, you hang a Vectograph upon the wall in an illumi-
nated frame and you can sit back and enjoy it from any angle and
from any part of the room, just as you would any picture.

174 THREE-DIMENSIONAL PHOTOGRAPHY

There has been a great deal of time and ingenuity spent upon
the integrated stereograin, and it would appear that this has been
done in order to achieve something which has already been
achieved in a perfectly satisfactory manner.

Finally there is the possibility that the integrated stereogram
will be used for published reproduction. Again we wonder why.
Right now we can publish excellent stereograms, anyone can
learn to see them in less than an hour and the quality is infinitely
superior to the best the integrated stereogram can ever hope to
match, in fact better than the Vectograph can produce. And it is
all set, ready for use by anyone at any time.

And the process is proven. For three years the Third Dimension
has regularly carried stereograms, usually transposed, at times un-
transposed, but always with a normal or less-than-normal separa-
tion. Guild members regularly view these stereograms without a
viewer and apparently have found the method fully satisfactory.
So, we have a proven method by which published stereograms can
be seen by the reader without benefit of viewer. Is this free vision
or not? Certainly it is an acquired skill, but so is reading.

We have no axe to grind regarding integrated stereo. Scien-
tifically, it is extremely interesting; there are many uses for it
where it will be extremely valuable. There is no denying the great
amount of genius which has gone into its development, and cer-
tainly stereo is richer for having it. But as a matter of practicality,,
most purposes served by it, or which may in the future be served
by it, are served better by a process existing today and freely avail-
ableif only the public would abandon its erroneous, precon-
ceived ideas and recognize the truth about this stereo which we
have with us right now.

The trouble with stereo is not so much faults of the process,,
which are extremely few, but with the public which condemns
without understanding. Truly "it isn't the process, it's the
people."

See Chapter 13 for historical and other discussions of integrated
stereoorams.

CHAPTER 12
STEREO PROJECTION

THERE ARE SEVERAL METHODS of stereo projection, the oldest
of which is the bichromatic anaglyph.

It is well known that filters cause objects of their own color to
appear as "white," while complementary colors appear to be
"black." Thus for anaglyphic projection of the stereogram a lan-
tern is used similar to the conventional one, but instead of polar-
izing filters we substitute red on one side and green on the other.
Thus the pictures are projected upon the screen in two colors.
Incidentally it may be added that only monochrome slides are
usable as the effect depends upon the purity of the two colors.

If we project the red-filtered image, which is a black image
upon a red ground, and examine it through the red filter, the
background seems to become white rather than red. But if we
exajnine it through the green filter, the red background becomes
black and the black image is lost in it. The reverse is true of the
green image. So if we use a red filter over one eye and a green one
over the other we shall have the desired condition of presenting
one image to each eye.

The images may be toned (red on white and green on white)
but in this instance the image alone is erased leaving a dazzl ing-
white space which tends to degrade the brilliance of the remaining
image, so it is preferable to use black slides of the ordinary type
and to color the background to provide a ''blackout'* rather than
a "whiteout."

Note that this is just the reverse of the printed anaglyph which
uses two colored images with white backgrounds.

A serious objection to the bichromatic anaglyph is the fact that
chromatic retinal rivalry commonly known as "color bombard-
ment" adversely affects many people and gives rise to distressing
headaches. This caused the abandonment of the method when
tried for professional motion pictures many years ago.

Polarization Anaglyphs. In seeking relief from color bombard-
ment, the solution of polarized light was proposed late in the

175

176 THREE-DIMENSIONAL PHOTOGRAPHY

nineteenth century, and a patent was issued more than a half
century ago.

It is a well-known property of polarizers, that although both
may be transparent, when superimposed in a particular position,
that of "crossed axes," they become as opaque as if some very dark
film had been interposed between them. However, the polariza-
tion of light is an effect which can be obtained only under certain
conditions. For example, the beams of polarized light may be
projected without altering them, but if they fall upon a non-
metallic surface the polarization is lost and they become ordinary
light again.

Thus if the images are to be preserved as polarized images, it is
necessary to project them upon a screen which has a metallic sur-
face. The most convenient is the familiar aluminum screen once
so popular for home motion pictures.

The two images are superimposed upon the screen with their
homologous separation at infinity, about two inches. One image is
projected through, for example, a polarizing filter with horizontal
axis and the other through one with a vertical axis. If we now
look at the screen we will see the two confused images. If we look
at it through a polarizer we may see the same two confused images,
but if the polarizer is rotated we will see only one image. If rota-
tion is continued we again see two images and then we see the
other image, and so on. At the zero and the i8o-degree positions
we see the left image, let us say, and at the 90- and syo-degree
positions we see the right image, while intermediate positions per-
mit both to be seen.

Although thin films of polarizing quality had been known for
many years, the development of Polaroid made it possible for us
to obtain unlimited quantities of a uniform, thin polarizing ma-
terial.

The polarized films are used in making the projector filters, and
for the viewing goggles as well. All that is necessary is to have the
axes of the projector filters at right angles relative to each other.
True, it is convenient to have these positions definitely vertical
and horizontal or at crossed 45-degree positions, but this is not
essential. The viewing goggles must have their filters so disposed
that the axes exactly cross the axes of the picture images so that

STEREO PROJECTION 177

the unwanted image is blocked. For example, the left goggle lens
is opposed in orientation to the right projector filter, thus block-
ing the vision of the right image by the left eye.

There is nothing more to it than this. The whole system is so
simple that any amateur can purchase a sheet of Polaroid and in
a single evening make the projector filters and the necessary
goggles. The latter are easily made by removing the glasses from
cheap sun glasses and replacing them with the Polaroid. If the
projector filter pair is set before a light, the goggle filters are
quickly adjusted by turning them in the frames until the comple-
mentary projector filter looks black. A drop of glue serves to keep
the filter in position. Any projector may be used which will ac-
commodate both stereo units in the slide holder. Slides made on
35mm film with a stereo reflector are the most convenient for this
type of projection.

If the stereo film has been made with a stereo reflector there is
no transposition problem, but if the image has been made with
two lenses (or two cameras) then transposition is required. In
projection, transposition is effected by simply revolving the two
projector filters in their mounts through ninety degrees.

It is not necessary to project through the camera reflector to
superimpose the images. In fact this is not practical unless the re-
flector is adjustable. The normal half-angle of projection is small,
so if we can shift the position of each image by this amount, we
shall have them virtually superimposed.

To this end, a filter holder is altered so that it will take a polar-
izing filter made with its two vertical halves at crossed axes posi-
tion, and in front of this two thin prisms are mounted with their
bases adjacent. Prisms to provide a deviation sufficient to produce
the required superposition may be used without introducing great
distortion in the image.

Another plan is to have the filter carried in the film gate along
with the film. This will do for still slides, but in motion-picture
projectors the great heat would soon ruin the filter.

Twin lenses, which consist of two lenses with a chord of each
element ground away and mounted side by side have been used,
but they are costly to produce and do not provide any great advan-
tage over the easier methods.

178

THREE-DIMENSIONAL PHOTOGRAPHY

Non-Anaglyphic Projection. The anaglyphic system Is not at
all essential to stereo-projection viewing. It is possible to view a
stereogram projected upon an ordinary screen through the use
o a simple, empty box. Holes are cut into this box so that the field
of view of each eye is restricted.

Figure 12-1 illustrates the basic principle as applied to viewing
a small stereogram. The two eyes EE' view the two images SS'.
Each eye is restricted to its own image by means of the shield M

S'

Fig. 12-1. Simple mask for viewing stereograms.

BOX

Fig. 12-2. Mask box substituted for simple mask.

through which two apertures are cut. This shield must be accu-
rately matched to the screen distance and image size. For actual
use, it is preferable to make a box mask such as is shown in Fig.
12-2. However, inasmuch as the centers of the images SS' cannot
be separated by more than 2i/ inches the method is limited to
small prints or to miniature projection.

To overcome this limitation, the stereo pair is projected in the
untransposed or pseudoscopic relationship illustrated in Fig. 12-3.

STEREO PROJECTION

179

Here a single aperture in the mask serves both eyes, and the con-
vergence of the axes directs each eye to its wanted image. Because
of the distance of the screen, usually several times its extreme
width, the actual convergence is far less than shown in the dia-
gram. This is a perfectly practical system, and can be successfully

S 1

Fig. 12-3. Simple mask for crossed axes screen viewing.

Fig. 12-4. Box mask for crossed axes viewing; b shows the shape of

the box when two front apertures are substituted for a single one,

making the box much shorter.

demonstrated by using a simple card. It has been used for years
for the purpose of viewing stereograms directly.

Figure 12-4 illustrates the application of the preceding figure to
a box viewer. The box may be made of cardboard, light wood,
plastic, or the like. It has the two eye apertures cut at the correct
interpupillary distance, while simple opaque covers serve to alter
the limits of the single aperture in the front so that the device may
be adjusted to various screen distances and sizes. In practice it will
be found that this box is not as successful as the simple screen of

180 THREE-DIMENSIONAL PHOTOpRAPHY

Fig. 12-3. The reason lies in the fact that for the correct conver-
gence, the length of the box would be excessive. Therefore this de-
sign is modified by moving the front wall backward, and sub-
stituting two apertures for the single one, as shown at b in Fig.
12-4. This design is closely similar to that of Fig. 12-2, the differ-
ence being that in Fig. 12-2 the outer apertures are centered with
the ocular apertures, namely at normal interpupillary separation.
In Fig. 12-4 the outer apertures are separated by less than the in-
terpupillary. In this model also it is advisable to provide sliding
shutters for the adjustment of the limits of the apertures. All of
these require an abnormal relationship of convergence to accom-
modation, and therefore cannot be used without considerable
practice.

The next step in the design introduces the deviating prism
shown in Fig. 12-5. Eye E observes image S normally. The dotted
lines from E f to S show the normal convergence upon image S>
but if the prism P is introduced, it deflects the visual beam and
the eye normally positioned to observe image S actually observes

Fig. 12-5. Convergent vision with single prism.

image S', thus providing optical superposition of the two images.
However, this is unsatisfactory in that the prism must have con-
siderable power (for example at 3 meters projection distance and
a 20-inch image, we have 16.6 diopters deviation) which tends to
introduce distortion. Instead, the prism is split between the two
fields as shown in Fig. 12-6. This seems to be a simple Brewster
stereoscope, but it is far different. The Brewster stereoscope has
ocular lenses and the fusion is a function of specific refraction, but
here we have plain prisms and the fusion is a function purely of
deviation. The dotted lines show the apparent perception of a

STEREO PROJECTION

181

bed

Fig. 12-6. Convergent vision with two prisms.

single central image. Viewers of this type are highly satisfactory
once the correct screen distance and image size have been de-
termined.

Figure 12-6 b, c and d shows the use of a rotating variable prism
to make the system illustrated more flexible. At b the two rotating
prisms are opposed and there is no deviation. At c the compensat-
ing prism is neutral and the lateral deviation is that of the single
prism. At d the prisms are in the position of maximum deviation.
By simply rotating the outer prism, and by having the outer
prisms for both eyes mechanically coupled to rotate in opposed
synchronism, a simple movement serves to fuse the images at a
great variety of distances and separations.

Figure 12-7 illustrates the Swan prism viewer which has been
widely used for X-ray viewing of side-by-side radiographic stereo-

Fig. 12-7. Screen viewer with image-inverting swan prisms.

grams. This prism is also known as an erector prism because it will
erect a normally inverted image. This fact in itself should serve
as a warning to the stereoscopist. When this viewer is used for
viewing a real scene (not a stereogram), it introduces a pseudo-

182

THREE-DIMENSIONAL PHOTOGRAPHY

scopic effect in which the farthest visible object appears to be
nearest the eye. Therefore, when this viewer is used, it is necessary
to project the images in un transposed order.

Figures 12-8 and 12-9 illustrate one of the most elaborate of
commercial viewers. The prisms i, 2, 3 and 4 are enclosed within
a body which resembles a small pair of prism binoculars. The in-

Fig. 12-8. Double prism viewer with prisms PI and P2 adjustable.

Fig. 12-9. Two mirrors for superimposing two stereo images upon a

screen.

ner prisms Pj and PJ are fixed, while Pi and P2 are mounted to
swing upon a vertical axis. They are connected by an external,
jointed bar which in turn is screw-actuated. Turning the screw
rocks the prisms in opposite directions and thus brings the two
images rapidly into the desired superposition.

There are many types of special viewers, but other than those
which have been described, there remains only one more signifi-
cant type, the shutter type. This shutter, as applied to still projec-
tion, consists of an alternating shutter in the projector synchro-
nized with a shutter supported before the eyes. The left eye is ob-
scured when the left image is blocked from the screen, and vice
versa. Thus the essential eye-image relationship is preserved, and
if the alternation is sufficiently rapid there is no annoying flicker.

STEREO PROJECTION 183

Although not a stereo projector, there is one novel projector so
closely associated with stereo and so often used in connection with
it, that it should be mentioned here. In Europe it is common
practice to provide single projector units for attaching to the bet-
ter cabinet-type viewers, to project one of the stereo halves.
Sawyer's Inc. has a similar projector which projects two-dimen-
sional pictures from the stereo reels of the Sawyer Viewmaster.
The usual Viewmaster reel has the title of each scene printed
in such a way that it is visible while the pictures are in the
viewing apertures. The projector has a prismatic system which
makes these titles visible at one side. Another feature is the micro-
controlled mechanical pointer which makes it easy to point out
individual features of a scene, even though these are made upon
i6mm film. This is one of the smallest practical projectors made,
and even though the present model is limited to two-dimensional
projection, owners of a Viewmaster library will find it a welcome
addition to their viewing equipment.

However, let us now turn our attention to stereo projection as
it is being practiced by thousands of amateurs throughout the
country.

This modern stereo projection is thoroughly practical and is
accomplished by the use of well designed, commercially made
projectors.

For individual viewing of the stereogram, most stereographers
prefer the hand viewer, but this is obviously inconvenient for
group use and when the group is large, a class or the audience at
a lecture, the use of hand viewers is impractically slow. It is for-
tunate that stereo projection is as simple and as satisfactory as any
type of projection. Even though it does fall short of orthostereo
results, the loss is probably more theoretical than practical. In
short, stereoscopic projection has been proven to be thoroughly
satisfactory.

Those who have seen modern stereo projection, now predict
that stereo movies will soon be developed; they do not know that
stereo movies were presented in a Broadway theatre a quarter
century ago, and in many other theatres throughout the land.
They do not know that polarized light stereo movies were featured
at both the Chicago (1933) and New York (1939) World's Fairs.

184

THREE-DIMENSIONAL PHOTOGRAPHY

There is little to be done in that field, it has all been done time
after time and any amateur can, with a minimum of ingenuity
make his own stereo attachments which will enable him to make
and project perfect stereo movies. This will be discussed in a
later chapter.

Not even the modern polar-stereo projection of still slides is as-
new as most of us think. Both Leica and SVE made stereo pro-
jectors upon this principle before World War II ? and shortly
after that the Depthro was made. However at the time this is
written the Stereo- Vivid projector is being widely sold and is giv-
ing satisfaction. The makers of the Realist camera have announced
a looo-watt projector which is scheduled to make a very early
appearance, and others are being planned.

Fig. 12-10, The Stereo Vivid projector takes the

standard 35mm stereogram, measuiing 1-5/8x4

inches.

The Stereo Vivid is a well made instrument, the body of sturdy
castings instead of the usual sheet metal. The lens housing moves

STEREO PROJECTION

185

backward and forward bodily for focusing while within the hous-
ing the separation and vertical relationship of the two lenses may
be adjusted by conveniently placed, large knobs. The lamps are
dual 5OO-watt, cooled by a fan set in the base of the lamp house.
The lenses may be had in different focal lengths, and of excellent
quality, coated type. In short the whole instrument is well de-
signed, well made and wholly satisfactory in operation.

We are sorry we cannot give all details of the Realist projector
as it is not yet available, but with the quality of the Realist camera
in mind, there can be little question about its quality. The dou-
bling of light output will be of value to those who appear before

Fig. 12-11. Realist projector, companion to the

Realist camera, projects the standard 35 nun

stereograms.

186 THREE-DIMENSIONAL PHOTOGRAPHY

large audiences, but we have found the soo-watt output quite ade-
quate for a 50x50 inch screen, as large as can be viewed with com-
fort in the ordinary living room.

We have received advance information concerning some of the
details of the projector. The general appearance can be seen from
the illustration, which also shows the top-insertion revolving slide
holder which the manufacturer tells us adds considerably to pre-
cision slide alignment.

The cooling consists of a low turbulence (quiet) blower of two
speeds, the high speed for use with 750- and looo-watt lamps for
large screen professional projection. The projector, as a matter
of economy will be available with f/2.8 lenses instead of the stand-
ard f/2.3, and with a single speed blower for home use. This
model uses only 5oo-watt lamps, but, because of a newly designed
optical system, we are told, even this home model will fill a six-foot
screen with considerably more than adequate brilliance. The pro-
fessional model with looo-watt lamps will fill a 1 2-foot screen
with full brilliance.

The "standard" home stereo screen is 50x50 or about 17 square
feet. The six-foot screen is 36 square feet in area or more than
twice as large, so the 5o-inch screen should be very brilliant. The
12-foot screen, of course, is four times as large as the 6-foot size
and about 12 times the size of the 5o-inch.

We are told that with this projector and permamounts, an un-
limited projection exhibition can be screened without any neces-
sity for projector adjustment, including focusing. The construc-
tion is cast metal of modern design as shown in the illustration.

PROJECTION PROBLEMS. Stereo projection itself is entirely sat-
isfactory, but those who use the projectors often contribute to the
dismal failure of the exhibition.

We certainly do not wish to appear as an advocate of careless-
ness or slip-shod technique, but the fact remains that the gravest
trouble encountered by projectionists in the stereo field is a direct
result of taking too much care.

Shortly after the stereo projector came into general use there
were repeated reports of people becoming violently ill at exhi-
bitions. It is true that spectators were made ill; it is easy to make
almost anyone ill by proper manipulation of the projector. But

STEREO PROJECTION 187

people are sea-sick and air-sick and car-sick every day and it does
not result in any permanent defect or serious condition; and be-
ing stereo-sick is very closely allied to these forms of nausea.

The explanation is simple. You are accustomed to a motionless
horizon, and when you find the horizon tipping insanely this way
and that, the eyes try to retain it as a fixed reference line with the
result that extreme dizziness followed by nervous upset of the
stomach follows. In stereo, the eyes see only one picture and that
in full relief. When the eyes are forced to constantly alter their
relationship while keeping the picture unaltered, the same thing-
happens, it is the exact reverse of sea-sickness. But as there is no
visible phenomenon to account for the sickness, the victim assumes
that he has been struck with some very serious visual affection,
and the fright simply increases the discomfort.

I have spent long periods with the projector, deliberately pro-
ducing these conditions. All I felt was a slight eyestrain and a
vague physical discomfort . . . but then I'm a fairly good sailor
so perhaps I should not use myself as an example. You will find
however that as a rule those who are most prone to sea-sickness
will be most easily affected by stereo.

This all comes as a result of the projectionist readjusting the
projector for the mounting discrepancies of each slide. He care-
fully spaces them laterally and carefully aligns them vertically
and by then it is usually time for the next slide. As a result the
spectators' eyes are continually on the jump without them being
aware of it.

This practice in turn results from the stress which has been
laid upon the necessity for super accuracy in projection slides.
Practically all of the stereo world has joined in this. I contributed
my own small share to the paean of praise for accuracy and
worked out systems for "projection mounting." Then through
accident I discovered the solution.

I was unexpectedly called upon to give a stereo exhibition be-
fore a group who had never seen stereo projection, and the slides
to be used were a series which would be familiar to this group.
These slides had been mounted as usual with no more than or-
dinary attention to "precision mounting." With great misgivings
I gave the exhibition, and then asked if anyone had experienced

188 THREE-DIMENSIONAL PHOTOGRAPHY

any perceptible discomfort. Everyone replied in the negative and
asked for more slides. Another series was run, a total of more
than 100 slides. Still no complaint, a fact I could not believe, be-
cause I had adhered to the first law of projection "Do not alter
any projector adjustment while the image is on the screen"; also
the second, which is, "Have no spectator nearer a 50-inch screen
than fifteen feet." The reason for this rule is that the discrepancy
of images upon the screen remains constant and the nearer the
spectator the greater the visual angle involved in such discrep-
ancy. It does assume, however, a full screen picture.

Later, the slides used were checked for alignment. Lateral
separation varied ten inches and the vertical alignment was out a
full three inches in one slide, and more than two inches in a few!
According to theory, such slides should have been very painful to
view, but not one spectator admitted the slightest degree of dis-
comfort after viewing more than 100 slides; and most of the
spectators wefe seeing their first stereo projection! At 15 feet a
three-inch difference means almost two prism diopters, when
i/> A is supposed to be the limit of vertical difference.

Precision mounting is certainly desirable, as is precision pro-
jection, but above all the important factors are those just stated
about never touching an adjustment with the picture on the
screen and not having spectators too close to the screen. It is
desirable not to have any spectator nearer the screen than the
projector itself, when using five-inch lenses.

So much for generalities. We can now discuss details.

Projection follows the general stereo law that the right eye
shall see its image, while in the corresponding field the left eye
will see the left image. However, the individual images are so
large they cannot be placed side by side in the usual manner.
(This would necessitate a wide divergence of the eyes, a feat
impossible to most people when more than a degree or so is in-
volved. It is true that side by side projection is possible by re-
transposition and using convergent vision or "crossed eyes/' but
this too is often painful and has only academic interest).

The first problem then is that of projecting two images in
approximate superposition in some way which will permit one
image to be extinguished without seriously affecting the other;

STEREO PROJECTION 189

and that this extinction can exist simultaneously in two opposite
phases. That is now done by the use of polarized light. If the
two pictures are projected by polarized light, with the two axes
of polarization at ninety degrees, and if the screen has a non-
depolarizing metallic surface such as aluminum, the first half of
the requirement is met. Viewed with the naked eyes the screen
will show the two images in confused superposition.

If the left eye is now provided with a polarizer (analyzer)
which will black out the right image, and the right eye with a
filter to black out the left image, each eye will see its own image,
and the slight discrepancies of position will be compensated by
slight movements of the eyeball. Because of this the left image
should never be more than two inches to the left of the right
image, otherwise the eyes would have to diverge, but the left
image can move to the right several inches (involving converg-
ence) without discomfort. The ideal position is that in which
images of objects at infinity shall have the left image two inches
to the left of the corresponding right image. Thus infinity vision
will be almost parallel, and all nearer objects will require greater
convergence, just as in direct vision.

If the polarizing axes are reversed, or if the slide used is not
transposed, relief will be reversed. To try this, project a normal
slide, then remove the viewing goggles and reverse them with the
ear bows pointing away from the face. In fact many slides make
very interesting subjects viewed in this way.

VIEWING EXPERIENCE. When someone unfamiliar with stereo
first looks into a hand viewer,* he will often either be unable to
focus the image or he will see two images. This exasperates the
stereographer who usually thinks it is pure stubbornness on the
part of the spectator. It is the result of a common condition, one
almost universal.

From our earliest visual experience, we are accustomed to focus
upon an ; object at some distance and to converge upon that ob-
ject at the same time. No matter what the distance there is a
degree of accommodation (focus) and a degree of convergence
which are fixed for that distance. You never have occasion to use
the one without the other. No matter what two muscles are thus
coordinated, if they are never used individually, habit soon makes

190 THREE-DIMENSIONAL PHOTOGRAPHY

the dual motion a single one. After perhaps 20 or 30 years of
unbroken duality of accommodation-convergence you are called
upon to keep accommodation relaxed (infinity) but to exercise
convergence normal for close objects. You simply cannot do it.
You either lose zero accommodation and focus with convergence,
thus blurring the single picture, or you find convergence follow-
ing accommodation to zero and you see two pictures, or at least
you do with pictures of close objects.

It is remarkable that most people can break this bond suffi-
ciently to view stereograms, within a few minutes, and a little
practise thereafter makes the two motions wholly independent.
Nor does this developed skill interfere with the normal coopera-
tion when it is required.

However in the projected stereogram, accommodation is not
for infinity, but for some 15 to 20 feet, and unless there is some
object nominally nearer than ten feet, the discrepancy is less ob-
vious so that many who are unfamiliar with stereo find projection
easier than the use of the hand viewer.

In any event, the more experience you have with stereo view-
ing the easier it becomes. This is true of both hand viewing and
projection. And the more you see of stereograms the more sensi-
tive you become to stereo relief. It is known to be a fact that
viewing stereograms improves your perception of depth in direct
vision.

SLIDE MOUNTING. The stereogram is supposed to have both
images parallel in every direction, laterally, vertically and all in-
termediates. This of course is nothing more than is assumed to be
true of any stereogram whether for hand viewing or projection.
In practise however it is often found that the two images are
slightly out of parallelism. One image may be slightly higher
than the other; the vertical axes of the two may lie at a slight
angle rather than being parallel, or in the case of an extreme
closeup, the two homologous images may be too close together.

There is another factor which may be present. The two images
may be perfectly parallel, but the horizontal axis may not be
parallel with the edge of the mount, which would, in projection
cause one image to be higher on the screen than the other.

To insure correct mounting a gauge is used. This is a trans-

STEREO PROJECTION 191

parent base upon which are inscribed vertical lines crossed by
horizontal ones. Although usually the gauge has four vertical
lines, only three are used for slide gauging. The fourth is used
for checking the homologous spread of closeup objects before the
film is cut.

An image of some object at infinity is placed over the guide
line 2. The homologous image in the second picture should lie
upon line 4. Some image of a closeup object is then placed over
line 2, and its homologue should lie between lines 3 and 4. If it
lies at the left of 3, that is between 2 and 3, the convergent dis-
placement is too great for comfortable viewing by inexperienced
people. The gauge is based upon a maximum convergence of
two prism diopters.

The alignment of objects along line 2 is noted and then the
right picture is examined. The same objects should be aligned
similarly. Suppose along line 2 there is a church steeple at in-
finity, and a telephone pole which is relatively near. In the right
frame the steeple should line along line 2, but the pole will lie
between lines 3 and 4, because of stereo parallax. Examine the
left picture carefully. Suppose the pole leans slightly to the left
at the top. Note carefully how much it leans away from line 2.
Now slide the stereogram so that the right pole is adjacent to
line 3 or 4, and see that the amount of leaning is the same as in
the left picture.

In judging verticals you cannot align several objects in the left
picture and then align the same objects in the right because their
relative positions are different. Therefore you must align by
reference to objects at infinity or by some part of a single object.

Aligning horizontals is much easier. You simply look to see if
any horizontal gauge line passes through identical images in both
pictures.

When you have carefully aligned a slide, project it and without
viewing goggles examine the image. Homologous objects should
lie at the same height upon the screen, and they should also be
separated horizontally not more than about six inches for closeup
objects. This of course assumes the projector to be in adjustment
before the test is made.

Many stereo projectionists make use of a set of control slides

192 THREE-DIMENSIONAL PHOTOGRAPHY

for preliminary adjustment of the projector. But it must be re-
membered that the controls themselves should be checked from
time to time to make sure the film has not become misplaced in
the mount. However as both images are printed upon the same
base, the only error which will occur is that the image base and
the mount base may not remain in parallelism.

Slide i is for horizontal spacing. The distance from the guide
line to the farthest indicator is 65mm and represents parallel axes
as for infinity. The indicators are spaced imm, which is equiva-
lent to a two-inch displacement on the 50x50 inch screen. Thus if
the index line is superimposed upon the second line, infinity im-
ages will have a screen spacing of about two inches.

These slides are designed to produce comfortable viewing of
a 5O-inch screen by inexperienced spectators at distances of 10
feet (or more) from the screen.

Slide 2 is for vertical alignment. The spacing is o.gmm which
will give a screen displacement of about 2/3 inch. This is very
near normal tolerance and even \vhen no visual discomfort is
noticed, the maximum tolerance should be kept to two divisions.

Adjust projector with these slides, then run through your col-
lection and check your pictorial slides against the standards.

Slide 3 is a combination slide for check testing. Take it with
you with your regular slides and just before the real exhibition,
check the positioning and adjustment of the projector with it. It
should not be substituted for i and 2 for the critical adjustment
of the projector.

Slide 4 is a personal test slide and projection demonstrator.
Note that no matter what your position in the room the line of
circles points directly at your chest, and apparently moves as you
move to preserve this alignment. If you move away from the
screen the line lengthens, if you approach the screen, the line
draws back. This extension and compression of space shows the
inevitable distortion of projection under ordinary conditions.

Anyone should be able to fuse the first four circles, one after
the other, from the projector position; a little practice will extend
this and experienced stereo observers should have no trouble
fusing all seven.

STEREO PROJECTION 193

Once the projector is aligned with the control slides, it is ready
for projection of aligned slides with no further adjustment. It
must be emphasized however that the remarks concerning mount-
ing which have just been made are all supplemental to the in-
structions given under the main discussion of mounting.

DEVIATION. In any discussion of the amount of error permis-
sible upon the screen, there is one factor which must be con-
sidered. This is the distance of the spectator from the screen. The
control slides have been based upon a spectator distance of 10
feet minimum. Of course at greater distances the effective error
grows less. The reason for this is that the error of significance is
determined by the arc through which the eyeball must rotate to
compensate for it.

For such purposes the unit of deviation is the prism diopter
(see Appendix), which has a value equal to the acute angle of a
right triangle whose short leg is 1/100 the length of the long one.
For our purpose we assume the interpupillary distance to be
65mm. Half of this distance is 32.5mm. One hundred times 32.5
is 3250, so at a distance of 314 meters (about 140 inches or just
under 12 feet), each eye deviates one prism diopter. Because both
eyes converge in opposite directions the total convergence is two
prism diopters.

However if the object viewed lies at 24 feet, the deviation is
only a half diopter per eye and at six feet it becomes two diopters
for each eye, in round numbers. For this reason it can be seen
that to cause the eyes to converge upon a point, the nearer the
point the greater the motion of the eyeball. In the projected
image, we have two separate points, so the eyes do not converge
to the amount normal for the screen distance, but their diver-
gence is controlled by the separation of the homologous points in
the image, and this is variable because we can move the images
upon the screen in relation to each other. Moreover, the differ-
ences in the homologous separations of images of various objects
in the same slide will of course present variations, so that the
spectator while looking at a screen whose distance is fixed, will
have to vary his convergence just as he does in looking at real
objects.

It is very easy to have such variation in a series of slides that

194 THREE-DIMENSIONAL PHOTOGRAPHY

the spectators' eyes have to range from parallelism to sharp con-
vergence, and as this is dissociated from accommodation, it is
tiring. Even worse is the fact that the eyes normally do not di-
verge vertically. The eyes see the same object, and while lateral
convergence is constant, there is no reason for vertical deviation.
If one stereo image is higher than another, there must be vertical
deviation to bring them into visual alignment, and being an un-
accustomed and unnatural movement, it is painful. Hence it is
necessary to keep the vertical error much smaller than the lateral.
And, as we have stated, it is essential that the permissible error be
computed for some minimum screen-to-spectator distance and
that no one be allowed to view the screen from any nearer dis-
tance.

Exceptions. It is assumed that stereo fusion will not take place
unless the eyes are so positioned that the two retinal images oc-
cupy exactly similar positions with regard to the macula* How-
ever, there are people who seem not to find excessive vertical
displacement uncomfortable and who can fuse stereograms badly
out of alignment. It has been suggested that these people have
the ability to fuse images which do not occupy similar retinal
positions and hence can fuse stereo images without the necessity
for compensating small errors of deviation by actual rotation of
the eyeball. This is another point about which we do not have
sufficient information to make a definite statement, but it is of
great potential interest.

i

Specific Dimensions. Errors of greater degree seem to be ac-
cepted in the stereoscope than in projection. No explanation is
offered. However in practice we have found that projection of
50-inch screen size is almost always successful when the minimum
spectator distance is 20 feet, that it is usually successful with the
minimum of 15 and at 10 feet we have found 5O-inch screen pro-
jection definitely not satisfactory. Experiments made with slides
compensated to show similar degrees of error at the three dis-
tances, did not yield comparable results. The same differential re-
mained. Therefore there must be a factor which involves the total
field angle as well as the angles of error. Reducing the screen size
made the 10 foot projection fully satisfactory when errors were
kept within specified limits.

STEREO PROJECTION 195

The actual tolerance, of course, depends upon the spectators
who view the pictures. We have found that 2.5 prism diopters
laterally occasionally gave trouble to individuals who spend most
of their time indoors, while those who spend a substantial part of
their time in the open were not disturbed by lateral deviations
as great as five prism diopters. As far as vertical deviation goes,
some individuals can tolerate three prism diopters, but no slide
should be projected which will call for a greater vertical deviation
than a half prism diopter.

Most of the Guild slides are based upon one-half and three
prism diopters, but many of you will perhaps find the 2.5 lateral
maximum better if your friends are largely indoor workers.

These limits are not those of actual tolerance, but those which
provide comfortable viewing for continuous periods of one to
several hours duration.

The following tables give the deviations of prism diopters as
measured upon the surface of a screen for spectator distances of
10 and 20 feet.

Deviation 10 feet so feet

1/2 prism diopter 0.6" 1.2"

i prism diopter " 1.2" 24"

2.5 prism diopter 3.0" 6.0"

3 prism diopter 3.6" 7.2"

Remember, it is not the projector-to-screen distance, it is not
the size of the screen, it is not the focal length of the projector
lenses which is important. It is the displacement of the two im-
ages in inches upon the screen surface,, and the ratio of those
distances to the distance from the SPECTATOR to the screen.

Close-up Compensation. We have seen that the use of the hand
viewer necessitates a separation of convergence and accommoda-
tion. The accommodation is that for the slide itself and if the
viewer is correctly adjusted this will be infinity. On the contrary
the convergence will be that normal to the original distance of
the object as reproduced by parallax, and normal to the parallax
itself.

In projection, we have the accommodation normal to the screen
distance. With the usual five-inch projector lenses and a 5o-inch
screen, this distance is about 22 feet. The convergence is not di-
rectly related to the distance of the object, nor is it related to

1% THREE-DIMENSIONAL PHOTOGRAPHY

parallax. In fact the convergence is variable and under the control
of the projectionist. This is one point of difference between the
stereoscope and the projection screen.

