Full text of "Journal of the Society of Motion Picture Engineers"

From the collection of the
z n

_ z m

o Prelinger

i a

Uibrary

t V

San Francisco, California
2007

. XXI

NO. 1

JOURNAL

OF THE SOCIETY OF

MOTION PICTURE ENGINEERS

JULY, 1933

PUBLISHED MONTHLY BY THE
OCIETY OF MOTION PICTURE ENGINEERS

The Society of Motion Picture Engineers

Its Aims and Accomplishments

The Society was founded in 1916, its purpose as expressed in its
constitution being the "advancement in the theory and practice of mo-
tion picture engineering and the allied arts and sciences, the standardi-
zation of the mechanisms and practices employed therein, and the
maintenance of a high professional standing among its members."

The membership of the Society is composed of the best technical
experts in the various research laboratories and other engineering
branches of the industry in the country, as well as executives in the
manufacturing and producing branches.

The Society holds two conventions a year, spring and fall, at various
places and generally lasting four days. At these meetings papers
dealing with all phases of the industry theoretical, technical, and
practical are presented and discussed and equipment and methods
are often demonstrated. A wide range of subjects is covered, many
of the authors being the highest authorities in their particular lines of
endeavor.

Papers presented at conventions, together with contributed arti-
cles, translations and reprints, abstracts and abridgments, and other
material of interest to the motion picture engineer are published
monthly in the JOURNAL of the Society. The publications of the
Society constitute the most complete existing technical library of
the motion picture industry.

iRAL 01

33 WEST 42iul
A YORK, N. V.

JOURNAL

OF THE SOCIETY OF

MOTION PICTURE ENGINEERS

Volume XXI JULY, 1933 Number 1

CONTENTS

Page

The New Cine-Kodak Special in Medicine 3

H. B. TUTTLE AND R. P. SCHWARTZ

Photoplay Appreciation in the Nation's Schools W. LEWIN 9

Report of the Non-Theatrical Equipment Committee 16

Some Properties of Two-Bath Developers for Motion Picture

Film. . .J. I. CRABTREE, H. PARKER, JR., AND H. D. RUSSELL 21

Sensitometric Control in the Processing of Motion Picture Film

in Hollywood E. HUSE 54

Officers 83

Society Announcements 84

JOURNAL

OF THE SOCIETY OF

MOTION PICTURE ENGINEERS

SYLVAN HARRIS, EDITOR

Board of Editors

J. I. CRABTREE, Chairman

A. C. HARDY
O. M. GLUNT F. F. RBNWICK

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers.

Publication Office, 20th & Northampton Sts., Easton, Pa.
General and Editorial Office, 33 West 42nd St., New York, N. Y.

Copyrighted, 1933, by the Society of Motion Picture Engineers, Inc.

Subscription to non-members, $12.00 per annum; to members, $9.00 per annum,
included in their annual membership dues; single copies, $1.50. A discount
on subscriptions or single copies of 15 per cent is allowed to accredited agencies.
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton
Sts., Easton, Pa., or 33 W. 42nd St., New York, N. Y.

Papers appearing in this Journal may be reprinted, abstracted, or abridged
provided credit is given to the Journal of the Society of Motion Picture Engineers
and to the author, or authors, of the papers in question. The Society is not re-
sponsible for statements made by authors.

Entered as second class matter January 15, 1930, at the Post Office at Easton,
Pa., under the Act of March 3, 1879.

THE NEW CINfi-KODAK SPECIAL IN MEDICINE*
H. B. TUTTLE** AND R. PLATO SCHWARTZt

Summary The new Cine-Kodak Special is briefly described, particularly in
connection with its application to photographing surgical operations both in color and
in black and white. In addition to the usual features of 16-mm. cameras, the ad-
vantages of being able to produce fades, dissolves, double exposures, multiple exposures,
single-frame exposures and to use masks of various kinds, increase the facility with
which the camera can be used for surgical instruction and records. The paper re-
fers briefly to some of the work done at the University of Rochester with this camera.

Progress in any field is dependent upon records. In medicine
the first great men gave us written records of their work. The
Science of Life as recorded by Sushruta in his Samhita is our source of
information relative to Hindu surgery six hundred years before the
Christian era. Two hundred years later, Hippocrates gave to the
world his observations on the practice of medicine and surgery. But
unfortunately, illustrations did not accompany their vivid and mo-
mentous descriptions.

With the development of art, drawings were used to visualize
descriptions. And of necessity, these sufficed for many centuries
until the progress of chemistry and physics gave rise to the
science of photography. This situation is evidenced by the fact
that as late as 1892 important books on medicine and surgery were
published without a single photographic illustration. Today, how-
ever, most of the medical literature is profusely illustrated with photo-
graphs made with regular equipment for clinical still photography.
And, as a result, the subjects are much more understandable.

The present importance of still photography to medicine is so
obvious that it hardly needs mention. But despite this fact, in many
situations its application has limits that can be exceeded only by the
photographic records of motion. Pathology alters normal function,
and treatment is prescribed to remove the limitations imposed upon

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Eastman Kodak Co., Rochester, N. Y.
t School of Medicine, University of Rochester.

4 H. B. TUTTLE AND R. P. SCHWARTZ [J. S. M. P. E.

normal activity. It is at once apparent that motion pictures provide
a very important means of supplying desirable records of (1) the
patient's condition before treatment; (2) the method of treatment;
and (3) the extent to which normal function is restored by the
treatment.

Cinematography, the most modern servant of visual expression,
first emerged usefully from the scientific laboratory of Thomas Edison
in 1894. In less than forty years an industry has developed that is
now vital to national and international understanding. Medicine
has been one of the departments of human endeavor that has been
influenced by the recording of motion.

To those who may not be familiar with the application of motion
pictures to the medical field, it might be well to outline briefly its
past history. From the advent of the motion picture, the possibility
of applying it to the field of medicine, as an educational medium and
a means of satisfactory record, has been recognized by the profession.

Since the 16-mm. motion picture camera was not introduced until
1923, all the early medical motion pictures were produced with 35-
mm. equipment. In most instances it was necessary for the indi-
vidual physician to bear the expense, a fact that undoubtedly retarded
early constructive pioneering. However, the work progressed as
various groups and institutions became increasingly interested in it.

The most influential event of this development was the comprehen-
sive program begun in 1927 under the joint auspices of the American
College of Surgeons, the Motion Picture Producers and Distributors
of America, Inc., and Eastman Teaching Films, Inc., for an organized
study 1 of the technic of making medical and surgical motion pictures.
The films that were produced under the auspices of these groups were
of excellent quality, and they represent to a large degree the recog-
nized technic and practice of the majority of the profession. They
also proved that 35-mm. originals, and 16-mm. reduction prints made
from them, are entirely economical and highly satisfactory where
a large number of prints are required.

As stated before, the 16-mm. camera the Cine-Kodak made
its appearance in 1923. Because of the relatively small cost of
both the apparatus and the film, the possibility of recording medical
subjects was greatly expanded. Individual physicians for the first
time found it possible at little expense to film their unusual cases and
technical methods for record and lecture purposes.

Each year has evidenced a steadily increasing effort to apply 16-

July, 1933] CINE-KODAK SPECIAL IN MEDICINE 5

mm. motion pictures to the problems of medical photography. The
degree of activity and the intensity of applications have increased de-
spite the fact that the films produced have not equalled the technical
quality of professional 35-mm. motion pictures. However, the ad-
vantages achieved were more than sufficient to overbalance the recog-
nized deficiencies.

Ten years have passed since the announcement of the Cine-Kodak;
and the increasing number of physicians personally interested in the

FIG. 1. The Cine-Kodak Special.

application of motion pictures to their work has created a large num-
ber of highly competent operators of 16-mm. cameras. In addition,
many new developments in films, lamps, and accessory apparatus
have facilitated the taking of medical pictures.

Kodacolor 2 made it possible for the first time to film an operation
in color. 3 ' 4 Later, supersensitive Kodacolor film, together with
photoflood lamps and a new type of ratio diaphragm cap, greatly ex-
tended the scope of medical films in full color. Supersensitive pan-
chromatic film made possible black-and-white cinematography in
the operating 5 room without additional lighting equipment.

H. B. TUTTLE AND R. P. SCHWARTZ [J. S. M. P. E.

Despite this widespread interest and the many advances, the medi-
cal films that have been made show that the limits of application of
the regular 16-mm. camera render it impossible to attain results con-
formable to professional standards. Therefore, it is obvious that a
special camera using 16-mm. film and capable of photographing any
subject under the most exacting circumstances will fill a definite need.
The Cine'-Kodak Special (Fig. 1), recently announced by the East-

man Kodak Company, is such a camera.

Among its features are a flexibility of opera-
tion and an accuracy of focusing and framing
heretofore unavailable to 16-mm. camera
users.

A reflex finder on the camera permits ac-
curate framing and focusing with lenses of
any focal length, at any distance. Supple-
mentary lenses are available for focusing ob-
jects from 4 niches to 24 inches away; they
permit close-up cinematography of small ob-
jects, with clear and sharp detail in the image.
The spring motor drives 40 feet of film
through the camera at one winding, and pro-
vision is made for attaching an electric motor.
The lens turret on the front of the camera
holds two lenses of selective focal length,
and it can be turned quickly into position.

The film chamber is of the removable
magazine type, which permits a rapid
interchange from a 100-foot to a 200-foot
chamber, or from Kodacolor to black-and-
white film.

Fades "in" and "out," dissolves, double

exposures, or multiple exposure effects can be produced by a
built-in dissolving shutter. Provision is also made for exposing one
frame at a time. With this feature, animation and time-lapse pic-
tures can be made. Masks may also be used in front of the film
aperture, so that many other effects can be produced.

The camera is geared so that changes of speed may be made while
it is running. These speeds, varying from 8 frames a second
to 64 frames a second, make possible slow-motion studies of subjects
wherever such an effect is desired. The necessary adjustment in ex-

FIG. 2. The Eastman
medical spotlight used
with the Cine-Kodak
Special to produce Koda-
color pictures of surgical
operations.

July, 1933]

CINE-KODAK SPECIAL IN MEDICINE

posure can be compensated by the variable shutter while the camera
is operating.

The appeal of the Cine-Kodak Special is assured because the basic
model includes all of these advantages. Moreover, it readily lends
itself to practically all modifications essential to individual require-
ments.

The quality of work obtainable with this new camera is revealed in
films made for William S. McCann, M.D.,
Professor of Medicine at Strong Memorial
Hospital of the Rochester University School
of Medicine. The work was done in mini-
mum time, with four Kodaflectors and with-
out the necessity of making retakes. The
composition of the film reveals perfectly the
sequence of procedures followed in determin-
ing lung capacities.

Kodacolor motion pictures of a cataract
extraction were made for John Gipner, M.D.,
of the same institution. An Eastman medi-
cal spot light (Fig. 2) was the only illumi-
nant required. Another film was made to
reveal various eye diseases. This work was
done in black and white, as well as in color.
For the black and white, a Bausch & Lomb
ophthalmic operating lamp (Fig. 3) was
found most satisfactory. In each instance
the excellent quality of picture produced can
be traced to the technical advantages of
the new Cine-Kodak Special.

A Kodacolor film for teaching purposes
was made of an autopsy on an infant with
a number of congenital deformities. This film provides the instruc-
tor with an excellent record that will be used from year to year to
accompany lectures to students.

Another most difficult problem solved with this camera is the
copying of radiographs. This task is now comparatively easy to ac-
complish. In order to obtain fine detail and good contrast, Cine-
Kodak panchromatic film was employed; all the detail and contrast
in the original radiographs are recorded with exactitude. There are
many possibilities in recording radiographs on motion picture film,

FIG. 3. The ophthalmic
operating lamp; used as
a source of illumination
for making photographic
studies of diseases of the
eye.

8 H. B. TUTTLE AND R. P. SCHWARTZ

especially in the study of bone and joint diseases, progress of disease
processes, and other situations in which the bony structure of the
human body is involved.

The results attained in all the previously mentioned instances have
been so satisfactory that there is no hesitation in recommending
the Cin-Kodak Special to any worker in the medical or general scien-
tific field. Its construction and design enable it to meet the most ex-
acting requirements.

REFERENCES

1 TUTTLE, C., AND MORRISON, C. A.: "Some Preliminary Experiments in
Medical Photography," Trans. Soc. Mot. Pict. Eng., XII (1928), No. 36, p. 1022.

2 CAPSTAFF, J. G., AND SEYMOUR, M. W.: "The Kodacolor Process for Ama-
teur Color Cinematography," Trans. Soc. Mot. Pict. Eng., XII (1928), No. 36,
p. 940.

3 TUTTLE, H. B.: "Some Experiments in Medical Motion Pictures in Color,"
/. Soc. Mot. Pict. Eng., XV (Aug., 1930), No. 2, p. 193.

4 BAKER, DR. H. H.: "Color Photography in Surgery," The Camera, 40
(March, 1930), p. 184.

& SCHWARTZ, R. PLATO, AND TUTTLE, H. B.: "Advantages of Using 16 Mm.
Cine-Kodak Supersensitive Panchromatic Film in Making Surgical Motion
Pictures," J. Soc. Mot. Pict. Eng., XVTH (May, 1932), No. 5, p. 609; also New
York State Journal of Medicine (Oct. 1, 1932).

PHOTOPLAY APPRECIATION IN THE NATION'S SCHOOLS*

W. LEWIN**

Summary. .4s chairman of the Committee on Photoplay Appreciation of the
National Council of Teachers of English, the author describes experimentation
undertaken by the Council during 1932 and 1933 for the purpose of determining
whether the movie habits of adolescents can be improved significantly through the
medium of the English classroom; and desirable ideals and attitudes can be de-
veloped through the medium of well-Selected current photoplays; and whether the
neighborhood theater can serve in part as an educational laboratory working in di-
rect relation to the public school.

During the academic year 1932-33 the National Council of Teach-
ers of English has been engaged in an experiment under the general
supervision of Dr. Walter Barnes of New York University, president
of the Council, to determine whether the movie habits of adolescent
America can be improved significantly through the medium of
the English classroom; whether, indeed, desirable ideals and atti-
tudes can be developed through the medium of well selected current
photoplays; and whether the neighborhood theater can serve, in
part, as an educational laboratory working in direct relation to the
public school, by providing enjoyable and worth-while "literature
experiences" for boys and girls of junior and senior high school
age.

The experiment was planned on the basis of suggestive units of
instruction in photoplay appreciation. It is proposed to introduce
these new units in the nation's schools as part of the revised course of
study being prepared by the curriculum commission of the National
Council of Teachers of English. This new development accords with
the present emphasis on social criteria in secondary education, and
is a phase of training in the right use of leisure.

Among the aims of the new instruction units are the development of
appreciation of basic themes and conflicts in selected current photo-
plays, skill in retelling the stories of photoplays as a part of social con-

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** National Council of Teachers of English, Newark, N. J.

10 W. LEWIN [J. S. M. P. E.

versation, appreciation of the literary sources of photoplays, and the
development of the vocabulary necessary to an elementary discussion
of screen art and technic. It is hoped that, before graduating from
the senior high school, students will be able to evaluate motives and
character traits as revealed in the varying patterns of conduct found in
popular screen dramas. It is hoped, too, that students will acquire
some degree of skill in conversing about current photoplays in the
light of definite critical standards.

The plan of the experiment has been to set up equivalent groups of
students in each of a series of schools in a rather extensive network
under the supervision of regional directors, in order to compare the
reactions of instructed and uninstructed classes. The groups have
been compared as to age, grade, sex, 'intelligence, home background,
and initial appreciation of pictures. The procedure thereafter has
been to arrange with local theater managers to admit both the experi-
mental (that is, the instructed) groups and the control (that is, the
uninstructed) groups in a body to see the same series of pictures at
the same time. The control groups were merely to see the pictures
without guidance and to have a good time. The experimental groups,
on the other hand, were to be prepared for an appreciation of the
pictures, and, after attending the theater, to evaluate the picture in
the classroom under the guidance of the teacher.

To accelerate the growth of appreciation among the experimental
groups, the activities in which they engaged included the dis-
cussion of such general topics as: What is the best photoplay you
have ever seen? How would you decide which of two photoplays is
the better? In how many ways may we consider levels of quality in
photoplays? What are the fundamental themes of some photoplays
you have seen? Story-telling "bees" were held, in which the plots
of interesting photoplays were retold in brief talks. The experimental
pupils were encouraged to start critical diaries of the photoplays
they saw, and to record their impressions for later discussion. Some
students were asked to start scrapbooks of newspaper clippings con-
taining interesting bits of information about photoplay technics.
Others made glossaries of motion-picture terms, with pictures pasted
in, where possible, to illustrate the means of the terms. Classroom
discussions considered how motion pictures were made, and pupils
learned to use such terms as close-up, long shot, and camera angle.

When historical interest was shown by older pupils, they were asked
to take notes on a formal talk prepared by the teacher on the de-

July, 1933] PHOTOPLAYS IN THE SCHOOLS 11

velopment of the photoplay from its crude beginnings to its present
status.

As the work progressed, the older experimental pupils prepared oral
and written reports about the photoplays they were seeing, as bases
for class discussions, with a view to pointing out striking examples of
screen story treatment, strong and sincere direction, unusual camera
work, subtle nuances of acting, and details of good sound recording.

As an approach to a more formal symposium by selected members
of the more advanced classes, there were impromptu discussions of
standards for judging photoplays and of methods of shopping for
pictures to see. Out of the symposiums of pupils' points of view, class
debates arose. The visits to the neighborhood playhouse were in
the nature of theater parties, with natural emphasis on enjoyment.

Among the materials used with a view of accelerating appreciation
and developing critical judgment were specially prepared study-guides
and rating-scales. Suggestive mimeographed questions and work
sheets were supplied for a number of specific pictures, including
I Am a Fugitive, Smilin' Through, Rasputin and the Empress, The
Son-Daughter, Madame Butterfly, Rebecca of Sunnybrook Farm, Tom
Brown of Culver, and Robbers' Roost. In addition, a general study-
guide was supplied in mimeographed form for use in discussing pic-
tures for which no specific guides were provided. Standard questions
were framed concerning the type and purpose of each photoplay seen,
its setting or locale, its characters and their motives, the elements of
the story, the logic of the ending, the best directed scenes, the scenes
that might have been omitted, and the lessons about life to be drawn
from the picture.