A set of experimental slides was made and used as a basis for
testing. One was a landscape showing clouds which made excellent
indicia for the infinity setting. This slide was used to set the pro-
jector so that there was a si/^-inch separation on the screen with
the left image at the left and vice versa. This arrangement pro-
vides, for a spectator with 2 14 -inch interpupillary, strictly parallel
vision for infinity.

The second slide w r as that of a child, full figure. There was a
separation of ii/ 4 inches upon the screen, but showing positive
convergence. In short this image was displaced 31^ inches from the
infinity position. This positive convergence was that demanded
for a point some 13 feet from a spectator at the projector position.

The third slide was the same child, but with the camera at such
distance that the frame line cut the figure at about the waist. This
slide gave a 3 14 -inch positive convergence on the screen or made
it necessary for the spectator to converge upon a point about nine
feet before him.

These computed convergences were checked by holding a white
rod between the spectator and screen and moving it until neither
rod nor screen image showed diplopia. The experimental and
computed distances were remarkably close.

The fourth slide was a close-up of a hibiscus made at a camera
distance of three feet. There was a 5 1/4 -inch positive displacement
which corresponds to convergence at about 61/9 feet in front of
the spectator. This was a noticeable departure from normal. View-
ing is easy enough for one experienced in stereo, but difficult for
a layman. For instance, we have:

Parallax distance 36 inches

Convergence 78 inches

Accommodation 264 inches

Of course in normal vision, all three would be for the same
distance. This brings up the problem of altering these close-up
slides so that the discrepancies are lessened.

The fifth slide was an identical shot of the hibiscus, but one in
which the mask had been cut in two and moved apart to give a

STEREO PROJECTION 197

minimum homologous separation of 64.5mm. The screen showed
a 2-inch negative separation, or one-fourth inch less than infinity
which involves a convergence normal to a distance of almost
standard stereo infinity. Still, this was much more "comfortable"
than the first close-up.

A sixth slide was made, just like the fourth and fifth, but this
time the compensation was such that the separation was Ggmm.
When this slide was projected, the point of actual convergence
was about 13 feet from the spectator and nine feet from the screen.
This was an extremely comfortable slide to view and analysis
shows:

Parallax distance 36 inches

Convergence 156 inches

Accommodation 264 inches

If all slides are so compensated that there is no convergence
point nearer than half the screen distance, slides may be projected
without projector adjustment for separation. Presetting the pro-
jector for substantially parallel vision at infinity gives greater lee-
way in slide compensation.

It may be added that there is always a wide variation in the
subjective interpretation of distance. With the hibiscus slide just
described, one spectator saw the flower about a foot in front of
the screen, another saw it at about four feet from himself (approxi-
mately parallax distance), and others saw it at other distances with
no close similarity among the group with a single exception. Two
spectators placed the flower at about 10 feet. Distances were de-
termined by coincidence with a marker target which was moved
along until one or another spectator would say it was in the posi-
tion of the flower. (Note that this is NOT the distance at which
there is only a single image seen of both flower and target. It is
purely subjective spatial orientation.) The distances from spec-
tator to target were then measured.

In working with either stereoscope or projector, one convention
of stereo must be mentioned because it has led to confusion more
than once.

It is customary to speak of the 'left" image and the "right"
image. There is nothing in an image which makes it inherently
right or left. Either film may be used as either the right or the left.
Of course there is that difference common to planar pictures as

198 THREE-DIMENSIONAL PHOTOGRAPHY

well as stereo, that when reading matter or other positive indica-
tion of right or left is included in the picture, the image should
be oriented to correspond. However, this is a matter of conveni-
ence and has nothing to do with stereo per se.

If you place a stereogram in the viewer, you see it normally. If
you remove it, turn it end-for-end and replace it, you have the film
which was originally at the right before the left eye and vice versa.
The stereo relief is unaltered. Therefore either image may be left
or right without disturbing the stereo relief. What then is the
distinction?

It is the relative positions of the two. The stereo position is that
which causes the eyes to converge when looking at a nearer object;
while in the pseudo position, looking at nearer objects necessitates
a divergence of the eyes. This must not, however, be interpreted
as suggesting that the convergence of itself is significant; the di-
rection of convergence is simply an index to the lateral distribu-
tion of parallax.

Therefore remember that when we speak of a right image or of
a left one, we are simply using a convenient, and brief, convention
which must not be translated too literally.

TOLERANCE. -It must be remembered that the absolute align-
ment of any slide is impossible. Man cannot achieve the absolute.
The goal is an error which is so minute that it ceases to exist as
far as visual perception is concerned. In fact, even that limit may
be passed because we do have a certain tolerance of visual error.
Because this tolerance varies among individuals and is greatest
among those who are familiar with stereo viewing, it is advisable
not to depend upon it, but to mount slides with all precision
possible. You will find some spectators, new to stereo, who may
object to a greater horizontal separation than six inches and a
vertical displacement of more than one inch as measured upon a
5o-inch screen. Thus a displacement of one inch is a ratio of 1:50,
which means roughly a half millimeter in the original slide which
has a height of about 24mm. (Twenty foot screen distance.)

As we have stated, we have found "new" spectators who accept
ten inches laterally and three inches vertically, but to use slides
of such error is a grave mistake. Just because it can be done in
some instances, do not use it as your standard.

STEREO PROJECTION 199

Before making an exhibition, project all slides with the aligned
projector. Do not make any adjustments. If any slide shows more
than six inches laterally (except for extreme close-up) or more
than H/9 inches vertically, remount it. Play safe!

MOUNTS. There is some grave question about the danger to
slides from projector heat. Even with a well-cooled projector, a
slide which has been projected for a half minute or more will be
uncomfortably hot to the hand.

Some stereographers believe that the glass mounted slide which
retains the heat for a long time, is bad. They use rigid open
mounts. Others believe that the glass absorbs the heat and keeps
it from the film. For what it is worth it may be said that in the
Guild laboratories, the open, rigid, cardboard mount is used, and
projection of any one slide strictly limited to a fifteen second
maximum.

RETRANSPOSITION. Remember, the dual lens projector retrans-
poses the images, so be sure to check your projection for right-left
position on the screen. Project a slide, close the right eye. Cover
one lens of the projector, then the other to determine the image
which is "left". This image should never be so placed that any
of the infinity images are more than two inches to the left of
the other image.

Many projectionists simply give infinity images coincidence.
This works very well when no close distances are present, but if
there is an extremely near object, this makes necessary a certain
divergence which is not advisable.

But above all, during the projection, readjust focus as little as
possible and do not touch the aligning adjustments. Should this
ever become imperative, ask the spectators to close their eyes
until the adjustment has been made.

PROJECTION SUMMARY-

1. Gage slides for parallelism and lateral continuity.

2. Mount films with the homologous images of the farthest
objects spaced 65mm or the normal infinity setting. This applies
even if the farthest object was really only five feet from the camera.

200 THREE-DIMENSIONAL PHOTOGRAPHY

3. Gage films to see that the range of separation does not exceed
i.5mm at most or preferably imm (3 to 4.5 prism diopters).

4. Gage films to see that the vertical error is less than 0.2 5mm.

5. Preset projector for focus separation and vertical alignment
before the spectators arrive.

6. Be sure projector is level and the two apertures aligned
vertically.

7. Do not allow any spectators to approach nearer the screen
than the distance for which you have compensated, preferably not
nearer than 15 to 18 feet from a 5o-inch screen.

8. Slide thickness may necessitate adjustment of focus during
the projection, but do NOT touch either of the stereo alignment
adjustments while anyone is looking at the screen. (Careful slide,
mounting and a preview for checking before the public perform-
ance will make such adjustment unnecessary.)

If you will do these things you will not have any difficulty with
projection and you will be assured that the spectators will thor-
oughly enjoy the projection with no uncomfortable results.

CHAPTER IS
APPLIED STEREOSCOPY

STEREOSCOPIC reproduction is not at all limited to the amateur
photographer. It finds a wide application in the diagnosis
and treatment of diseases and maladjustments of the visual ap-
paratus. It is used to excellent advantage in military work, par-
ticularly in aerial photography. In advertising it has already
gained high favor. Students in schools obtain visual impressions
of places and objects which could not be obtained in any other
way. Scientists have found it to be the only photographic method
which meets their demands fully. In the medical profession it
is widely used for teaching purposes, and hardly an X-ray installa-
tion is without its Wheatstone viewer for stereo X-rays. Stereo-
scopic photography is now being used to record installations in
factories, keep check upon field work by engineers, record results
of scientific work in the field, and of course in the more elaborate
stereogrammetry it is used for actual surveys.

New applications are appearing almost daily, not the least in-
teresting being its use in recording evidence for both civil and
criminal trials.

THREE-DIMENSIONAL VOGUE. It is now the fashion to use the
term "three-dimensional" in both apt and inapt connections. No
phrase used in the discussion of pictures has been more over-
worked or more abused. Very serious efforts have been made to
convince the public that three-dimensional pictures are something
quite new, and totally different from the old stereogram. They
either ignore or do not know the fact that any truly three-dimen-
sional picture must be stereoscopic, for the latter term is specifi-
cally descriptive of "depth vision." It might be remarked in pass-
ing that the phrase "three-dimensional" is often applied to systems
which are unsound, and even to some whose only pretense to
relief consists in the name itself!

It must be added, however, that there are three-dimensional
pictures which are certainly not ortho stereoscopic.

Although "three-dimensional" is less than stereo, it is more
widely known and it is recognized by those to whom the word

201

202 THREE-DIMENSIONAL PHOTOGRAPHY

''stereo" means nothing. At the same time the almost universal
practise of advertising slide viewers, projectors, lenses and the
like on the basis of "three-dimensional effect" (when they are
not truly stereoscopic) is to be deplored.

No other pictorial art has ever been subjected to the false pub-
licity, the inoperative systems and inventions and the intentionally
false representations which have dogged stereoscopy throughout
its life of more than a century. Hardly a year passes without the
announcement of the discovery of some revolutionary three-di-
mensional process, only to prove it has nothing < three-dimen-
sional" in it other than the name. This has been one of the most
popular of pastimes in the motion-picture field.

SINGLE IMAGE RELIEF. It is well known that when a single
picture is viewed in a concave mirror or through a convex lens it
does take on an appearance of "unflatness," but so vague is the
relief that it cannot by any stretch of the imagination be called
stereoscopic, yet that is just the word used to sell these inoperative
devices. Even some of our reputable magazines permit the use of
the term for selling concave mirrors, probably because of editorial
ignorance of true stereo.

A stereoscopic effect is one in which the true, natural quanti-
tative relationships of all objects and planes within the field of
view are normally reproduced. Nothing less than this can truth-
fully be called stereoscopic. As we have seen that this effect can
be obtained only when the two eyes each receive a differentiated
image, which in turn necessitates two photographs, it is obvious
that a true "single image" would send identical stimuli to both
eyes, and that therefore stereoscopic perception of a single image
is fundamentally and basically impossible.

INTEGRATED STEREOGRAMS. However, the single image ideal is
approximated by the integrated type of stereogram. Loosely ap-
plied, this may be used to describe the anaglyph and the Vecto-
graph as well as the bar-screen type of stereogram, which is or-
dinarily called the Parallax stereogram. But inasmuch as all
stereograms are made possible only by parallax, the term becomes
vague and meaningless. "Bar-screen" or "grid" describes the proc-
ess specifically and leaves no room for confusion. In all of these
processes the two images occupy substantially the same positions

APPLIED STEREOSCOPY 203

upon the supporting base, and are differentiated by color (an-
aglyph), polarization of light, (Vectograph), or by an opaque
grid (bar-screen) or a lenticular grid (grid). In each instance
there is a sharp differentiation into two distinct and different
images.

FREE VISION. The aim, of course, has been to introduce some
system which would make the viewer unnecessary, and to a cer-
tain degree success has been achieved, but this degree is slight.
All of these experimenters have overlooked the essentially vital
factor that the full realism of true stereoscopy can never be real-
ized in free vision nor by any system which dispenses with the
stereoscope for viewing.

The essential factor involved is that of visual-psychological iso-
lation. This factor is so fundamentally important that it is in-
credible that it could ever be overlooked. The only excuse is that
those who work toward this goal seek only an appearance of solid
relief, not the realistic recreation which is the sum and substance
of stereo.

Visitors to New York (for example) often take home as sou-
venirs, small metal models of the Empire State Building. One of
these small models is the full equivalent of the free-vision stereo-
gram of the same building. Both are tiny, solid appearing models,
nothing more. The dual stereogram viewed in the stereoscope is
not anything like this. It provides the visual isolation which gives
to the spectator every visual stimulus he would receive in looking
at the real thing. In short, provided with a good stereo outfit, the
spectator does not look at a picture, he sails away upon a magic
carpet to New York and actually looks at the building itself. He
sees the original, not a picture.

This is not an exaggeration for emphasis, because when you
view stereogram, you actually do see the real original, even though
you are not in its physical presence. The explanation has been
called metaphysical. It is not. It is based upon an accepted fact
of vision which is not widely known.

You say that you "see an automobile,' 7 inferring some kind of
direct physical contact with the automobile. What actually hap-
pens is that you do receive a direct optical image of the automo-
bile in the eye. That image is perceived, as is the one from the

204 THREE-DIMENSIONAL PHOTOGRAPHY

other eye. But in the brain these two images are mixed and from
the mixture is produced a third synthetic image, which is produced
wholly within the brain itself. This image is then psychologically
projected into space so that it occupies the same position as the
real object. Only thus can we have stereoscopic vision. Incidental-
ly this is accepted theory and will be found in authoritative text-
books of physiological optics.

So if you ''see" an automobile, what you actually see is an arti-
ficial image projected from your own mind. If you see a stereo-
gram, you see an identical artificial image, synthesized in your
brain and projected in exactly the same way. The image is placed
in space at exactly the same distance. In fact, the only difference is
that existing between two different external stimuli, but from the
retinal reception on the two processes are identical.

The true fact is that you never actually ''see" any object stereo-
scopically, you only see the artificial "mockup" which you have
created yourself. Thus in stereo, although you will not want to
accept the fact, you actually and literally see exactly the same thing
you would see in observing the original object.

The difference between the free-vision stereograms and those
seen in the stereoscope will always remain the difference between
a tiny model and the real thing. As a matter of fact, free-vision
stereograms have very little advantage over the ordinary flat pho-
tograph, aside from their novelty. Make them commonplace for
five years and people will not care whether they have them or the
old planar type.

The stereoscope is a fundamental and essential part of that
stereoscopy which actually re-creates realism in its fullest detail.
Second is the use of a good projector in a fully darkened room.
Other systems fall too far short of realism to have more than
novelty appeal, with the exception of the Vectograph whose stereo
quality makes it a true member of the stereo family.

THE BAR-SCREEN. Ives invented the system in 1903, and four
years later Lumiere Autostereo plates were marketed. These plates
were backed with a grid of fine lines spaced from the emulsion by
the thickness of the glass base. A dual diaphragm in the lens
caused the images to fall upon this grid from the two sides so that
the "shadow" of a grid line from one lens aperture fell exactly

APPLIED STEREOSCOPY 205

between two grid shadows from the other aperture. Thus, al-
though the whole image was composed of alternate fine strips
from the two stereo images, the image as a whole would appear to
be normal, except for a suggestion of a line pattern caused by
adjacent parallactic differences. The process worked all right, but
the picture was of course crossed by vertical black lines which
were visible and marred the picture. Later the use of lenticular
elements embossed upon a plastic base was suggested by both
German and French experimenters, and in 1925 such grids were
successfully used in France. Among the details introduced were
specially shaped diaphragm apertures and a moving diaphragm
which caused a slit aperture to move across the base aperture. And
these things are now almost a quarter century old, and of course
the essential characteristics have long since become public prop-
erty.

Briefly, the lenticular grid makes the stereoscopic depth appar-
ent throughout a greater viewing angle than the line grid, but
nevertheless the junction of the elements produces a line effect.
The only way to overcome the line effect is to make the grid so
fine that the lines are invisible as such. With a screen as fine as
this, registration becomes difficult and the images are processed
by reversal, a complication which removes any process from full
practicality, even though it is widely used. No process can hope
to be truly successful until there is complete freedom of repro-
duction, preferably by the usual negative-positive process.

The equipment necessary for making the bar-screen stereo-
grams consists of a camera; a lens whose diameter permits the use
of two apertures with a center-to-center separation of 65mm;
and the essential grid screens which may be either integral with
the sensitive material, or separate. (Eventually the large-diame-
ter lens may not be necessary as the writer has been informed that
.applications for patent have been made covering a system which
permits the use of any normal small camera lens, and which also
maintains the vital grid angle with accuracy.)

Under present day conditions, unless one is lucky in having
old equipment from which to draw parts, the initial cost will run
to about five hundred dollars, and the results are not worth itl
The writer has made a great many bar-screen stereograms using

206 THREE-DIMENSIONAL PHOTOGRAPHY

screens of various types, both line and lenticular, and it is his
considered opinion that as far as good quality stereoscopic photog-
raphy is concerned, the process is of little interest.

On the contrary it must be said that as far as commercial appli-
cation is concerned, particularly if these pictures eventually can
be printed in periodicals, the method has perhaps greater po-
tential commercial value than any other. But it has only the
slightest resemblance to that result which the stereographer knows
as stereoscopy.

The stereoscopic effect in the bar-screen is considerably less
realistic than it is in the Vectograph, for example. The Vecto-
graph presents an unmistakable, an unchanging, and normally
undistorted true stereoscopic image. The only objections are the
resemblance to a miniature model, the absence of orthostereo-
scopic dimensions, and the absence of visual isolation. For these
reasons not even the Vectograph can replace the conventional
stereogram and stereoscope among stereographers. When one
surrounds his eyes with the hood of the viewer, it is like opening
a door to distant places. There is almost a bodily transportation,
something which the hand-held free-vision image can never
imitate.

Still, the bar-screen is an ingenious method, and one follows
the history of the work by Ives, Lumiere, Bessiere, Draper and
Winneck with deep interest. The mere fact that for our specific
purposes the method is inferior, is no reason why these men
should not receive our admiration and respect.

The original suggestion for the use of the grid came from
Berthier in 1896. Acting upon his suggestion, Ives in 1903 in-
vented the system, giving it practical design. This was greatly
improved by Estanave in 1906, and upon his improvements were
based the Autostereo plates of 1907. To overcome the necessity
for viewing the print from the one correct viewing angle, Lipp-
mann suggested the lenticular base in 1908, and proceeded to
make grid stereograms, using the lenticular grid.

Estanave elaborated the use of the bar screen and in 1910
introduced the "magic picture'' which by a slight side shift
changed its nature, but later he went back to stereoscopic effects
and greatly improved Lippmann's lenticular process in 1925, the

APPLIED STEREOSCOPY 207

same year in which Bessiere made use of the same process.

Brewster in 1 860 suggested that the parallax involved in images
made by different portions of a large lens was sufficient for stereo-
scopic effects. Lehmann in 1878 produced such effects, and in
1900 Boisonas made some improvements upon the dimensions of
the apertures used. Thus the lens was waiting when the grid
process was introduced, the final steps in the optical development
almost coinciding with the initial efforts to produce the grid
stereogram.

The parallax found in all images formed by lenses is often
overlooked, but it is one of the fundamental factors in all photog-
raphy. Only the pinhole has practical freedom from parallax, all
other photographic images having a certain degree of inherent
stereo parallax.

As first pointed out by Brewster, if a lens whose diameter is
more than 65mm be provided with two apertures situated at the
ends of a horizontal diameter of the lens, and two images are
made by alternately covering the apertures, those two images form
a normal stereo pair.

It follows that if this is true of two portions of the lens separated
by the normal stereo base, then any two points in a lens will form
images which will exhibit parallax corresponding to the loca-
tions of the two apertures. It is known that when a stereo pair is
made with a separation as small as a half inch, the pair will
exhibit a visible degree of relief, so it follows that any lens in
ordinary use must form an image in which there are definite
stereoscopic pairs of images.

If three test objects are set up in a line receding from the cam-
era, and if the lens is focused upon the central object, the images
of the other two objects will be blurred. This is normal experi-
ence. However, if the lens be provided with two relatively small
apertures and the same picture made, the central image is still
sharp, but the other two now show two overlapping, relatively
sharp images instead of the original blur. Thus the photographic
image really consists of an infinite number of infinitely sharp
images, which appear collectively to be sharp or diffused depend-
ing upon the degree to which they are superimposed in exact
register.

208 THREE-DIMENSIONAL PHOTOGRAPHY

It can be seen then, that among the other images, a pair ex-
hibiting stereo parallax is included in every photographic image.
It is also apparent that even where the image is "in focus" that
is, exactly superimposed the superimposed images have parallac-
tic differences, hence cannot be superlatively sharp. The larger
the diameter of the lens, the greater these differences and the
greater the departure from the ideal monocular, pinhole defini-
tion.

This phenomenon gave rise many, many years ago, to a very
clever device. A lens cap was made which included a vertical bar
across the center of the lens. The width of this bar was from 1/4
to 1/3 the total lens diameter. This gave a clear differentiation
between the two stereo images when placed over any lens. When
focusing the camera, it was only necessary to bring the two images
of any object into close coincidence to produce a sharp focus upon
that object. This was the forerunner of the rangefinder. It is as
good as any focusing device when used with any camera equipped
for visual focus, although the larger the lens diameter the better
the result.

The writer introduced his o\vn variation of that device by
applying filters of complementary colors to the two sides so that
the image which is being focused not only comes into super-
position but exhibits natural color, while other objects are tinged
a distinct filter color, red or green, for example.

Thus it is that an image, although not visibly diffuse, does have
a dual, even multiple, character. The presence of this funda-
mental difference of stereo parallax, although in no way giving to
the planar photograph any degree of stereoscopic relief, does give
to the image a degree of planar realism never achieved in the
small-lens image. It should be emphasized that this effect is one
of diameter alone, not of relative aperture, nor of focal length,
nor of degree of correction. It is simply the presence of images
whose natural parallax includes that degree of parallax normal to
vision. The "realism" of the portrait is simply an unconscious
recognition in the planar image of those fundamental elements
which are normal to vision prior to the synthesis of the stereo-
scopic image.

STEREO PARALLAX AND SOFT Focus. Soft-focus portraits have

APPLIED STEREOSCOPY 209

always been popular, but their origin is not widely known. The
admired absence of razor-sharp outlines should not be the pres-
ence of actual image diffusion. It should be the quality imparted
by the old-fashioned, huge portrait lenses which had such aper-
tures that they did include this normal degree of stereo parallax.
The "soft focus*' is simply the difference between the parallactic
images at the point of best registration. It is a visual effect normal
to everyday vision, and it has a characteristic quality quite differ-
ent from the conventional soft focus, and one which cannot be
imitated by uncorrected lenses, or aberrating glass discs. You can-
not obtain it from a sharp negative during the enlarging process.
The only \vay in the world to obtain this original type of "soft
focus" which is still sharp enough to reproduce every hair on the
head is to use a lens whose actual aperture exceeds two and one-
half inches. If you use that aperture you cannot help obtaining
the soft-focus effect. It is a matter of stereo parallax, not loss of
definition in the usual sense.

This fact has been recognized from time to time by "inventors"
who reason that if the picture includes the necessary differentiated
images of stereo parallax, then the image must be stereo. Some
of them went so far as to place a vertical bar across the front of the
lens to produce two images differentiated in the same print, and
then insisted that these were stereograms. All the time they
neglected the one fundamental point: In any true stereogram, the
eyes must be able to fuse the images of any object within tfre en-
tire field, and at the same time the "double" or "ghost" images of
other objects must be separated as in normal, direct stereoscopic
vision. Although this fact is elementary, not less than a dozen in-
ventors have at one time or another "invented" this half-way
stereo which is not stereo in any sense of the word.

PSEUDOSCOPIC PICTURES. One interesting phase of stereoscopy
is the pseudoscopic picture. This is a stereogram in which the
left and right pictures are left in their original negative positions,
that is, untransposed. The effect is to reverse all depth dimen-
sions, bringing the far distance close to the eyes and removing
nearby objects to a remote position in space.

This phenomenon is interesting, but it also has a practical
value. Through familiarity, we are prone to fail to appreciate true

210 THREE-DIMENSIONAL PHOTOGRAPHY

spatial relationships and take them more or less for granted. We
may grasp the fact that the stereogram is faithfully reproducing
three-dimensional space, and still fail to appreciate the various
relationships which go to make up this spacing. The pseudoscopic
picture so emphasizes these differences that planar separations
unperceived in the stereogram become very distinct. This makes
the pseudoscopic picture of definite value when examining cer-
tain microstereograms and other stereograms of scientific interest.
No stereographer should neglect the study of pseudoscopic prints
of various objects. The pseudoscopic picture has also demon-
strated unusual potentialities in the pictorial field. Many objects,
a rock strewn seashore, for example exhibit in pseudo a wholly
abstract form, yet through the retention of natural dimensions
and color, such pictures have a quality only equalled by the best
of our modern artists,

A characteristic of the pseudoscopic picture usually ignored is
that, contrary to the stereogram, those parts of the picture which
appear nearest have less parallax than those farther away, which
results in a peculiar emphasis of depth, a gradually increasing
depth distortion which adds to its value rather than detracts.

These pseudoscopic pictures are true stereo in reverse, and
should never be confused with the false stereograms sometimes
called pseudo-stereograms. These false stereograms are pictures
with absolutely no stereoscopic relief, but which by some device
or other have been given a vague appearance of roundness or
false solidity. Although often widely heralded as "stereoscopic,"
"three-dimensional," and the like, such claims are definitely false,
and should be so recognized. They should never be tolerated,
and they owe any measure of acceptance to the fact that the gen-
eral public is wholly ignorant of the realism of the true stereo-
gram. It seems strange that any pictorial medium which has once
known universal acceptance and which is so widely popular today
should be so wholly forgotten that the greater part of our popu-
lation should never have seen an example. Yet this is true of
stereoscopy. Fifty years ago a home was not complete without its
stereoscope. Today there are hundreds, even thousands of people
who sincerely believe that stereoscopy was an invention made
during the late war,

APPLIED STEREOSCOPY 211

The term <4 stereoscopy" is used advisedly, for it is a part of the
new teaching that stereoscopy pertains only to pictorial reproduc-
tion. This, too, seems incredible, yet there are war-time students
of stereography who insist that stereoscopy has no place in normal,
direct vision!

The fact is that in the year 1870 every photographic dealer sold
stereo equipment, and every photographic amateur (of which
there were even then more than fifty thousand in this country)
was familiar with the process even it he did not practice it. It is
well known that the vogue for daguerreotypes began to die out
before 1860, but nevertheless stereo-daguerreotypes were made
and in demand! And since that time, while equipment has been
improved, there has not been one fundamental addition to the
known physical laws of stereoscopy. Basically we practice today
the same system that has been used almost a century.

Throughout the full life of stereoscopic reproduction, dating
from the first observations of Wheatstone in 1833 and the work of
both \Vheatstone and Brewster through the following decade,
stereoscopic photography has been in unbroken and consistent
use. Since the beginning of the century it has been of great value
in X-ray work; and it has proven to be the most successful photo-
graphic medium for medical reproduction. It has been used in
engineering and construction as well as in many other professions.

In schools, stereograms have been used for visual education for
years, and it is curious to note that although the average adult is
wholly unfamiliar with the art, there is one company whose busi-
ness in stereograms is said to amount to more than a million dol-
lars a year, and has been at a corresponding level for years past.

Just a year or so ago stereo was generally acknowledged to be
dead and buried, a relic of the horse-and-buggy days. A few months
ago, the photographic world was amazed by the rapidity with
which its popularity was growing, surpassing the famed rise of the
miniature camera. Today those not in stereo are seriously con-
cerned, and one commonly hears conjectures as to how soon planar
photography will pass out of existence except for specialized work.
That is an event which, of course, will come to pass eventually,
because no system of photography can expect to be permanent un-
til its reproduction of reality is perfect. Stereo itself will pass on

212 THREE-DIMENSIONAL PHOTOGRAPHY

before the advance of superstereo in which we shall have the
scene projected in space, not with an appearance of space as we
now have, but an actual projection in space around which we can
walk and examine from every angle. Oh, yes, that will come-
sooner than many of you think. But when it does, it will have
color, motion, sound and perhaps odor as well. When that is
accomplished there remains but one more step, the re-creation of
the scene in such a way that the human spectator can enter the
scene while retaining normal spatial relationships which would
exist had he entered the real scene. There is only one major
problem in the way of that. The camera sees, not a solid, but a
shell which has no rear half. This sounds like fantasy, but it is
child's play compared with the fantastic things among which we
now live, radio, television rockets and atomic power!

CHAPTER 14
POLARIZED LIGHT APPLIED TO STEREOSCOPY

THE POLARIZATION OF LIGHT is one of the most interesting phe-
nomena encountered in all optics. To go into the details of
the actual procedure would take far more space than we have
available. Moreover, many of the more beautiful of these phe-
nomena pertain to some of the more complex reactions of polar-
ized light and have no bearing upon our subject. For that reason
we shall content ourselves with a very simple explanation of the
one polarization phenomenon which has to do with stereoscopic
observation.

Light travels as a "ray" from the source to some given point
which may be illustrated by stretching a cord between two posts.
The cord represents the line of travel of the light. The light
itself is an electro-magnetic vibration whose direction is outwardly
from (perpendicular to) the line represented by the cord. If now
the cord is replaced by a string of Christmas tinsel, we might
assume the central core to be the line of travel and the out-thrust
tinsel strands the direction of vibration. This illustration is at
some variance from the actual theory, and wholly neglects the
projection of the wave front, but its simplicity more than out-
weighs its faults from the standpoint of advanced theory. We are
not students of advanced physics, but only trying to arrive at an
acceptable, if crude, explanation of the polarization of light.

Suppose you spray the tinsel cord with glue and then pass a
coarse comb along it with the teeth held vertically. The tinsel will
be "ironed out" and instead of a large round rope, it will be a
thin ribbon standing vertically. If you treat a similar cord with
a comb whose teeth are horizontal, you will produce another rib-
bon which lies horizontally. These ribbons now represent two
light rays which have been polarized, with their axes of polariza-
tion at ninety degrees. The polarized ray retains its full ampli-
tude of vibration only in one plane in which its own axis lies. A
ray may be polarized at any desired angle.

Now if the tinsel is allowed to dry so that it will retain its rib-

213

214 THREE-DIMENSIONAL PHOTOGRAPHY

bon form against a considerable force, we can illustrate the polar-
ization phenomenon with which we are concerned.

Imagine one of the combs held with its teeth perpendicular and
spaced about an inch from it, another w r ith its teeth horizontal. If
you pass the vertical tinsel ribbon through the teeth of the first
comb, which can be done easily, it will not pass the teeth of the
second comb because the edge of the ribbon lies across the teeth
of the second comb. However the second (horizontal) tinsel rib-
bon can be passed through the second comb but not through the
first.

The action of a polarizer is crudely analogous to that of the
comb upon the glued tinsel. It permits the passage of a ray of
light which is vibrating in only one major plane. If a second
polarizer is held before the ray in such a position that it would
polarize the ray at right angles to its first polarization, the ray of
light cannot pass.

We have the apparent paradox of two substantially transparent
bodies which freely transmit light in a certain position, but when
given another relationship they become opaque and will not trans-
mit light.

This cancellation of light by a second polarizer when light has
already been polarized, is widely used. Inasmuch as light is polar-
ized by reflection at some angle from any non-metallic surface,
much of the daylight about us is polarized. Thus the glare from
a paved street is largely polarized and the use of polarizing goggles
eliminates this glare without a corresponding reduction in the
normal light intensity. The same thing is done by the photog-
rapher who often eliminates undesirable reflections, as in a win-
dow, by using a polarizer over the camera lens.

Polarized light is used widely in the study of crystals and min-
erals under the microscope, but it is used with certain accessories
so that the polarization may be that just described and known as
"plane" polarization, or it may be circular or rotary polarization.
The light may be split into two rays which are differently re-
tarded by the specimen under examination. The two rays are
mixed, and in mixing they interfere, producing very beautiful
color effects which are of great value in the study mentioned. But
these effects are outside our province.

POLARIZED LIGHT APPLIED TO STEREOSCOPY 215

Polarized light may be reflected and if the reflection is from a
metallic surface, the polarization is retained, but if the surface is
non-metallic the polarization is lost. This fact is of the first im-
portance when polarization is used for anaglyphic stereo pro-
jection.

Many years ago, indeed about a half century ago, when stereo
was highly popular and when polarized light was a prime favorite
among both amateur and professional scientists, a patent was
granted upon polarization anaglyphic stereoscopic projection.

According to the specification, this patent called for a double
projector which projected two superimposed images upon the
screen. Each projection lens was fitted w r ith a polarizer. Those
who were to see the projected image in stereo relief w^ere pro-
vided with a pair of polarizers to be supported before the eyes.
The axes of the viewing polarizers were to be crossed as w r ere those
of the projector so that each eye saw nothing but its appropriate
image. That is exactly the nature of the polarization projection
which we have today to the last detail. And the patent is dated
before 1900!

However, the drawings and specification both were made
around the use of Nicol prisms as the polarizers, which made the
system both bulky and costly. These prisms were not specified
because polarizing films were unknown at that date, but simply
because the Nicol was the acknowledged supreme polarizing agent
of its day.

POLARIZING FILMS. The general use of polarizing agents today
is largely due to the development of a system by which polarizing
films may be made to a given standard, and in quantity. The idea
that the polarizing film itself is new is erroneous. In the Quar-
terly Journal of Microscopical Science for January, 1854, Dr.
Herapath gives directions for making the films of iodosulfate of
quinine, markedly similar to the commercial films which we have
today. Many attempts have been made to produce these films,
often successful ones, but because of the quality and uniformity
of their product the Polaroid Corporation produces most of the
polarizing films used in this country today.

The fact that a single film mounted between two relatively thin
glasses will perform in all major respects similarly to the very

216 THREE-DIMENSIONAL PHOTOGRAPHY

bulky and costly Nicol and similar prisms, has placed in our hands
the availability of unlimited variations of polarization of light.

THE POLARIZATION ANAGLYPH. The use of the polarized an-
aglyph has become commonplace, and practically every stereo
amateur has seen examples of such projection. Inasmuch as we
have already described this procedure in some detail in an earlier
chapter, we shall not repeat the description. It will be remem-
bered that the two stereo unit images are projected through two
polarizing films whose axes lie at right angles; that they are pro-
jected in approximate superimposition upon a metallic surfaced
screen; and they are viewed through spectacles composed of two
polarizing films whose axes are at right angles and which are
oriented with the projection axes to produce the necessary can-
cellation.