For example, the study-guide for Madame Butterfly, dealing with
the consequences of an American naval officer's desertion of his
Japanese wife, suggests that the restraint of brave little Cho-Cho-San
be compared with the pathos of Oenone when deserted by Paris in
Tennyson's poem Oenone, or contrasted with the furious passion of
Dido when deserted by Aeneas in Virgil's great epic. Life-problems
considered were : Should a girl marry a man she loves, even though
he be of different race, with different ideas and customs? What
should a girl do when deserted by the man to whom she entrusted all
her happiness ? How seriously should a naval officer take his romance
with a geisha girl? What is the most touching scene? Is the picture
convincingly directed ?

In the study-guide for I Am a Fugitive, the marvelous effect of

12 W. LEWIN [J. S. M. P. E.

suspense was studied. In the guide for The Son-Daughter, the graceful
acting of Helen Hayes and the subtle nuances in her use of the fan
were mentioned, as well as the ideals of the character she portrayed.

Each teacher in the experiment was supplied with a manual on
how to appreciate motion pictures, prepared by Dr. Edgar Dale, of
Ohio State University. This manual is now being tried out inten-
sively in Ohio under the auspices of the Payne Fund and in connection
with special radio lectures by Dr. Dale. Radio, however, was not used
in the English Council experiment. The manual was used in about
fifty cities and is still being used in a few cities that have not yet
reported.

A set of three experimental rating scales was used to obtain objec-
tive data as to the pupils' judgments. On these scales the student
indicated whether he considered the fundamental idea of the picture
trite or original; the story structure logical or illogical; the charac-
ters life-like or unnatural ; the settings appropriate or inappropriate ;
the photography effective or crude ; the dialog apt or dull ; the acting
sincere or artificial; the voices of the actors clear or blurred; the
direction intelligent and imaginative or weak and uninspired; the
social value of the production constructive or destructive. In judg-
ing the picture as entertainment, a weight of 50 per cent was given
to basic story elements and 50 per cent to direction, photography,
acting, and speech delivery.

To measure specific differences in appreciation, teachers and pro-
fessional reviewers set up specific criteria by rating and ranking the
photoplays involved in the experiment about fifty pictures in all.
Some of the newspaper critics objected that they were required to
write practically an essay on every picture. The teachers did not
feel that way. The ratings assigned by the pupils are being com-
pared with the ratings assigned by the teachers and professional crit-
ics. Sectional differences are being noted, as well as age and sex dif-
ferences.

To measure general growth in appreciation, the reactions of the
various experimental and control groups to preliminary and conclud-
ing questionnaires are being tabulated. These questionnaires serve
the purpose of initial and final tests of appreciation.

In the preliminary questionnaire the student was required to indi-
cate which of ten worth-while photoplays of 1931, suitable for high
school pupils, he had seen: Tom Sawyer, A Connecticut Yankee,
Huckleberry Finn, Spirit of Notre Dame, Merely Mary Ann, Skippy,

July, 1933] PHOTOPLAYS IN THE SCHOOLS 13

Cimarron, Trader Horn, Street Scene, and Alexander Hamilton. He
was required to indicate which of the following things he was in the
habit of doing before attending a photoplay: finding out who di-
rected the picture; seeing whether the story is by a good author;
reading what a critic says about the picture ; asking a teacher about it
(which is generally the last thing he thinks of doing) ; discussing it
with a member of the family; and discussing it with a friend. He
was asked also whether in his home the question of what photoplay
to see was brought up at a family discussion at least once a month ;
whether he belonged to a photoplay club or similar group interested
in motion pictures; whether he owned or had the use of a movie
camera (it was found, incidentally, that this experiment has caused
motion picture cameras to be purchased by the experimental groups,
due to arousing their interest) ; whether he could mention a book in
which he had become interested as a result of seeing a photoplay;
whether he considered the director more important than the camera-
man in making a photoplay; whether the story is more important
than the star as a basis of choosing a photoplay to attend; whether
he enjoys going to the movies so much that he considers it one of
his favorite leisure occupations; whether he usually discusses with
his friends the photoplays he has seen; whether he has ever pre-
pared a theme or a talk on a photoplay in connection with school
work; whether a photoplay, to be a good one, must end hap-
pily (in Utah, for example, a group overwhelmingly said, "Yes, it
must end happily to be good;" in Brooklyn the opposite was the
answer) ; and whether he could recall the names of any characters he
had admired in photoplays for ideals of bravery, honesty, devotion,
or self-sacrifice for a great cause.

In the final questionnaire the student was required to indicate
his reactions to similar queries, with a few changes needed to
maintain the validity of the test, and to make the final data
comparable with the preliminary data. One interesting question
added to the final questionnaire, "Can you mention a photoplay
that has influenced your conduct in any way?" is providing very
interesting data.

The reactions of 1500 representative boys and girls involved in the
experiment are now being compiled by thirty teachers in sixteen states
and the District of Columbia, representing nearly all sections of the
country California, Colorado, Florida, Georgia, Illinois, Indiana,
Kansas, Michigan, Minnesota, Montana New Jersey, New York,

14 W. LEWEST (j. s. M. P. E.

Pennsylvania, Tennessee, Virginia, West Virginia, and Washington,
B.C.

The results of the experiment will be formally announced at the
annual convention of the English Council at Detroit, next November.
Meanwhile, it is safe to say that comparative data now being as-
sembled indicate significant trends. One of the outcomes of the pro-
ject has already been the establishment of a strong national preview-
ing committee of teachers, with Dr. Stella S. Center, past president
of the Council, as chairman. The function of this Committee is to
determine what current productions are worthy of consideration in
the classroom. For such productions study-guides will be prepared.

Not the least interesting observation to be made meanwhile is
that progressive teachers everywhere are enthusiastic about doing
careful research work in this field and are eager to continue the ex-
periment.

After all, the motion picture, whether it be in the theater or in
the classroom, does not concern the English teacher alone. All
knowledge is interrelated. Any given picture may be viewed from
many angles. The teacher of social sciences, the teacher of geog-
raphy, the teacher of physical sciences, looking at a picture, sees it
from his point of view. The home-room teacher, the parent, the
principal, the board of education member, by the same token, looks
at the motion picture from his angle. Each seeks not only cleaner
pictures, but also better pictures. Therefore it is evident :

(1) That visual education must widen the scope of its activities
so as to include a consideration of the neighborhood theater and its
programs as a community problem. This problem is destined to
become less and less a problem and more and more an opportunity as
time goes on. The problem has already been defined by the National
Council of Teachers of English. A problem well defined and well
understood is already half solved. Visual educators, at any rate,
can not escape this cinema problem. They must face it squarely
and constructively, with a view of arriving at some consensus which
will lead to united effort.

(2) That the visual education movement must accept research and
experimentation as part of its daily work. Education is an on-going,
creative evolution. It is not enough that we teach our teachers how
to use charts, slides, and films. We must teach our teachers how to
do research. It is at once the advantage and the disadvantage of
visual education that visual aids are in a continual state of obsoles-

July, 1933] PHOTOPLAYS IN THE SCHOOLS 15

cence. Engineers, it seems, have found a way of inventing inven-
tions, so that a problem need but be denned, and lo, its solution is
but a matter of time. The classroom teacher must keep abreast of
science. She must learn how to evaluate her methods, her materials,
her equipment. The English Council experiment demonstrates
that progressive teachers enjoy motion picture research work. Let
us give them more of it.

(3) That progress for the director of visual instruction lies in
making his job less and less a physical problem and more and more a
mental problem. The work of distributing charts, slides, and films,
and caring for machines no longer presents any real problems. The
machines have been simplified until now they are so easy to operate
that even children can handle them. In high schools, boys and girls
enjoy learning how to show slides and films. The work of classifying
and distributing pictures and objects should be part of the training
of every good library assistant. Slides and films should, further-
more, be correlated with books by a unified system like the Dewey
decimal system.

The director of visual instruction needs time to devote himself to
the major problems of selecting the best visual aids, planning experi-
ments in methodology, and organizing the many interesting activities
that teachers and pupils enjoy and that are of practical benefit to
the community. It should not be the function of the English special-
ist to take the initiative in regard to motion picture theaters, but
rather of the visual educator, who can see the problem in all its phases.

(4) That visual education must embrace radio, as is done by Dr.
Charters and Dr. Dale in teaching photoplay appreciation in Ohio,,
because inevitably visual and auditory aids will be merged in the
classroom through the medium of television.

REPORT OF THE NON-THEATRICAL EQUIPMENT
COMMITTEE*

The Non-Theatrical Equipment Committee is one of the newest
committees of the Society, having been formed a little more than a year
ago. Its function and purpose are to investigate all matters relating
to 8- and 16-mm. film cameras and projectors, 35-mm. film portable
and semi-portable projectors, stereopticon and film slide projectors,
and accessories such as screens, film splicers, reels, etc. It is the
function of the Committee also to investigate proposals of standardi-
zation for submittal to the Committee on Standards and Nomen-
clature on matters that can be more advantageously studied by this
Committee because of its close contact with the non-theatrical field.