THE VECTOGRAPH. A different type of polarized stereogram,
and one which is both deeply interesting and commercially prac-
tical, is the Vectograph. We are not going into the detailed theory
of the Vectograph, because it is somewhat involved for those un-
familiar with the simpler phases of polarization, and because the
stereographer has no call to make use of such knowledge.

Vectograph in Projection. The Vectograph, as first made, con-
sists of the substantially transparent base with a different image
upon each side of this base. It can easily be understood that it can
therefore be projected, just as any transparency or lantern slide
is projected. We also know that polarization is not affected by
projection per se, so if we use a standard aluminum-surfaced pro-
jection screen (which does not depolarize the light), we have in
this stereogram a slide which may be projected normally and
viewed with the standard 3-D goggles. Cloth, beaded screens,
plaster, and the like will show both images even when the correct
viewer is used.

Inasmuch as machines were long ago developed for printing
motion-picture films in two colors for the early color processes,
there is no reason why Vectographic motion-picture films could
not be made, thus providing stereoscopic films which could be
projected with any standard motion-picture projector and viewed
with spectacles as easily worn as ordinary sun glasses.

Vectograph as a Transparency. Because there is no reasonable

POLARIZED LIGHT APPLIED TO STEREOSCOPY 217

size limitation, large Vectographs can easily be made and dis-
played as transparencies in public places. These would be espe-
cially suited for demonstrations at industrial exhibitions, for ex-
ample, and in museums. Instead of individual viewers, a series of
strips of polarizing film about two and a half inches wide running
vertically would enable anyone simply to look through them to
see the transparency in full relief. Of course the pairs of strips
would have to be separated to prevent the use of the opposed
strips which would produce a pseudoscopic effect.

Vectograph as a Photograph. It the Vectograph is to be used as
a normal photograph for mounting in a book or the like, one side
of the finished product is flowed with an aluminum paint of spe-
cial grade. This provides a non-depolarizing reflecting backing
for the print. This is in every way similar to the doretype which
was at one time highly popular as a portrait medium.

Vectograph in Medicine. One of the principal roles of the Vec-
tograph is in medicine. At present stereo X-rays must be viewed in
special viewers which are extremely large and cumbersome. The
usual viewer will occupy a floor space of about three by six feet,
and stand five feet high. The surgeon who is to follow the X-ray
as a guide in operating must study it beforehand and then take
only a "flat" film into the operating room. If the X-ray is a Vecto-
graph and the surgeon is provided with half-goggles of polarizing
films, he can observe the X-ray in full relief while actually oper-
ating.

GENERAL APPLICATION, In fact, there are many more applica-
tions of this process, some of which were proven during World
War II when hundreds of Vectographs were made from aerial
films. The process is attractive to stereographers, even though it
does lack the orthostereo quality as do all "viewerless" processes.
It is an interesting type of stereogram and for many purposes it
has certain advantages over the conventional double stereogram.
There is hardly a question of superiority, rather a question of the
suitability for any specific purpose. When orthostereo is de-
manded, a viewer must be used because it is an integral part of
the ortho technique, but when it is simply desired to obtain
stereo relief without ortho proportions, the simplicity of the Vec-
tograph has much to recommend it. The fact that the viewer

218 THREE-DIMENSIONAL PHOTOGRAPHY

(which resembles a pair of sun glasses) and several prints may be
carried in the pocket is an unquestionable advantage.
Also see Chapter 19.

CHAPTER 15
TYPES OF STEREOSCOPY

ONE ORDINARILY DOES NOT THINK of asking of stereoscopy,
"What kind is it?", simply because stereoscopy is not some-
thing which we think of as being capable of division and classifi-
cation. One thinks we have stereo or we do not have stereo, but
this is not entirely true.

When the term is not qualified, we consider stereoscopy as be-
ing orthostereoscopy, but already we have had occasion to discuss
stereoscopy which is really the genuine thing but which is cer-
tainly not ortho.

ORTHOSTEREOSCOPY. This, the normal form, actually repro-
duces upon the retinas images which are identical with those
which w T ould have been produced by the original objects, disre-
garding of course the non-essential factor of motion which may be
present, but not necessarily so.

The orthostereogram, or rather the stereogram viewed ortho-
stereoscopically shows us the appearance of the original object in
full life size at full natural distance and in full, natural color.

It is true that color is not usually included in the definition of
orthostereoscopy, but it is hardly less important than other fac-
tors, because it is an essential factor of stereo relief, wholly aside
from the added attractiveness it gives the picture. This has been
fully discussed elsewhere in this volume. Inasmuch as ortho-
stereoscopy is by definition, fully correct stereoscopy, the term
can hardly be accepted if it does not include color.

HETEROSTEREOSCOPY. This is a mixed form, such as using a
viewer unrelated to the camera, or any of the " free- viewing" de-
vices. The projected image (in the absence of controlled specta-
tor position), the Vectograph, and the bar-screen processes all fall
into this class. The heterostereogram shows us true parallactic
relief which bears a definite relationship to the real relief, but
not necessarily in normal proportions. The relief may be exag-
gerated much more than normal, as in some aerial views.

Ordinarily the reproduction presents neither full life size nor

219

220 THREE-DIMENSIONAL PHOTOGRAPHY

true natural distance. In using viewerless methods, the perception
is that of a small relief model held in the hands.

HYPERSTEREOSCOPY. This is a form of stereoscopy which is, as
pointed out by Helmholtz, to all practical intents and purposes
telestereoscopy. It is the result of using a stereo base which is
greater than the normal separation, sometimes as much as several
feet. In any event, the increase in base is disproportionately large
compared with the increase in the focal length of the lens (if any).
It usually denotes a wide base with the normal lenses.

HYPOSTEREOSCOPY. This is just the opposite of hyperstereo and
denotes a base less than normal. Some stereographers erroneously
use a narrow base when making any stereogram nearer than 10
feet. The technique is of value when making very close-up photo-
graphs, at distances of 20 inches or less, when otherwise there
would not be sufficient overlapping field to produce a good stereo-
gram. Hypostereoscopic bases as little as a tenth millimeter have
been used in microscopic work, but for that purpose the method is
less satisfactory than the convergence or rotary method.

METASTEREOSCOPY. This is commonly referred to as orthostere-
oscopy, but it must be remembered that orthostereoscopy not only
presents objects in full life size and at full natural distance, but
it does this for every object within the field. Metastereoscopy re-
produces one original object in its natural proportions, usually
much larger than life size; but when more than one plane is in-
troduced, objects lying in planes other than the principal plane
are not reproduced in the same proportionate relationships. It
might be said that metastereoscopy is the orthostereoscopy of a
single plane. It is typified by the carefully produced stereomicro-
graphs usually referred to as "ortho." It might also be called the
normal stereoscopy of the microscopic field.

PARASTEREOSCOPY. This is the peculiar composite of com-
pressed space and normal relief characteristic of the stereo-tele-
scopic field. Although the field appears to have only a portion of
its true depth, objects within that field appear to have their full
natural relief. Thus a three-times parastereoscopic field would
appear in such a manner that a wall 300 feet distant appears to be
only 100 feet away. That is, the space is compressed to one-third its
true value. But at the same time every object within that space

TYPES OF STEREOSCOPY 221

appears to have its full degree of relief, and does not appear to be
compressed to one-third its depth. Here we have a distinct separa-
tion of spatial relief and solid relief.

Parastereoscopy is the nearest approach to orthostereoscopic re-
sults we can obtain combined with any effective telescopic effect.
It results from increasing the focal length of the taking lenses and
increasing the base in exact proportion to the focal length. A two-
times k "para" involves a base of i3omm and lenses of twice normal
taking length. Any departure from this proportionate increase
throws the result into the general heterostereoscopic class.

Any true stereogram may be classed as one or the other of the
foregoing, and many of course meet the characteristics of two
classes, but at least it is a working classification which eliminates
much confusion and which has proven sufficiently definite for
practical use.

CHAPTER 16
TRICK WORK

TRICK AND STUNT PHOTOGRAPHY with the stereo camera is di-
vided into two parts; that which is conventional camera
trickery applied to the stereo camera, and that which depends
wholly upon stereo principles and has no counterpart in planar
photography. Naturally there are some tricks which are more or
less borderline, as for example the false impression given by using
an abnormal camera angle. This is common practice in planar
work, but when so used it lacks the unreal appearance which
makes it a part of trick work.

DOUBLE EXPOSURE. This is done too often by accident to be
unfamiliar to the stereographer, but the accidental shot is merely
confusion while the planned double exposure can be made highly
effective.

There are two kinds of straight double exposure. The simplest

Fig. 16-1. Wide angle lenses may be used in stereo as easily as the

close-up type of supplementary.

222

TRICK WORK 223

is that which involves two uncontrolled subjects. For example
you may make an exposure of a model seated in a chair. The pose
must be comfortable so that it can be held for several seconds
while preparing for the second exposure. It is essential that the
camera be mounted upon a rigid tripod.

After the first exposure, which should be less than normal but
not less than 1X> normal, a second model is introduced, and
a second exposure made. The sum of the two exposures should
approximate the full normal exposure.

The first shot records the first model and the background and
the second exposure repeats this. However the second exposure
has the second model in place. The background behind the sec-
ond model will receive a short exposure as will the second model.
The result is one which shows the second model as a ghostly
figure through which the background is seen. The background
is darkened just enough to suggest the semi-solidity of the second
figure.

This is an old trick in ordinary photography, but it does not
prove very effective in planar work because the second figure ap-
pears to be painted upon the background. In stereo the ghostly
figure stands out in space with the background behind her. This
gives the complete illusion of a fully transparent figure standing
out in space. A true ghost. This is not difficult and provides
possibility for many amusing shots.

It might be added that the sum of the two exposures may equal
1 1/ 9 times normal with no ill effects.

MASKED DOUBLE EXPOSURE. This is a somewhat more ambi-
tious version of the double exposure. The simplest "mask" con-
sists of arranging some part of the background so that it will be
in deep shadow. The exposure is made, for example, with a male
model who is smoking. A cloud of smoke before the deep shadow
provides an efficient setting for the second shot. This shot may
be made freehand, but it is essential that the position of the
deep shadow be remembered with some accuracy.

The second model is now posed against a dark background, and
the second exposure made. The figure of the model is placed in
the position occupied by the deep shadow, and the finished result

224 THREE-DIMENSIONAL PHOTOGRAPHY

will show the second model in line with the cloud of smoke.

The worst difficulty in this type of work is that the dark back-
ground for the second shot must be large enough to fill the whole
film area. The actual size of course depends upon the distance
between camera and background, and this in turn determines the
size of the model.

In the first type of double exposure, the second model is in-
troduced into the scene, and the position is recorded automati-
cally. But in the separate exposure method, it must be remem-
bered that distance will be faithfully reproduced. For example if
the man is seated 10 feet from the camera, and the second model is
also placed 10 feet away, the figure will appear to be in the smoke
cloud. But if the second model is only 5 feet away, then the figure
w T ill be much larger than life size and will be suspended in air
between the observer and the smoke cloud. This is the easiest and
least effective way.

If the model is 20 feet from the camera, the size of the figure
will be in miniature, but it will be seen behind the smoke cloud,
and of course a very large black background will be required for
this purpose. The position in space can be controlled as explained
later in the chapter.

One way to eliminate the large background is to make use of
camera masks, but as there is no precision finder or ground glass
focusing available on the usual stereo camera, the arrangement
must be made by approximation. These masks are made of black
card or heavy black paper. An area is cut or torn from them to
leave an opening corresponding to the position of the model. If
the masks are situated about one inch before the lenses, the block-
ing will be effective and the nearer the mask to the lens, the
broader will be the vignette border. As the mask is moved away
from the lens the open area grows smaller and the edges become
sharper, but even when using distances of 3 to 4 inches the edges
will still be satisfactorily softened. If the mask is placed very near
the lens surface, almost in contact, there will be no masking, but
a variation in exposure can be seen. In short, any portion of a
lens theoretically covers the entire film, and practically the effect
is nearly enough the same to prevent effective masking so near
the lens.

TRICK. WORK 225

A variation of the old motion picture mask box can easily be
made. This is a rectangular box with flaring sides. It may be
made 2 inches wide by 6 inches long, extending 2 inches from the
camera. The masks are cut from pieces of 2x6 black paper. If
the paper is laid out in two squares with centers 7omm apart, the
corresponding areas may be cut with little trouble. But if for any
reason these masks are to be used with a background which shows
any detail, that is, a background not solidly black, the relative
positions of the apertures must be cut to place the aperture where
you want it.

With the black background the mask has no spatial position,
and the same care is not necessary. Inaccuracies will result in a
certain stereo discrepancy at the edges of the space where only one
image appears, but these may be disregarded. No one w r ill see
them unless attention is directed to the error.

Such masks make it possible to reproduce distant figures in
miniature without the use of an impossibly large background.

Another method is to have the model and black background in
one room, the camera in another and to shoot through a doorway,
using the non-illuminated room in which the camera stands, as an
extension of the black background.

DUPLICATION. This is still another variation of double expo-
sure. The mask box will be needed, or a special lens cap mask.
It is necessary that the masks be carefully adjusted, something
which can be done when the camera is empty, placing a strip
of matte celluloid in the film track and using it as a ground
glass. If the mask segment is too large there will be a double
exposed strip down the center, that is a light strip which spoils
the effect. If the segments are too narrow there will be an un-
exposed strip in the center which is even worse. The straight side
of the mask opening must be cut so that the dividing line comes
exactly at the film center. Once this has been determined the
masks may be used at any future time with full assurance. If the
mask box is used, the matte celluloid is used to mark the central
dividing line on the edges of the mask box.

Set the two masks at the same side, left, for example. Pose your
model at that side of the normal film area, in some pose suggest-
ing a companion. An old favorite is to have the model seated as

226 THREE-DIMENSIONAL PHOTOGRAPHY

though playing bridge. Make the exposure. Then place the masks
to cover the opposite side of the lenses and pose your model again
on that side of the table. Make the exposure. If the masks are
correctly adjusted you will have a perfect stereogram of the model
playing bridge with herself.

The duplication trick requires more care than the usual double
exposure, and is rarely used except when it is desired to show a
person twice in the same stereogram. The result is so obviously
impossible that it always arouses great interest.

A variation consists of making the first exposure as usual, then
reverse the masks, make a halftime exposure without the model
and a second halftime exposure with the model to show the model
playing bridge with her own ghost.

It will be seen that double exposure involves (A) exposing the
same film area twice, as in the ghost effects or it may involve (B)
exposing two different areas of the film in sequence. The latter
requires a considerable amount of care in matching the edges of
the two areas, and both demand a rigid tripod to support the
camera immovably during the two exposures.

There are times when two casual, freehand exposures may be
made upon the same film. One of these which we saw showed an
attractive girl seated within a showcase in front of a shop, but
because of the confusion of backgrounds this type of work de-
mands the greatest care of all in selecting the setting for both ex-
posures and hence is the most difficult.

There is no limit to the possibilities of double and even mul-
tiple exposure, but you must expect disappointments at first. You
must learn the limitations as well as the possibilities, but once you
get past this preliminary obstacle you can have a lot of fun with
the process.

The fact has been mentioned that size differences are faithfully
reproduced in stereo, so that the well-known shot of a girl seated
in a wineglass, so familiar in planar photography, cannot be ac-
complished by the methods which have been described. However
such shots are easy in stereo, but as they involve the fundamental
problem of space control we shall discuss it later in this chapter.

CAMERA ANGLES. Unusual camera angles have been worked to

TRICK \VORK 227

death by the planar photographer, but when used in stereo the
results are so novel that the phrase Bangle shot" takes on a wholly
new meaning.

To make this fully clear, one or two related factors must be
explained.

It is well known that stereographers are warned repeatedly to
keep the camera level from side to side. The stereo camera may
be pointed upward or downward somewhat, but never from side
to side, that is with the two lenses off level.

This rule is deliberately violated in making angle shots, but
other conditions must be such that this can be done without spoil-
ing the shot.

A second factor is a physio-psychological one. If you see a tree
cut down it does not look as it did when standing. You may stand
at the butt and look along the length of the trunk but the ap-
pearance is wholly changed. There is a certain psychological-
visual effect gained by lifting the head to look upward or bending
forward to look down. Undoubtedly this has some close connec-
tion with the matter of balance.

This, too, is taken advantage of in making shots upward and
downward. If you stand near a tree and shoot directly upward,
the slide has little effect when viewed naturally, but if you lift the
viewer so the head is tilted backward and the eyes turned upward,
the vertical appearance is astonishingly convincing. However, as
this demands cooperation on the part of the observer we shall
leave it with this mention, and return to the true angle shot.

An excellent slide of this type was sent to us recently. The
location was in a Western National Park. The terrain was of
rugged hills and immense boulders scattered about. The stereog-
rapher (F. W. Kent) had posed his model so that she leaned
toward a huge boulder at an angle of some 45 degrees, supporting
herself by an arm stretched out against the boulder. The camera
was tilted so that her body was parallel with the sides of the
finder and the exposure made. The result was astounding. There
was a perfect stereogram of a young girl, nonchalantly supporting
with one hand a boulder some fifteen to twenty feet in diameter
which hung over her with no other support than her hand! It
was wholly convincing.

228 THREE-DIMENSIONAL PHOTOGRAPHY

Of course there is a trick to it, and it is this: The subject must
be so arranged that there is nothing whatsoever in the picture to
indicate the camera tilt. In the stereogram just described, model
and boulder were silhouetted against the sky. Had there been a
tree or a mountain cabin or other normally vertical line in the
composition, its tilt would have spoiled the whole thing.

Another shot of the same type was made of a cat walking along
a tree limb. As you know a cat usually starts down a tree trunk
head first and almost immediately swings about and backs down.
This tree limb was almost horizontal, but its smaller branches all
tended outwardly. The cat was walking upright along the limb.
Nothing but other limbs appeared in the field of view. The cam-
era tilt gave the limb a direction about 10 degrees off vertical and
a position chosen which caused the smaller branches to tend up-
wardly. The result was a cat walking upon fully extended legs
down a tree trunk, an impossible thing to do because the position
would give no purchase for the claws.

This has been hardly more than a suggestion of the tricks which
can be done with the stereo camera, but it should be enough to
get you started and to free you from the bonds imposed upon the
use of the stereo camera by the conventional rules. Once you know
when you can violate the rules, you will find endless opportunities
to make shots which will combine full stereo realism with an ap-
parent disregard of physical laws like gravitation.

Most striking of all will be the tremendous effectiveness of the
stereograms as compared with the usual trick shot in planar
photography.

PARALLAX INVERSION. There was a time when the photographer
who did not fully understand the effects of parallax, and of trans-
position, simply did not make stereograms. Today it seems that
more than half our stereographers have their slides mounted for
them, and are not even aware of the importance, or even the
existence of the step of transposition.

Perhaps that in itself would not be important, except for the
fact that those who are unfamiliar with transposition are not
familiar with the appearance of an untransposed slide. Several
times now we have received slides with letters requesting informa-
tion as to the cause of the curious appearance exhibited. This

TRICK WORK 229

appearance is variously described as ''fuzzy/' "blurred/' "double
exposure/' "eye-straining" and so on. In fact the slides examined
were all sharp, but the trouble with all of them was the absence
of transposition. In other words, the two unit images were
changed left for right and vice versa.

This has a direct bearing upon the subject of stereo vision it-
self, and there seems to be little doubt as to the value of the study
of such untransposed slides. Such slides are called "pseudograms"
to distinguish them from "stereograms." This term is often con-
fused with "pseudo-stereograms," The pseudogram (picture in
the method of pseudoscopy) is a differentiated pair mounted in
reverse. The pseudo-stereogram is a pair of identical pictures,
made from one negative which of course exhibit only that peculiar
indefinite relief common to all pictures viewed through a lens and
which so many people consider to be truly stereoscopic. So, to
prevent confusion, we shall refer to the pseudogram as an inverted
stereogram.

The basic effect is that of transposing distances. Those objects
far away seem to be near; while nearby objects seem to be far
away. Of course, nearby objects obscure parts of those objects be-
hind them, so we have silhouettes of odd shape hanging in the
air nearby. But because of the absence of perspective diminution,
the fact that the distant objects conform exactly to the outlines of
these patches is not at once apparent. This effect has been briefly
discussed in the chapter on pictorial work.

In fact, unless there are marked planes and a fairly great
variety of them, the effect is appreciable only to the experienced
stereoscopist. There is great excuse for the beginner who cannot
see what is wrong with one of these slides.

One of the worst types of subject is that which includes several
planes, closely spaced, and with a considerable distance between
the camera and the first significant plane. Subjects of this kind
have confused even experienced stereo technicians for a time, be-
cause the inversion of distances is not immediately apparent. Very
often the stereoscopist states that the slide is made of two identi-
cal prints.

So, if you find among your slides one which for some obscure
reason does not look just right; which appears to be somewhat

230 THREE-DIMENSIONAL PHO'I OGRAPHY

"jumbled" without being fuzzy just try changing places with the
two films and see if that does not solve the problem.

It has been said (and written) that pseudoscopic vision is purely
a phenomenon of photography, but this is definitely untrue. It is
quite easy to see inverted distance by direct vision. One of the
easiest ways is to remove both ocular and objective lenses from a
pair of prismatic binoculars, and to look through the prisms alone.
There is, of course, no magnification, but the pseudoscopic relief
is considerably more definite than in a pseudogram, although the
image is upside down.

A better, if more elaborate, method is to make a twin barrel
telescope and mount in the tubes a pair of Dove prisms. A Dove
prism is similar to a simple right angled prism with the apex cut
off. In fact an ordinary 90 degree prism may be used but the
unused apex takes up more room than the truncated Dove form.
The bases of the prisms are parallel, with the apices facing each
other. Light rays strike the first oblique surface, are bent toward
the base, reflected there, fall upon the second oblique surface and
thence to the eye. The reflection at the internal base of the prism
has produced a mirror type inversion, that is changed right to
left and vice versa. (This is also an excellent w r ay to make a trans-
posing viewer.)

In short, the only thing required to produce pseudoscopic
vision is to exchange the right and left elements of visible sur-
faces! In the pseudogram, the right-eye image is seen by the left
eye and vice versa, but in direct vision the reversal is simply that
of the scene itself, as though reflected in a mirror. NO! You can-
not do this by simply looking into a mirror, each eye must have
its own independent reflection.

The action of the Dove prisms is shown here:

TRICK WORK 231

The tubes of your pseudoscopic binocular need not be round.
In fact for experimental purposes you can make square cardboard
tubes. Or you can make cork mounts to slip into short lengths
of mailing tube. You can cut two holes in each of two pieces of
cardboard to hold the tubes parallel, and there you are. If you
wish a more permanent instrument you will find plastic as good
as metal and a lot easier to work with.

If you just want one look, you can lay the prisms on a table,
then kneel down and look through them. Be sure the shorter of
the long faces are turned toward each other. If the shorter face
is upward or downward you will see objects inverted, but if in
correct position you will see right and left interchanged.

It is an interesting experiment, but of what practical use is it?
This experiment proves the importance of parallax, in fact it
proves some other facts which have been frequently disputed.

First. Experts in the field of vision are often inclined to dispar-
age the importance of parallax in stereoscopic perception. This is
given as one of several factors which result in distance perception.
Such things as perspective diminution, the overlapping of far
objects by those near at hand, and the effect of atmospheric haze,
to mention but a few, are often said to have quite as much effect
in producing relief perception as the parallax of stereo.

But what happens in pseudoscopic vision? Here every factor
other than parallax remains unaltered. Even when there is a
strong light to give the contour effect of shadow forms, no differ-
ence is made. The inverted parallax overrides all other factors
and produces an obvious effect of inverted distance. Inasmuch as
the factor of parallax is the only one altered, it seems we are
justified in saying that the distance perception due to stereo par-
allax is greater in its influence than all other contributory fac-
tors combined!

Second. If the foregoing is accepted, we are then forced to
accept the statement . . . "The apparent distance of an object is
determined by its stereo parallax/' And that too is true. But once
more:

Third. If it is possible to alter the parallax without altering
the distance of an object, that object will appear to lie at the
distance corresponding to its parallax rather than at its true dis-

232 THREE-DIMENSIONAL PHOTOGRAPHY

tance. Again this is demonstrable, for it is the very factor upon
which we base space control. BUT this statement is attacked often
by writers about human vision; some of whom state that other
factors, principally that of the diminishing size caused by per-
spective will prevent the object from being visually accepted at
an altered distance. All stereographers who have experimented
with space control know how utterly false this is. The realism of
space control is the reality of life and results in one of the few
pictures of any kind which can and do produce genuine astonish-
ment in the spectator.

Now let us carry this reasoning to its logical conclusion. It
might be said that the external stimulus of stereoscopic percep-
tion is parallax; but this is not enough. For any one given dis-
tance there is a corresponding parallax. But if the stereogram
shows us just one plane in the strictest sense it is not perceptibly
stereoscopic; at the same time if we do have a single plane, the
parallax normal to that plane is present. Therefore it is not
enough simply to have parallax.

If the subject consists of a group of objects all in precisely the
same plane, each exhibits the parallax of its position, but still we
do not have stereoscopic relief, as the image is still confined to a
single plane.

Therefore the mere existence of multiple parallax is not suffi-
cient, although it is obvious that multiple parallax is essential.

If there are objects in different planes, each will exhibit the
parallax normal to its distance, and these various distances will
create a series of different parallax values. Thus we have the
establishment of depth in the sense of the establishment of several
planes, and we are approaching our goal.

Now I must ask you to bear with me in a bit of repetition: If in
viewing a stereogram (or real objects for that matter) we fix our
eyes upon a single point and hold it steadily for several minutes
(which by the way is almost superhumanly difficult, because most
people do not realize that "steadily" means without shifting the
gaze as much as 1/100 inch) the eye becomes conscious only of
the spot upon which attention is focused. The side images are
as visible as ever, but they remain motionless and demand no
attention whatsoever. The result is that we return to the condi-

TRICK WORK 233

tion of a single parallax, and there is a definite loss of stereo per-
ception. The instant the eyes are moved from the target spot, the
scene jumps back into full relief. In fact this experiment is more
strongly marked when direct vision is used rather than a stereo-
gram.

Thus we come to our final factor, the shifting of the eyes from
point to point, the focusing of attention, no matter how briefly,
upon point after point. Our eyes are constantly in motion, some-
thing of which we are unaware until we gaze fixedly at some ob-
ject and so produce eye strain! Therefore parallax must be dy-
namic if depth is to be perceived.

To result in stereoscopic perception then we must have par-
allax . . . this parallax must be multiple ... it must be differen-
tiated ... it must be dynamic . . . then and only then do we have
that stereoscopic vision which because we always have had it, we
regard as a simple and unremarkable thing.

If you wish to know just how important this is, get a black patch
and tie it over your eye and leave it there for as brief a period as
a single hour. Then remove it. You will then begin to appreciate
the fact that stereoscopic vision is something far more important
in our lives than being merely the phenomenon which makes
stereo photography possible!

These principles will serve to make the explanation of space
control more easily understood.

SPACE CONTROL. Space control is a type of trick work which is
based wholly upon stereoscopic principles and has no counterpart
in planar photography. Incidentally, nothing equals it as a demon-
stration of the full realism of stereo and of the shortcomings of
planar photography.

Visual judgment of distance is based upon parallax. To repeat,
it is possible to list and describe a half dozen other factors which
are supposed to have some bearing upon it, but the fact remains
that the actual perception of distance is solely a matter of the
parallax involved. In fact, when the parallax is such that the
resulting size is unreal or even impossible, the visual perception
is unaltered and we actually see objects in impossible size rela-
tionships, regardless of all of the other so-called "distance factors"
combined.

234

THREE-DIMENSIONAL PHOTOGRAPHY

Therefore, it is obvious that if an abnormal parallax can be
introduced, an abnormal distance will also be introduced. The
whole system of space control lies in this simple manipulation.

In the above figure, L and R represent the two camera lenses.
T is a table in the foreground and O is a living model twice as
far from the camera as the table.

The lines LA and RB are the parallactic rays which cross in
the center of the table T. The dotted lines LO and RO which
meet at O are the parallactic rays for the model.

If the two images are exposed separately, and if the model O
can be placed in position X during the right exposure, then
moved in the direction of the arrow to position Y during the
exposure of the left film, it is obvious that the parallax of the
model will be identical with that of the table. As a consequence,
the model will appear in half natural size, in that spatial position
coinciding with the tabletop. If the vertical positions of the table-
top and the feet of the model are placed in correct alignment, the
finished stereogram will show a half size, living model standing
upon the tabletop. Such is the basic mechanism of space control.
Further than this there are only operational details to consider.

In the above figure (2), a similar problem is shown, but this
time the table is in the background, and the desired result is to
show a double sized giantess seated at the table. To increase the
effect and make it more definite, a second living model may be
seated at the opposite side of the table.

TRICK WORK

235

Here rays LT and RT are the parallactic rays for the chair
position beside the table T. There is no chair there, but the large
chair will appear there. LA and RB are the parallactic rays for
the model O, who is seated midway between the camera and the
table. In making the first exposure, the model is moved to the
position X and for the second exposure to the position Y.

In the first figure the model moves to the left for the right ex-
posure and to the right for the left exposure. In the second ex-
ample, the model moves to the right for the right exposure and
to the left for the left exposure.

It might be argued that instead of moving the model in the
second example, which is somewhat difficult, it would serve the
same purpose to move the table according to the first example;
and the table is much easier to move than the model.

But consider! If this were done, the result would not be a
giantess seated at a normal table, but would simply show a normal
human being seated at a miniature table, and miniature tables
are not astonishing.

The object which is moved is the object which will alter in size,
while all other objects remain in their natural sizes.

The question will naturally arise as to why there is any size
change. A camera lens will form an image of definite size, and
that size depends upon the distance of the object. Then how is
the size altered?

In any photograph the actual size of an image, let us say the
height of the image of a six-foot man, is definite for a given lens
and a given distance of the man from the camera.

B

In the above figure (3), F is the camera film, L the lens, AA'
and BB' the limiting rays. At Mi a man's figure just fills the field

236 THREE-DIMENSIONAL PHOTOGRAPHY

but at Mz twice as far away the height of the image is just half
that of the film. So the image height in Ms is just half of that of
Mi. There is a direct relation between the actual size of the image
upon the film and the distance of the object from the lens.

M2

In the figure (Fig. 4) above, the man who occupies the full
camera field at Mi, is pushed back to Ms>, but is increased in size
as he moves backward, so that at all times his image fills the
film. It will be seen then that the height of the image upon the
film is directly related to the distance from the camera, and nor-
mally, (because we cannot actually increase size with distance)
the size on the film is directly proportional to the distance.

How r ever, the parallax positively defines the distance. If the
parallax is for 10 feet and the image size is the diminutive one
normal for 20 feet, there is no confusion in perception. You will
simply see the miniature figure at a distance of 10 feet. Parallax
is in full control of distance.

Therefore, the problem is simply that of moving the object
nearer the camera optically (by increasing parallax) while re-
taining the image size normal for the true distance (and this is
automatic). Such a thing is impossible in planar photography be-
cause we have no absolute standard of distance judgment, because
we have no definite perspective-distance perception and because
the planar photograph contains no parallax, the sole source of
definite distance perception.

However, let us return to the first part of this discussion where
it is stated that parallax is the sole distance determining factor in
stereo. Then the whole thing becomes quite clear. We leave
(Fig. i) the model in the distance, and we obtain the small size
normal to that distance; at the same time we alter the parallax to
that of a nearby object. The stereo result combines the size nor-

TRICK WORK 237

mal to the original perspective, and shows the distance normal to
the parallax used.

Let me emphasize that this effect is not one which you have
to persuade yourself to see. It leaps to the eye, just as stereo relief
did when you first saw it. Almost always, the first question asked
xvhen a space control shot is shown, is "How on earth did you do
it?" If the size reduction is only about half, and the model a
stranger, you will be asked where you found a midget to pose for
you.

But one of the most interesting forms of this control is when
a human figure between 6 and 12 inches tall has a place among
normally sized people. For example, with a little patience you can
show a man seated at a table and standing by his plate a girl
holding up a dinner knife as tall as she is!

You have seen the old, old combination shot in which a girl is
shown seated in a wineglass. It has been used over and over again
in publicity pictures, and the reaction of the laymen is a mild
curiosity as to where the huge glass was obtained. When well
done it is characterized by reality without conviction and when
poorly done there is no reaction. In stereo it is all so different.
You see the wineglass a couple of feet away with human fingers
grasping the stem, the distance is normal, the size is normal . . .
yet there sits the girl, maybe four inches high when standing, but
beyond any doubt a real, living human being! And you can do
it easily.

You will spoil some film, but the result is worth it.

You will need a mask box which has been described, and you
will mask out all but a portion of the field in the center laterally
and just above center vertically. Pose the model. It is advisable to
have the camera upon a tripod with tilt head.

Memorize the position of the model in the finder, or better yet
make a dummy camera with ground glass to mount on top the
stereo camera, using simple lenses of 35mm focal length, and out-
line the model's position with a pencil in both frames. Remem-
ber that these exposures must be made separately and the parallax
must bring the model to the distance of the glass.

Now make the glass setup, align the glass and the model by
means of the memorized position or the dummy camera and ex-

238 THREE-DIMENSIONAL PHOTOGRAPHY

pose for the glass. If you use the memory system you will be
lucky to get one good result from twenty shots, but with the
dummy camera you should get it within the first three or four.

Computation of parallax is easy, but failure often is a result of
failing to understand the importance of accuracy of measurement.
However working by computation is, as we have said, untrust-
worthy and tedious. It is much better to work under visual
control.

DUMMY CAMERA. You will need a cardboard box about 2x5
inches in size and at least 2 inches deep, of the telescoping variety
or a more permanent dummy. You will also need a piece of
ground glass the size of the box or a sheet of matte celluloid. If
you wish to make a more perfect job you can use plastic sheet for
the box. You will need a pair of simple lenses of about 35mm
focus, plus or minus a millimeter. Make holes for these in the
front of the box exactly yomm apart (center to center). (Focal
length and separation should match those of the camera you use).
Cut out two squares about 1 1/ 2 inches square from the bottom of
the box. These squares should also be yomm apart center to cen-
ter. Place the ground glass over the openings, ground side out,
and slide this into the part of the box holding the lenses. By slid-
ing the bottom into the top you can bring the dummy to focus.
You will then need to make a light framework or other connector
so that this dummy can be held in fixed relationship to the photo-
graphic camera. The only real difficulty is that the dummy must
be made to tilt forward and should have this tilt calibrated for
each foot up to 10, after which the vertical parallax can be ig-
nored. (As a matter of fact little trouble will be involved at dis-
tances greater than 5 feet.) Vertical parallax is involved only in
such alignment as the feet of the model upon the tabletop and
does not affect space control. Because this dummy is only neces-
sary when the exposures are made in different places, and as that
requires a mask box, it would be more to the point to build the
mask box and dummy camera as a unit into which the camera
could be clamped when desired. *

ANGLE SIGHTS. For the simpler type of space control, in which
the whole thing is shot from one camera position, the best control
is a pair of angle sights. These are simply plain sights, such as

TRICK WORK 239

a plastic or metal tube with cross wires fixed at one end, the other
end capped with an eyehole. A simple telescope can be made
which makes the cross wires easier to see. Two such sights are
required, and a metal strap to support them, one directly over
each camera lens. A pivot over each lens permits the sights to be
swung laterally and to be clamped in position.

Refer to Fig. i, the right exposure to be made first. The sight
over the right lens is turned until the center of the table lies
across the vertical cross wire, and the camera with sights tilted
until the far edge of the table coincides with the horizontal cross
wire.

The model O is then moved until the vertical cross wire bisects
her figure and her feet rest upon the horizontal wire, for example.
The exposure is made.

Without moving the camera, the left sight is brought to bear
upon the center of the table, and the model moved until the
vertical cross wire again bisects her figure. The horizontal -wires
serve the same purpose in both. The second exposure is now
made, and you may be assured that if the model did not change
pose between the exposures, you will have the effect you want.

It is advantageous to have a napkin or the like crumpled on
the table so the model's feet can be hidden behind them, or make
use of a dark table and place the feet in the center of the table
allowing the lighter flesh tone to override the dark top.

For best results it is advisable to have the model pose upon a
panel or sheet of metal which can be dragged into the new posi-
tion, as it is almost impossible for a pose to be duplicated with
the exactitude demanded in parallax work. If the subject permits
a bicycle, wagon, sled or the like to be a part of the picture, the
task is much easier.

We have made stereo registration easier by recording the first
pose with a Land-Polaroid camera placed just above the center of
the stereo camera.

Such work does demand some preparation, a loss of time and
film in acquiring skill and you can never count upon perfect re-
sults. However, with patience you can acquire sufficient skill to
be reasonably sure of making four hits out of five tries. Of course

240 THREE-DIMENSIONAL PHOTOGRAPHY

the cooperation of the model is essential, and it is advisable to
explain in detail just what is required.

When you, for the first time, see a living miniature, you will
probably burn up more film than you should in making these
fascinating freaks of nature. And by going a bit to extremes, you
can make use of a preserved butterfly to make a subject such as a
butterfly with 24-inch wing spread carrying on its back a ten-inch
human. Fairy tales come to life . . . and the beauty is that the
"faking" is nothing more than a warping of natural law in such
a way that the layman can never guess how it is done. All he
knows is that his intelligence is affronted by an utter impossibility
which is undeniably fact.

As we go to press we are engaged in some new work on space
control. It is obvious that because of the successive exposures em-
ployed, a single camera will serve as well as a stereo type. The
existence of reflex cameras makes alignment much easier than
with the usual optical finder. For 6x13 we have used the Rolleiflex
with satisfactory results, but in the 35mm field, for making color
stereograms to match modern stereo format, some difficulty has
been encountered.

At first we used a reflex attachment, but this involved the use
of long focus lenses, which was not advisable. More recently we
have been using the 35mm Alpa Reflex and have found it highly
satisfactory for this as well as other types of successive exposure,
35mm stereo. The extra width of the frame permits gage crop-
ping, and still leaves sufficient film for mounting to correspond
to any of the three accepted widths now in common use.

CHAPTER 17

CLOSE-UP STEREOGRAPHY

(Macro - Micro)

THERE HAS BEEN A GREAT DEAL of confusion regarding close-up
work in stereo. Indeed there have been many "laws" cited
which are not laws at all, but simply principles arising from a de-
sire to present the most attractive appearance in the finished
stereogram. For example, there are the many rules which give the
minimum working distance at practically every conceivable dis-
tance between 3 and 10 feet. In reviewing such statements, let us
consider the optics of orthostereo. The angular values of the orig-
inal direct vision, including that of convergence, are substantially
duplicated in the camera. Of course, exact duplication is not pos-
sible for several reasons, including the difference among individ-
ual interpupillaries, the greater base of several types of cameras
and the like. The important factor is that these slight discrepan-
cies exert no appreciable visual effect.

Therefore it is obvious that the actual minimum working dis-
tance is the minimum distance of comfortable direct vision, or
approximately 16 inches. It is well known that two widely fa-
vored cameras, the Realist and the Verascope, have focusing
scales calibrated to two and one-half feet and two feet respectively;
and that hundreds of owners use the camera at these distances
with full satisfaction. There is a reason for avoiding extremely
close work, but it is nothing more than an incidental limitation
imposed by the camera.

In direct vision we have learned to concentrate upon that ob-
ject which is the center of interest, and largely to ignore other
objects. This "attention" angle is very small, normally, and the
duality of other images is automatically suppressed to conscious
vision although they are essential in the subconscious perception
of the stereo image, that is to the synthesis of the stereo image.
Outside this minute angle of acute vision, visual perception ex-
tends rather widely and eventually falls off. The sharpest cutoff
we have is the diffused image of the brows and the bridge of the
nose and the outer eyelids.

241

242

THREE-DIMENSIONAL PHOTOGRAPHY

In the camera we have the sharp cutoff of the edge of the aper-
ture, and this is duplicated in the sharp edge of the mount aper-

!3

I

IB

IA

T

I A. The normal lenses produce images in which the close-up object is

recorded upon the film outside the center,
IB. When transposed, this stereogram shows the close-up images nearer

than the centers, causing the loss of a wide band at each end.
2 A. With prisms before the lenses (or prismatic close-up lenses), the
rays are so deviated that the nearby object is recorded at the film

center.
2B. This centering of the images avoids the loss of bands at sides.

ture. This is the real reason for avoiding extreme close-ups with-
out compensation. When such a close-up object is photographed,
the plane area occupied by that object is not at all duplicated in
the two images, so there is a relatively wide ' 'ghost" band at the
sides of the viewer field. If objects at a distance are also included,
it will be seen that these side bands shrink to an imperceptible
degree when looking at the distant object, but return when the
vision returns to the close-up object.

Those familiar with stereo disregard this phenomenon as in-

CLOSE-UP STEREOGRAPHY 243

evitable when both extreme close-up and distance appear in the
same slide, but there are many who regard such an appearance as
a "distortion." It is nothing of the kind, simply a phenomenon
which is part and parcel of the differential homologous distance
which in turn is the essential factor of stereo.

Thus, the decision sinks to the level of personal preference, but
in stereo there is a habit, and a bad one, of condemning every-
thing which violates personal preference, as a "distortion/' There
can be no true laws or rules which do not apply equally to all
individuals concerned. Thus the establishment of a minimum
working distance is akin to the problem of the window. Many
workers insist that the window MUST be nearer the observer
than any part of the subject; others find that having the window
in midplane actually exaggerates the appearance of relief, so there
again, as we have said, the matter is not one of rule, but one of
personal preference and nothing else.

However, if you prefer to avoid the characteristic appearance of
the close-up object, you can do so easily, but it will be necessary
to keep the background within a distance which does not exceed
the object distance by more than 5 to 10 times. Using a single
camera, you may center the object for both exposures by rotating
the camera and moving it laterally between shots. There is no
limit to the object distance in this case other than that imposed
by the working capacity of the camera used. If you use a stereo
camera, you may place prisms over the lenses which deviate the
optical axes so that they intersect at or near the object.

The background limit of five to 10 times the object distance is
maximum and should not be used if the very best results are
desired. Actually the deviation of the principal object should be
determined, and the background placed at a corresponding dis-
tance. According to one school the farthest background should
have a deviation of not less than half that of the principal object,
but we prefer to keep the background within three prism diopters
of the nearest foreground. This will permit the principal object
to be placed, for example, one diopter behind the immediate fore-
ground and two diopters in front of the background.

The difference can be illustrated by examples. Assume the ob-
ject distance to be 20 inches. According to the first method (five

244 THREE-DIMENSIONAL PHOTOGRAPHY

to 10 times object distance), this would place the maximum back-
ground at 100 to 200 inches, roughly 8i/4 to 17 feet.

In the second method, assuming the camera has a lens separa-
tion of 7omm (half base 35), we may consider 20 inches as
equivalent to 50 centimeters. This gives us a deviation of 7 diop-
ters. If we use one-half of this or 3.5 diopters, we place the back-
ground at one meter or tw r ice the distance of the object. (40").

In the third method, if the object lies at 50011 or 20 inches, the
nearest foreground should be not less than 43.75011 and the back-
ground not farther than 70011. To substitute approximate equiva-
lents \v r e have: Foreground 171/4 inches, principal object 20 inches,
background 271/2 inches. We have found such a distribution to
produce extreme close-ups which are as easy to view r as the normal
long distance shot.

The vital point is not the absolute convergence, but the restric-
tion of deviation range within one slide to a value which is not
excessive for the scene of normal distances. A total range of three
diopters in normal slides equals infinity dow r n to 1.16 meters, or
just under four feet.

But this can be misleading. The distance of 1.16 meters is the
distance equivalent of three prism diopters for a camera with
jomm lens separation, but for average human vision, three di-
opters indicates only about 42 inches. Therefore the range in-
volved in the close-up is that which would appear in a slide to
embrace distances from 451^ inches to infinity, but which is
equivalent to direct vision from 42 1/ inches to infinity (provided
the human interpupillary is the average 65mm).

It can be seen from this that if you wish to make exact compu-
tations, you will have to consider both the separation of the lenses
of your camera and your own interpupillary separation. As a mat-
ter of fact the examples given are not precise because the camera
used actually has a lens separation of about Gg-f-mm, but to make
computation easy, the round figure of 70 was substituted.

If you make any such computations you will be surprised at
the diversity of result, (as the three inch difference in the three
diopter limit between camera and eyes). In actual experiment you
will be equally surprised to learn the great divergence from theo-
retical limits which can be made without the introduction of any

CLOSE-UP STEREOGRAPHY 245

visual distortion. In fact most stereographers use the 65mm base
as normal for all computations regardless of the camera used. The
only exception is that made for extended or diminished base as in
hyperstereo, and in micro work where a fraction of a base is used.
For anything between 62 and 70, the practical result is satisfactory.
Rotary parallax. The usual stereo technique involves the use
of parallel lens axes. However this rule is violated often enough
to make the exceptions interesting. When parallel axes are not
used, convergent axis are substituted so that the camera positions
resemble those of the converging eyes. The explanation as to why
this convergence is inferior to parallel axes is given elsewhere in
this book.

We have seen that physically the convergent camera and the
rotated subject are essentially the same, so that the convergent
camera becomes essentially a phase of rotary parallax.

In speaking of the convergent camera, it has been assumed that
the camera is used for successive exposures and bodily rotated be-
tween the exposures to converge upon the subject. However, with
the rigid stereo camera and simultaneous exposure the technique
would seem to be impossible because the stereo camera will not
bend. (It may be of interest to know that a convergent camera
was made commercially which was adapted to the binocular micro-
scope objectives made for the Greenough type of stereoscopic
microscope.) It is possible however to adapt rotary parallax or con-
vergent axes to the conventional stereo camera. This is done by
bending the light beams rather than the camera body.

In the paragraph above supplementary prisms are indicated.
In practice the magnifying lens commonly used for close-up work
is combined with a prism in a single piece of glass. Thus when
such a supplementary is applied to the stereo camera lens, there is
a simultaneous change in focal length and a deviation of the axial
ray. The prism power is so adjusted to the magnifying power that
when the object is focused it will also be centered in the two films.
Such supplementaries are made in three close-up powers and a
fourth weak prism is supplied for those who wish correction at
distances less than 10 feet but within the focal capacity of the
unassisted camera lens.

As is true of all "rotary" parallax, objects at infinity will be

246 THREE-DIMENSIONAL PHOTOGRAPHY

abnormally displaced, due to the crossing of the axes at the object,
but this does not interfere if the background is kept reasonably
near.

However, there is one factor which must be considered, one
which varies with the stereo skill of the individual. When the
object distance is small, 16 inches or less, and when it has con-
siderable depth, there will exist a condition of excessive difference
in parallax as the eyes move from one part of the object to an-
other. This is known in stereo work as an excessive depth of
parallax, to some extent analogous to depth of field.

When there is an excessive parallax depth, the spectator is
conscious of the change in convergence in looking from the near-
est to the farthest point of the object. Some, particularly those
who are new to stereo, find this actually painful. Those who have
had more experience in stereo viewing and have developed some
stereo skill do not find it uncomfortable, but the sensation re-
mains, including that of a definite time interval in moving from
one position to the other. Whether this is objectionable or not
and the degree to which it is objectionable, depends upon the
individual. As a rule, however, it is advisable not to make full
base stereograms at a distance of less than 12 to 14 inches, even
when prisms are used to eliminate the ghost side-images. For
nearer distances, macro techniques are advisable.

MACRO. The stereo camera is easily adaptable to macro photog-
raphy, and with suitable supplementaries consisting of prisms
with positive correction ground in, it is possible to use the normal
stereo camera at distances as close as 10 inches. This is approx-
imately 25omm, and as the usual focal length of the camera is
35mm, the ratio will be approximately 1:7. However it must be
remembered that the parallax remains normal. That is, if the
object is at 8 inches, the parallax will be that of an object actually
viewed at 8 inches; and, as explained above, the relief will appear
to be exaggerated and parallax depth will be excessive.

Many workers object to relief which results from any parallax
greater than six or eight feet. Personally the writer does not agree.
He finds the parallax of 16 to 20 inches quite acceptable, and for
many objects even preferable to a compressed relief. It is advis-
able to limit this technique to a minimum of 12 inches.

CLOSE-UP STEREOGRAPHY 247

Parallax increases as distance decreases. "Greater parallax" re-
fers to nearer objects.

SUPERMACRO. For that type of macro shot which merges into
low power micro, it is ordinarily necessary to use two cameras.
Zeiss make a stereo ophthalmic camera for making macro-stereos
of the retinal surface, but aside from such complex and costly
instruments, the work requires two cameras or a single camera
used to make successive exposures.

The two cameras should naturally have sufficient bellows capac-
ity to make possible images at least i : i in size, and it is preferable
if this can be increased. The ideal equipment is bulky, because
it consists of two cameras of great bellows capacity equipped with
lenses of long focal length. Eight-inch lenses on cameras of some
go-inch bellows capacity are excellent, but the lens boards should
be small, if a diagonal is to be used.

The Diagonal. This is some kind of beam splitting device. It
may be a normal beam splitting cube, but unless it has great size,
the camera lenses must be in extremely close proximity, and there
is only a minimum base available. For this reason, the beam
splitter is usually restricted to special instruments used for one
type of subject only.

A better device is a 50-50 mirror. This mirror transmits 50
percent of the light and reflects 50 percent (actually there is a
slight loss in transmission, but due to the coating used this may
be ignored). The better mirrors are dichroic, however, and in
color work this is somewhat objectionable as there is a pink cor-
rection applied to one and a blue to the other. It is said that this
slight color difference is merged visually by the observer, but the
writer has found a discernible effect.

However, the mirror does permit the base to be altered at will
and a 1 1/9x3 inch mirror will permit from zero to almost 2 inches
base.

In using the diagonal, the cameras are positioned at right
angles. One of the cameras accepts the image through the mirror
(nominally a direct image), and the other accepts the image re-
flected from the surface of the mirror at right angles. Displacing
either camera laterally will alter the base, but the motion is

248

THREE-DIMENSIONAL PHOTOGRAPHY

DOUBLE DIAGONAL

3. The beam splitter is used with two cameras in macro work when the

base is very small. The setting here is zero base.

4. For greater stereo bases the 50-50 mirror permits one camera to be

moved sidewise to adjust base. Drawing shows zero base.

5. When small bases are not required, the double diagonal pictured
here is easily operated and gives excellent results. The base is changed

by moving one camera away from the other.

6. When the camera is used with convergent rays (objects centered), the
actual relationship is that shown above. Camera is moved from C to C'

between exposures; object at O.

7. A conventional "darkfield" diaphragm used in microscopy.

8. A stereo dark field diaphraghrn corresponding to Fig. 7, but as

used for stereomicrography.

9. A bright field diaphragm, complementary to that of Fig. 8.

usually given to the camera which lies normal to the object, and
that is the camera which receives the transmitted image. The mir-
ror, of course, is placed between the lenses at 45 to both o them.

Double Diagonal. This is the method favored by most workers,
although it has the disadvantage that zero base cannot even be
approached. Each camera has a right angle prism attached to the
lens, and the cameras point directly at each other. Thus BOTH
cameras receive a reflected image. However the smallest base is
one which is about 50 percent greater than the lens aperture, the
exact amount depending upon the lens mounting, the angle of
field and so on.

CLOSE-UP STEREOGRAPHY 249

The lens mounts make it impossible for the lenses to work in
approximate contact with the prism, so relatively long focus
lenses are necessary to increase working distances. This, in turn,
indicates long bellows. The method is most easily applicable to
moderate degrees of macro, that is to a size ratio of about 1:1.
For enlargements up to about 2:1, the double diagonal may be
used with small lenses and small prisms, but for extreme work the
prism interferes with the orientation and lighting of the object.
Thus, for greater magnification it is necessary to work with a
single camera and successive exposures.

ROTARY DISPLACEMENT. The argument is often made that be-
cause the human eyes converge upon an object, the stereo camera
should also be made to do the same. We have explained this
fallacy elsewhere, but when the object is limited to a group of
closely associated planes, the method is quite satisfactory.

In Fig. 17-6, the object lies at O and the two camera positions
are indicated as C and C'. It is easy to see that what actually hap-
pens is that the camera is moved along the arc XY even though in
practice the camera is moved laterally and then turned to bear
upon the object. What happens is that the camera has two posi-
tions, the difference between which is equal to the parallax used.

When this is considered, it will be seen that, insofar as the ob-
ject is concerned similar results will be obtained by rotating the
object through the same arc. The object is often easier to move
than the camera, and this is a method widely used in higher power
macro and in micro stereo. It will be noticed, of course, that
when the camera is moved, the parallax involves all objects with-
in the field, while with object rotation only the object itself is
affected. Thus, with camera rotation the effect is closely similar
to that obtained by normal stereo displacement, but when the
object is rotated, only the object will have relief and the back-
ground and surroundings will be astereoscopic. Therefore, object
rotation is limited in most instances to a single object mounted
upon a neutral background.

MICRO. When the image is some ten times the size of the ob-
ject or more, we enter the field of photomicrography. This as-
sumes the use of a microscope in connection with the camera. In
fact, the fields of macro and micro overlap. It is possible to ob-

250 THREE-DIMENSIONAL PHOTOGRAPHY

tain many magnifications with macro methods which can be ob-
tained with the microscope.

There was a time when photomicrography involved the use of
only the objective of the microscope, and to that extent it was no
more than an additional degree of macro. However, modern
photomicrography assumes the use of the complete compound mi-
croscope, although at times the visual ocular is replaced by a spe-
cial photographic lens system.

Therefore, it is impossible to draw a sharp line between macro
and micro so far as size is concerned. We shall, for that reason,
consider the micro field to include all conditions in which the
combination of camera and compound microscope are used. It
may be added that with very few exceptions, the camera is the
normal, single, photomicrographic type and not a dual stereo
camera. This, of course, indicates the necessity for making suc-
cessive exposures a technique which is classic in stereomi-
crography.

THE MICROSCOPE. Photomicrography, and consequently stereo-
micrography, is hardly popular among amateurs. Probably the
reason is that the compound microscope must be mastered before
really good results can be obtained. The possibilities can be
shown when it is stated that among those who use the microscope
daily in their work, not one in one hundred has ever seen a really
well defined microscopic field. There is no instrument in common
use, not even the camera, which is so universally abused by its
users. The reason is not hard to find. Anyone can buy a camera,
shoot one roll of film and get results good enough to make vaca-
tion souvenirs; anyone can be shown how to focus a microscope
and be taught to handle it well enough to obtain passable results
in a few days, so why go further?

As a matter of fact, it takes months or years of practice before
one can use the camera with complete confidence and competence,
and it takes at least twice as long to learn correctly to manipulate
the compound microscope.

Among other elementary things, one must learn how to illumi-
nate, how to control the cone of illumination, how to set the tube
length, how to compensate adjustable objectives and how to focus
the condenser. If one is interested in making photographs, the

CLOSE-UP STEREOGRAPHY 251

apochromatic objective is just as essential as is the anastigmat
upon a camera.

Volumes larger than this whole book have been written about
the correct manipulation of the microscope, and the ground has
not been fully covered. Obviously, we cannot be expected to cover
that ground in a single chapter. We must, therefore, fall back
upon purely stereoscopic technique, leaving the purely micro-
scopic and the purely photographic techniques to be learned from
other sources. We shall, then, assume that the reader has, or can
acquire from other sources, the necessary data concerning these
techniques.

LOW-POWER STEREOMICROGRAPHY. The lower magnifications
are obtained without the use of the compound microscope, al-
though the conventional micrographic camera is used. This, of
course, is really a variation of macro technique. This is a camera,
usually mounted upon a vertical bed, with the lens pointing
downward. Beneath the camera there should be some kind of
stage or raised platform to hold the subject. This stage should
have a surface plate of glass, plain and transparent. It should be
so arranged that black or white or gray underplates can be placed
beneath the stage. There should also be a swiveled mirror be-
neath the stage for providing strong transmitted illumination.
While these things may be purchased, some of the best outfits
have been either homemade, or have been based upon a com-
mercial unit to which* homemade accessories have been attached.
A convenient film size is 314x41/4' although anything between
214x314 and 5x7 may be used.

A great deal of work has been done in this field during the past
several decades, but naturally most of it has been in black and
white. Because the emulsions and the processing is unusual, any
discussion of the work has heretofore necessitated a prolonged
treatment of these subjects. Fortunately, stereo is almost wholly
color work, and as color is just as essential and just as attractive
in this field as in normal stereo, we will not give any space to
processing or to a discussion of the various special emulsions used.

It will be necessary for you to establish your own exposure data.
This will depend upon the light source used as to color, intensity
and distance and upon the color film you use. Remember that

252 THREE-DIMENSIONAL PHOTOGRAPHY

sheet color film usually has a slightly different color balance than
Kodachrome A. The easiest way is to find some panchromatic
emulsion which has a sensitivity two or three times greater than
color film. This is used to determine the exposure by test, and
from that data you get the color exposure. The first color ex-
posures should be made with i/ and 2X the computed exposure
as well as the computed one itself. These three will give you an
accurate check upon your computation. If you change your light-
ing setup, you must change your exposure. Meters can be used,
but you must learn how to use the meter for such work, particu-
larly how to compute the increase due to the extremely small
effective aperture used.

However, despite these minor difficulties, color is infinitely
preferable to black and white.

HIGH-POWER STEREOMICROGRAPHY. It must be borne in mind
that the term "high-power" is relative only, and actually refers to
those stereomicrographs made with the aid of a compound micro-
scope. Actual high-power stereomicrography (200 diameters and
more), is disappointing because the depth of focus of the ob-
jectives used is so slight, and because most specimens have so
little thickness. It is not at all unusual for the microscope speci-
men to be cut to a thickness of two or three microns, and for many
types of tissue five microns is the maximum. As this equals
roughly 1/5 of 1/1000 of an inch, it can be understood that there
is very little depth to be revealed.

Stereomicrographs can be made of these subjects, but such work
requires great care, and at best the relief is so slight that results are
disappointing. Therefore, even in the ' 'high-power " field, the
magnifications of one hundred to two hundred diameters are re-
garded as the useful maxima, although these powers are ordinar-
ily referred to as "low-power" in ordinary microscopic termi-
nology.

As we have said, we have no space in this volume to discuss the
technique of microscopy, and if the interested reader has no
knowledge of the subject, it is suggested that he obtain a reliable
reference book ("The Microscope" by Belling is one of the most
practical), and acquire a working knowledge of microscopic tech-
nique before undertaking photomicrography.

CLOSE-UP STEREOGRAPHY 253

The technique of conventional photomicrography is followed
throughout, with the exception of those steps introduced to pro-
duce the desired stereo relief. Assuming the reader can produce a
good planar photomicrograph, we shall consider the methods used
in obtaining the stereo relief.

Stereomicrography is closely related to ordinary photomi-
crography, just as normal stereo is closely related to normal pho-
tography. The same equipment is used and the fundamental
methods are the same. Parallax may be obtained either by lateral
displacement of the object, by object rotation or by control of
illumination. The last method renders a peculiar appearance
which could be called pseudostereoscopic if that term were not
already used to designate a different appearance. There is a
definite relief. The relief is associated with true relief, but does
not accurately reproduce it. False relief is sometimes introduced.
Most important of all, the effect obtained by lighting control
falsifies the true appearance because in stereo, particularly in
stereo close-up, there is a difference in the lighting caused by the
different angles of reflection to the two eyes, an effect which we
call illumination parallax. Just as false stereo parallax will exhibit
some kind of relief (spurious), so will false illumination parallax
produce spurious relief.

Therefore, it is advisable to use illumination control with re-
straint and only as an adjunct to physical displacement.

Lateral Displacement. Due to the great magnification, care
must be used in making the displacement or the prints will have
to be carefully aligned later. The most important thing to remem-
ber is that the displacement is lateral and only lateral. If the
object is moved diagonally trouble will ensue.

Some kind of guide bar against which the slide may be moved
is practically essential. Fortunately most modern instruments are
equipped with a mechanical stage. If this is set to traverse the
line of displacement, all that is necessary is to move the stage. The
calibrated scales make it easy to move the object through just the
distance required.

The motion is controlled by examination of the focusing screen,
otherwise the displacement might move the object out of the
field. Also because in both exposures the object lies toward the

254 THREE-DIMENSIONAL PHOTOGRAPHY

edge rather than the center of the field it is highly advisable to use
apochromatic objectives with special photographic (flat field) ocu-
lars. If this is not done, the curvature of field will introduce image
distortion which may be serious if some parts of the object ap-
proach the field edge. For this reason, and because the micro-
scope field is very shallow, it is preferable that rotary displace-
ment be used.

Rotary Displacement. Rotary displacement is not difficult with
the microscope. An ordinary slide may be rotated by propping
the end upon a match stick, with the opposite ends being raised
for the two exposures. However, this is too crude for good work.
Some English microscope makers (e.g. Watson) make stereo stages
in which the slide lies in a cradle pivoted in the optical plane
(surface of the slide). A screw adjustment is provided for con-
trolled motion. This stage is clamped to the regular microscope
stage, and rotation obtained by tilting the stage between the
screw-set limits between exposures.

The writer uses a somewhat different setup for this work. The
microscope used is a petrographic instrument with high stage
clearance. Upon this is mounted a petrographic "universal stage,"
which permits the object to be moved in all directions. Thus the
final orientation so difficult to obtain when mounting a speci-
menis obtained and the position fixed by locking the stage rings
involved, and leaving the main axis free. The required tilt is then
obtained by using the main axis, which is calibrated in steps of i/
degree. The only objection is that such a setup is far too costly for
any but the serious worker.

It might be added that the petrographic microscope is used
solely because it has working distance to accommodate the uni-
versal stage. Polarization is not normally employed, and apochro-
matic objectives from a biological instrument are used.

Many biological microscopes can be raised high enough to ac-
commodate the universal stage, provided objectives are used
whose focal length does not exceed gsmm.

The illumination is provided by special "light funnels" which
concentrate the beam upon the surface of the object. Most ob-
jects transparent enough to be used with transmitted illumination

CLOSE-UP STEREOGRAPHY 255

have insufficient relief for stereo. It is obviously difficult to re-
produce a relief which does not originally exist.

Another factor must be considered when the object is "mount-
ed" by being sealed in glycerine, balsam or the like. These sub-
stances have a greater refractive power than air, so the apparent
relief is diminished, sometimes to the extent of 2/3 of the true
depth. When making stereos of mounted specimens the rotation
should be increased by at least 50 percent to compensate for this
loss, if image definition will permit.

The tilting stage is placed upon the microscope stage and fixed
to it. Using the centering adjustment of the main stage, and the
adjustment of the auxiliary stage together, center the tilting stage
so that the crosslines are centered in the microscope field, and
remain centered when the main stage is revolved.

Set the main stage so that the crosslines of the tilting stage are
accurately "up-and-down" and "right-and-left." Focus upon the
crosslines with the objective to be used. Tilt the stage from side
to side. Note that the vertical line remains motionless and in
focus. If it seems to sway to one side or the other or to go in and
out of focus, it is not correctly oriented. Readjust until the tilting
does not affect the position of focus of the vertical line.

Observe the lateral crossline. Note that it rapidly loses defini-
tion as the stage is tipped. Tip the stage until the good definition
and diffusion are about evenly distributed along the line. Near the
center the crossline will be sharp, but halfway to the edge of the
field it will begin to soften, and at the edge of the field it is
definitely diffused. Note the angular displacement as shown upon
the scale. Now tip the stage in the opposite direction an equal
amount and see if the distribution of focus is the same. If it is not,
reset the stage to the correct diffusion and read the scale.

If, for example, the right tilt is 6 degrees and the left tilt is 5
degrees for equal lateral diffusion, the true "zero" would be the
mean of +6 and 5, or +0.5. Then if you wish to make use
of a six-degree convergence you would add 3 degrees plus and
minus to the true zero to obtain the two stage settings, 3 + 0.5 =
3.5 and 3 + 0.5 = 2.5, so your settings would be 3.5 degrees
right and 2.5 degrees left. With this preliminary adjustment, you
are ready to proceed with the photography.

256 THREE-DIMENSIONAL PHOTOGRAPHY

In macro work it is customary to use tilts of from four to six
or even seven degrees at each side of zero, but in micrography
with the microscope the amount of tilt is limited by the depth of
field, so that it will often be necessary to limit the tilt to as little
as two degrees each side of zero, or a total convergence of four
degrees.

The microscope field is always considered to be at the standard
reading distance corresponding to a total convergence of some
nine degrees, but if we are limited to a four-degree total we place
the object stereoscopically at the equivalent of some three feet or
a little more, which means a relatively flattened relief, while the
flattening effect of any mounting medium increases the flattening.
Thus there is a definite stereo limitation imposed by increasing
magnification. The greatest tilt allowed by the limits of diffusion
imposed by the lateral line diffusion should be used.

Illumination Control Illumination control, as developed in
the laboratories of the Stereo Guild, is based upon the use of a
variation of the Zeiss stereo diaphragm which is used beneath the
condenser. It might be added that it is important to use a good
achromatic condenser and to be as careful of condenser focus as
of objective focus. (Figs. 17-7, 17-8, 17-9.)

The Guild controls consist of a substage diaphragm in which a
90 degree sector is open. This sector does not extend clear to the
center. In the full illumination (bright field) the solid central
disc is about gmm diameter, but when used for dark field, the cen-
tral solid disc is of such size that the full field of the objective is
blocked. In short, this diaphragm is just like the usual dark-field
diaphragm, but instead of having almost full annular openings it
has only the 90 degree opening. Similar but smaller diaphragms
are made to be inserted just above the rear of the objective, inside
the tube. The position of the tube sector in relation to the con-
denser sector is reversed for changing from bright to dark-field.

Although some improvement seems to be apparent in some in-
stances, the value of such diaphragms is questionable owing to the
great difference in lighting effect in the two images, far greater
than is normal for binocular vision.

In general, in all macro and micro stereo, as in normal photog-
raphy, the photographic technique is not altered. The stereo ele-

CLOSE-UP STEREOGRAPHY

257

ment lies in making two exposures of the object with parallactic
differentiation produced by alteration of physical relation to the
camera lens.

DEGREE OF DIFFERENTIATION. In all of these methods there
arises but one factor of importance; the computation of the degree
of change. In general, it may be said that the base changes in
proportion to the distance of the object from the lens, but any
such basis, although it is widely used, is at best vague. In normal
stereo we use a fixed base for distances between about five feet or
less and infinity. Thus the first question which arises is whether
you will use infinity or five feet as your reference base. That, in
itself, introduces too great a variation. Therefore we make use of
a definite reference base, and this is derived from the value of the
parallactic angle.

In any anortho stereo, the first question to be decided is, "At
what distance shall the object appear?" You will keep in mind that
the true distance factor in all stereo is the parallactic value. There
is ONE and only ONE parallactic value for any given distance.
Thus, when you decide the distance at which you wish the object
to appear, you have this value. Ordinarily we should choose the
arbitrary "reading distance" of 16 inches, but because 15 inches

M

10. The micro arrangement for lateral displace-
ment. Object S is displaced first to O, then to O f
for the two exposures. XY is the plane of motion

and OBJ the microscope objective.

1L The "half angles" ordinarily used in stereo

computation as explained in the text

258 THREE-DIMENSIONAL PHOTOGRAPHY

is very close to a half angle value of 5 degrees, we shall adopt
it. Therefore, when you wish the object to appear to lie at 15
inches from the eye, the deviation on each side of zero will be
5 degrees. In the figure, A and B are the eyes and O an object at^
any distance. Assuming the distance AB to be 65mm, as is usually
done, than AB/2 will be the half base, and by marking the center
of AB by X, we have either AX or BX as the half base. Then
AOX or BOX will be the visual half angle or the parallactic value
for one eye (for 15 inches this will be very nearly 5 degrees). By
solving the problem for any distance XO, the parallactic value is
obtained which will permit you to place the object at any de-
sired apparent distance.

It is inadvisable to make use of any value for XO, which is less
than eight inches, as the depth of parallax may be disturbing to
fuse.

Using the 5-degree value, a tilt stage will be inclined to 5 degrees
each side of center for the two exposures, or a total of 10 degrees.
If the specimen is "permanently mounted/' the swing will be
about 7.5 degrees each side of center.

For lateral displacement, the distance moved is one which will
correspond to the length of the short leg of a triangle of 5 de-
grees apical angle and whose long leg has a length equal to the
focal length of the microscope objective used.- This is true be-
cause when a microscope is correctly focused, it is focused for in-
finity and the lens distance is actually the focal length. (The
free working distance is less because the optical center lies with-
in the lens.) Thus with a i6mm objective, the long leg RS
would be a i6inm. The angle ORS is 5 degrees (or 7.5 for a
mounted object), and the motion through OS is the desired dis-
tance. Remember that OS is the distance from center. O' is the
second position of the object and OS O'S.