The word "equipment" in the name of the Committee is definitive
of the real purposes of the Committee ; although a few of the members
have felt that the Committee should concern itself also with matters
relating to visual education and other non- theatrical activities.
The personnel of the Committee includes representatives of practically
all the manufacturers of non-theatrical motion picture equipment.

The Committee is now engaged with the problems relating to
the standardization of 8-mm. film, stereopticons employing very large
slides, advantageous in the projection of color plates, and a further
study of the entire field.

The large variety of types of projection lamps and the general lack
of standardization, due to the rapid growth of this field, have been
perplexing problems for some time for both the equipment and lamp
manufacturers. The matter was dealt with at considerable length
by E. W. Beggs 1 at the Washington, D. C., Convention. Owing to
the great activity in this connection, particularly in the field of 8-
and 16-mm. film projectors, the Committee felt it imperative to direct
its activities toward an improvement of existing conditions.

While, of course, the development of a suitable group of projection
lamps is largely the problem of the lamp manufacturers, nevertheless
cooperation of the Committee members has greatly facilitated the
results that have been achieved. A line of projection lamps has been

* Presented at the Spring, 1933, Meeting at New York, N. Y.
16

NON-THEATRICAL EQUIPMENT COMMITTEE 17

made available which includes interchangeable high and low priced
lamps for ordinary and super service, fewer voltage ratings, and 25-
hour life for the newer lamps. All may be operated on either a-c. or
d-c. circuits.

In analyzing the problem of projector lamp standardization, the
equipment now in service was considered as well as that yet to be de-
signed. The particular point of distinction between the two groups is
that the older projectors are only moderately well ventilated, whereas
the newer ones are greatly improved in this respect, and contain also
improved optical trains. Thus, lamps designed to utilize these ad-
vantages can not be used in projectors now in service. The following
types of projectors must be considered in developing a group of pro-
jection lamps :

8- Mm. motion picture

home application
semi-proi
portable
semi-porl
pocket size

16-Mm. motion picture

semi-professional

35-Mm. motion picture

semi-portable

Film slide projector

large size

low power
Lantern slide projector , . ,

[ high power

Fortunately, it has been found that equipment of more than one
type can be served by a single lamp, thus somewhat reducing the
number of types necessary.

Fig. 1 shows briefly the essential data pertaining to a group of
lamps now available for use in non-theatrical equipment. Referring
to the item "service," it will be noted that in most cases two types of
equipment are given, the upper of which is the more important, the
lower being of secondary importance. It will be noted that all
lamps are of the 100- volt class, obviating the expense and weight of
auxiliary transformers or large resistances used in the past with
low-voltage lamps. It appears that except on the lower -priced pro-
jectors the practice will become general of using 100- volt lamps in
series with a small resistance and in combination with a voltmeter,
which will permit the adjustment of the resistance so that the lamp
will receive 100 volts on all circuits. Thus the full advantage of the
highlight output of the lamps will be combined with satisfactory
lamp performance.

The three short T-20 bulb lamps, namely, the 500-, 750-, and

NON-THEATRICAL EQUIPMENT COMMITTEE [J. S. M. P. E.

&*" E
s--s

ifl ji; -3 <u

tn <->
o *>

a.2?
E"

4^ ? O

i =
II

in O

E o

rt-rj

w *->
'- es
*j

U 4)

~ J2
>'3

"8 SS

"S P

=i 3

i
en

y *o o *i *o o ?>

iO u js.p o d *

^is^is

^iljsg 1

E
I

1000-watt lamps, form an interesting group. The 500-watt lamp
is the original member of the group, and was intended for 35-mm.
portable projectors as well as for lantern slide projectors. It was

July, 1933] NON-THEATRICAL EQUIPMENT COMMITTEE

found possible to use a 750-watt filament in this bulb, and with the
ventilation available in most 35-mm. portables the performance of
the lamps is satisfactory. If the projection machine manufacturers
are desirous of increasing the illumination still further and are willing
to improve further the ventilation in the lamp house, there is available
a 1000- watt lamp. The medium powered lantern slide projector must
continue to use the 500-watt lamp, since the ventilation of such pro-
jectors is not forced; hence, the retention of this lamp in the group.
The two 1000- watt long T-20 bulb lamps are another interesting
pair. The lamp at the left is the original member, and was intended
for both 35-mm. semi-portable and high powered lantern slide projec-

1000

600

1931 1932 1931 1933 1931 1932 1931 1933 1931 1933

FIG. 2. Screen illumination from best available lamps and best
motion picture projectors.

tors; the lamp at the right provides about 40 per cent more light in
motion picture projectors because of its greater efficiency and con-
centration of its source. This source concentration is of little
value when used with a stereopticon optical system; hence the user
of such equipment should not be penalized by having to pay the
higher price of the biplane filament lamp.

The three T-10 bulb lamps and one T-12 bulb lamp form another
important group. The 300- watt lamp will replace many lamps now
used in the moderately priced 16-mm. projectors, such as the 250- watt,
20- and 50- volt types; and affords better screen illumination when
used in place of the 200- watt lamp.

The 200-watt lamp is retained for film slide projector service, as
the substitution of a 300-watt lamp in many of these devices will cause
the film to blister. With adequate ventilation, as in the case with

20 NON-THEATRICAL EQUIPMENT COMMITTEE

many 16-mm. projectors now in use, the 500- watt lamp can be used.
The 750- watt lamp is available for manufacturers of 16-mm. equip-
ment who require still more light and are willing to provide the special
optical system and ventilation necessary to obtain the 30 per cent
increase of illumination possible as compared with the 500- watt lamp.
It is thus apparent that this move toward projector lamp standardi-
zation has made it possible to comply with practically all projector
requirements with eleven lamps, and at the same time to produce
increases in available screen illumination, as shown in Fig. 2, when
used in the improved equipments and with the better optical ele-
ments that the industries have provided.

R. E. FARNHAM, Chairman.

A. A. COOK E. R. GEIB J. H. KURLANDER

W. B. COOK N. B. GREEN R. P. MAY

H. A. DEVRY H. GRIFFIN R. F. MITCHELL

E. GALE L. A. JONES A. SHAPIRO

REFERENCES

1 BEGGS, E. W.: "Standardization of Projection Lamps," /. Soc. Mot. Pict.
Eng., XIX (July, 1932), No. 1, p. 817. MILI, J. T.: "Biplane Filament Con-
struction A High Intensity Incandescent Lamp Light Source for Motion
Picture Projection," J. Soc. Mot. Pict. Eng., XIX (July, 1932), No. 1, p. 829.

DISCUSSION

MR. E. W. BEGGS: In his presentation, Mr. Farnham referred to my last
year's paper on standardization of lamps, and perhaps some will think of the
Westinghouse Lamp Company as being entirely responsible for this revolution-
ary change in lamps. I should like to emphasize that it is a joint venture of the
General Electric Company and the Westinghouse Lamp Company.

Some projector manufacturers have recently designed projectors around a
400-watt biplane filament lamp, which is omitted from the new list. However,
the 400-watt lamp will always be available for those who need it. The number
of projectors that can not use a 500-watt type, but can use a 400-watt, is small.

This line of lamps that the Committee has presented represents a goal. It is
the duty of projector manufacturers to adhere to it. A line presented here will
do no good unless the projector manufacturers use it fully.

About a year ago one large manufacturer said to me, "How can I standardize
my projectors when every year or every six months the lamp manufacturers
bring out a new and improved device? Can we ever hope to adhere to one design
or standard line?" As to the answer: the design features of these lamps are such
as to indicate a stopping point beyond which marked improvements can not be
made within a short period of time. We have reached a point where we can
stand on the merits of the achievement, possibly, for some years.

SOME PROPERTIES OF TWO-BATH DEVELOPERS FOR
MOTION PICTURE FILM*

J. I. CRABTREE, H. PARKER, JR., AND H. D. RUSSELL**

Summary. It is well known that the properties of developers change during
use as a result of depletion of the developing agents and accumulation of reaction
products. This change is greatest for low-energy developers such as those of the
borax type, and results in (1) a reduced development rate, and (2) a lowering of
effective emulsion speed. These two effects may be offset to some extent by increasing
the development time and by adding booster solutions to revive the developer. Another
method of securing a more uniform degree of development throughout the life of a
developer consists in using two developer baths in succession.

Three types of two-bath developers have been investigated, as follows: (1) bath A
contained all the developing agents plus sodium sulfite; bath B, all the alkali plus
the balance of the sulfite. (2) Both baths contained developing agents. (3) Both
baths were of identical composition, the first bath being replaced by the second an it
became exhausted. The results of the investigation showed that Type I is the most
satisfactory developer combination and with this method it is possible to obtain an
almost constant gamma with only a slight loss of emulsion speed over a fairly wide
range of time of development. A formula is also suggested for the development of
variable density sound negatives. The application of two-bath developers to machine,
and rack and tank systems is described.

The idea of developing a photographic image by successive im-
mersions in two separate solutions is by no means new. A large
number of workers have investigated the effect of starting develop-
ment in a weak developer and then, according to the manner of
appearance of the image, continued the development in either a
restrained, normal, or very active developer. 1 In the case of nega-
tives exposed under widely varying conditions of lighting, such
a method appears to have merit.

Other workers 2 ' 3 have claimed that in the case of an underexposed
negative of a contrasty subject, if the negative is first immersed in a
developing solution which does not contain alkali (A), and is then
placed in a second solution containing only alkali and sulfite (B), the

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Eastman Kodak Co., Rochester, N. Y. Communication No. 517 from the
Kodak Research Laboratories.

22 CRABTREE, PARKER, AND RUSSELL [J. S. M. P. E.

quantity of developer absorbed by the film in A is sufficient to
assure adequate development of the shadows in B; whereas in the
case of the highlights, the developing power of the developer is
exhausted before full development is attained. In this way, good
shadow detail is obtained while the highlights are not overdeveloped.

At the outset it was considered that the two-bath method of
treatment might permit the development of motion picture film to a
fixed degree of development or gamma irrespective of the time of
development beyond a certain critical value; that is, if the gelatin
emulsion absorbed a definite volume of developer in the first bath,
this portion would only be sufficient to develop the image to a
definite maximum contrast in the second bath, provided the latter
contained enough alkali.

So far as is known, the two-bath method of development has
been given very little consideration by motion picture laboratories,
although some of the advantages of this method have been pointed
out by Dundon, Brown, and Capstaff . 4 The merits of this method as
applied to motion picture film, therefore, seemed worthy of more
thorough investigation.

I. EXPERIMENTAL PROCEDURE

General methods of testing the photographic properties of de-
velopers have been described by one of the authors, 5 and the inter-
pretation of results obtained by the sensitometric method of Hurter
and Driffield (H&D) has been discussed by Jones and Crabtree, 6
and more recently by Jones. 7

The sensitometric measurements in this investigation were made
with Eastman motion picture supersensitive panchromatic nega-
tive film, exposed through step tablets on a Bell & Howell continuous
printer to a tungsten light source diffused with ground glass for
an exposure time of Vs2 second, thus receiving an intensity-scale
exposure.

Development was carried out in a miniature duplicate of the rack and
tank apparatus used for the commercial development of motion
picture film. Small racks holding six 10-inch strips of motion picture
film were placed in glass battery jars containing one-half gallon of
developer, which jars were immersed in a water bath to control the
temperature. It has been found that with the proper manipulative
technic this method of development gives more consistent results than
any of the other usual small-scale processing methods.

July, 1933]

TWO-BATH DEVELOPERS

For the exhaustion and graininess tests, 50-foot lengths of motion
picture film were developed in 1 -gallon tanks of solution, and the
fine-grain developer formula D-76 was used as the standard with
which the various experimental developers were compared.

II. TYPES OF TWO-BATH DEVELOPERS INVESTIGATED

This investigation has been limited to a study of three types
of two-bath developers as follows: (1) Those of Type I, in which
the first bath contained all the developing agents plus sodium sul-
fite, while the second bath contained all the alkali plus sodium
sulfite; (2) those of Type II, in which both the first and the second

zo
i. a

i.fe

1.4
I.Z
1.0
0.8
0.<o
0.4
O.Z

SHAPE:

OF" TWO Bf\TH OEVEUOPE.R ON
THE: CHARACTER I?>T\C CURVE:.

NORMAL.

MlQ>Hl_lCWT

EXPOSURE.

FIG. 1. Effect of two-bath developer on shape of the characteristic

curve.

baths contained developing agents; and (3) those of Type III, in
which the first and second baths were complete developers having
identical compositions.

A few of the developer formulas investigated which represent
the various types are given in Table I. The concentrations of the
constituents are given as grams per liter, and the relative speed and
fog values were measured at a gamma of approximately 0.7.

III. PHOTOGRAPHIC CHARACTERISTICS OF DEVELOPERS OF TYPE I

A . Shape of the Characteristic H&D Curve

At the outset it was considered that by the two-bath method of
development it might be possible to increase the steepness of the
toe gradient of the characteristic H&D curve, which would result

CRABTREE, PARKER, AND RUSSELL [J. S. M. P. E.

puoosy

EqC^.E
p

PPV
auog

XBJOQ

ninipog
anoninb

mnipog

anoninb
-ojpk H

O> CD * Oi

O O O ^ O CO CO

00 O> 00 00

vt C^ O '* W OS O5

CO CO Oi 00

00 00 00 00 00 00 t~

-* O <N

o o d

ooo ooooo

8 SS :

80 o ooooo
CO O ~ O O O *

ts. CD CO

Tj* CO CO 00 ^
CO CD CO CO CO

o o o

OOO O

O CO O 00

S S8SSS

O O

ooo ooooo

00 00 00 00

000000 *fTt 4 CDCO^J <

2 ic o o o o
S c c "-" >c '-'

^3 *^
Q W

Tj< O 00 ^f

O >Q o o

. <N . . t-i

So o o
o o o

888 88888

s ~f> o
' H -2

1C C 1C

|88 8

8 888

888 88888

I 1

1C 1C O CO N

CO CO ^^ ^C

1-H .1

iCiC>C 1C iC 1C CD C

nV*

-I ~S ~-( "<

July, 1933]

TWO-BATH DEVELOPERS

i >

o
o

c
o

-t

CD
f^

00
1C

o
o

1 :

o
o

Q
O

Q
c:

T 1

in '

O <M

2 >-?

loo

b. r

u
u
k

U 1- I

IMS

K L (D

z z

26 CRABTREE, PARKER, AND RUSSELL [J. s. M. P. E.

in greater picture shadow detail and an increase in the effective
emulsion speed. Tests showed that the developers with a high
concentration of alkali in the second bath gave slight increase in
speed, but the increase was not great enough to be noticeable in
camera exposures.

In order to test the effect on the latitude, strips of film were given
comparatively heavy exposures through the step tablets, and de-
veloped in two-bath developers and in the single-bath developer
D-76.

As illustrated in Fig. 1, the strips treated in the two-bath developer
(formula 13} showed a suggestion of a shoulder at the upper end of
the curve while those treated in the D-76 developer were perfectly
straight. However, since the highlight exposures in an average
normally exposed camera negative were equivalent to an exposure
value A, which is well down on the straight-line portion of the
curve, it would require more than eight times the normal exposure
before this effect could be detected in a picture negative. For this
reason, the slight loss in latitude is of very little practical importance.

B. Effect of Various Times of Treatment in the First Bath on Time-
Gamma Curves

The effect of varying the times of immersion in both the first
and second baths of formula No. 13 at 65F. is shown in Fig. 2 at A.
The shape of these time-gamma curves is characteristic for all
developers of Type I, although the increase in gamma with increasing
times of treatment in the first bath varies with the different formulas,
and is a minimum with those containing restraining agents.

From the curves it is seen that for a short time of immersion
in the first bath the degree of development increased rapidly during
the first 2 minutes of treatment in the second bath, but increased
only very slowly between 2 and 4 minutes and became practically
constant after 4 minutes. As the time of treatment in the first bath
increased, there was less and less increase in the degree of develop-
ment after immersion in the second bath; that is, the time-gamma
curve for the second bath became more nearly flat. In all cases
there was little increase in the degree of development after 2 minutes
in the second bath. This means that this method is not suitable in
cases where it is desired to control the gamma over a considerable
range, but that it is advantageous where, as is normally the case in
modern laboratory practice, a fixed gamma is desired.

July, 1933] TWO-BATH DEVELOPERS 27

C. Effect of Restraining Agents in the First Bath on the Time of
Appearance of the Image

With an ordinary developer, its properties change during use as a
result of the accumulation of the by-products of the development
reaction. With two-bath developers it is not necessary that any
development reaction occur in the first bath but only that the
emulsion absorb enough developing agents to carry out the develop-
ment process when placed in an alkaline second bath. Under such
conditions the solution of the first bath should last indefinitely,
requiring only the addition of enough liquid to replace that carried

TABLE H

Effect of Restraining Agents on the Time of Appearance of the Image

Restraining Agent

Quantity
(Grams per Liter)

Time of
Appearance

None

2 Min.

Sodium sulfate (desiccated)
Sodium sulfate (desiccated)
Sodium sulfate (desiccated)

25
100

2Min.
3 Min.
6 Min.

Cane sugar
Cane sugar
Cane sugar
Cane sugar

5
25
100
200

2 Min.

2 Min.
3 Min.
4 Min.

Cane sugar
Sodium bisulfite

100 \

5 Min.

Cane sugar
Sodium bisulfite

100 \

10 f

7 Min.

out by the film. However, developers containing elon, either with
or without hydroquinone, and a high concentration of sodium
sulfite, were quite active and produced an image in less than 2
minutes. From practical considerations it appeared that the film
should remain in the first bath at least 4 minutes, and an attempt was
therefore made to increase the time of appearance of the image of a
normally exposed negative to at least 4 minutes by the addition of
restraining agents.

When sodium bisulfite was added to decrease the alkalinity of
the first bath of formula No. 5, it was found that 10 grams per liter,
which decreased the alkalinity to a pH of 7.6, were necessary to
delay the appearance of the image sufficiently. This concentration

28 CRABTREE, PARKER, AND RUSSELL [J. S. M. P. E.

of bisulfite, however, gave low speeds when the second bath of the
developer had a low alkalinity; and in all cases it tended to lower
the alkalinity of the second bath on exhaustion, and so cause an
excessive change in the photographic properties. Other restraining
agents were tested, therefore, which gave the results shown in Table
II. The developer used was No. 12 (Table I).