LIMITATIONS. There are always limitations to be considered
and in micro work these are somewhat severe. For example, when
a stage is tilted, the sides of the field are out of focus because one
side is too near the objective (inside focus) and the other is too
far from it (outside focus). Therefore, the amplitude of swing is
limited and the greater the magnification the less is the tolerance.
Even with the i6mm objective, one commonly used for stereo

CLOSE-UP STEREOGRAPHY 259

micro, it has been found advisable to obtain a special mount
which incorporates a small iris.

It has been argued that this fault proves the superiority of the
lateral displacement, but the loss of definition and distortion at
the edges of the field, even with apochromatic objectives and flat
field oculars, is just about as bad as the violation of field depth
limits.

Secondly, high-power stereo, on the order of 1000 diameters, is
almost worthless because the field is too shallow to permit the
variations in depth which make up relief. The work then de-
scends to the level of making a stereogram of a printed page. If
the aperture of the immersion objective is lowered to a degree to
make possible any penetration worth while for stereo, the loss of
definition is too great. In our work, while we have made stereos
at relatively high power, we have found that a good stereo
is rarely obtained at powers above 200 diameters.

SUMMARY

We have deliberately refrained from trying to give any details
of normal photomicrographic technique as that is a field in itself;
but we wish to emphasize the fact that normal photomicrography
should be mastered before working with stereo. Stereomicro is
nothing more than the stereo technique superimposed upon nor-
mal photomicrographic technique. There is nothing new in it for
those who know both stereo and micro, other than the manner of
the control of parallax. The fundamentals which have been given
are adequate to enable the experienced photomicrographer to
adapt stereo to his work. The stereographer who would add photo-
micro to his work will find it highly advisable if not essential, to
master photomicrography first . . . and we would" also like to add
that before venturing into photomicrography he should become
thoroughly familiar with the technicalities of microscopy.

It is true that a box camera upended over a microscope ocular
may occasionally provide an image which bears a remote re-
semblance to a microscopic field, but that is a far cry from being
a good photomicrograph.

If the micro field is not as clear and sharp and distinct as nor-
mal objects seen by direct vision, the microscope is not in adjust-

260 THREE-DIMENSIONAL PHOTOGRAPHY

ment and cannot give a truly good micrograph. Also remember
that the condenser focus is NOT provided to enable you to con-
trol the intensity of illumination. It should be focused as care-
fully as the objective and stopped down just enough to cut into
the full cone for the objective used. Then use neutral filters or
crossed polaroids to control intensity. You will never get a good
micrograph with the condenser stopped down to a pinhole. In
short you must KNOW the microscope and use it correctly. Then
you must learn to make photomicrographs, a very easy step if you
have photographic experience and have truly mastered the mi-
croscope. Then you add stereo. Assuming you are a competent
photographer, it may take you three months to a year to really
master the microscope, but from 2 to 10 days after that should
give you fair ability in photomicrography and stereomicrography.

For those who are competent in both photography and micro-
scopy, the little volume "Photomicrography" by Eastman Kodak
Company, will prove adequate for the initial attempts. It is un-
fortunate that really capable instruments are so rare and so costly,
but many amateurs have done excellent work with only the usual
laboratory microscope. However, it is a serious error to try to
start with an instrument which has no condenser, and if at all
possible the condenser should be achromatic. For the benefit of
photographers who are inclined to laugh at "achromats" we will
add that the achromatic condenser is the finest made. It is desir-
able, but not so essential, that the objectives be apochromatic. In
any event, obtain the very best instrument which is possible.

In closing we should like once more to remind the beginner to
learn how to use the condenser. It is relatively easy to see when
focus has been obtained, but the condenser is too often ignored or
misused, and very, very rarely used as it is supposed to be used. It
may help to remind the reader that it is impossible to obtain a
really excellent photomicrograph unless the condenser is given
fully as much attention as is given to the objective focus.

Second only to this in importance is the nature of the illumina-
tion. Simply to see an illuminated field is far from 'enough. The
light must be uniform, it must be accurately centered, and the
image of the source must be brought to its focus in the correct
position, substage iris or slide, depending upon the type of illu-

CLOSE-UP STEREOGRAPHY 261

mination used. The difference between an amorphous smudge
and a crisp, brilliant photomicrograph lies in a well controlled
condenser, condensing a beam of critically adjusted light.

CHAPTER 18
HYPERSTEREO

HYPERSTEREO WAS BRIEFLY MENTIONED in an earlier chapter,
but because of its importance and because the subject has
been the cause of so much bitter argument, it will now be dis-
cussed in detail.

It will be remembered from the preceding discussion, that par-
allax is the determinative factor in the appearance of distance. The
greater the parallax, the closer the subject appears to be. Al-
though not in accord with the classic theory of stereophotography,
parallax is the sole determinative factor, any telescopic effect re-
sulting from the use of long-focus lenses being insignificant. This
fundamental fact was recognized years ago by Helmholtz who
constructed an instrument closely similar to the stereo reflector,
but made for direct visual use. This instrument he called the
telestereoscope. It gives the absolute distance effect indicated by
the proportion of its base to the normal base. Thus if such a re-
flector is made with a 65011 base, objects will appear to be 10
times nearer than when it is not used. However, when used for
direct vision, we have no telescopic means to enlarge the per-
spective value of the scene, therefore it is obvious that the objects
viewed will appear to be small, only one-tenth natural size in our
example. The whole thing will give the effect of looking at an
accurate miniature model very near at hand.

This fact has led, together with sound stereo theory, to the state-
ment that the use of hyperstereo produces a stereogram which has
the appearance of being a miniature model, a short distance away.

That argument is based upon a fallacy which is not obvious
until it is closely studied in the light of what little we know about
subjective reactions.

In the instance of the direct vision telestereoscope, we have the
direct comparison with the natural aspect made possible by look-
ing through the instrument, then looking at the same scene with-
out it. The comparison is made with the minimum time interval,
and for this reason, the appearance described can be observed.

The stereogram offers no such comparison with the natural

262

HYPERSTEREO 263

aspect, and for that reason we can only guess at the normal appear-
ance. Our ability to perceive differences among the distances and
the relief of objects in space, is a most acute one. Because this
sense is so well developed, most stereographers assume that it
applies as well to size perception when it is not based upon direct
comparison. This is not true. For example, if, in a single stereo-
gram, two objects are shown, one of which lies only an inch or so
farther from the camera than the other, that difference can easily
be perceived, even though the objects lie at about 10 feet from
the camera. But if one stereogram shows an object at 8 feet and
another shows a similar object at 10 feet, it is difficult to differen-
tiate between them when viewing the stereograms in succession.
This inability to carry over standards of comparison from one
period of time to another is familiar to those who try to match
colors by memory rather than by direct comparison.

The perception of relief and distance is purely a relative thing
and not at all absolute. Therefore if in a stereogram the relative
positions and proportions are normal, the lack of absolute dimen-
sions cannot be recognized. The sole exceptions are those in-
stances where the skilled stereographer has some extrinsic factors
to guide him in the perception of the departure from absolute
size and distance.

Because of the great confusion which exists regarding this sub-
ject, an experiment was performed under highly exaggerated con-
ditions. The subject was a roadway lined with trees and shrubs
which gave a continuous recession of planes. The nearest object
within the scene was the foreground only twenty feet away, and a
limb of a tree at substantially the same distance. It must be re-
called that the stereoscopists who conform to classic rules insist
that 10 feet is the minimum distance for using normal base sep-
aration, and that anything nearer must be made with reduced
base. Therefore 20 feet represents a fair, normal near point for
the stereogram.

This scene was stereographed three times, using normal, two-
times normal, and four-times normal base. The four-times base
is that which would be used theoretically for a subject about one
mile distant.

The stereograms were shown to a group of people, laymen, who

264 THREE-DIMENSIONAL PHOTOGRAPHY

were familiar with the original of the scene, and who are accus-
tomed to viewing stereograms, although they were not stereogra-
phers. The usual comments upon the attractiveness of the scene
were made, but no criticism was offered. Then they were asked
to point out any visible differences. None did so. Then it was
explained that the distances within the picture were condensed
by two, and by four, compared with the standard. Still only a few
even tried to classify the three prints, and all who tried failed. It
would appear quite obvious from such experiments that the seri-
ous distortion of the wide base, and the insistence upon the nar-
row base have little factual foundation.

It must be admitted that the writer was astonished by the result
of the test. It seemed incredible that the four times base should
not have been seen to be abnormal. Experienced stereographers,
of course, would have detected such gross exaggeration.

If there is no foreground object which requires more than about
six to eight prism diopters convergence, and if the mounting is
based upon homologous separation rather than the standard cam-
era negative format, the really valid arguments against hyperstereo
disappear. It might be added that every competent stereographer
watches both factors in every print he makes.

FLAT STEREOGRAMS. Every year there are hundreds of stereo-
grams made with excellent stereo cameras by competent stereog-
raphers which are not at all stereoscopic. These are .views of dis-
tant scenes, particularly in mountainous country where the sub-
ject matter is of such huge dimensions that to get it upon the film
at all demands a great distance between subject and camera.

When such a subject is stereographed under normal condi-
tions, the distant subject lies beyond stereo infinity and becomes
just as flat as in a planar photograph. To compensate for this, the
usual procedure is to include some object in the immediate fore-
ground. The classic foreground subject is a human figure looking
toward the subject. Of course this figure, by its position and its
relief, takes upon itself the major importance of the composition
so that the true subject is not only flat, but sinks to the level of
an accessory to the truly insignificant foreground.

The only solution is to use such a base that the true subject
takes on realistic depth, and is then permitted to make up the

HYPERSTEREO 265

pictorial composition without the use of the irrelevant foreground
figure.

This is the practice which is condemned by classic stereo pro-
cedure, wholly upon unsupportable grounds, namely that the
rules of orthostereo are violated and that the subject is presented
as an insignificant miniature instead of in its original grandeur.
Neither argument will stand searching inquiry.

In all fairness however it must be admitted that the normal
stereogram exhibits all of the relief in such distant scenes that
would be perceptible to the eye if the scene were actually viewed.
Because we are accustomed to "read" into distant scenes the re-
lief indicated by the extrinsic depth factors, the normal stereo-
gram is about as satisfactory as direct vision. However, the pur-
pose of the hyperstereogram is to reproduce the scene in a better
manner than it could be seen by direct vision. Any such attempt
to improve upon nature is condemned by many, but any such
reasoning would rule out the extremely attractive field of photo-
micrography, so the objection is hardly tenable.

As for the laws of orthostereo, it is quite true that these are vio-'
lated, but it is significant that those who are the most bitter an-
tagonists of hyperstereo upon these grounds habitually violate
the principles of orthostereo even when there is no reason for
doing so. Although the necessity for having the focal lengths of
the lenses of both camera and viewer identical is recognized, if
orthostereo is to be achieved, this is rarely found in practice. Even
when allowance is made for permissible variation, it will be found
that practice violates the principle to an extent far beyond the
permissible. Specifically, there is a widespread habit among stere-
ographers, of using a 6 by igcm camera with 3-inch lenses and
then mounting contact prints from these negatives for viewing in
the Brewster stereoscope which ordinarily has lenses whose focal
length is between 6 and 8 inches. Fortunately, the widespread
use of 35mm cameras and associated viewers is eliminating this
abuse.

The fact is that, as shown by repeated experiment, violations of
the ortho principle of considerable degree have no effect which is
obvious when the stereogram is viewed. This is unfortunate as

266 THREE-DIMENSIONAL PHOTOGRAPHY

far as the law is concerned, but as it is true, the insistence upon
orthostereoscopic conditions should be limited to those special
instances where it is highly desirable, as in scientific record. Ob-
jection should be limited to those instances in which the viola-
tion has resulted in a space distortion which can be seen when the
stereogram is viewed.

As for the second argument, that of the nearby miniature, or
the giant's eye view, it, too, is far more important in theory than
in practice. As has been stated, our perception of absolute size is
as undependable as our perception of relative size is acute. The
optics of stereo will prove conclusively that the subject made by
hyperstereoscopic methods is shown as a miniature near at hand.
However, when the stereogram is viewed, this aspect is not seen
except by those stereo experts who know the stereogram is a
hyperstereo, and who see what they have been taught to look for.
This is a common occurrence in many activities other than stereo,
so no more need be said about it.

When stereograms are shown to laymen who are familiar with
the original, and when the subject is such that hyperstereo is
warranted, the universal comment is that the hyperstereogram
best displays the realistic, natural aspect of the original. This
alone would warrant its use, even if the arguments against it
were not so wholly invalid.

ESTIMATING BASE SEPARATION. The reason for using hyper-
stereo is that the maximum distance at which we have stereoscopic
perception is limited. Although the actual limits vary with indi-
viduals, it has been accepted that among those of highly developed
stereopsis, the limit of stereo differentiation is that due to the
parallax of 20 seconds of arc. Based upon the normal interpupil-
lary, this means that objects beyond 670 meters are not seen in
stereo relief. As a matter of fact, to most people stereo infinity lies
at about half that distance. Thus for the normal interpupillary
the maximum object distance is 670 meters, and as that is the dis-
tance at which a round object appears flat, it is not a useful dis-
tance. At half that distance, a cylinder has the appearance of hav-
ing an oval section, one of whose diameters is half the length of
the other, the side-to-side dimension being the greater.

On the contrary, it is not at all uncommon to find among out-

HYPERSTEREO 267

door people a definite stereopsis two or three times this value,
occasionally more.

Thus, if an object is to be seen in readily appreciable stereo
relief, it must have at least that parallax of an object seen by
normal vision at a distance not exceeding 250 to 300 yards, and
the parallax for an object at 100 yards is preferable.

Parallax is angular measurement, and a given amount of par-
allax is that corresponding to a certain angle. It can easily be un-
derstood that for some definite angle, let us say one minute of
arc, the base at any given altitude (distance) must have a definite
value. If we double the value of the base, we either double the
value of the angle, or we double the value of the altitude.

This means that if we double the normal stereo base, the dis-
tance and relief of all objects in the field of view are given double
value, objects appear to be only half as far away, and appear to
have double their usual depth. It requires but a moment of con-
sideration to see that when the apparent distance is halved and
the apparent relief doubled, the specific relief of any object is
exactly the normal degree of relief for the new apparent distance.
That fact, so often overlooked, is the reason for the great success
of hyperstereograms. There is no distortion of specific spatial
values within the stereogram itself.

The practice of hyperstereoscopy is simple. Most stereo cameras
are provided with a single lens cap for this purpose. The left lens
is covered and an exposure is made. The right lens is then cov-
ered and the camera is moved to the left through the required
base distance minus one base length, and the second exposure is
made. Inasmuch as the motion is often a matter of some feet, the
exposures are made freehand, aligning the camera as accurately
as possible for the two shots. When the negative is developed, it
should be cut apart and the bases aligned carefully. This is the
one really difficult factor in hyperstereo technique. Naturally, as
the two exposures are successive, the subject should be motion-
less, as it almost always is in hyper shots. If a single-lens camera
is used, it is moved sidewise through the base length desired.
When using 35mm cameras and color film, the camera must be
carefully leveled and some key object placed in exactly the same
position in the finder for each exposure. Cameras with positive

268 THREE-DIMENSIONAL PHOTOGRAPHY

interlocks are used by making the usual dual exposure, moving
the camera and making the second. Later mount the two lefts to-
gether, then the two rights. This gives two complete hyperstereos.

This brings us to the critical point of the discussion, that of the
determination of the degree of base increase to be used. Most
stereographers who try hyperstereo use two or three times normal
base even when the object is far away. The results are of course
disappointing, Hyperstereo is without value unless used cor-
rectly.

One school of thought bases the increase upon stereo infinity.
That is, assuming stereo infinity to be 670 meters, and given a
subject whose distance is 6700 meters, a ten-times base would be
used. The fallacy is apparent. If hyperstereo is to be useful the
subject must be moved to some distance within the stereo limit.
By moving it just to this limit the whole effort is rendered useless,
for the subject continues to appear flat.

It is often said that a five times normal base is the absolute
maximum permissible. It might be cited against this that a very
successful stereogram has been made with a base of some 8000
miles, as was used in making stereograms of the moon where the
successive shots were made at a considerable time interval so that
the base was substantially the diameter of the earth.

At the same time you do not want a distant mountain to have
the appearance of being only 10 feet away, for then indeed there
would be an appearance of a nearby miniature. The base must
always be selected by the consideration of the subject, and this
cannot be limited by any rule-of -thumb regulation.

The first step is to determine the apparent position in space the
subject is to occupy. But in all the computation it is necessary
that approximate values be used so that the whole thing may be
easily and quickly done mentally without recourse to involved
calculations with pencil and paper.

Our hyperstereo computations are based upon the following
assumptions which are not strictly accurate, but which have been
found from experience to serve admirably. Because of the varia-
bility of individual stereopsis, accurate computation would be a
loss of time, to say nothing of its inconvenience.

It is necessary to know the distance of the object to be photo-

HYPERSTEREO 269

graphed. It is necessary to know the distance at which this object
is to appear. That is about all there is to it. The second factor
of course may give trouble. You can assume stereo infinity to be
1000 yards, and half that the distance of best general relief. There-
fore you will wish your distant object to appear at a distance of
500 yards more or less, or 1500 feet. As all of these distances are
roughly approximate, you may consider 1500 feet to be 1/3 mile
without seriously affecting the result.

Of course, if the object is a mountain, it would be wrong to
show it only 1/3 -mile distant, but still you must have it within
1/2 mile or it will be very flat. Therefore, the chosen apparent
distance is controlled by the object to some extent.

Assume the mountain is five miles away, and you wish to bring
it to within i/% mile. Parallax is a directly proportionate factor.
Double the parallax and you halve the distance, treble it and the
distance becomes 1/3. Therefore to make the distance 1/10
(1/2 mile from 5 miles), the required base will be 10 times nor-
mal or about two feet.

The only other factor is that of foreground. Keep the nearest
foreground at least seven feet distant. As 7x10=70, the nearest
permissible foreground would have to be 70 feet away when you
use a lox base.

Keep these two factors in mind and you can use hyperstereo
freely and with excellent results. Remember that you can use
a 3 foot foreground in normal stereo, so you could use a 30 foot
foreground with lox base, but the greater distance is advisable in
hyperstereo.

In practice a reflecting camera, either stereo or mono-lens, will
be found the most satisfactory, although excellent work has been
done with the usual 35mm camera. The ground-glass is ruled in
crosslines, if this has not been done by the maker. These lines
serve to align the subject in the two successive exposures, and if
key objects at the extreme limits of the screen are observed and
exactly duplicated as to position, a continuous base perpendicular
to the axes of the lenses may be maintained with surprising accu-
racy. Because of the great distance of the subject, there is prac-
tically no observable difference at the edges resulting from stereo
parallax.

270 THREE-DIMENSIONAL PHOTOGRAPHY

LONG-FOCUS LENSES. It may be asked why a long-focus lens
should be used i base alone determines apparent distance.

The original purpose of hyperstereo is to impart relief to some
object whose size is such that a remote camera position is essen-
tial. As long as the subject has this characteristic, hyperstereo
alone is the solution. In fact, to add long-focus lenses would de-
feat the purpose, as the large object would have its image spread
beyond the limits of the negatives.

It must be remembered that the conventional telescope does
not, as so commonly stated, bring objects nearer. It simply in-
creases the perspective values, the angular value of the field of
vision filled by the object. Therefore we shall make use of long-
focus lenses only when the object is too small to be satisfactorily
recorded from the nearest accessible camera position.

If we maintain a proportional increase between base and magni-
fication, the type of reproduction closely approximates that of
orthostereo. We reproduce the objects at their full natural tele-
scopic size, and with that distance and degree of relief which is
fully natural for that size. Thus if the lenses are doubled in focal
length and the base is also doubled, objects are shown in the size
and relief normal to a distance which is just half that of their true
distance.

If the object is small and far away, the proportionate increase
just described, that is, the use of parastereoscopic technique, will
have much the same effect as that of using a telephoto lens upon a
planar camera except that we have stereo relief. That is to say,
we are shown the object as it would appear from a point of view
midway between the real camera position and the object. The
same distortions appear as in the planar telescopic record, but like
them, these distortions are extremely difficult to perceive even
when they are pointed out in detail. Any distortion as subtle as
this may be completely disregarded without detriment to the qual-
ity of the stereogram.

Therefore, in making stereograms of objects at a great distance,
we first determine the base necessary to introduce the desired de-
gree of relief. The next step is to select the lens which will cause
the image to occupy the desired space upon the film, and if it is
desirable that strictly normal proportions be preserved, the first

HYPERSTEREO 271

two steps are compromised so that the two increases are of the
same degree.

For example, suppose the scene you wish to photograph is a
group of chimney rocks, tall pillars of stone as found in some of
our great National parks. These are huge things, and when you
are among them, the general effect is lost because your effective
vision is held by the immense bases or the height of one or two
pillars as you look upward. It is only when you go to some ele-
vated position at a considerable distance that you can see the
characteristics of the formation.

If this scene is photographed by hyperstereo, the result will be
a stereogram including the whole of many of these pillars, and
will thus give an impression of their true nature which is superior
even to direct vision. Here we maintain the wide field of the nor-
mal lenses used at a distance, plus the exaggerated relief of the
increased stereo base.

But if there is a man standing on top of one of these pillars, and
you want to make a stereogram showing him, the original hyper-
stereo is of no value because the tiny figure is lost in the distance.
This situation calls for parastereoscopy with the combination of
telephoto lenses and telestereo base.

We cannot approach closely, because then we should have
to shoot upward and the top of the rock would hide the man.
The only point of view possible is the edge of a cliff 1000 yards
from the man. In this instance, we wish to obtain at least a rec-
ognizable image, which means that we should use, at most, an
apparent distance of 200 yards. Thus we should make use of a
lens whose focal length is five times normal, and we should use a
five-times normal base,

It will be seen that the choice of the base in hyperstereo, or the
choice of the lens-base combination in parastereoscopy, is de-
termined, not by the traditions of old time stereo, not by some
complex mathematical computation, but solely by the exigencies
of existing circumstances.

Increase the stereo base (hyperstereo) to bring the object nearer,

Increase the focal length of the lenses to make the object larger.

The two functions are not at all identical.

Let it be emphasized that there is no valid reason why the

272 THREE-DIMENSIONAL PHOTOGRAPHY

stereographer should not vary both base and focal length of his
lenses if by so doing he is enabled to obtain a result which is
visually superior to that which would have resulted otherwise.
The freedom of choice extends beyond that limit. If the stereog-
rapher wishes to make a stereo caricature through the use of ex-
aggerated relief, that is perfectly good stereography. In fact, if
well done it constitutes expert stereography, because understood
theory has been used to exert a wanted control.

If the stereographer can obtain more realistic effects through
hyperstereo, as he most assuredly can, then that is fully satisfactory
evidence of the very real value of hyperstereo as a recognized ele-
ment of stereo technique.

Too much cannot be said against the stringent and wholly un-
reasonable limitations placed upon stereography by those whose
sole interest is theory rather than the making of stereograms, and
who condemn work for the method used rather than for the re-
sult achieved.

Any statement that so-and-so many times normal is the limit for
hyperstereo, simply indicates that the speaker is not familiar with
fundamental stereoscopic laws. The best results are obtained by
those who experiment, and who find, not the limit in base units,
but the apparent distance at which certain large objects appear to
the best advantage.

It is one of the beauties of stereo that you can control it to an
almost infinite extent. You can place objects at any desired dis-
tance and by careful work you can intermingle natural and fan-
tastic distances in a single stereogram.

Finally the question arises, "Why not approach so near to the
subject that natural stereo relief is seen?" The answer is that
many of the most promising hyperstereo subjects are so huge that
any such approach eliminates most of the subject and leaves only
a restricted area, often of little beauty or interest.

Above all, avoid the common and regretable practice of using
normal base for such shots with some bit of extraneous fore-
ground, usually a wholly inappropriate human figure looking
away into the scene, just to introduce enough stereo relief to show
that it is truly a stereogram. If the stereogram is not self-evident
in the principal part of the subject, then it had better not be

HYPERSTEREO 273

made. If there is a genuine foreground subject of real interest,
and the mountains are to form simply a theatrical backdrop, then
the normal base may be used; but when the distance itself is the
subject, then by all means make a stereogram of it, not a double
flat photograph!

CHAPTER 19
STEREO MOVIES

THIS CHAPTER is so BRIEF and so uncomplex that a word is nec-
essary in explanation. The chapter is brief because stereo
movies are so extremely simple there is really nothing much to
be said about them. You can assemble the beginner's outfit with-
in ten minutes or so, and you can depend upon it that the results
will astonish you. Movie projection, stereo or otherwise, is no
more than a mechanical elaboration of still projection. Optically
the two are identical. Therefore, everything which has been said
about still stereo projection applies equally to movie projection.

The utter ease and simplicity and the thoroughly satisfactory
results of even elementary stereo movies are so great that the
hesitation of the amateur to try it is incredible. Everyone seems
to be awaiting some very mysterious, very complex, very magical
(and costly) method for doing this simple thing. I only wish I
could think of some way to convince you that this hope is all
nonsense. You already have available everything you need to
make perfectly beautiful stereo movies with very little expenditure
and with very good assurance of success in the first roll of film
you expose.

The method outlined here is that used successfully in the
Stereo Guild laboratory experiments, and the results were just as
successful as those obtained from still stereo projection. This is
one time when the expression "there is nothing to it" is true. It
is so simple stereographers cannot believe it, but one trial will
convince you.

It is perfectly easy for any movie amateur to make excellent
stereo movies, although unfortunately he will have to provide his
own equipment, as there is nothing really suitable on the mar-
ket. It may be said, however, that for preliminary experimental
work, the stereotach or the cine-stereo attachment of Mattey can
be employed. At the same time it must be emphasized that wholly
satisfactory results cannot be obtained from a fixed reflector.

You will need a stereo reflector, which fundamentally resembles
the Stereotach, but which has two adjustments. (A) The angle of

274

STEREO MOVIES 275

the outer mirrors must be adjustable and (B) The distance be-
tween the outer mirrors must be adjustable. Before World War I,
about 1910, Burke and James sold a stereo reflector which had the
first adjustment, and this is the reflector used in the preliminary
experiments of the Guild. Many amateurs make their own, but if
you have no skill with tools, you will surely have, among your
acquaintances, an amateur metalworker who will be glad to make
the attachment for you for a moderate sum. ^ ,

L

Fig. 19-1. The Stereo reflector. This diagrammatic plan view shows no
mechanical supports, only optical elements. The two inner mirrors,
RI t and RI S set at 90 y join opposite the vertical diameter of camera
lenSj L. At the right and left are the two outer mirrors, RO^ and RO 2 >
Al and Ar indicate the axial rays for right and left, while LO, LI, RI
and RO indicate the approximate field limits. LI and RI indicate the
gradual convergence of these rays with distance.

276 THREE-DIMENSIONAL PHOTOGRAPHY

We shall not give detailed working drawings, because there is a
wide latitude of detail design permissible. The actual require-
ments are few and simple.

1. The outer mirrors must be adjustable through about ten de-
grees, swinging upon a vertical axis.

2. This movement must not affect the precision of the vertical
plane (the reflecting plane) of the mirrors. In short, a sturdy pivot
must be provided which is exactly perpendicular to the base of
the framework.

3. The outer mirrors should move laterally so that the base may
be varied between two and one-fifth and six inches or a movement
of one and three-fourths inches for each mirror. This is to con-
trol parallax, especially when telephoto lenses of moderate focus
are used. This motion must be along a rigid base so that the ver-
tical planes of the mirrors will not lose parallelism.

The inner mirrors should have as sharp an apex as possible, but
it is not at all necessary that it be an "invisible line" joint. One
amateur made use of a right angle prism with the sides surface
silvered (aluminum vaporized). This is a sensible solution and
one not too expensive.

However, it is imperative that some means be provided by
which the image as formed in the film plane can be subjected to
visual examination. The equipment can be calibrated when once
made, but the initial calibration demands visual inspection in the
film aperture. There are "focusing prisms" and "focusing attach-
ments" available which will provide this and which are applicable
to most cameras.

In calibrating, remember that if the base of the reflector is not
parallel with the base of the camera, one image will be higher
than the other. If this happens do not conclude that your mirrors
are out of parallel. Just raise or lower one end of the reflector
base. That is, twist it with the lens axis as center of revolution.

Unless you are experienced in this kind of work, you should by
all means make your first attempts with a Stereotach, which may
be used without visual checking and calibrating, but you must
use extreme care to get the reflector base parallel with the cam-
era base.

FORMAT. One serious objection to making stereo movies in this

STEREO MOVIES 277

way is the fact that the image is duplicated in the normal frame
which has proportions 3:4, and as a result the screen picture is a
very narrow, very tall image with the horizontal to vertical ratio
2:3.

There is a solution to this problem also, one which is used in
making Guild stereo movies, but which will involve some trouble
in shopping. About 20 years ago, lenses were sold to movie ama-
teurs which produced an imitation of the "wide screen" which
was popular in theatres of that time. This was a corrected cylin-
drical lens, which compressed the image laterally, and in projec-
tion expanded it again so that the width of the picture was in-
creased by some 50 percent without affecting the vertical dimen-
sion. When this lens is used for stereo movies, the screen format
is practically square and wholly satisfactory. . However, it will be
necessary for you to locate one of these lenses, because they are
not now an article of regular commerce, this may be difficult. It
is to be hoped they will be re-introduced.

MAKING THE STEREO MOVIE. The camera is set up as usual.
The compressor lens (if used) is placed before the camera lens
and the stereo reflector mounted before the compressor. Thus the
light strikes the reflector first, passes through the compressor,
through the camera lens and to the film. The exposure factor is
about 4x (in our test outfit it was about this, but the factor may
vary somewhat).

Note that there is no polarization involved in the camera work.

Once the accessories are correctly adjusted and the diaphragm
set for the exposure factor, the rest of the procedure is that of
normal movie making. Objects within 30 feet of the camera, or
so, will provide better results. The film is processed as usual.

PROJECTION. The two images of the film must be projected in
superposition, just as in regular stereo projection. When a fixed
reflector is used, something is necessary to provide adjustable su-
perimposition. This may be a pair of mirrors at slight angle, or a
series of weak prisms which are placed over the lens base to base.
However, the serious amateur will have an adjustable reflector
made and when this is done, the same reflector used on the camera
will be used with the projector and the mirrors adjusted to pro-
vide the superposition. This is by all odds the best method.

278 THREE-DIMENSIONAL PHOTOGRAPHY

The whole accessory unit of compressor-reflector t is removed
from the camera and supported before the projector. In front of
the reflector openings are placed the two polaroid filters with their
axes at 45 degrees left and right so the standard g-D viewing
goggles may be used.

When superposition is obtained, no further adjustments are
necessary and projection proceeds in a normal manner.

Filter position is easy to check. If the axes are reversed, dis-
tances will appear in reverse. When correct, relief will be nor-
mal. If a filter is "between" extinction positions, two images will
be seen upon the screen when using the g-D viewers. If you have
the filters in rotating mounts, you can easily make the adjustment
while the picture is on the screen.

Remember, the screen MUST have a surface which does de-
polarize light, the most convenient being the aluminum surface.
Ordinary beaded screens will not work. You will see both images
no matter whether you use the goggles or not.

SUMMARY OF ESSENTIALS

The most satisfactory stereo accessories for movie work must be
home made. This is a condition we hope will soon be remedied.

The compressor lens, if one is found, is simply placed before
the camera lens. Be sure to use the guide line so that the com-
pression is accurately lateral, and be sure to focus the compressor
using the engraved scale. Otherwise it is simply an accessory lens.

The reflector must be set for the lens in use, and so adjusted
that the images are equally and symmetrically divided between the
two halves of the aperture.

Use a wide base when using lenses of more than normal focal
length. Use twice normal base for a two-inch lens and so on.

Use exposure compensation because of losses in both reflector
and compressor. About three times to four times is advisable.

Use the whole unit, compressor and reflector upon the projector
and add the polarizing filters.

Superimpose images upon the screen, preferably with the adjust-
able reflector.

Do not get too close to the screen.

Otherwise, use normal movie procedure.

STEREO MOVIES 279

Many amateurs who have expected some very difficult and very
complex procedure may think this too easy, but like all stereo, it
is as easy as normal photography once the necessary equipment is
available. If you consider the general laws of stereo and the re-
marks concerning still stereo projection together with this chap-
ter you will be fully prepared to embark in making stereo movies,
and very often your first film will be fully successful. Starting
from scratch some 14 years ago, we made one film which w r as
slightly distorted as to depth-width relationship, but the second
was good and no more trouble has been encountered.

PRACTICAL DETAILS

SMM STEREO. We have not actually made 8mm stereo movies in
the Guild laboratory, but there seems to be no reason why, within
the usual size limitations, they should, not be satisfactory. There
is nothing to change in the general technique. Unfortunately, al-
though we are now preparing to make such tests, they will not be
completed in time for a report to appear in this volume.

OTHER METHODS. -There are several other methods of making
stereo movies, but all of them demand more complex preparation.

First of all, there is the obvious method of using two cameras
and making two complementary films. This, then, requires two
projectors, but that is not the great objection.

Such a system demands exact synchronism in the projection,
and it demands double film expense. It demands the greatest care
in editing to keep both films identical, for even one frame mis-
match would cause trouble. Each film break means the same pains-
taking care. All in all it is simply too much trouble for amateur
use when the simple reflector method is available.

CONVERTED SMM. This is an excellent method, as it eliminates
the need for the compressor. But it requires some very accurate
camera conversion, and makes the camera unfit for planar use.

Any standard 8mm camera may be used as the basic equip-
ment. The aperture plate is removed and widened so that it is
full width (lomm) just as in a i6mm camera, BUT the height is
not changed. This provides the full size of an 8mm frame for
each of the stereo units, and when the reflector is used, the stereo
pair is recorded in normal format.

280 THREE-DIMENSIONAL PHOTOGRAPHY

The camera will work after a fashion with no more alteration
than this, and it is simple. But for really good results it is neces-
sary to shift the lens center 2.5mm toward the film center so that
the lens is centered on the dividing line between the two images,
and this is difficult. The lens must retain its distance from the
film to maintain correct focus. The only practical solution is to
dismount the lens, make a new eccentric base to replace the old
one and substitute it. This requires the attention of an expert
instrument maker or camera repair man. The projector requires
a similar aperture plate conversion, but the lens shift is not as
critical as in the camera. Nevertheless, for a first class job, the
projector lens should also be shifted. When this is done, you will
have a complete and highly satisfactory 8mm stereo movie outfit.

i6MM CONVERSION. The same thing can be done with a i6mm
outfit with less trouble because the lenses do not have to be
shifted.

The top or bottom half of the aperture is blocked off. For tem-
porary use, you can do this by pasting a bit of black paper over it,
from the shutter side. In most cameras, this can be done by re-
moving the lens and revolving the shutter until it opens. Do not
paste in the film side. The film will be scratched and will soon
pull the paper from its position.

Finally, mask the finder to correspond, BUT if you mask the
bottom of the aperture, mask the top of the finder. Then add side
masks to the finder so that 14 of the field is cut off at each side.
The area left is equal to 14 of the original.

The reflector is used with no compressor.

Only half the film will be exposed, and if you choose you can
use the same roll for a second series of exposures, if you remember
that sequences will be scrambled and editing will be impossible!
The reflector is used with the projector too, of course. In this
method, the actual film image has just the same size as in the
converted 8mm outfit, but the i6mm involves either double film
cost or the sacrifice of editing.

COMPROMISE. There is one compromise which is perhaps the
best of all solutions for the problem of format. The i6mm frame
is approximately 7.5 x lomm. If you wish to use a square aper-
ture, it will be 5x10 because there must be two pictures side by

STEREO MOVIES 281

side. Therefore instead of masking half the aperture you may
mask only 1/3. For the best possible results, you will divide the
2.5mm mask into two parts. Mask off i.25mm at the top and
i.25mm at the bottom. You will now have a centrally spaced
aperture measuring 5mm high and lomm wide. Your effective
aperture is 5mm square or 25 sq. mm. area. With either of the
former conversions, the effective frame measures 3.75x5 or about
18.75 sc l- mm -

For the utmost gain commensurate w r ith acceptable format,
mask imm from top and bottom leaving the total aperture 5.5x10
or an effective aperture 5 wide by 5.5 high, a ratio of 10:11 or
roughly the same proportions as the Realist frame. This gives an
area gain of almost 1/3 over the 8mm conversions, and provides a
highly satisfactory format. For rapid comparison of effective aper-
ture areas we have:

8mm conversions, 3.75x5 nr 18.75 S( l- mm -

Compromise, 5.5x5 = 27.5 sq. mm.

i6mm 7.5x5 = 37.5 sq. mm.

The area ratios are roughly 234.

For those who do not object to the slightly smaller picture, the
compromise is the best format to use. It differs from the straight
8mm frame in being slightly higher than its width, while the nor-
mal 8mm is wider than high.

In all of these conversions the width of the frame is that of the
normal 8mm frame. This can be increased ONLY by (A) using a
35mm camera to obtain masked i6mm screen size, (B) using a
compressor to provide an effective frame width of about 7.5mm,
or by using two separate cameras.

When using the compromise system, it is not necessary to mask
the projector. When the camera aperture is masked, the unex-
posed film will be blackened in processing thus making an auto-
matic mask for the film. This method together with simple prisms
to deflect the projected images is the simplest of all stereo movie
methods.

The two prisms are placed base -to base and then placed over
the projection lens so that the junction of the prism bases lies
vertically. The strength of the prisms depends upon the screen
distance and is best determined by trial. For example, obtain a

282

THREE-DIMENSIONAL PHOTOGRAPHY

Fig. 19-2. A projector prism

which has been cut to circular

shape.

Fig. 19-3. A simple projector
prism made by placing the thick
edges of two prisms together.

pair of plane, 5 diopter prisms. Move the screen back and forth
until the images superimpose. Measure the screen-to-projector
distance. Also note the leeway that is the distance the screen may
be moved without separating the images more than two inches
right or left. Then try 3 diopter prisms and compare the results.
This will enable you to determine the prism strength needed for
your projection conditions.

Practical Conversion. Of all the methods recommended, the
compressor is best provided you can obtain the lens. (Original
cost around $80.) If the lens cannot be found, the next best is
the compromise masking which provides a square picture only
about 25 percent smaller than normal i6mm. The compromise is
of course by far the simplest of all methods and will probably
prove most attractive when all factors are considered.

Only as a curiosity do we mention the alternate system. This
provides full frame i6mm size, but it necessitates some kind of al-
ternating reflector, reciprocating or rotary, which could be made
only by an expert mechanic and would require connection to the
camera drive which adds to the motor load. Similar provision is
necessary for the projector. It is feasible but not entirely practical
for the average amateur. Film speed is 32 or more, so the camera
and projector mechanisms are subjected to excessive wear. Two
successive frames are required for one picture, so film cost is
doubled. It has been done and stereoscopically it is satisfactory,
but it requires special equipment as mentioned.

STEREO MOVIES

The compromise method does not permanently disable the cam-
era for planar work. If your camera has provision for aperture
masks, such as the Cine Kodak Special, the changeover can be

Fig. 19-4. Stereo-cine camera Cine Kodak plus compression lens plus
adjustable reflectoris the equipment used in the Stereo Guild experi-
ments in stereo movie making.

made in a few seconds so that stereo and planar can be alternated
in the same film. The only significant difference is that the camera
must be moved back somewhat from the subject. A compromise
film, in any modern projector will fill a 50x50 inch screen with a
good quality image.

Thus, while methods are used which do not include the re-
flector, those systems which are most practical are based upon
the use of the reflector.

Corf. Reflector systems for the movie camera must be of pre-
cision quality. If you have the tools and are capable of fine ma-
chine" work as many stereographers are, the cost (without com-
pressor), may be kept to less than $15 or $20. If you must have the
reflector built by someone else, a competent amateur mechanic

284 THREE-DIMENSIONAL PHOTOGRAPHY

might do it for anything from the material cost up, but a profes-
sional will charge not less than $100 and from that up to perhaps
five times that amount. I saw one camera which represented an
outlay of $2500, but that included approximately $1000 for the
Cine Kodak Special which was the basis of the outfit.

Stereo movies are perhaps even more attractive than motionless
projected stereograms. That, of course, is a matter of personal
opinion. Certainly, however, one thing can be said, which has
often been said and proven regarding still stereo: Once you have
the pleasure of seeing your color movies in the full realism of
three dimensional space, you will not be content to return to the
flat screen type.

In closing it may be said that anaglyphic movies, as well as still
slides, have been offered in limited quantity. These are for use in
normal projectors and require red-green viewers instead of the
polarizing type. They are fairly satisfactory, although some peo-
ple see a color bombardment when the red-green anaglyph is used.
They do provide stereo projection for those who have no regular
stereo equipment. Unfortunately they are limited to commercially
available subjects which is a grave disadvantage.

It is rumored that similar slides and movies are to be made
available using the Vectograph principle. This system uses polar-
ization extinction and has no color bombardment effect. They
are highly satisfactory, but again the "commercial subject" limita-
tion is a disadvantage. It is said that when the Vectograph films
do make their appearance they will be in color, whereas the red-
green are perforce monochromatic.

In the absence of specific information, no more will be said
about these stereo projection slides and films.

However, there are certain factors regarding the recent history
of stereo motion pictures which will be of interest, so we shall
give them passing attention. The first involves the often dis-
cussed "free vision*' viewing, or the viewing of stereo motion pic-
tures : without the aid of any kind of viewing device, and the great
publicity which has been given the subject, to say nothing of the
multitude of spurious methods which have been proposed and
which have actually been used.

Free-Vision Screen. This is based upon the principle of a mul-

STEREO MOVIES 285

titude of intersecting stereo beams, so disposed that the spectator
will receive one left and one right beam in the corresponding-
eyes. It can readily be understood that as long as the essential con-
ditions prevail, it. will also be possible for a spectator in some
position to receive alternate beams, thus producing a pseudoscopic
relief rather than stereoscopic.

It will also be obvious that there will be certain positions of the
spectator where neither the stereoscopic nor the pseudoscopic re-
lief will be apparent.

Such a screen has been reported from Europe and has received
a great deal of publicity. However, the technical report reveals a
significant condition. The seats in the auditorium are specially
positioned in groups, and the spectator must move his head about
until he locates the correct position for relief viewing. These two
facts reveal the nature of the screen to be that which was long ago
developed in this country and later abandoned as impractical.

It must be kept in mind that stereo projection per se is an ac-
complished fact, but the practical objection lies in the fact that in

Fig. 19-5. The simplest stereo movie outfit, A Stereotach used in con-
junction with a 16mm Cine Kodak.

286 THREE-DIMENSIONAL PHOTOGRAPHY

America the system will never be considered commercially prac-
tical as long as the spectator has to do anything to aid in the re-
ception of the stereo images. This means that individual viewers
cannot be used, no matter what their nature. It means that special
positions cannot be tolerated.

It seems strange that anyone should object to wearing a pair of
light spectacles which are no more uncomfortable than any pair
of lightweight spectacles or sunglasses, yet that is the only thing
which stands in the way of the general adoption of wholly suc-
cessful commercial stereo motion pictures!

One older system made use of a rotating shutter for each spec-
tator, either fastened to the back of the seat in front of the spec-
tator, or held in the hand. These shutters were synchronized with
the projector shutter and gave an excellent stereo result, but were
never used because the spectator would not cooperate by using
the viewer. In fact, the list goes right back to the original red-
green anaglyph, also a somewhat successful method, but one in-
volving spectator cooperation.

Stereo movies, according to the producers, will never be com-
mercially successful until some method is devised whereby the
spectator cannot possibly see anything other than true stereo re-
lief and without aid of any viewing device. Thus, when we dis-
cuss the failure of stereoscopic motion pictures, let us lay the blame
where it belongs not upon stereo, not upon the methods used,
but upon the non-cooperation of the spectator. However it is our
belief that the widespread and successful use of stereo still pro-
jection which we now enjoy may change the professional attitude.

Both the dual projection by polarized light and the Vecto-
graphic methods are available, although the latter is hardly prac-
tical for the amateur. The production of long lengths of motion-
picture films in Vectographic relation would require the use of
elaborate production machinery, but this machinery is available,
as it has been used in printing two-color natural color films.

For the amateur, the polarized and non-polarized methods are
both available, and with but slight alteration in the projector, the
successive type is also readily available.

This alternate method was introduced in 1928 for amateur mo-
tion-picture use. Through a curious combination of circum-

STEREO MOVIES

287

stances, a most unusual result was achieved. This was partly due
to the fact that the inventor was not at all familiar with stereo-
scopic principles.

The films were made by two cameras, spaced 65mm and oper-
ated in synchronism. Thus the right and left images had no

Fig. 19-6. Adjustable reflector.
By pivoting, the rear edges of
the outer mirrors may be swung
as shown by the broken lines.,

thus making it possible to place
the two stereo images in any de-
sired relationship upon the film.
Compare with Fig. 191.

difference due to motion, which produces much better quality
than when the films are made in succession. The positives were
designed for the color anaglyphic method, and the inventor,
through confusion, assumed the cameras must be equipped with
complementary filters. The result was a pair of negatives repre-
senting two-color separations.

The films were printed in skip printers, which advanced the
positive two frames to each single frame of the negative. This left
room for printing the other negative in the alternate spaces. The
finished positives were dyed in the complementary red and green.

When the film was projected and viewed with the color spec-
tacles, the images were not only seen in relief, but there were other
and unexpected results. The relief was entirely satisfactory, there
was more than a suggestion of natural color effect, and because of
the alternation in time, the color bombardment effect common to
still-color anaglyphs was missing. The films were surprisingly sat-

THREE-DIMENSIONAL PHOTOGRAPHY

isfactory, and, except that they were positive-negative and re-
quired elaborate equipment, the system could be regarded as
highly successful.

In this chapter we have not emphasized the distinction between
motion-picture and still projection, for one very good reason.
Optical projection remains the same no matter whether the pro-
jected images are changed twenty times a second or twenty times
an hour. A system which will work with one, will, with a few ex-
ceptions, work with the other.

The outstanding exception is the motion-picture system just
described in which the right and left pictures are projected alter-
nately upon the screen. This is done simply to preserve the orig-
inal format of the picture. The stereoscopic control may be by
means of the polarized anaglyph, in which case the polarizers are
shifted in synchronism by an oscillating support, or a mechanical
shutter and an auxiliary synchronized shutter is held before the
eyes of each spectator so that each eye sees only its own picture.
Inasmuch as each eye sees only half of the projected pictures, it
is necessary to operate both camera and projector at double nor-
mal speed to prevent flicker. For that reason if for no other, the
method may be considered obsolete as far as motion pictures are
concerned.

FLICKER PROJECTION. In the rotating shutter previously de-
scribed, a left picture is projected to fill the whole screen and is

Fig. 19-7. The variable base re-
flector. The broken lines indi-
cate the normal positions of the

outer reflectors, while the solid

lines indicate the . "wide base"

position.

STEREO MOVIES 289

seen only by the left eye. This is removed and a right picture
fills the screen and is seen only by the right eye, and so on. The
shutter which alternately blocks the vision is a rotary (occasion-
ally oscillating) one so supported in front of the spectator, and
synchronized with the projector, that the open sector comes be-
fore the eye with accurate timing. Thus the viewing shutter is
synchronized with the projector film advance. Of course polariza-
tion is not necessary when synchronous shutters are used.

A similar device has been used, with far greater success, in still
projection. The two stereo pictures are projected upon the screen
in superimposition, just as when the anaglyphic method is used.
A shutter is built into a dual lens projector so that during one-half
revolution the left beam is passed, and at the next the right beam
is passed. Thus upon the screen there is a rapid alternation of the
two. However, because of the speed of the alternation, to the un-
aided eye the screen will present both images.

Each spectator is provided with a similar shutter which is syn-
chronized with the shutter of the projector, and when the screen
is viewed through it, the left eye sees only left pictures and the
right eye sees only right pictures. In one commercial adaptation,
that of Stereovision, Inc., the spectator shutter is enclosed in a
small oblong case which is very easy to hold.

One is inclined to ask why such a system should be used when
the polarization system is available with its light-weight goggles
which may be worn without discomfort.

The reason most commonly advanced is that the absorption of
light passing through two polarizing filters requires such a high
light intensity that a special lamp must be used in the projector.
This argument is quite valid from a theoretical point of view, and
would of course give tremendous advantage to the flicker system
which has no abnormal absorption losses. However, in actual
practice we have found that i6mm polarized projection with a
75O-watt lamp and a 4O-inch screen is wholly satisfactory, while a
still projector equipped with 5oo-watt bulbs will also give a good
50-inch picture, using polarizers.

On the contrary, the flicker system utilizes only 50 percent of
the light and the screen is not uniformly illuminated. We prefer
polarization.

290

THREE-DIMENSIONAL PHOTOGRAPHY

SUMMARY. i. The film is made normally, using a stereo re-
flector of the type used for still photography with a single-lens
camera. The individual images are a half frame in width and
hence the screen image will be abnormally high and narrow un-
less use is made of a system just devised whereby optical or mask-
ing means are utilized for overcoming this narrow format. The
film is projected either through a prismatic lens or through the
stereo reflector, each image being polarized at right angles to the
other. Polaroid spectacles are used for viewing the images, which
are projected upon an aluminum screen.

Fig, 19-8. The projector-reflec-
tor. The polaroid filters, P-P,
have been added to the stand-
ard reflector and the outer mir-
rors set for sharp convergence.
Compare with Fig. 19-L

The amateur will find it to advantage to use the polarizing
filters set at 45 degrees from the vertical so that the normal Polar-
oid g-D spectacles may be used for these polarized images, and for
Vectographs. This eliminates the necessity for two sets of spec-
tacles.

2. The film is made with the reflector and projected normally.
This also provides a narrow picture, and it, too, may be optically
expanded or masked to proportion. The images are viewed by
prismatic viewers which eliminate the unwanted image by angu-
lar displacement.

3. The camera is operated at double speed and the projector is
speeded up to the same speed. The camera is equipped with a
stereo reflector and a half-speed oscillating selector shutter which
impresses the rights and lefts alternately upon the film. The pro-
jector has a similar oscillating shutter with polaroid films set at
right angles in the two halves, thus the right and left images are

STEREO MOVIES 291

projected in succession at double speed. When viewed with the
polarizing spectacles, the stereo relief is experienced.

4. The projector and each spectator is supplied with a flicker
shutter which mechanically obscures the two stereo images alter-
nately. The shutter may be rotary, rocking, or sliding, and may
be actuated by rotary motors, magnetic motors, or solenoids. A
shutter geared to the fan motor of the projector and actuating the
individual viewers through solenoid control, is simple and effec-
tive.

CHAPTER 20
SPECIAL PHASES

STEREO is NOT LIMITED to the making of personal pictures. In
fact, it is rapidly moving into all industrial, commercial and
scientific fields.

A dvertising. Advertising photography, which includes illustra-
tions used for reproduction in the advertising space of various
media, remains planar because we have not at this time accepted
any adequate method for reproducing quality stereograms at a
cost commensurate to that of the planar.

However, this is a temporary condition. The fundamental key
to the whole thing has been overlooked because of a very human
psychological quirk.

Everyone who reads has spent years in learning to read fluently,
and it is assumed that one simply looks at a picture and grasps its
meaning. This is not true. The youngster has to learn to see pic-
tures just as he has to learn to walk and to speak. It has been
demonstrated that a savage, seeing his first picture, helplessly
turns it upside down, upon its side, right side up and then gives
up. It is meaningless to him, even when the subject is his own
village!

Some training, some experience is necessary before we can in-
terpret anything at all. But when it is suggested that we spend
just sixty minutes in learning to see three dimensional printed
pictures, the suggestion is brushed aside with the typical comment,
"Too much trouble!"

Anyone who has normal vision can learn within one hour to
see a non-transposed stereo pair by the use of convergent vision,
as explained elsewhere in this book.

All that is necessary to bring three dimensional printed pictures
into reality is the sustained use, on the part of three or four pub-
lications, of untransposed pairs not exceeding a two inch base!
And the small size is not detrimental, because the subjective size
will be considerably larger than the printed area, making for eco-
nomical use of paper space. It may be years, it may be only
months, but eventually we shall recognize the utility of taking

292

SPECIAL PHASES 293

the short time necessary to teach people to "read" three dimen-
sional pairs in books and periodicals.

It has been reported that one widely used physics text, using
just this kind of free vision stereo, has been in print some years.
Unfortunately it has been impossible to obtain details before go-
ing to press.*

Sales. Makers of bulky and heavy products commonly equip
their sales representatives with albums of planar photographs
from which sales are made. This method is admittedly unsatisfac-
tory, but a salesman can hardly carry a five ton truck with him.
Manufacturers have recently experimented with stereo and have
found new possibilities. Stereo proved instantly successful, because
it enables the customer to see, not a worked-up photograph, but
the real thing.

One manufacturer of dresses, whose saleswomen used to carry
two huge trunks of samples, now send them out with a stereo set
and a half dozen samples in a suitcase to demonstrate material
quality and fine details of workmanship. This has resulted in a
sharp upswing of sales because the buyer can" actually see all of
the real models.

Manufacturers of specialties for physicians have found stereo
the first practical substitute for personal demonstration. Full
techniques for using new products and instruments are shown in
stereo.

A manufacturer of power shovels sent out a set of stereos
showing various models, small and large, actually engaged in op-
eration. Again stereo scored a huge success in a demonstration.

Throughout the sales field, from the smallest products to the
largest, the use of stereo has proven successful, and is so rapidly
advancing that already most large cities have stereo specialists at
work.

Industrial Applications. Stereo is rapidly gaming favor for
making industrial and laboratory records. Mining and engineer-
ing firms make field records; scientists record laboratory experi-
ments; factories have wiring and conduit circuits recorded. In
planning, miniature scale models properly photographed show
the space form of the ultimate full size structure. Whenever a visi-

* "From Gallileo to the Nuclear Age," Harvey Brace Lemon, University of Chicago
Press.

294 THREE-DIMENSIONAL PHOTOGRAPHY

ble record of concrete fact is desired, stereo has proven superior.

Education. For teaching in schools as well as for adult educa-
tion and specialized technical training, stereo has proven to be
free from the disadvantages of planar photography. It is the only
photographic process which can rival handmade drawings in
such fields as surgery and pathology.

Educational stereo, however, lags. For some reason which can-
not be understood, many educators are against visual aids in
school work. They seem to have the same idea which rules medi-
cine in primitive civilizations. The more unpalatable education is
the greater its value. That this is false has been demonstrated time
after time. Lasting knowledge comes only from study which is in-
teresting. The forced cramming of ordinary book study fades
from memory within weeks or months; while that which has been
learned visually is a permanent acquisition.

Reading is a comparatively recent acquirement of man, but
vision has come up with him from primeval time. It is easy to
forget what we read and hear, but it is almost impossible ever to
forget what we have seen. And as has been demonstrated time
after time, the artificiality and convention of the planar photo-
graph makes it akin to reading, but the stereogram is in every
way equivalent to the sight of the real object. Those who have
seen a stereogram remember it fully as long and as distinctly as
do those who saw the original object.

It must be admitted, however, that in this discussion we speak
of the stereogram as we have it today, the true replica of the orig-
inal in gradation and color, coupled with the benefit of stereo-
chromatism. The monochrome paper print is not nearly so effec-
tive and holds a position midway between the true stereo and
the planar.

Even so, stereo is gaining in education, and it will be perhaps
only a matter of a few years before it will have been fully accepted
as a part of our educational program.

Forensic Stereo. The acceptance of a photograph as evidence
in a court of law is hedged about by many proper limitations. The
photograph as we ordinarily know it is subject to too much alter-
ation by skilled hands and to far too great a diversity of interpre-
tation. True forensic photographs are made by trained operators

SPECIAL PHASES 295

and always in the presence of witnesses who can swear that the
picture is truly representative of the original.

Since the stereo revival, stereograms have been received favor-
ably by many courts, and the effect upon a jury is great because
there is no room for interpretation. The original is seen as though
it were present. Although, as far as we can learn, stereo started in
the legal field with the presentation of evidence in a civil case,
police photographers are now taking it up and using it to supple-
ment or replace planar photography in the routine of criminal
investigations.

One disadvantage at present is the delay in color film process-
ing but with the steady improvement in "home" processed color
film, this disadvantage is temporary.

Restricted Subjects. Much of the subject matter of forensic,
biological and medical work is such that it would not ordinarily
be returned from the laboratory; but when it is fully understood
that such films are of serious, professional nature this will not be
the case.

On the contrary there is one class of subject matter which is in-
herently restricted records of experiments and progress of secret
developments. Ordinarily these would be safe enough, but there
is always the slight chance that someone who is not authorized
might see the pictures. At the same time, ordinary home process-
ing involves the loss of color which is often of the first importance.
Then there is only one course open. That is the use of home
processed color such as Ansco color or Ektachrome. Both of these
are widely used in stereo and have given highly satisfactory re-
sults. Of course, it is necessary to learn the routine, but that is
true of any kind of photographic processing when it is first
undertaken.

Stereo Radiography. One of the first types of technical stereo
was the stereo X-ray, first made before 1900 and used to some
extent ever since that time. The manufacturers of X-ray equip-
ment have produced special tube mounts and calibrated tables
to facilitate the use of stereo in X-ray, but the technique has
not been widely used for one reason. The radiographers do not
take sufficient time to master the fundamental principles of
stereo. In fact, from time to time articles have appeared in profesr

296 THREE-DIMENSIONAL PHOTOGRAPHY

sional journals which were based upon fallacy. As a result, there
are a few specialists who have done excellently well in stereo X-
ray, some who obtain satisfactory results and the great majority
who fail to obtain usable results.

The technique involved ordinarily makes use of two exposures
from two tube positions. These tube positions are based upon the
theory of convergence, but by the very nature of the X-ray emis-
sion they are fundamentally comparable to the parallel ray tech-
nique of the standard dual camera. This is because the emission
of the tube is in all directions within the scope of the mask used;
or, in other words, almost the exact reverse of the rays impinging
upon the camera lens. However, there are no truly axial rays, so
even with the actual tube converged the result is substantially that
of parallel axes.

The result, of course, is a highly variable relief depending upon
the tube distance versus spacing. It is a simple problem to one
who knows stereo theory thoroughly, but the fact that operators
believe the process to be extremely complex makes them commit
errors.

Recently, a new technique has been introduced which promises
to clear up a great deal of this confusion. This "new" technique is
nothing more than the adaptation to X-ray of the long and favor-
ably known rotary displacement, so common in stereo micro work.

The subject is placed upon a table which may be slightly ro-
tated, or the exposed hand, foot or head is simply rotated between
the exposures. This gives a full control of relief through angular
displacement, and the distance factor has little significant effect.

If the films are to be viewed in the normal Wheatstone viewer
with an eye-to-film distance of three feet, the angular value of the
rotation will be normal to human vision at three feet. Nor does
the rotation have to be exact. Twice the normal rotation will not
produce too much exaggeration. For locating foreign bodies, of
course, more precise control of angle is desirable.

Thus, while a great deal of space could be devoted to a discus-
sion of the older technique, it is suggested that those who are
interested in stereo X-ray work try this new technique, and for
experimental purposes try rotating the subject through approxi-
mately two degrees each side of center, a total excursion of four

SPECIAL PHASES 297

degrees, which will place the image in space at a distance of ap-
proximately three feet. It is understood, of course, that the axis
of rotation is perpendicular to the central ray of the tube.

By the use of a single fundamental technique, all object sizes and,
distances normal for any part of the human body can be used.

Vectograph. The clinical .X-ray is greatly increased in value
through the use of the Vectograph. The Wheatstone stereoscope
has never been really satisfactory as a stereo viewer. It is used
simply because, until the advent of the Vectograph, it was the only
form of stereoscope suitable for use with the large image sizes
incident to X-ray work.

When Vectographic prints are made, it is possible to hang the
X-ray film over the illuminator in the operating theatre. Then, by
using polaroid $D viewers with the bottom sectors cut away to
provide unobstructed direct vision, a surgeon has the full three
dimensional guide before him at all times, and it can be seen at
any distance which would be practical with the conventional
X-ray. It is a single film, as easily handled as the conventional
type, and it may be used with the same illuminators and is in
every physical way comparable to the single, conventional X-ray
film.

One of the greatest of the advantages of the Vectograph is the
fact that the three dimensional aspect is clearly seen at any dis-
tance at which the image itself may be seen; and it may be viewed
at any angle without disturbing the full stereo relief.

As has been described elsewhere in this volume, the Vectograph
process is easy, and does not require too much time. In emergen-
cies, of course, the time factor might be important, but ordinarily
the Vectograph should be available within one hour after the
original film is dry.

Another value of the Vectograph is that by using a ruler of
transparent material with one edge covered with Polaroid along
one axis and the other covered with the opposed axis, the two
component views are revealed. This makes it possible to make
direct measurements between the two positions of any embedded
object. This is very important in locating foreign bpdies or even
tumors. This simple device makes it possible to locate such inclu-

298 THREE-DIMENSIONAL PHOTOGRAPHY

sions in the most direct (and accurate) manner and with no loss
of time or involved computations.

STEREO DRAWING

Because the characteristic stereo-subjective reaction has its
stimulus in definite geometric relationships, it is possible to pro-
duce drawings based upon such relationships and to have these
drawings appear in full three dimensions. Once no more than a
novelty, it is now somewhat commonplace to construct true stereo
elevations from floor-plans.

Geometric Drawing. This should not be misunderstood. The
geometric drawing itself is easily constructed by stereo methods,
but many freehand artists make a "net" or "cage" in stereo by
geometric methods and then use this net as a guide for the stereo-
freehand drawing. This is similar to the use of cross-section paper
for the purpose of reducing or enlarging a freehand sketch.

Stereo drawing is based upon the principles of geometric per-
spective. Inasmuch as full sized textbooks have been devoted to a
discussion of this art, we do not have the space to develop the
theory fully. However, it is possible to give a superficial explana-
tion of the process.

Perspective. Perspective drawing is based upon a horizon line;
and somewhere along that line there are two vanishing points.
Usually one vanishing point is nearer the working center of the
line than the other, so that (assuming the object to be a cube), the
perspective view will show more of one side of the cube than of
the other. If the vanishing points are equally spaced, one edge of
the cube will be directly before the eyes and an equal portion of
both sides will be seen.

There are really two distinct steps in the drawing. The first
step is one which starts with the undistorted plan and converts
this into a plan in perspective.

The pivotal point is the station point or point-of-view. This is
the assumed position of the spectator who is looking directly at
the nearest point of the object.

If we wish to draw a cube in perspective, we start with the plan,
which is a simple square. The first step is to convert this plan into
the shape which it would have if we were above it and looking

SPECIAL PHASES

299

Fig. 20-1. Simple geometric figures illustrate principles of stereo

drawing.

at it from an angle, downward and in front of us, with one corner
of the plan nearest us. Lay a square sheet of paper upon the
ground, five or six feet in front of you and with one corner point-
ing toward you. That is the visual shape the perspective plan will
have. The nearest point of the plan (the corner) will lie in the
horizon line.

The horizon line should be near the center of the drawing
paper. Then, between that horizon and the bottom of the paper
and somewhere near the center, place the station point. The selec-
tion of this point is wholly arbitrary and your skill, as you pro-
gress, will be reflected in the pleasing appearance of the objects,
which is largely controlled by the selection of this point.

The details of drawing the plan are somewhat complex but are
not at all difficult to learn. There are several textbooks available
which treat the subject of perspective in a manner easily under-
stood by everyone, even those unacquainted with either drawing
or mathematics.*

In the second step, the perspective plan is substituted for the
station point. Verticals are erected from each significant point of
the plan. These verticals, which in the cube consist of one vertical
from each corner of the plan, represent the actual verticals in the
finished drawing. In our example these would comprise the four

* An excellent book for the beginner is "Perspective, An Elementary Textbook/*
by Ben J. Lubschez (Van Nostrand),

300

THREE-DIMENSIONAL PHOTOGRAPHY

vertical edges of the cube, one of which is at the rear of the cube
and so invisible, but its position must be known nevertheless.

Assuming that our eyes are level with the bottom of the cube,
the point of intersection of the nearest edge vertical and the
horizon is used as a starting point. The length of the edge is
measured upward from the horizon. From the upper end of this
line, other lines are drawn to the vanishing points. These form
the two front-upper edges of the cube. The side edges are the
projected verticals between these oblique edges and the horizon.
This is the simplest aspect of the cube of two sides; just one hori-
zontal base, three verticals and two oblique upper edges. With
the eye above or below the cube, the upper or lower face would
be seen, with edges running to the vanishing points.

Such is the very superficial description of the perspective draw-
ing. It is to be hoped it is sufficient to identify the steps in rela-
tion to a more detailed treatise.

Stereo Drawing. This brings us to the important phase of pro-
ducing a drawing which will have three dimensions when proper-
ly viewed.

The plan is put into perspective exactly as if the conventional
perspective drawing were to be made. Then the station point is
moved to one side a sufficient distance to permit a second plan to

Fig. 20-2. Target figures as used for testing and inclusion in
certain stereo-optical instruments.

SPECIAL PHASES 301

be placed in perspective without overlapping the first. For a draw-
ing to be viewed in the normal stereoscope, it is advisable that the
first plan should be kept to a size of two inches or less and that the
station point be moved 214 inches. In practice this usually means
that the drawing will be made in a size sufficiently large to permit
easy drawing and then the completed stereo pair reduced to cor-
rect size by photography.

When the dual plan is transferred to the second sheet and the
two perspective drawings made, they will automatically be in
stereo relief. It is understood that, although the station point is
moved, the vanishing points must remain the same for both
drawings.

Geometric Freehand Lattice. for reproducing a freehand
drawing it will be necessary to draw a geometric lattice (using the
preceding technique). This lattice surrounds the drawing like a
cage. Significant points in the drawing are given points in space
within this cage. Then the second cage is drawn in stereo relation-
ship and the corresponding (homologous) points are produced.
These points permit the correct stereo orientation of the second
freehand drawing.

Freehand Net. The freehand net is a type of graph paper in
which there occur vertical lines at intervals, usually of one centi-
meter. At a distance of one millimeter on each side of each line
are drawn broken lines, and spaced i.5mm from the outer sides of
the broken lines there are dotted lines.

The solid lines represent mid-distance, the other lines represent
varying distances from infinity to near foreground. One freehand
drawing (on tracing paper) is placed over the net. A correspond-
ing drawing is made over the other side of the net and the signifi-
cant points moved to one side or the other according to the dis-
tance of the point. For example, the nose is placed upon a solid
line, a similar nose outline is drawn upon a dotted line which is
nearer the first drawing. The first drawing is moved so that a dis-
tant hilltop is coincident with a solid line. In the opposite draw-
ing, for example the right one, the second hilltop is drawn coinci-
dent with the dotted line which is adjacent to and at the left of
the homologous solid line. Thus the drawings are shifted back
and forth to locate various parts of the drawing. It is necessary to

302 THREE-DIMENSIONAL PHOTOGRAPHY

work only with homologous guide lines, which are numbered to
facilitate identification.

Freehand Drawing. The best freehand drawings are actually
made freehand. The first is drawn just like any freehand drawing.
The second drawing is then made upon a piece of tracing paper
placed over the first and during the tracing the lines are separated
to right or left according to the spatial orientation of the detail
being shown.

It has been reported that there are a few artists who can make
the first drawing, and then, by using dissociated convergence, they
can draw the second unit, being guided by the actual stereo relief
developed by the drawing itself. If this report is true, such people
must be very few, although there is no logical reason why the
faculty should not be developed.

As far as we have been able to ascertain, no material covering
this phase of stereoscopy has been published other than the Stereo
Guild Manual of Stereoscopic Drawing.

Stereo Therapeutics. A great many people refuse to engage in
stereo for fear of injuring eyesight. Much of this fear can be traced
to very definite injuries of the kind in the past, because stereo
viewing can be injurious if the stereoscope used is of inferior de-
sign or badly out of alignment due to age or abuse.

Even though it does constitute repetition, we must say again
that stereoscope quality is of vital importance and that there is no
excuse for the existence of cheap viewers which are poorly made.
If you use a g8-cent viewer you are making exactly the same error
you would make if you were to buy your spectacles at the ten-cent
store instead of having them made by a competent optician. The
stereoscope is an accessory to your vision, and it should be of un-
questionable optical quality. This does not mean that it has to be
extremely costly. Good imported viewers still run around $50
and more, but we have domestic 35mm stereoscopes which sell
as low as $10 and which are of satisfactory quality.