Sodium sulfate reduced the solubility of the developing agents,
and the addition of only 25 grams per liter caused a slight precipita-
tion of the developing agents at 65F. The effect of glycerin was
found to be practically identical to that of sugar, which was con-
sidered preferable for economic reasons.

For the succeeding tests the following formula was used with
variations in the concentrations of developing agents and alkali
and the addition at times of other substances, such as antifogging
agents. It is formula No. 13 in Table I.

Two-Bath Developer, Formula No. 13

First Bath Second Bath

Elon 5 grams

Hydroquinone 2 grams

Sodium sulfite (desiccated) 100 grams 100 grams

Sodium carbonate (desiccated) 10 grams

Sugar 100 grams

Sodium bisulfite 5 grams

Water to make 1 liter 1 liter

D. Effect of the Alkalinity of the Second Bath on the Maximum
Gamma, Fog, and Emulsion Speed

The relation between the alkalinity of the second bath and the
maximum gamma, fog, and emulsion speed was determined for
elon developers with formulas Nos. 5, 6, 7, and 8, and for elon-
hydroquinone developers with formulas Nos. 9 and 10.

From the data in Table I it is seen that for the pure elon developers
the maximum gamma was independent of the concentration and
nature of the alkali present in the second bath, but when hydro-
quinone was present the alkalinity of the second bath had a decided
influence on the maximum gamma. Tests on numerous other
developers showed that this was due to the hydroquinone rather
than to the restraining agents, which were also present in the de-
velopers listed.

In general, an increase in the alkalinity of the second bath had a

July, 1933] TWO-BATH DEVELOPERS 29

tendency to increase the emulsion speed, but it also caused an in-
crease in the fog. A number of antifogging agents such as potassium
bromide, potassium iodide, and several organic compounds 8 were
tested; and it was found that for the developers containing 10 grams
per liter of sodium carbonate, the addition of 0.5 gram of potassium
bromide and 0.01 gram of potassium iodide to each liter of the second
bath reduced the fog sufficiently without affecting the speed ex-
cessively. For the higher gamma developers containing 50 grams per
liter of carbonate in the second bath, it was desirable to double
these quantities.

E. Effect of Temperature of Baths on Gamma and Speed

Since the gamma obtained by this method of development de-
pends on the quantity of developing agents absorbed by the emul-
sion, which is determined by the degree of swelling, it was desirable
to ascertain the relation between the temperature and the degree of
development. The data in Table III show that with developer No.
13 the change in gamma with temperature, at least between 55
and 75F., is not excessive.

TABLE III

Effect of Temperature on Maximum Gamma

Maximum Gamma after Maximum Gamma after

Temperature 3 Min. in First Bath 5 Min. in First Bath

55F. 0.39 0.43

65F. 0.48 0.55

75 F. 0.56 0.63

The variations in temperature also caused variations in the emul-
sion speed, a temperature rise of 1 degree raising the speed about 2
per cent, but this increase was not sufficient to make the higher
temperatures desirable.

F. Method of Controlling the Degree of Development

With the ordinary single-bath developers, the degree of develop-
ment is controlled by varying the time of treatment. Two-bath
developers of Type I, however, were so designed that variations in
the time of treatment would have a minimum effect on the degree of
development. For this reason, the usual method of controlling the
gamma was not applicable to the two-bath developers, and it was
necessary to seek some other means of regulating the gamma ob-
tained. Temperature changes and changes in the degree of agitation

30 CRABTREE, PARKER, AND RUSSELL [J. S. M. P. E.

caused variations in the degree of development, but the range
covered was too narrow to be of practical use. Variation in the con-
centrations of the developer constituents caused a wide variation in
the gamma obtained. Since in the cases where this type of develop-
ment would be applicable the gamma desired is known definitely
in advance, it was considered that control of gamma by adjustment
of the composition of the solutions would be quite satisfactory.

The following variables in the formula directly affect the
gamma :

(a) The concentration of elon.

(b) The concentration of hydroquinone.

The data in Table IV show the effect of changes in these three
factors with developer No. 13 as a base.

TABLE IV

Relation between Concentration of Developing Agents, Alkalinity of Second Bath,
and Maximum Gamma

Elon

(Grams per
Liter)

2
2
2
2
5
5

Changes in the alkalinity of the second bath had little effect
on the pure elon developer as noted above, but had an increasing
effect as the quantity of hydroquinone increased. Increase in the
elon concentration caused the gamma to increase by an amount
which was apparently independent of the hydroquinone concentra-
tion. Increasing the hydroquinone concentration from to 5 grams
per liter caused a large increase in gamma, but an increase from 5 to
10 grams per liter caused little further change. For practical work,
satisfactory concentrations would be: for gammas from 0.5 to 0.7,
5 grams per liter of elon, 2 to 5 grams per liter of hydroquinone,
and 10 grams per liter of carbonate; and for gammas from 0.7 to
0.9, the same elon and hydroquinone concentrations with 50 grams
per liter of carbonate in the second bath.

Hydroquinone
(Grams per Liter)

Maximum Gamma:
10 Grams per Liter
Carbonate in
Second Bath

Maximum Gamma:
50 Grams per Liter
Carbonate in
Second Bath

0.25

0.27

0.34

0.44

0.62

0.73

0.69

0.83

0.50

0.70

0.90

July, 1933]

TWO-BATH DEVELOPERS

Effect of Exhaustion on the Time-Gamma Curve, Emulsion Speed,

and Fog

Practically all the changes occurring in a developer of Type I
on exhaustion take place in the second bath. The first bath changes
slightly by virtue of the solvent action of the sodium sulfite on the
silver bromide in the emulsion and the reduction of this silver bromide
to metallic silver. From practical considerations, however, these
changes are negligible. In the second bath, however, a number

I 050

0.40

IOO

% 80

t i oo

EXHAUSTION Or

(p-oRt^vji-K No.iV)

SPEED <\T

F-OR C,\VES

EXPOSURE.

FIG. 3. Chart showing exhaustion of two-bath developer, Type I (Formula

No. 13).

of important changes occur. There is an accumulation of bromide
and iodide and of the oxidation products of the developing agents
which are by-products of the development reaction, and these
products tend to restrain development. 9 On the other hand, there
is a gradual accumulation of the excess developing agents carried
over by the film from the first bath, which tends to continue the
development after the developing agents absorbed by the emulsion
have been exhausted. These two opposing effects cause a change in
the shape of the time-gamma curve. During the first few minutes of
treatment in the second bath the effect of the restraining forces
predominates, and the time-gamma curve falls below that for the

32 CRABTREE, PARKER, AND RUSSELL [J. S. M. P. E.

fresh developer; but later the accumulated developing agents pre-
vent the flattening of the curve, so it continues to rise and crosses
the curve for the fresh developer. In order to maintain the most
uniform properties in the developer, the quantity of developing
agents carried into the second bath should therefore be kept as low
as possible. In Fig. 2 are given the time-gamma curves obtained at
two stages in the exhaustion of the baths of formula No. 13 containing
0.5 gram of potassium bromide and 0.01 gram of potassium iodide
per liter in the second bath.

Data showing the changes in other properties of this developer
are given in Fig. 3, where the gamma, relative emulsion speed,
density, and fog are plotted against the number of feet of film de-
veloped per gallon of developer. The exhaustion was carried out in a
small tube developing machine, and the film squeegeed by means of
cotton as it left the first bath so as to reduce the volume of the first
bath carried over into the second. The film was flashed to give a
uniform density of 0.9 in this developer. The first bath was ex-
hausted with 900 feet of film per gallon and the only revival was the
addition, every 300 feet, of sufficient fresh solution to replace that
carried out by the film. The second bath was changed every 300
feet per gallon, and the bromide and iodide were omitted after the
first 300 feet had been processed.

It is seen from Fig. 2 at B that over a developing range of 4 to 6
minutes in the second bath the gamma changed only from 0.52 to
0.56, even after exhaustion with 200 feet of film per gallon. With the
fresh developer there was no change in gamma. It is considered that
with air squeegeeing, the change in gamma on exhaustion would be
even less than indicated above. From Fig. 3 it is seen that the speed
loss was only 20 per cent of the original speed after exhaustion with
300 feet of film per gallon, and that the first bath was practically
unaffected, since the renewal of the second bath brought the speed
back almost to the original value. After 600 feet per gallon had
been processed, the first bath began to be affected, as is shown by the
slight drop in gamma.

H. Use of Two-Bath Developers with Sound Recording Films

For the development of variable density sound records, relatively
low gammas (around 0.5) are required for the sound negatives, and
relatively high gammas for the combined picture and sound positives.
With regard to the latter, even with the use of caustic alkalies in

July, 1933] TWO-BATH DEVELOPERS 33

the second bath, it was not possible to obtain sufficiently high
gammas (around 2.0) with existing sound recording films.

Preliminary tests indicated that in general the factors to be con-
sidered in the development of sound recording films were similar
to those which applied in the case of picture negative films, although
since the former as a class are relatively fine grained, the concentration
of the sulfite in the developer solutions was found to be important.

High sulfite concentrations caused distortion of the characteristic
curve of Eastman motion picture films, emulsions Nos. 1301 and
1359, manifested by the total absence of a straight-line portion.
When the sulfite concentrations were lowered to 10 grams per liter
in the first bath and 25 grams per liter in the second bath, this
distortion disappeared, and these sulfite concentrations were suffi-
cient to prevent excessive aerial oxidation. The following formula
was chosen for further tests :

First Bath Second Bath

Elon 7 grams

Sodium sulfite (desiccated) 10 grams 25 grams

Sodium carbonate (desiccated) 10 grams

Sodium bisulfite 2 grams

Sugar 100 grams

Water to make 1 liter 1 liter

With the above formula, the following properties were investigated :
(7) Time-Gamma Characteristics. Table V shows the gammas
obtained for various times of treatment in the second bath after 3
minutes in the first bath, and also the gammas obtained with the
same film in D-76. Practically no action occurred during the time of
treatment in the first bath but immediately on immersion in the
second bath, the development proceeded rapidly and was practically
complete within 1 minute.

TABLE V

Comparison of the Effect on the Gamma of Negative Sound Records of Varying
Times of Development in D-76 and in a Type I Two-Bath Developer

Two-Bath Developer

Time of Treatment D-76

First Bath Second Bath Gamma Time Gamma

3 Min. Min. 0.0 2 Min. 0.29

3 Min. 1 Min. 0.49 3 Min. 0.49

3 Min. 3 Min. 0.52 4 Min. 0.64

5 Min. 0.74

(2) Effect on Latitude and Resolving Power. There was very
little difference between the latitudes obtained in the D-76 and

CRABTREE, PARKER, AND RUSSELL [J. S. M. P. E.

the two-bath developers, although a slightly greater speed was
obtained with the two-bath formula. The resolving power was
equal to that observed with formula D-76.

(3) Changes during Exhaustion. Fig. 4 shows the changes
in gamma, speed, and highlight density during the exhaustion of
D-76 and of the two-bath developers with Eastman motion picture
film, emulsion No. 1301. The D-76 exhaustion was carried out in a
small tank apparatus, while the two-bath developer was exhausted
in a small tube developing machine in which the film was squeegeed

I .4

Q .oo

U80

Si CO
40

EFFECT OP EXHAUSTION OF O-7G

AND TWO BATH OEVE.LOPE.R.

. AJ 'S.MIN.
-6MIN. \N O-lfc
(NO REVIV/kLJ)

GAMMA,

RELATIVE SPEED

DENSITV FOR

EXPOSURE

> ao

64.

100 ZOO 3.00 400 BOO C.OO TOO 800 3OO

FIG. 4. Effect of exhaustion of D-76 and two-bath developer

(Type I).

between the first and second baths. The sensitometric strips were
developed in the machine.

In the D-76, the speed, gamma, and density all dropped rapidly
on exhaustion, while in the two-bath developer, the gamma and
density remained approximately constant and the speed dropped
gradually.

(4) Control of Gamma. Table VI shows the effect on gamma
of changes in the concentrations of developing agents. Gammas up
to 0.6 are readily obtained with elon alone, but for gammas above
0.6 it becomes advisable to reduce the elon concentration and add
hydroquinone.

July, 1933] TWO-BATH DEVELOPERS 35

IV. PHOTOGRAPHIC CHARACTERISTICS OF DEVELOPERS OF TYPE II

Developers of Type I were considered suitable when a definite
predetermined gamma was required, but in order to be able to vary
the degree of development in the case of negatives exposed under
widely varying conditions, a further study was made of developers
of Type II which contained developing agents in both baths.

The two-bath developers of Type II were divided into two classes,
A and B. Those of class A contained developing agents and sodium
sulfite in both baths with all the alkali in the second bath, and those
of class B contained developing agents, sodium sulfite, and alkali
in both baths but in different proportions. Although with developers
of class A such as No. 14 it was possible to control the degree of
development in the second bath, the emulsion speed decreased

TABLE VI

Effect of Varying Eton and Hydroquinone Concentrations on the Gamma of a Two-
Bath Sound Record Developer

Elon Concentration Hydroquinone Concentration

(Grams per Liter) (Grams per Liter) Gamma

5 0.44

7 0.52

10 0.60

5 2 0.69

5 4 0.81

5 10 1.05

rapidly on exhaustion for gammas less than 0.5. For this reason, the
class A developers of this type were not investigated further.

Previous experiments 9 have indicated that the by-products of
development which accumulate during exhaustion of the D-76
developer rapidly decrease not only the effective emulsion speed but
also the rate of development. In order to prevent to a certain extent
this change in the photographic properties of the developer, it was
recommended that after 80 feet per gallon of motion picture panchro-
matic negative film had been processed, the bath should be revived
with one-half the original quantities of elon, hydroquinone, and borax
dissolved in hot water, with the addition of enough sodium sulfite
to make its concentration in the reviving solution equal to 10 per
cent. By this procedure it was possible to maintain the rate of
development fairly constant, and also to maintain a higher emulsion
speed than without revival, but there was an inevitable loss in speed
as the developer aged. At the outset, it was considered that the

36 CRABTREE, PARKER, AND RUSSELL [J. S. M. P. E.

emulsion speed could be maintained by the two-bath method of
development according to the procedure suggested by Dundon,
Brown, and Capstaff, 4 and that possibly the rate of development
could be maintained constant by so adjusting the composition of the
first bath that the developer carried over into the second would
serve as a replenishing solution for the latter. Accordingly, de-
velopers of Type II, class B, were compounded.

A . Factors Influencing the Choice of Formula for First Bath

In the choice of a formula for the first bath, it was necessary to
consider the following :

(1) The quantity of developer carried over by the film.

(2) The limiting solubility of the constituents.

(3) The formula of the second solution.

(4) The restraining action of the reaction products accumulated in the second
bath.

A consideration of these factors indicated that the first bath which
is to act as a replenisher for the second should increase the alkalinity
and concentration of the developing agents in the second bath
during exhaustion. On the other hand, it should not be so concen-
trated that any considerable degree of development occurs during
the time of treatment in the first bath, since in such a case the first
bath itself would become exhausted too rapidly. Also, since it is
desirable to be able to use desensitizers in the first bath, and since
the presence of hydroquinone tends to cause the precipitation of
many desensitizers, this reducing agent was omitted from the formulas
tested. The study of developers of this type was also limited to those
in which the first bath contained elon, borax, and sulfite, while the
second bath consisted of the regular D-76 formula.

B. Exhaustion Tests with Typi II Developers

In the following exhaustion tests sensitometric strips were de-
veloped in the fresh baths and after every 50 feet of film per gallon
had been processed. The baths were exhausted by developing
lengths of motion picture panchromatic negative film, flashed so
that a density of 1.0 was obtained when developed for 12 minutes
in fresh D-76. The film was bathed for 3 minutes in the first bath,
and for the remainder of the time in the second.

In order that the results might be compared with those obtained
with the D-76 developer under the most favorable conditions, the

July, 1933]

TWO-BATH DEVELOPERS

D-76 developer was revived in a manner similar to that recommended
in a previous investigation 9 described above.

The chief factors considered in comparing the properties on
exhaustion included (1) the change in the gamma and shape of the
characteristic curve; (2) the loss in emulsion speed and change in
fog for a given gamma; and (3) the decrease in density for a given
exposure.

The effects of exhaustion of the single-bath D-76 developer with

<: O.T
fa*

UIOO
B

t 10
(/>

0.18

00.14

EXHAUSTION OF
0-T<b WITH XMSO WITHOUT
t?EVIVA>U.

EXHAUSTION Of
TWO BATH DEVELOPER
F~ORt--\UL.A> NO. IB

0.-7
O (o
O "b
IOO
80
<SO

1.0
OB

0.16
0.14
IO

^ h-llN. IN F-IRVT
It " " SE.CONO

FIG. 5.

O IOO

FT. PER G*M_. OV FIR^T

FIG. 6.

FIG. 5. Chart showing exhaustion of D-76 with and without revival.
FIG. 6. Exhaustion of two-bath developer, Type II (Formula No. 18).

motion picture negative film are shown in Fig. 5. The results indi-
cate, as has been previously shown, 9 that the products of exhaustion
rapidly affect the photographic properties of the developer. The
properties are revived by additional quantities of developing agents
and alkali, but even then they decrease rapidly on further exhaustion.
The effect of exhaustion to 200 feet of negative film per gallon
on several two-bath developers is shown in Table VII. The con-
stituents in grams per liter are given for the first bath only, since
D-76 was used as the second bath in all cases. The first bath was
compounded with 100 grams per liter of sodium sulfite, and through-

CRABTREE, PARKER, AND RUSSELL [J. S. M. P. E.

out the exhaustion the film was bathed for 3 minutes in the first
bath and for the remainder of the time of development in the second
bath.

Since very little development occurred in the first bath, it did not
exhaust rapidly, and therefore furnished a fresh supply of developing
agents to the second bath.

A study of the results in Table VII indicates that it was necessary
to increase the concentration of both the developing agents and alkali
in the second bath in order to maintain the rate of development
constant, and in no case was it possible to obtain a constant emulsion
speed. Other experiments were made to obtain a developer with
which the emulsion speed did not decrease on exhaustion, but this
was not possible with the use of elon-borax formulas tested for the
first developer.