Provided the stereoscope used is good, there is no question
about stereo being definitely beneficial. Ophthalmologists pre-
scribe definite stereo exercises which are carried out by means of a
standard type of stereoscope, and which make use of stereograms
of the usual type. The subjects are selected, of course, and the slide

SPECIAL PHASES 303

originals carefully prepared for this purpose, but there is no sig-
nificant factor involved which is not present in all stereo.

Visual Comfort. Strangely enough, this is a factor almost uni-
versally disregarded. Unless one is vain to the point of foolishness,
a pair of shoes must be comfortable, but a pair of spectacles as
often as not are far from comfortable. This is not the fault
of the ophthalmologist but results from our own failure to co-
operate during the examination. A pair of spectacles should bring
the same comfort to our eyes and to vision that a well fitted pair
of shoes bring to our feet.

It is also true that physical comfort demands a certain amount
of physical activity. Otherwise we become stiff and sore. The same
is true of vision. Eyes which are used normally for looking at ob-
jects at a variety of distances are more comfortable than those
which are strained by long periods of fixed distance vision. Many
of us have only rare opportunities to exercise true distance vision,
and as a result infinity vision becomes impossible to us without
strain. That condition which should be the "rest" position be-
comes an abnormal one. Stereo viewing provides opportunity for
true distance vision with resulting relaxation and the maintain-
ance of more nearly normal visual conditions. This does not make
close vision more difficult. On the contrary, the exercise itself
makes close vision more comfortable through the occasional re-
laxation of the strain of constant near vision. Near vision necessi-
tates a continued tension for both accommodation and conver-
gence, and there is the same strain which would result if any
muscle were kept under tension for hours at a time. If you want
to know just what happens, hold your arm extended in front of
you and in a fixed position for five minutes!

Prescription stereograms, if necessary, include some of positive
divergence. Each patient has prescribed those slides which will
provide the type of exercise he needs. Thus, by the simple act of
looking at attractive stereograms for a certain period each day,
visual comfort is tremendously increased and very often quite re-
markable improvements are made in serious malfunctions of the
visual apparatus.

Visual Skill This is even less familiar than visual comfort. Of
course we know that some people are unfortunate enough to have

304 THREE-DIMENSIONAL PHOTOGRAPHY

defective vision. We know that even among "normal" visions some
are better than others, usually with the comparison based upon
resolving power. But the idea of one person being able to "do
things" with vision which others cannot is wholly strange.

In fact, it is just as possible to have skill with the eyes as with
the hands or body. Two people can look at a field of vision. One
person will see much more than the other. If they match colors,
one can do so far more accurately than another. If they look at a
tree, one can estimate the distance and height much more accur-
ately. And if both look at three posts in the distance set very
closely together, one will be able to say which is the nearest and
which the farthest, while the other may not be able even to guess
at the relative positions.

These are a few examples of visual skill. It is more largely sub-
jective than physical, in most instances. However, the point in
which we are interested is that although stereo reproduces the con-
ditions of direct vision, those who are accustomed to viewing
stereograms develop such visual skills more rapidly than do those
who see the same objects in reality. Just why the artificial stimulus
should produce greater skill seems to be in doubt, but the fact it-
self has been too often demonstrated for there to be any doubt
whatsoever.

Stereo Supremacy. -It is very easy to become highly enthusiastic
about stereo. Therefore the writer would like to point out that his
stereo experience is more than 30 years old, quite old enough for
the sharp edge of enthusiasm to be worn off. He should also like to
emphasize the point that the statements of stereo superiority
which have been made are not expressions of personal opinions,
but facts. They are facts which are subject to demonstration, facts
which are matters of record.

CHAPTER 21
STEREO PROCESSING

COLOR is DEFINITELY one of the factors of stereo relief. But at
the same time stereo in monochrome has had a life of more
than a century and it would be foolish to expect monochrome to
disappear over night. In fact, there will be a demand for mono-
chrome stereo for years to come, for reproduction, for record
work, for experimental purposes and certainly among the old
guard who have given years to the gold toned paper print and to
the thiocarbamide toned transparency.

It is demonstrably true that no stereo paper print can equal the
quality of the monochrome transparency; and it is equally true
that the monochrome transparency can never equal the quality
of the color transparency. Despite this, there will be hundreds, if
not thousands, of paper stereo prints made in this country during
the next 12 months and for years to come.

It is a fact that diminished parallax is deliberately used with
only moderately close-up subjects to diminish the relief we could
never understand why. It is just as difficult to understand why
anyone should still be content with the diminished relief, the
silhouette form, which is common to the paper print because it
lacks the gradation of the color transparency. But just as long as
these things are done, there will necessarily be interest shown in
stereo processing.

It may be added, however, that if monochrome is used, -it is
highly advisable to do the processing yourself because the tech-
nique is much different from that used in the average commercial
processing laboratory.

In modern photography it is usually assumed that the emulsion
sensitivity, as indicated by the ASA rating, the time and tempera-
ture of the developer are all unalterable factors. This is not true.
Of course, the actual reaction of the emulsion to the impact of
light is substantially stable, but the amount of such reaction neces-
sary to produce a usable negative varies widely, according to the
development used.

It is known that in the usual developer, all tones both light and

505

306 THREE-DIMENSIONAL PHOTOGRAPHY

dark appear at about the same time, and that the highlights grow
more rapidly than the shadows. Thus, the normal negative pro-
gresses in both contrast and density as development proceeds. It
is also known that the density grows more rapidly after an in-
creased exposure. Thus it follows that if the exposure is increased,
printable density .will be reached coincident with a lower degree
of contrast than would otherwise be the case. Thus, by increasing
the exposure and decreasing the development, it is possible to de-
crease the overall contrast of the negative. This is the essence of
control development, and stereo development is above all control
development.

The possibilities of control development are not fully under-
stood by the average amateur. Consider an emulsion normally
rated at ASA 25. By the use of a high-energy developer, a print-
able negative may be obtained with 14 normal exposure. This in
turn means an exposure based upon a sensitivity of ASA 100*. But
if a fine-grain, superficial developer is used, four times normal
exposure is necessary, or an effective sensitivity of ASA 6.25. Yet
only one film is under consideration, and while it does have a
fixed primary sensitivity to light, its sensitivity in relation to the
developed image actually ranges from ASA 6+ to ASA 100, or a
range equivalent to that from a slow positive emulsion to a high-
speed panchromatic!

Degree of Deviation. The exposure is that required to record
the deepest shadow detail of significant size . . . and that means
anything larger than a pinhead in the negative. The deep shadow
is read by holding the meter close to it. The highlight is read at
its brightest exclusive of the rest of the scene. The normal photo-
graphic range from shadow to highlight is 1:100 (the "normal"
negative has less than gamma 1,0, so such a recorded range is de-
veloped to about 1:60 to 1:80 and can be printed on paper). Sup-
pose your shadow reads 4 and the highlight 800. The range is
1:200. You multiply the minimum reading by 50, the "latitude"
factor, 4 X 50 = 200 and give an exposure indicated for 200 maxi-
mum highlight (meter = 50) regardless of the 800 highlight. This
is only a rough rule but is usually near enough. But with 200 as
the maximum, highlights over 400 will be blocked.

It requires a bit of experience to learn the exposure control,

STEREO PROCESSING 307

but the exposure indicated for a maximum highlight of 50 times
the deepest shadow usually works out.

In development, the ratio of time to density will depend upon
the composition of the developer and the temperature. Some de-
veloping agents have steeper progression curves than others. But
as a rule of thumb, use normal development when the meter
range lies within 1:100, and try % to % development time when
the meter range is 1:200 or more. Your own experiments will
soon indicate the correct factors. It is better to have the negative
too soft than too hard. You can use a high contrast paper, but if
the highlights are really blocked, you can do nothing short of
chemical reduction.

But nothing systematic can be done unless both extremes of
light intensity in the original scene are measured separately.

Most modern meters are good, but we have been using the G.E.
because it has a triple sensitivity range which is convenient for
deep shadows and because it fits the H&H color meter attach-
ment, which has been discussed in the color chapter.

Control development is said to result in distorted tonal values.
Of course it does. How else can you reproduce an original range
of perhaps as much as i : 800 upon a medium which at best will
give you 1:80? You either "distort" the gradation and represent
every tone of the original in reduced form, or you obtain a print
in which most of the highlights are pure white, most of the shad-
ows are ink black and a small middle range is represented as a
succession of harshly stepped tones. It may be said that every
photograph made by sunlight either has such distorted tonal
values or it is worthless as a picture! The distortion, of course, is
but a matter of compression with some true but unnoticeable dis-
tortion of relative values in the extreme tones.

Developer. Among the developers not commonly used by ama-
teurs, both pyro and glycin are excellent, but both require experi-
ence. Old fashioned staining pyro is really superior to the non-
staining types, but care must be used to avoid spots.

However, the mild alkali developers such as the borax types are
good; fine grain developers are also good, although the surface
types are not recommended. However, you will probably not care
to try processing unless you are familiar with ordinary planar

308 THREE-DIMENSIONAL PHOTOGRAPHY

processing. You will, therefore, have some favorite developer, and
if it is one which gives you good printing negatives you can use it.

All stereo manuals emphasize the necessity for very soft nega-
tives, and this is misleading. In the old days of stereo, the normal
printing method in photography was by contact, and contact
printing requires a negative which is considerably more contrasty
than one which enlarges well. Today we deal mostly with nega-
tives for enlargement and they are just about right for stereo
contact printing. Therefore, you may discount most of the re-
marks about general softness provided you have a negativ^ from
which you can make a print which does not show either a dense
shadow without detail or a glaring highlight which also lacks de-
tail. But the detail need not be strong. The most delicate hint is
enough provided it is unmistakable. As a matter of fact, most
modern negatives do have unblocked highlights so the pivotal
factor becomes simple, but be sure the deep shadow has detail.

Formulae. There is no good reason to give formulae here.
Those used are standard; and occasionally a new formula is pro-
duced. There is no such thing as a specific stereo formula, par-
ticularly now that fine grain is a standard procedure. If you al-
ready do your own processing, use what you normally use. If you
have had no experience, you require guidance in greater detail
than can be given here.*

Fixing. We do have a suggestion to make regarding fixing.
There is a tendency for fixing solutions, particularly hypo, to at-
tack the image. This action is not vigorous, but it is a mistake to
develop negatives with very delicate shadow detail and then to
permit them to soak in the fixer for an hour or so, or as we re-
cently found one amateur doing, to soak them all night long in
fixer! Use the solution fresh and use it according to directions.

Rapid fixers are available which have proven satisfactory. In
the Guild laboratory we still use hypo, and make it according to
the ancient 1:1:1 formula. This uses a hardener made up of i
ounce each of sulfite, glacial acetic acid and potassium alum to
each half gallon of 20 percent hypo. This is more acid than the
usual formula, but we can develop normally at 85 and above,

* For the beginner we heartily recommend "How To Make Good Pictures" and the
"Kodak Reference Handbook," both published by Eastman Kodak.

STEREO PROCESSING 309

and do not have pinholes, reticulation, blisters and the like. It
may also be said that repeated comparative tests have shown that
this fixer does not degrade print quality as so often stated.

Washing. EG sure to wash thoroughly. If your negatives have
been properly hardened you can wash two or three hours without
danger, but one-half to one hour in some device which thorough-
ly agitates the films or prints, or in a rack which keeps them posi-
tively separated, is sufficient. We have stereo negatives on hand
which were made in 1918 and were fixed and washed this way.
They are as printable as the day they were made.

Swab the negatives off after washing, squeegee to remove sur-
plus water and dry in a place where drying occupies from one to
three hours, neither too rapidly nor too slowly.

As soon as they are dry, cut the negatives apart, transpose them
and place in individual protective envelopes. Negatives left lying
around collect dust and dust makes scratches even upon a proc-
essed negative. If a negative is worth making it is worth protecting
and a scratched stereo negative might as well be destroyed. ' *

THE STEREO POSITIVE

For almost a century, stereo negatives have been made princi-
pally upon glass, and printed upon either paper or glass. The re-
cent revival of stereo is only a few years old, and is based upon
the use of 35mm reversible color film. The physical differences
have resulted in confusion and in the inevitable attempt to modi-
fy old processes by modern techniques.

Thus today we have adherents of glass for both phases, of film
negatives and glass positives, of film negatives and positives, of
glass negatives with paper positives and film negatives with paper
positives. Only in the modern phase do we find a definitely stand-
ard procedure, that of 35mm color film.

It is difficult to obtain glass' sensitive material, and only a few
dealers carry it. The general trend is toward the use of film in
the camera, paper for prints and film for transparencies. It has
been our experience, however, that glass diapositive plates make
possible a quality which we have not been able to duplicate with
sheet film. However, that is a minor matter as the difference is
slight.

310 THREE-DIMENSIONAL PHOTOGRAPHY

One thing is important. There is nothing available in the
35mm field to compare with the automatic cabinet viewer or
"classifier." The result is that many users have adapted the old
cabinets to modern mounts. A viewer which makes use of side
guides, such as the Taxiphot, presents quite a problem in rebuild-
ing, while the Mattey, with its loading weight resting on top of
the slide, and already adaptable to 6x13 or 45x107, requires only
shortening of the trays. For glass mounted 35mm slides, color trays
are used ( 1 2 slides to the tray) and for rigid cardboard the normal
25 slide trays are used. The semi flexible "folder" type of card-
board mount is not rigid enough for either projection or auto-
matic viewer use.

The same thing is true of film positives made from the older
negatives. The film is usually unmounted and is too flexible for
anything but the hand viewer. If glass mounted, the slide is too
thick for anything but color trays in the cabinet. There have
been some successful experiments in the use of thin plastic for
binding, but for projection it must be remembered that some
plastics are optically active and produce rainbows over the screen.
If slides are to be projected it will be necessary to use some plas-
tic material which is inactive. The acrylics are satisfactory, but
the cellulose plastics are, as a rule, very active.

We shall of course ignore the 35mm color positive as it is proc-
essed for you.

Paper Prints. There is nothing out of the ordinary in making
the paper print other than the problem of keeping the tonal range
within the range of the paper with no burned out highlights or
coal black shadow areas.

The short tonal range of paper prevents the full tonal repro-
duction of the original and hence tends to diminish the degree of
stereo relief and produce stereograms in which round objects pre-
sent a decidedly flattened appearance. Human figures resemble
"gingerbread men." This defect is less noticeable when soft, full
range prints are made and increases as the print becomes harsh
in contrast.

Masking. Prints may be masked as you choose, but the film
negatives should be carefully spaced when transposing. The usual
3x6 stereogram has no space between the units, but it is sug-

STEREO PROCESSING 311

gested that the amateur slit adhesive tape to a i/" width and
leave this much space. The mask should be cut with a 3/16" bar
between opening so that any slight irregularities of the i/" tape
can be obliterated.

Prints 6x13 and larger appear best with rectangular masks hav-
ing slightly rounded corners; 45x107 and smaller with sharp
cornered, rectangular masks. If you choose you may follow the
old convention of using a domed top with a sharp cornered bot-
tom. This is almost universal in the 3x6 size, although no one
has been able to cite a good reason for it. Perhaps the suggestion
of an arched window seemed more "elegant" to the Victorian
stereographer.

It is also known that commercial paper prints in 3x6 are
mounted upon cards which have a decided curve. We have been
unable to discover any reason for this practise which is certainly
optically inferior to the flat mount.

Window trimming or masking is useful when open type viewers
are used, but of less value in the better grade stereoscopes which
have the field masked by the visual tunnels in the instrument.

Transparencies. It is difficult to discuss this subject, because it
is one photographic art which has relatively few adherents today.
The stereo transparency, aside from the duality of image, is iden-
tical with the lantern slide. The stereographer can do nothing
except get some materials and work at it until he has mastered
the knack of doing it well. We know of no better guide for the
beginner than the lantern slide brochure in the Kodak Reference
Handbook.

The difficulty lies in the fact that in transparency making, the
fixed exposure and development are abandoned, and the contrast
of the positive controlled by a sliding scale relationship between
exposure and development.

If the highlights have the correct delicate tone but the shadows
are weak, decrease exposure and increase development.

If shadows are strong, but highlights fully clear, increase ex-
posure, decrease development.

If the image is weak throughout with clear highlights, increase
both exposure and development.

312 THREE-DIMENSIONAL PHOTOGRAPHY

If highlights are gray and shadows black, decrease both ex-
posure and development.

That is but a rough guide, and only experiment and experi-
ence will enable you to make a good transparency. It is not, how-
ever, difficult to learn if you really try.

The good transparency is almost black as it lies in the developer,
far darker than a good paper print. The final judgment can only
be made after the dry slide is examined in the viewer, but you
can learn, by looking through the slide at the yellow-green safe-
light to judge very closely.

Because the transparency will reproduce many times the tonal
range of the paper print, the negative should be made for trans-
parency printing. You will find no negative which works at its
best in both media. The negative for the transparency will be
more vigorous than usual and will, in fact, be fully as vigorous
as the hardest negative considered suitable for planar projection
printing under modern methods; harder in fact than most ama-
teurs ever use.

Project! on .Projection printing has already been mentioned.
If the image is enlarged two diameters, the viewer should have
lenses whose focal length is twice that of the camera lenses. This
is the only significant factor in projection printing. Extreme care
must be used, because the slightest softness of focus becomes visi-
ble in the stereoscope.

Ton in g Perhaps no division of amateur photography lays such
stress upon toning as does the older stereo field.

Among the old guard a stereogram simply isn't ready for view-
ing until it has been gold toned. Again no one seems to know
why. The technique apparently was born at the period when a
commercial photographer couldn't sell a print unless it was
"sepia." For some reason the richness of black did not appeal.
Planar photography recovered from the fad, but it remained a
stereo requirement. It is true that occasionally a gold toned print
is really beautiful, usually it is a ghastly "half-baked" brown, not
greatly different from the usual "sepia" tone.

If you grow to the point where you wish to make use of gold
toning, you can purchase the necessary materials ready mixed at
any photographic store. Just follow directions. It is a tricky process

STEREO PROCESSING 313

but eventually you can master it. I do not suggest you take the
time to make up your own solutions.

The transparency makers also lay stress upon toning, but with
better reason. The tf ansparency does have an extended tone range,
and often it is visually harsh when the gradation is good. The
toner most favored is even more temperamental than gold toning,
but yields a beautiful blue-gray range of tones which is the most
realistic monochrome for outdoor scenes I have ever seen. This is
the thiocarbamide toner, and is used as a part of the developer,
not as a subsequent treatment.
SOLUTION A

Metol 22 gr 2.5 g

Sodium sulfite 130 gr 15 g

Hydroquinone 90 gr 10 g

Sodium carbonate 130 gr 15 g

Water to make 20 oz 1000 ml.

SOLUTION B

Ammonium carbonate 2 oz 100 g

Ammonium bromide 2 oz 100 g

Cold water to make 20 oz 1000 ml.

SOLUTION C

Thiocarbamide 26 gr 3 g

Ammonium bromide 9 gr ig

Water to make 20 oz 1000 ml.

Color is controlled by varying the proportion of A, B } and C.

Relative
Color ABC Exposure

Pure black 7 i/ 2 i/ 2 i

Blue black 6 n/ 2 i/ 2 2

Blue 51/2 2 1/2 4

Violet 5 2i/ 2 i/ 2 8

Finishing. Glass positives should always be flowed with a good
negative or print lacquer as soon as thoroughly dry. They should
then be further protected by filing in individual envelopes or in
a filing cabinet which is provided with slots or other means for
keeping the slides separated. With the cabinet viewer, the slides
are filed in the viewer "trays."

Those of you who wish to get into stereo processing seriously
will not find it difficult, although it does demand careful tech-
nique. Carelessness and stereo do not mix. But with reasonable
care there is no great difficulty, and most of the procedure is iden-
tical with that of planar photography.

APPENDIX A
THE PRISM DIOPTER

In stereo, we are vitally concerned with two visual axes which
converge. Parallel visual axes do not give any stereo parallax,
hence are useless in stereo. That is, objects at infinity lie beyond
stereo infinity. For that reason we are concerned only with con-
vergent axes and with light rays which have other corresponding
angular relations.

It is commonplace to use the prism diopter as the unit of meas-
urement, rather than the conventional degrees, minutes and sec-
onds.

The prism diopter is a prismatic power which will cause a
beam of light to deviate to the extent of one part in one hundred.
Thus if you project a target upon a screen at a distance of 100
inches from the projector, and mark the position of the center
of the target; and if you then place a prism over the lens of the
projector which causes the center of the target to fall one inch
at one side of the original position, you have a prismatic devia-
tion of one diopter, and that prism is said to have a power of one
prism diopter.

Of course, if the prism causes a deviation of i inch in 100, it
will produce a deviation of i centimeter at a distance of 100 centi-
meters or a deviation of i foot at a distance of 100 feet.

It is customary to give the deviation as applied to one ray only,
that is, if each eye deviates one prism diopter, we have actually a
total convergence of two prism diopters, but we speak of the con-
dition as a deviation of one prism diopter. Note that "deviation"
is the movement of one eye, "convergence" is double the deviation
and applies to both eyes. This is not a fixed rule and often in
optical literature you will find the terms used interchangeably,
with considerable resulting confusion.

THE HALF ANGLE

This is similar and refers to the monocular deviation. The base
is the distance from the pupil to the center of the nose, or one-
half the interpupillary distance. Conventionally the base has a

314

APPENDIX A 315

value of 32.5mm, which is equal to 3.250111. As one meter con-
tains 100 centimeters, the deviation normal for a distance of one
meter is 3.25 prism diopters, or a total convergence of 6.5 prism
diopters.

There is no fixed angular value for the prism diopter. One
prism diopter has an angular value of about o34'23", so we
should expect 100 prism diopters to have 100 times this value or
57i8 / 2O // . However, by definition the distance through which
the beam is deviated is compared with the distance of the sur-
face upon which the deviation is measured, so at a distance of 100
units we should find a lateral deviation of 100 units giving us
two equal legs of a right angled triangle, so the angular deviation
must be 45 for 100 prism diopters. The reason for this is that
each additional degree intersects the measuring plane at a greater
distance than the preceding one, so that each succeeding prism
diopter has a smaller angular value than its predecessor.

This method of angular measurement is considerably more
convenient in our work than the conventional angular measure-
ment.

PRISM BASE

Another term often encountered in reference to steroscopic
vision is "base-in" and "base-out," with certain prism diopter
values. Thus we might encounter "base-in 13 diopters." This is
actually the normal convergence for looking at an object at a
distance of 50011. That is, if two parallel rays are deviated toward
each other by the use of a pair of prisms, each of 6.5 power, the
two beams would be deviated toward each other 6.5 prism diopters
and would have a total convergence of 13 prism diopters. If these
parallel rays are separated by 65mm, this 13 diopter convergence
will cause the rays to meet 50011 in front of the prism.

On the contrary, if you have a standard therapeutic stereoscope
with soomm lenses and gomm base, one prism diopter will have
a lateral value of 2mm. If you prepare a stereogram for this in-
strument in which the maximum homologous separation is g6mm,
you have the condition of 3 prism diopters, base-out.

Thus "base-in" indicates convergence while "base-out" repre-

316 THREE-DIMENSIONAL PHOTOGRAPHY

sents divergence, both related to the parallel vision of the normal
condition o rest.

Most people can converge 32.5mm in 200, that is, look at an
object 8 inches in front of the nose. This is a deviation of 16.25
prism diopters and a convergence of 32.50 prism diopters for both
eyes. On the contrary, most people find difficulty in fusing with
as little as 2.5 prism diopters divergence from the parallel.

DEVIATION IN STEREOGRAPHY

Deviation is important in stereography, because of the variations
in convergence necessary for fusing images of close-up objects and
also variation at a distance.

For example, when slides are to be projected, there should be
a maximum of 3 prism diopters variation between the nearest and
farthest images, and 2.5 prism diopter is even more advisable.
But it is important to note that this is a differential value, not an
absolute one. For example, assume that the camera lenses are
spaced 7omm, with a 35mm half base. Then a deviation of i prism
diopter occurs at an object distance of $i/% meters, although to
average vision (65mm) it would be 314 meters. A deviation of 3
prism diopters would therefore be encountered at a distance of
3.5/3 ~ 1.166+ meters.

Under the 3 prism diopter rule, objects which lie between 1.166
meters and infinity can be fused comfortably. If the 2.5 deviation
is used, the minimum distance will be 3.5/2.5 = 1.4 X !-i66 or
1.63 meters, about 51^ feet.

However, if we compute the deviation for 6 diopters, we find
it to be 0.585 (58.3 centimeters) and 81 centimeters for the 5
diopter distance (31/^x2).

Therefore, if we stereograph an object at a distance of 58.3011
(23 inches in round figures), and have objects at infinity in the
same picture, the range of fusion is excessive and the stereogram
will not be comfortable to view. However, if the object at 58.3011
is placed before a wall whose distance is 1.166 meters (46 inches
more or less), the range of deviation will be from +3 to +6, and
the stereogram will be just as comfortable to view as the one whose
range is from o to +3.

Using the 2.5 maximum the near distance would be 32 inches

APPENDIX A 317

with the background at 64 inches, giving a range from +2.5 to
+5 diopters-

The adverse criticism levelled against the very close close-up in
stereo may be justified in those instances where there is a distant
background, but the criticism is not warranted when the total in-
cluded range lies within limits. Most of the criticism against the
close-up originates in the fact that many stereo technicians have
inherited a tradition but are unfamiliar with its technical nature.
A close-up is, to them, inherently poor regardless of any and all
other circumstances,

DEVIATION IN PROJECTION

When projecting a series of slides, it is advisable to have them
pre-adjusted, because if the lateral separation is set for distance
subjects, the projector will maintain the infinity separation and
close-up subjects will require compensating adjustment.

Slides for projection should be arranged so that the separation
of the farthest images should be uniform. For example, in the
usual 35mm slide, assuming it has been made with lenses of 35mm
focal length, we may assume one prism diopter to have a lateral
value of 35/ioomm, which is roughly one-third millimeter. Let
us also assume that in standard mounting, the infinity-homologous
distance is 65mm. Slides which include infinity, have images of
objects at infinity separated by 65mm. But such slides should have
no images separated by less than the limiting deviation of 3 prism
diopters or a total of 6 prism diopters for the two units. Therefore
the images of the nearest objects should be not less than 65 2.1
or 62.9mm apart.

If, however, the widest separation (image of most distant ob-
ject) is 62.9, then the nearest may be as little as 60.8 in separa-
tion. In that instance, the films themselves are moved apart
2.1 mm. This slide which represents an original field with a depth
extending from 23 to 46 inches, will project just like the one
which represents an original range of 46 inches to infinity.

The whole technique lies in keeping the images of the most
distant object at the nominal homologous separation for infinity.
This applies to slides made at a range of i H/% to 23 inches, as well
as to the other two examples cited.

318 THREE-DIMENSIONAL PHOTOGRAPHY

SPECIFIC DEVIATION

I you will divide the focal length of the camera lens by 100,
you will have the nominal lateral value of one prism diopter
under taking conditions (variation in value caused by extension
of lens in focusing is ignored). Dividing the focal length of the
viewer lenses by 100 will give you the nominal value for the visual
prism diopter. The ratio of these two will give you a numerical
evaluation of the inherent distortion. For example, if the camera
has a prism diopter value of o.gsmm, and the viewer has one of
0.50 diopter, the inherent stereo distortion amounts to 35:50 or
70:100, which may be interpreted as a 30% distortion.

It is significant that only when the distortion exceeds 50%
(that is with the viewer twice the focal length of the camera),
or when the viewer has less than two-thirds the focal length of the
camera, does the distortion start to become visible.

PRACTICAL APPLICATIONS

A little experiment will soon show you the tolerance which you
have and the range through which you can deviate images with-
out detriment insofar as your specific audiences are concerned.
Once this knowledge is available, you can compute the character-
istics of your slides and mount them with accurate compensation
for deviation. This produces an excellent slide for direct viewing
and for projection.

APPENDIX B
STEREO GLOSSARY

There is nothing very difficult or different about stereo, but it
is a distinct branch of photography and has its own terminology.
These specific words and phrases are not intended to confuse,
they simply make it possible to say some specific thing in a few
words and without which it would require many more words to
make the same statement, and in less satisfactory manner.

Although you will soon become familiar with these terms, it is
convenient to have a glossary available while you are still un-
familiar with the subject. This glossary is offered for reference.
It is far from complete, but it does include the most useful words
common to stereo but less familiar outside the field.

As you progress you will meet many terms not included here,
but for the most part they will be terms dealing with more ad-
vanced phases of the subject.

American stereoscope. The common par-
lor or Holmes version of the Brewster
stereoscope. Also Mexican stereoscope.
Anaglyph. A. stereogram of two superim-
posed images in which the images are
blocked to their respective eyes, so each
eye sees only its own proper image. The
original form was the bichrome in which
the images are printed in two comple-
mentary colors and viewed through cor-
respondingly colored niters. A half cen-
tury ago the process of using polarized
light instead of color was patented, and
today is favored over the color type for
projection. In motion picture work, alter-
nate frames are dyed in complementary
color and the film run at double speed to
produce the alternate anaglyph. Both bi-
chrome and polarized types are now in
use. Books and magazines, for practical
reasons, are usually illustrated in bi-
chrome anaglyphs.

Anaglyphic restoration. Reconverting an
anaglyph to its original form of two sep-
arate photographs to be viewed in a con-
ventional stereoscope.
Autostereo plate. An obsolete plate with
a parallax grid on the back, used for
making direct parallax stereograms. Re-
versal processing.
A utotransposer. Either a printing frame

or machine in which provision is made
for transposition by simply pulling two
levers.

Auxiliaries, stereo visual Factors in ster-
eo vision which enhance the stereo relief
but which are not essential to stereo
vision. The two most important are the
effects of perspective and overlapping of
distant objects by those nearer.
Base. The stereo base is the distance be-
tween homologous optical points, usually
the optical axes of the camera, but also
applied to the separation of viewer lenses
and at times to human interpupillary.
Box stereoscope. A. stereoscope in the
form of an enclosed box as contrasted
with the skeleton form of Holmes. Stand-
ard in Europe.

Brewster,, Sir David. Probably the true
inventor of stereoscopic reproduction, al-
though the claims of Wheatstone are also
favored by some. Brewster was the first to
point out the overwhelming importance
of the subjective in stereo. He also in-
vented the prismatic stereoscope (semi-
lenticular) which is properly known by
his name, but often called Holmes, Mexi-
can, American and the like. These are
only one type of Brewster construction.
C. T. meter. Color temperature meter.
See Color meter.

319

320

THREE-DIMENSIONAL PHOTOGRAPHY

Cham stereoscope. A. cabinet stereoscope
with the views hung in chains which pass
over twin sprockets thus bringing into
view one stereogram after another. More
familiar in Europe than here.
Chromatic perspective. See Stereo-chro*
matism.

Chromatic relief. See Stereo-chromatism.
Color correction The use of faintly tint-
ed filters with color film to compensate
for off standard color of incident light.
More important in stereo than planar
photography as normal color is an im-
portant factor in panortho stereo. Also
used to adapt color film to a type of light
for which it is not designed as the con-
version filter to use Type A kodachrome
with daylight. Also color control.

C. C. Filter. One type of filter used in
color control. Also L. C. filters.
Color meter. A special meter which indi-
cates the spectral balance of any existing
light of photographic level, and by doing
so indirectly indicates the proper correct-
ing filter which should be used. Among
the most popular are the HScH attach-
ment which converts the G.E. exposure
meter into a color meter and the Rebi-
koff precision color temperature meter,
made in Switzerland.
Color perspective. See Stereo-chromatism.
Complementary stereoscope. A. stereo-
scope matched to the camera as is essen-
tial in orthostereoscopy.
Compressor, image. A cylindrical lens
used in making stereo motion pictures to
permit a rectangular screen image instead
of the narrow vertical format of a half
frame.

Contour reversal. A type of space con-
trol in which certain elements of the scene
are shown in pseudoscopic relief while
others are shown in proper stereoscopic
relief, although both elements are com-
bined in a single stereogram.
Convergence. The meeting of visual (or
other optical) axes when produced from
two separate stations to a common object.
Absolute convergence is the actual angle
between the convergent axes and is identi-
cal with the parallactic angle. Relative
convergence is the convergence upon one
object within a scene compared with that
upon another object. Relative convergence
is the more important in stereo photogra-
phy. Convergence is often said to be the
primary stereo stimulus, but this has been
disproved.

Convergence camera. A method of mak-
ing stereograms by altering the direction
of the camera between exposures so the

object is centered in both films. Impracti-
cal for any subject except one limited
largely to one plane. Otherwise more or
less serious distortion is introduced.
Conversion filter. A specific type of color
control filter designed to permit the use
of artificial light film by daylight or vice
versa.

Coronet (Trade name). An inexpensive
stereo camera and viewer using 828 or
equivalent size film. Comparable to a de-
luxe box camera.

Deviation calculator. Dial calculator for
determination of the range of deviation
in any scene before photographing.
Diagnostic stereo.~-The use of stereograms
to aid in the diagnosis of visual disorders.
Diopter. Ophthalmic lenses (and viewer
lenses) are designated by power rather
than focal length. The power in diopters
is equal to 100 divided by the focal length
in cm. Thus a soon lens is 5 diopter, a
locm is 10 diopter, a 5 cm is 20 diopter
and so on.

Diplopia (Double vision). Ordinarily an
abnormal visual condition, but normal to
all stereo vision outside the macular area.
Drawing, stereo A drawing made in
duplicate but with stereo differentiation
so that when viewed stereoscopically it
will have stereo relief.
Equivalent distance. In stereomicrogra-
phy, a photograph made at a very short
distance is viewed as if it were at some
greater arbitrary distance, usually "read-
ing" distance. This apparent distance is
the equivalent distance. Also applied to
other distortions, deliberate and acci-
dental, of stereo relief.
False stereogram. (See Spreader, Twister,
Infusible.) A true stereo pair so mounted
that viewing is uncomfortable or im-
possible.

Flicker projection Stereo projection in
which the two images are projected upon
a screen in rapid alternation, and viewed
through a shutter which alternately ob-
scures the right and left eyes in synchro-
nism with the projector.
Frame masking. Masking the top and
bottom of the aperture in a motion pic-
ture camera (less commonly in the pro-
jector), to produce the desired square
stereo screen format.
Free vision Any system which permits a
true stereoscopic effect to be seen without
the use of any kind of viewer.
Fusion. The visual registration of the
two images in the eyes. This is distinct
from the phenomenon of subjective syn-
thesis to which the term is often applied.

APPENDIX B

321

Purely a physioneurological factor.
Ghost image .Any visual or photographic
image which permits objects behind it to
be seen; a transparent image. The condi-
tion common to images of objects not the
subject of direct vision.
Glyphoscope (Trade name). A simple de-
vice now obsolete, used as both camera
and viewer.

Graflex (Trade name). The stereo-gra-
flex uses 5x7 sensitive material to make
stereo pairs 31/^x5. Has stereo viewer
lenses in focusing hood.
Grid stereo. Narrow strip elements from
two images alternate. Viewed through a
grid spaced before the print, the two eyes
see different images due to angular vision
through the grid. The least satisfactory
of the general integrated type of free
vision stereogram.

GuildmountA special, rigid, cardboard
stereo mount made for the Stereo Guild.
Homologue. That which exactly corre-
sponds in position, proportion or struc-
ture. In stereo applied to corresponding
points or parts in the two images.
Homologous distance The distance be-
tween two homologous points. Identical
with base or separation when the homolo-
gous points lie at infinity.
Homologous points. The two point im-
ages in the stereo pair which exactly cor-
respond. The two images of the same
nailhead in a board, for example, are
practically homologous points.
Hyperstereo. Stereo with an extended
base. Results in theoretical error which is
practically imperceptible. Reduces appar-
ent distance of objects, which appear as
scale models of the larger object.
Hy poster eo. Stereo with diminished base,
Used for close-up and stereomicrography.
Iloca. Stereo version of the 35mm Iloca
camera.

Infusible stereogram. A stereogram so
made or so mounted that it cannot be
fused.

Integrated stereogram. A free vision ster-
eogram of the basic grid type but oper-
ating upon different optical principles to
provide stereo vision from integrated ele-
ments of the two images. A lenticular
(spherical or cylindrical) screen replaces
the grid. Peristereoscopy.
Interocular. Basically the distance be-
tween the two eyes; also applied to the
separation of camera or viewer lenses.
May be variable.

Inversion. The complete reversal of posi-
tion exhibited in the stereo negative as

compared with the true positions. The
optical phenomenon which makes neces-
sary the transposition of stereo prints.
Kelvin meter. Color temperature meter.
See Color meter.

Lens, stereo-prismatic. Supplementary
lenses for stereo close-ups which combine
both the positive refraction of the con-
ventional close-up lens and a prismatic
deviation which permits the image of the
close-up object to be centered in the film
area.

Light control See Color control.
Macular vision. That part of vision which
is characterized by maximum definition
and resolution. Also known as "sharp"
vision. The object of macular vision is
that object upon which the visual atten-
tion is fixed.

Magnification of viewer. In orthostereo
the magnification is unity, as the taking
and viewing lenses compensate. A viewer
with lenses of less than the camera focal
length results in undesirable magnifica-
tion of the image. Magnification as a pri-
mary function of the viewer may be ig-
nored when the viewer is a complemen-
tary one.

Metastereoscopy.'R.epToduction of origi-
nal in normal proportions but in abnor-
mal size. Common in stereomicrography.
Mexican stereoscope. Holmes stereoscope.
Mirror stereoscope A stereoscope which
involves reflection from a mirror. The
Wheatstone stereoscope largely used in
X-ray work is one type. See Reflecting
stereoscope.

Monocular viewer. This is not a stereo-
scope. It is a single lens which is used to
give to a single photograph a three-dimen-
sional but not stereoscopic appearance.
Pantoscope.

Normal stereoscope. Any stereoscope
which uses stereograms whose infinity-
homologous distance is substantially equal
to the average interpupillary separation,
i.e. 65mm. Contrasted with the Brewster
in which the I-H points are much farther
apart than the normal human eyes.
Orthostereoscopy. That tyj>e of stereo-
scopy in which the image is reproduced
in full natural size as seen at full natural
distance and in full normal degree of re-
lief. Not possible with projection nor free
vision; only when an individual viewer is
used, and this must be complementary to
the camera.

Panortho-stereoscopy. The nearest ap-
proach to perfection in stereo. This is an
orthostereogram in which full advantage
has been taken of the auxiliaries (q.v.),

322

THREE-DIMENSIONAL PHOTOGRAPHY

and which is in full, corrected, natural
color. By all means the most perfect type
of photograph yet produced.
Parallactic distortion. A. degree of paral-
lax too small or too great for the object
size. Objects appear as thin (i.e., silhou-
ettes) or elongated in the direction of the
optic axes.

Parallax. A difference. Specifically the
difference of the visual images which re-
sult from the two eyes having different
points of view. The fundamental external
stimulus of stereoscopic vision.
Absolute P. The parallax of a given
point as determined by the angle be-
tween the visual axes meeting at that
point.

Circular P. Stereo relief produced by
rotating the object rather than by hav-
ing two separate points of view.
Differential P. The variation between
the parallax of two points upon the
surface of an object. The fundamental
external stimulus which enables us to
see roundness or spatial difference.
Dynamic P. The ever changing paral-
lax involved when the eyes shift from
object to object. The specific external
stimulus of stereoscopic vision.
Illumination P. In stereomicrography
the production of parallax by opposed
light beams rather than by physical
difference.

Relative P. The parallax of one object
or point as compared with that of an-
other. Closely similar to differential
parallax but used as a basis of meas-
urement.

Rotary P. See Circular parallax.
Stereo P. Parallax as specifically relat-
ed to stereo vision and in contrast to
strictly geometric parallax.
Parallax stereogram. Commonly but in-
correctly used to designate an integrated
stereogram. Because parallax is the vital
factor in stereoscopy, any stereogram of
any type might properly be called a
"parallax" stereogram. A free vision ster-
eogram, see Grid stereo.
Parallel vision. The normal condition of
vision at complete rest. In stereo the
ability to fuse a stereogram of conven-
tional type without the use of a viewer.
Accomplished by separating the habitual
co-action of convergence and accommo-
dation.

Parastereo. A stereogram which is ap-
parently ortho in quality but which can
be demonstrated to possess the anomalous
spatial relationship of the stereo tele-
scopic field. To all intents and purposes

an orthostereogram of telephotographic
type.

PePax. Coined term to indicate the rela-
tionship between the apparent size and
the apparent relief of any object. Normal
PePax is characteristic of the ortho-
stereogram. Abnormal PePax indicates
parallactic distortion.
Personal. A miniature stereo camera
using standard 35mm film but making 70
stereo pairs of substantially i6mm size
upon a standard 36 exposure film. Also
applied to camera and mounting acces-
sories made by Sawyer's, Inc.
Perspective. The disposition of lines to
represent a solid object upon a plane sur-
face. In stereo regarded as the comple-
ment of parallax. Perspective has to do
with "size-in-width" while parallax has
to do with "size-in-depth."
Planar. Restricted to a single plane. Or-
dinary photography in contrast to stere-
ography.

Polarized lightLight which vibrates sub-
stantially in a single plane. Used in stereo
largely because of its property of extin-
guishing an image when planes of polar-
ization are crossed.

Polaroid (Trade name). A synthetic ma-
terial which has the property of polariz-
ing light. Used in stereo projection of
polar anaglyphs.

Prismatic stereoscope. A stereoscope in
which the lenses are really half lenses,
thus forming spherical prisms. The Brew-
ster type of stereoscope. Also a reflecting
stereoscope using prisms. Often an auto-
transposing viewer.

Prism diopter. The unit used in meas-
uring the amount by which a beam of
light is deviated by passing through a
prism. One prism diopter deviates one
unit in 100.

Projection gage A film gage to facilitate

mounting stereo films with the correct

separation to compensate for various

ranges of deviation.

Pseudoscopic.The effect of untransposed

stereo images. Depth values are reversed

with far objects seen nearby and vice

versa.

Pseudostereoscopic.A. "stereogram" made

up of two identical images. As there is

no parallax, no stereo relief is seen. Such

pseudostereograms have actually been

marketed as stereograms.

Realist (Trade name). A widely popular

35mm stereo camera. Also generic name

for a projector, viewer and other 35mm

stereo accessories made by David White

Company.

APPENDIX B

323

Separation. Stereo parallax results from
the existence of two optical (visual) sta-
tions. The distance between these is the
separation. Normal ocular separation is
assumed to be 65mm, but in fact varies
widely through several millimeters upon
each side.

Slide. Common name for the stereogram.
S. commercial. Stereograms made for
sale to the general public rather than
for private use.

Sliding base. A. camera base, usually tri-
pod head, which enables a conventional
camera to be moved laterally 65mm so
that stereo negatives may be obtained by
successive exposure.

Space control. A method of stereography
in which objects can be given an appar-
ent position in space which is not the
natural one. A miniature human figure
for example can be posed upon a table-
top with every appearance of realism.
Spreader. A stereogram with unit images
too far apart necessitating divergence of
the optic axes, painful at best and im-
possible for most.

Stereo. Abbreviation for stereoscopic, also
as prefix for many compound words.
S. book A book illustrated by stereo-
grams.

S. calisthenics. Exercises for the eye
muscles obtained through viewing spe-
cial stereograms. Intended to produce
visual comfort and skill rather than to
relieve definite visual malfunctions.
S. chromatism. The extrinsic depth
factor provided by the alteration of
chroma and saturation of color with
distance. Of vital importance in repro-
ducing stereo depth.
S. copier. A special support for the
stereo camera in making copies of other
stereograms and in photography of
small objects.

S. diplopia.The diplopia or double im-
age vision normal to all objects outside
the macular field, in stereoscopic vision.
S. drawing See Drawing.
Stereogram. The "picture" used in
stereoscopy. Two images either separ-
ate, side-by-side, superimposed or in-
tegrated.

Stereogrammetry.A -specialized form of
stereoscopy used for making stereo-
grams which are used as a basis for
measurement rather than for visual
examination.

S. Guild. An international association
of stereographers which circulates col-
lections of 35mm stereograms among
members.

S. infinity That distance beyond which
stereo relief can no longer be distin-
guished. Highly variable. As little as 50
yards for some individuals and as much
as n/2 miles for others. Normal average
670 meters or roughly three-eighths
mile.

S. macro bench. See S. copier.
S. macro graphy. The stereoscopic pho-
tography of small objects at magnifica-
tions of 5 to -j-io magnification.
5. micrography. The stereoscopic pho-
tography of small objects at magnifica-
tions in excess of ten diameters.
S. microscope. A compound microscope,
usually of the Greenough type which
gives a stereoscopic field.
S. mount A physical support for the
stereo pair. Usually cardboard in the
case of paper prints. Of cardboard, plas-
tic, metal or glass or combinations of
these materials for the 35mm films.
S. prism (A). The free vision stereo-
prism is a stereo novelty in the form of
a cubical solid in which a stereo image
is seen. Improved by HCM through in-
troduction of distributed critical angle.
(B). A stereo reflector or element there-
of, using prisms instead of mirrors.
S. projector. A projector used for pro-
jecting stereograms upon a screen in
such a manner that they may be viewed
in relief.

S. radiogram. An X-ray stereogram.
S. relief. The appearance of solidity,
roundness and space seen in the stereo-
gram.

S. Society. An international (British)
organization of amateur stereographers
which circulates collections of stereo-
grams among members, primarily of
the 3x6-inch paper type.
5. synthesis. The subjective process by
which the stereo image is built up from
the two dissimilar planar images trans-
mitted to the visual centers from the
two eyes.

S. Tach. (Trade name). A device to
produce stereograms with an ordinary
camera.

S. telephotography. Telephoto stereo-
scopic photography. Applied loosely to
any stereogram made with long focus
lenses, but can be correctly applied
only to paras tereograms.
S. telescope. An instrument invented
by Helmholtz which provides a magni-
fication of relief rather than the magni-
fication of size common to the conven-
tional telescope. The Battery Com-
mander's binocular (scissors type) com-

324

THREE-DIMENSIONAL PHOTOGRAPHY

bines the double telescope and the
stereo telescope.

5. therapeutics The treatment of visual
disorders by stereoscopic methods. In
wider sense it includes stereo calisthen-
ics and diagnostic stereoscopy.
S. typogram.An artificial stereogram
produced by differential spacing of
printer's type, notably ornaments.
Stereography. In this field, identical with
stereoscopic photography,
Stereomatics.The art of planning and
exposing a series of stereograms in such
a way that they will tell a coherent story.
The planning of a series of stereograms of
definitely related interest and in definite
sequence.

Stereopsis Stereoscopic vision. The abili-
ty to perceive depth visually, and speci-
fically the degree to which such perception
is possessed.

Stereopticon.This word is included to
point out the fact that it has absolutely
no relationship of any kind with any
phase of stereoscopy although it is com-
monly used when "stereoscope" is meant.
Actually, a stereopticon is a dissolving
lantern slide projector as used for project-
ing song slides in the old nickelodeon
days.

Stereoscope. A device or instrument used
for viewing stereograms. Not ordinarily
applied to the colored or polarizing filters
used with anaglyphic stereograms.
American S. The Holmes (or Mexi-
can) open type of viewer.
Bates 5. Bates took over the very crude
Holmes skeleton viewer and produced
what was substantially the same parlor
viewer which we know today.
Box S The enclosed type of viewer.
European type.

Brewster S.A viewer incorporating two
half lenses which thus form spherical
prisms, mounted base out.
Brumberger S. Plastic, self illuminated
viewer for 35mm stereograms.
Busch S.~Plastic, self illuminated view-
er for 35mm stereograms, specifically
those made in the Verascope F-^o.
Cabinet S. A stereoscope built as a
piece of furniture; a finished cabinet
enclosing the automatic or semi-auto-
matic mechanism.

Holmes S Open type Brewster viewer,
also American or Mexican.
Mattey 5. The firm of Mattey of Paris
is one of the few which makes viewers
exclusively in all grades from folding
cardboard viewers to the finest inlaid

wood models of automatic classifiers.
Mirror S. Usually the Wheatstone or
Pulfrich type of viewer.
Pedestal 5. An automatic cabinet view-
er placed upon a cabinet pillar or
pedestal which forms a cabinet for the
slide collection. Either one or two piece.
Pocket S. Folding stereoscope for carry-
ing in the pocket. Usually of inferior
quality.

Pulfrich S.A small viewer of the
Wheatstone type for tabletop use. See
Ryker.

Realist S. Plastic, 35mm viewer self il-
luminated. Companion to the Stereo
Realist camera.

Reflecting S.A viewer of (A) the gen-
eral Wheatstone type or (B) an auto-
transposing viewer.

Richard S. Companion viewers to the
Verascope cameras, usually self trans-
posing type in 35mm sizes.
Ryker S.A tabletop stereoscope of the
general Wheats tone-Pulf rich type but
equipped with magnifiers for the study
of detail. Military-aerial instrument.
One of the best examples of the Wheat-
stone type.

Stereo-Vivid S. Self-illuminated plastic
viewer, companion to the Stereo-Vivid
camera. 35mm.

Transposing S.A viewer which by an
arrangement of reflectors or a refracting
system makes it possible to view stereo
images which have not been transposed.
Used with film-strip stereo views.
$D S. This is actually a "viewer" rather
than a "stereoscope" inasmuch as it
consists of a pair of goggles equipped
with polarizing lenses. Used to view
stereo projected pictures, Vectographs
and the like.

Universal S. (A) A viewer which may
be used either as a normal or as a
Brewster instrument. (B) A viewer, such
as a classifier, which will take a variety
of sizes of stereogram.
Wheatstone S.A viewer which uses
separate stereo units, usually facing
each other, which are viewed as reflect-
ed images. Largely used for stereo X-
ray and large print stereo.
Stereoscopic. Having to do with stereo-
scopy.

S. balance.A term used in stereo com-
position in differentiation from the pic-
torial balance of a plane surface.
S. camera. A camera used for making
stereoscopic negatives or images, speci-
fically a dual chamber camera.

APPENDIX B

325

S. vision Normal, binocular human vi-
sion is stereoscopic. Many beginners
seem to think that "stereoscopic" ap-
plies only to stereo reproduction. Stereo-
scopic vision is normal, everyday vision
enjoyed by most of us.
Stereo-Vivid. (Trade name, also T.D.C.)
Generic name for a polarizing stereo pro-
jector, a 35mm stereo camera and a com-
panion 35mm stereoscope, made by the

Three Dimension Company.
Subjective. In stereo, the final step of
stereo perception, the consciousness of the
image, the actual step of "seeing" as op-
posed to the purely physical, physiologi-
cal and neurological steps.

S. anomaly. The failure of subjective
interpretation to conform to stereo the-
ory. Two identical prints (pseudostereo-
scopic) mounted as a stereogram when
viewed^ in the stereoscope should ap-
pear as images of gigantic size located
at infinity. In fact there is no such ap-
pearance at all. There is no depth of
any kind or degree visible. There are
many such effects which can be predict-
ed but which rarely are experienced as
a part of the subjective interpretation.
S. stereo. That part of stereoscopy
which has to do with the interpretation
of stereo synthesis; the actual "seeing"
as contrasted with geometric and physi-
ological stereoscopy.

Televiewer. (Trade name). Stereoscope of
Brewster type in metal construction.
5 -D. Popular abbreviation for Three Di-
mensional. Also polarizing goggles made
for viewing projected stereograms and
Vectographs.

Three dimensional Having an appear-
ance of depth or solidity, but not neces-
sarily of stereoscopic cjuality. Monocular
viewers give a three dimensional but not
a stereoscopic appearance.
Telestereoscope.This is a viewer which
resembles a pair of opera glasses and is

used for the examination of large size
stereo pairs, (aerial, X-ray and the like)
or stereo pairs projected side by side.
Fusion is accomplished through the use
of rocking prisms.

Twister. A stereogram so mounted that
the eyes^are strained in trying to view it.
Unit magnification. The principle which
states that any photograph should 'be
viewed from a distance equal to the focal
length of the taking lens. When extended
to stereo this principle becomes that of
orthostereoscopy.

Vectograph.A stereogram made by plac-
ing polarizing images upon both sides of
an optically active base. Viewed with po-
larizing goggles.

Verascope (Trade name). Generic name
for cameras made by Jules Richard of
Paris, also extended to the 35mm model.
Verascope F-jo.An improved 35mm Ver-
ascope camera distributed by Busch Cam-
era Corporation.

Videon.A. 35mm stereo camera.
Viewer. See Stereoscope,
Anaglyphic V. (A) A mask or goggles
containing two differently colored lenses
or filters for viewing bichrome ana-
glyphs. (B) A similar device contain-
ing polarizing filters set at 90 degrees
to each other for viewing polar ana-
glyphs.

Flicker V.A hand held shutter for
viewing flicker projected images.
Viewmaster (Trade name). Small stereo-
scope for viewing specially mounted
i6mm stereograms. Including those made
with the Sawyer Personal camera.
Window. In a stereoscope the stereo-
gram appears as if seen through a win-
dow. The "window" which may be a mask
or simply the way the unit prints are
trimmed, may be placed in any desired
plane; but it is usually placed nearer the
observer than any other unit of the pic-
ture.

APPENDIX C
FREE VISION FOR ILLUSTRATION

Although the subject of free vision has been explained in this
volume, we wish to present, at the risk of repetition, a resume of
practical free vision as it can be learned by anyone interested in
acquiring a new and fascinating visual skill.

The illustration of books and periodicals by three dimensional
photographs and drawings would enormously enhance the value
of all illustrative material. Now that color reproduction is com-
monplace, the use of color stereograms would approach the ideal.
In fact, the desirability of the process has been recognized for
years. The only fact which has been ignored is that a fully practi-
cal process has been available for years. One which demands no
more than three or four periods of 10 to 15 minutes for its com-
plete mastery.

It is not necessary at this time to review the theory and methods
which already have been presented in this book. It will be re-
called that the process involves:

(i) Printing the normal stereogram by any of the conven-
tional processes of reproduction.
(a). Transposed stereograms
(b). Un transposed stereograms

Either (a) or (b) to have a homologous separation not
exceeding 6omm.

The transposed stereograms are more difficult to see without
benefit of viewer. It would be more accurate to say that this
method is the more difficult to learn. Once mastered it is as easy
and natural as reading.

The use of untransposed stereograms (b) offers the easy method
so far as learning is concerned, but whereas (a) leads to a visual
relaxation which is restful, method (b) demands a greater-than-
normal degree of convergence and tends toward increasing a pres-
ently unpleasant national condition, namely, a tendency toward
"squint" (cross-eyes) which is already aggravated by our habits
of continued close vision.

326

APPENDIX C 327

It would be far easier to initiate method (b), but if criticism
and future trouble are to be avoided, it is definitely advisable to
standardize upon method (a).

Introduction. -For some time, perhaps as much as a year or
more, it would be necessary to carry a small "box" in each issue
of a publication which would briefly explain how the ability to
see the three dimensions may be acquired. But if two or three
national publications were to make a practise of running a single
page of stereos in each issue, public demand would not permit
them to be abandoned after a year.

Learning. General training would present difficulties. Elderly
people would have more difficulty than young ones. Because
ophthalmologists still differ as to the value of stereo, there would
be professional opinions, pro and con. Many people, feeling the
strain which accompanies the initial divorce between accommo-
dation and convergence, would be convinced that their eyes were
being injured. As a matter of interest, I had to change ophthal-
mologists not long ago because the one I had insisted that my
"lack" of normal co-action between accommodation and conver-
gence was pathological and insisted upon treatment to restore it.
I patiently told of the trouble I had taken to achieve it, and was
then told I had probably ruined my eyes! I went to another, and
considerably better qualified man and he laughed at the idea. He
told me that a number of his colleagues were bitterly opposed to
any form of stereo and even refused to use the professional instru-
ments based upon stereo principles. He agreed with modern stereo
researchers that stereo skill is valuable and well repays any slight
difficulty incident to acquiring it. He then pointed out that my
range of vergence was considerably greater than would be expect-
ed in a patient less than half my age, with presbyopia consider-
ably delayed.

Method. There is perhaps no better method than that of using
the mask to start with. If the two eyes cannot converge upon a
common point, they will try to fuse the two similar images even
under unusual conditions of convergence. I have found the great-
est success in teaching stereo viewing by starting with small pic-
tures with 2omm separation, and gradually substituting others of
increasing separation until 6omm was reached. This is the end

328 THREE-DIMENSIONAL PHOTOGRAPHY

point, because there are many people who have less than 65mm
interpupillary. The individual with widely spaced eyes can view
the narrow stereogram comfortably; but the one with closely set
eyes cannot see the wide stereogram except by painful divergence.

The worst obstacle is the seeming impossibility of the thing at
first. Try as one will, the fused image is out-of-focus, or the sharp
image is seen double. It is best to let the focus (accommodation)
go and learn to hold fusion. With fusion locked, accommodation
is tried. The images fly apart, and one starts over again. Repeti-
tion and more repetition slowly brings results. When the 2omm
picture can be held and focused the battle is over. Spreading the
ability to wider and wider separations is a matter of a little prac-
tise. The big difficulty comes in holding the first 2omm picture.

Just how practical is it?

My wife was mildly irritated because I would glance at stereo-
grams without a viewer and comment upon them. She wanted to
see "if it were possible." You will find this skepticism quite com-
mon. Although at that time she was not particularly interested in
stereography, I undertook to teach her. It required five periods
of five minutes each, with five minute rest periods between. At the
end of that time she was fusing and focusing a normal 6x13 stereo-
gram. (No graded series was available at the time.) Since that
time there are intervals of six or eight weeks during which she
has no occasion to use the ability, but when it arises she views the
stereogram as easily as a page is read. Not only that, but she fuses
a standard 3x6 paper print with homologous separation up to
8omm! The latter I find difficult, but she does it easily. And her
vision is excellent.

Similar results have occurred in teaching a number of students
the same skill. The time varies, but once the first free fusion has
been accomplished the rest follows easily; the acquired skill is
permanent and general visual comfort is usually increased in that
headaches from visual fatigue and the like are less common,
"tired eyes'* almost unknown and the ability to consciously relax
convergence gives the physical comfort incident to relaxation of
any muscles.

Problem. The single problem is to persuade any publisher that
the thing is feasible. He is sold on the stereo idea, he can learn

APPENDIX C 329

it himself but he cannot be persuaded that his readers share his
own ability to acquire the skill.

When the time comes that one or two courageous publishers
make a serious attempt to introduce this natural type of free vision
stereo, publishing will experience the greatest revolution it has
known since the invention of the typesetting machines.

INDEX

Abstract stereo, 145
Accommodation, 53, 190
A.G. viewer light, 32
Adapter, Rexo, 7
Adjustable reflector, 26, 288
Advertising. 292
Alignment, rotary, 85

slide, 83

vertical, 85

Alternate projection, 282
American stereoscope, 36
Anaglyph, 167

Dichromatic, 168, 175

integrated, 171

lenticular, 172

motion picture, 284

polarized, 169, 175,216

reproduction of, 169

spacing, 171
Angle sight, 238
Angular parallax, 267
Aperture, 94

Apochromatic objective, 254
Apparent relief, 255

B

Bar-screen stereogram, 202
Base, continuity of, 4

long, 61

out (or in), 315

variations of, 69
Beam splitters, 247
Book, stereo, 49
Book viewer, 32
Borax developer, 307
Box viewer, 32

Brewster stereoscope, 36, 40, 180
Brumberger viewer, 31
Busch Verascope, 21

C

Cabinet viewer, 35, 310
Camera angle, 226
Camera mask, 224
Camera, stereoscopic, convergent, 245

dummy, 237

folding, 24

Heidoscope, 25

Iloca, 23

level, 93

Personal, 13

press, 24

reflex, 25

Rolleidoscope, 25

simple, 4

Stereoflektoskop, 25

Stereo Realist, 16

Stereo Vivid, 16

35mm, 27

3-lens, 25

Verascope, 21
Chain stereoscope, 36
Cine-stereo attachment, 274
Classifier, 35
Closeups, 96
Color, 119

additive, 167

balance, 120

bombardment, 168, 175

control, 123

error, 119

filters, 121

Kelvin degrees, 121

meter, 123, 127, 128

perspective, 133

temperature, 121

visual meter, 123

"white", 120
Composition, 138

compensation, 195

emphasis, 141

selection, 141
Compression lens, 277
Control slides, 191
Contour gradation, 132
Contour interference, 132, 136
Convergence, 53

absolute, 244

range, 84

Convergent axes, 73
Cryptograms, 58
Cyclopean center, 76
Cyclopean vision, 42
Cylindrical lens, 277

D

Daguerreotype, 211
Daylight flash, 104, 111
Definition, 9, 94
Depth of parallax, 233
Depth perception, 131, 140
Developer, 307
Development control, 306
Deviation, 82, 243

in projection, 193, 317

range, 244

stereo, 316
Diagonal, 247

double, 248

Differential parallax, 134, 233
Diminished parallax, 305
Diopter, prism, 193, 314
Diplopia, stereo, 50

dynamic, 51

331

332

INDEX

Distance, limit of, 96, 147, 243

perception, 197

subjective, 197
Drawing, 298

freehand, 301

lattice, 301

net, 298

Dummy camera, 237
Duplicators, 225
Dye transfer, 168
Dynamic parallax, 233

E

Educational stereo, 294
8mm motion pictures, 279
Elongation, 68
Emulsions, 10

sensitivity, 305
Engineering, 293
Enlarging, 312
Exercise, stereo, 45
Extension flash, 111
Extension reflector, 288

F

Fantasy, 146
Field records, 293
Figure studies, 143
Filing and storage, 99
Film positive, 310
Filters, control, 121
Fine grain, 307
Fixing, 308
Flashlight, 104

balance, 110

blue bulbs, 110

blue shield, 111

color, 113

color factor, 109

daylight, 112

daylight level, 117

exposure, 105

extension, 111

filters, 110

guide numbers, 105

intensity, 116

local color, 115

mist and fog, 115

multiple. 111

pinkeye, 108

position, 106, 116

rain, 116

staring eyes, 109

stop motion, 116

synchronization, 104

synchronizers, 104

technique, 114
Folding viewer, 32
Follow shots, 10
Forensic stereo, 294
Formulae, 308
Free vision, 203, 325

in motion pictures, 284

G
Glycin developer, 307

Gold toning, 305
Grid stereo, 202

H

Half angle, 314
Hand viewer, 36
Haze, 132
Headache, 46
Heterostereoscopy, 219
Holmes viewer, 36
Hyperstereo, 50, 72, 220

Illumination, micro, 256

parallax, 253

viewer, 31
Illustration, 325
Indexing, 101
Industrial stereo, 293
Integrated stereo, 202
Interlock, 20

Interpupillary distance, 4
Iridescent stereo, 57

Jig, mounting, 81

K
Kelvin degrees, 121

L

Leica projector, 184
Legal stereo, 294
Lens cap, 268
Lenticular grid, 205

viewer, 268
Life size, 60

Light and shade, 132, 137
Lighting, 137
Long base, 61

M

Macro stereo, 246
Mask, box, 237

choice of, 98

fancy, 98

negative, 98

positive, 98
Masking, 310
Mattey viewers, 35
Mechanical stage, 253
Merger, tonal, 136
Metastereoscopy, 220
Mexican viewer, 36
Microscope, 250

petrographic, 254
Micros tereo, 249

field, 256
Mining, 293
Mist and fog, 115
Models, 159
Modern treatment, 144
Monochrome stereo, 305
Mount, cardboard, 81

Permamount, 81

projection, 199

reel, 81

rotating filter, 278
Mounting, 190

INDEX

333

errors, 83

gage, 190

kit, 81

precision, 188

projection, 187
Moving objects, 10, 115
Motion picture stereo, 183

adjustable reflector, 287

alternate, 282

anaglyphic, 284

camera mask, 281

Cine Kodak Special, 283

compromise form, 280

converted 8mm, 279

8mm, 279

extension reflector, 288

free vision, 284

projector mask, 281

projector prism, 281

reflector, 275

rotary reflector, 282

rotary shutter, 286

16mm conversion, 280

stereotach, 274, 285

Vectograph, 284

N

Nicol prism, 215
Nudes, 144, 154

O
Orthostereo, 60, 219

P

Paper print, 305, 309
Pantoscope, 37
Parallax:

angular, 267

controlling factor, 48

depth of, 246

differential, 134, 233

diminished, 305

dynamic, 233

fixed, 56

hyperstereo, 262

illumination, 253

in stereoscope, 52

inversion, 228

multiple, 232

rotary, 245, 249

stereo, 208

Parastereoscopy, 61, 72, 220, 270
Parlor viewer, 36
PePax, 61
Permamount, 81
Personal camera, 13

viewer, 28
Perspective, 61, 75

drawing, 298

geometric, 131, 135

tail buildings, 136
Pictorialism, 139
Pink eye, 108
Polarized light, 176

anaglyph, 169, 175,216

extinction, 278

filter position, 278
Polarizing films, 215
Polaroid-Land camera, 8, 26, 239
Presbyopia, 46
Principles of stereo, 2
Print lacquer, 313
Prismatic lenses, 242
Prism base, 315
Projection, stereo, 175, 183

free vision, 178

motion picture, 277

mounting, 82, 187

printing, 312

prism, 281

problems, 186

screen, 21 6, 278

viewer, 180
Projector, stereo, 184
Pseudoscopy, 56, 145, 209
Pseudo stereograms, 50, 210, 264
Pulfrich viewer, 38
Punch, film, 15, 81
Puzzle pictures, 56, 145
Pyro developer, 307
Pyrograph, 146

R

Radiography, 295
Rain shots, 115
Realist camera, 18

mounting kit, 81

projector, 185

viewer, 28

Re-creation in stereo, 74
Reels, Viewmaster, 15
Reflector, stereo, 7, 177, 274

adjustable, 26

extension, 288

motion picture, 275

rotary, 282

Restricted subjects, 295
Retinal rivalry, 57
Richard viewer, 35
Rolleidoscope, 25
Rotation, 78

S

Sales stereo, 293
Scenic views, 142
Screen distance, 193
Separation, 84, 266
Shift head, 4
Side bands, 242
Size, stereo, 12, 49, 60

abnormal, 72

diminution, 132, 136
Soft focus, 208
Space control, 49, 240, 233
Spacing, lateral, 84
Stereo-chromatism, 134
Stereograms:

abstract, 145

bar-screen, 202

characteristics, 47

control, 191

334

INDEX

flat, 264

grid, 202

integrated, 202

iridescent, 57

non-stereo, 46

parallax, 202

printed, 206

pseudo, 50

pseudoscopic, 56

puzzle, 56, 145

surrealist, 145

35mm, 81

trick, 56, 145
Stereoly, 7
Stereomatics, 102
Stereomicro stage, 254
Stereoscope:

American, 36

automatic hand, 34

book, 32

box, 32

Brewster, 36, 40

Brumberger, 31

cabinet, 35, 37, 310

chain, 35

choice of, 42

classifier, 35

folding, 32

hand, 36

Holmes, 36

illuminator, 31

illuminator, A.C., 32

importance of, 38

in use, 42

lenticular, 39

Mattey, 35

Mexican, 36

pantoscope, 37

parlor, 36

Personal, 28

projection, 180

Pulfrich, 38

Realist, 28

Richard, 35

spacing, 28

table, 36

Taxiphot, 35

T.D.C., 28

Telebinocular, 37

therapeutic, 35

35mm, 28

variable base, 33

variable focus, 34

Verascope, 31

Viewmaster, 28

Wheatstone, 37
Stereoscopic:

diplopia, 50

dynamic diplopia, 51

effect, 202

exercise, 45

idiosyncrasy, 44
Stereotach, 7, 26, 274, 285
Subjective projection, 43
Supermacro stereo, 247
Supplementary lenses, 242, 245
S.V.E. projector, 184

T

Taxiphot, 35
T.D.C. viewer, 28
Telebinocular, 37
Telestereoscope, 262
Telestereoscopy, 60
Therapeutic stereo, 36, 201, 302
35mm cameras, 27
3-D viewers, 170
Time and temperature, 305
Tolerance (proj.), 198
Toning, gold, 312

thiocarbamide, 313
Transparencies, 311
Transposition, 75

film base, 79

glass, 79

methods of, 79

prints, 80

Trick work, 56, 145
Tripod, 94, 223

V

Variable base, 33
Variable focus, 34, 63
Vectograph, 169, 216, 297
Verascope, 21
Viewmaster, 28

camera, 13

mounts, 81
Visual angle, 12

comfort, 303

skill, 303
Vivid camera, 16

projector, 184

viewer, 28

W

Washing, 309
Wheatstone theory, 52
Window trimming, 85, 311

placement, 87

118908