TABLE VII

Exhaustion of Two-Bath Developers, Type II (65 F.")

First Bath*

Fresh Developer

Exhausted Developer

Time of
Bathing

No.

(Grams per
Liter)
Elon Borax

Gamma
(12Min.)

Relative ,,
Speed Fog

Gamma

Relative ~
Speed Fo

(First Bath)

(at Gamma - 0.7)

(at Gamma - 0.7

2 5

0.72

100 0.16

0.64

72 0.12

3 Min.

5 10

0.74

100 0.15

0.62

60 0.12

3Min.

2 20

0.74

100 0.16

0.66

71 0.10

3 Min.

5 20

0.74

100 0.16

0.70

71 0.11

3 Min.

* Second bath = D-76.

The change in photographic properties of formula No. 18 (Table
VII) on exhaustion is shown in detail in Fig. 6. A comparison of the
curves with those in Fig. 5 shows that except for the loss in emulsion
speed, the photographic properties of this developer are maintained
more uniformly than those of the D-76 developer revived. The loss
in the emulsion speed is practically equal in both cases.

V. PHOTOGRAPHIC CHARACTERISTICS OF DEVELOPERS OF TYPE III

The only experiments made under this classification were with
two baths of formula D-76. Dundon, Brown, and Capstaff 4 have
stated that the loss in emulsion speed in an exhausted developer is
due to the solvent action of the sulfite on the latent image during
the period of induction, which is greatly lengthened by the presence
of accumulated bromide. If this is so, there should be no loss in
emulsion speed if development is started in a fresh developer until

July, 1933]

TWO-BATH DEVELOPERS

the image appears and is then continued in an exhausted or bromided
developer. Since tests made in the above investigation seemed to

1.4
l.fc
1.0
0.8
O.C,
0-4
O.Z.

TWO BATCH O-HQ>

L.OG EXPOSURE,

FIG. 7. Characteristic curves of motion picture panchromatic negative film
with two-bath D-76 developer, neither bath being bromided.

1.4
l.
1.0
0.6
O.C,
0.4

TWO BATH D-

FIG. 8. Characteristic curves of motion picture panchromatic negative film
with two-bath D-76 developer, the two baths being bromided to the same

degree.

substantiate this conclusion, further practical tests of the method
were undertaken.

A . Effect of Bromide in Two-Bath Developers
Since the changes in the properties of a developer during exhaustion

CRABTREE, PARKER, AND RUSSELL [J. S. M. P. E.

are undoubtedly due largely to the influence of bromide, iodide, and
other reaction products of development rather than to an actual

TWO aATTH O-1>

\ &R*t^ PC. LITER KBf

FIG. 9. Characteristic curves obtained with motion picture panchromatic
film, the development being started in an unbromided developer and completed
in a bromided developer.

1.4
I.Z

10
0.6

O.fc
0.4
O.Z.

TWO BATTH

FIG. 10. Characteristic curves obtained with motion picture panchromatic
negative film, the development being started in a bromided developer and
completed in an unbromided developer.

exhaustion or oxidation of the developing agents, and since the effect
of bromide in single-bath developers has been carefully investigated,

July, 1933] TWO-BATH DEVELOPERS 41

a study of the effect of bromide in two-bath developers was under-
taken.

Previous investigators 10 have found that with certain emulsions
and certain unbromided developers, the tangents to the straight-
line portions of a series of characteristic curves for various times of
development meet in a common point of intersection on the log E
axis. The addition of bromide to the developer causes a downward
displacement of this intersection point. Motion picture panchromatic
negative film, with developer D-76, gives curves of this type. Similar
results are obtained with two-bath developers when both baths are
unbromided and when both baths are bromided to the same degree,
as is shown by the curves in Figs. 7 and 8. The effects on this dis-
placement of the intersection point when (a) development is started
in an unbromided developer and completed in a bromided developer,
and (6) when development is started in a bromided developer and
completed in an unbromided developer, are shown in Figs. 9 and 10.
The curves in Fig. 9 were obtained when the film was first placed
in a D-76 developer for 2 minutes and then transferred to a D-76
developer containing 1 gram per liter of bromide for the completion
of development. It is seen that the tangent to the straight-line
portion of the curve for the strip treated only in the plain developer
intersects the log E axis at the normal point, but the strips which
were treated for increasing times in the bromided second bath gave
tangents showing a progressive lowering at this value of log E, until a
maximum depression was obtained equal to that obtained with full
development in a bromided developer. In Fig. 10, curve A shows
the density lowering due to bromide, while the curves for longer
times show a progressive elevation of the intersection point with the
above log E value until, for long times of development, the effect of
the bromide practically disappears. This shows that the effect of
the bromide is not permanent, as it would be if a destruction of the
latent image were involved, but is rather in the nature of a mass
action effect where the bromide lowers either the solubility or degree
of ionization of the silver bromide, thus retarding development.
Tests were made with different concentrations of bromide and with
developers of Type II, and all gave curves similar to those illustrated.

Several series of tests were then made, using fresh and exhausted
D-76 developers for the two baths. Except for slight deviations
which were no greater than normal experimental error, all these
tests gave results similar to those obtained with the bromided and

CRABTREE, PARKER, AND RUSSELL [J. S. M. P. E.

unbromided developers. The results obtained in one of these series
are given in Table VIII. One series of tests was conducted in 120-
gallon tanks with developers which had been exhausted in the routine
processing of motion picture film, and gave results parallel to those
given in Table VIII.

B. Advantages of First Immersing the Film in a Fresh Bath

From the data in Table VIII it is seen that there is only a slight
loss in emulsion speed if a fresh developer is used during part of

TABLE VIII
Effect of Using Fresh and Exhausted Developers in Two-Bath Development

Solutions

Time in
First
Bath

Gamma in Total
Development Time

Total Time
for Gamma of

Relative
Speed at
Gamma of

8 Min.

12 Min.

15 Min.

20 Min.

0.5 | 0.7

Fresh D-76

(I)

0.54

0.70

0.78

0.91

7 Min. 12 Min.

100 100

D-76 Ex-

hausted to

200 feet

per gallon
(II)

0.55

0.64

0.78

17 Min.

First Second

Bath Bath

II I

2Min.

0.49

0.71

0.78

0.91

8 Min. 12 Min.

76 85

II I

4 Min.

0.45

0.60

0.73

0.85

9 Min. 14 Min.

68 85

II I

6 Min.

0.48

0.69

0.77

0.90

8 Min. 12 Min.

49 66

I II

2 Min.

0.60

0.65

0.78

9 Min. 16 Min.

72 74

I II

4 Min.

0.51

0.64

0.67

0.78

8 Min. 15 Min.

78 78

I II

6 Min.

0.46

0.64

0.70

0.79

8 Min. 15 Min.

89 79

the development time. The question arises as to whether the fresh
bath should come first or last. The table shows that for short times of
treatment in the first bath, there is a slight advantage in using the
exhausted bath first. Practical considerations, however, make it
desirable that the fresh solution be used as the first bath. When the
film is transferred from the first bath to the second, it carries a
considerable quantity of the solution with it. Therefore, if an
exhausted bath were used first, the bromided developer would be
carried into the fresh second bath, which would therefore rapidly
become exhausted. The fresh bath should be exhausted no more
rapidly than necessary, because when the exhausted bath is dis-

July, 1933]

TWO-BATH DEVELOPERS

carded, the partially exhausted developer may be used with a fresh
bath through another exhaustion cycle. Also, if the fresh developer
is used as the first bath, the portion of it carried into the exhausted
bath will have somewhat of a reviving action and retard its ex-
haustion. For these reasons, in all the exhaustion tests made with
this type of developer the fresh developer was used as the first bath.

C. Comparison of Properties of Two-Bath Type III Developers with
a Single-Bath Developer on Exhaustion

Exhaustion tests were made to compare the behavior of developers
using two baths of formula D-76 with that of a single bath of D-76.

< 0.-7
O.fe
*OA

5 100

6 60

BATMS INTERCHANGED

SECOND BATH
DISCARDED

_ GAMt-IA AT It MIN.TOTAU T\h^E

U 60

t. 10

O 014

EXHAUSTION OF
TWO BXVTM D-Tfc

. RELATIVE SPEED AT OAf--M^A

DEfSSITV FOR GIVEN EXPOSURE

ATT GAT-lt^KV =

-RELATIVE. SPEED AT GAMMA =

O 18
OIA

NO REVIVAl-

AT VZ- ^^ IN. TOTAL- TIME.

FCE.T PER G.AUUON OF

FIG. 11.

FIG. 12.

FIG. 11. Chart showing exhaustion of two-bath D-76 developer, the baths
being interchanged. FIG. 12. Chart showing exhaustion of two-bath D-76
developer, with no revival.

In all the two-bath tests the film was treated for 2 minutes in the first
bath, then transferred to the second bath for the completion of the
development. The procedure used during the exhaustions has been
described under the discussion of developers of Type II.
Three different sets of tests were made as follows :
(1) Two baths of formula D-76 were exhausted with 100 feet of

44 CRABTREE, PARKER, AND RUSSELL [J. S. M. P. E.

film per gallon. The second bath was then discarded, the first bath
used as the second, and a fresh solution supplied for the first bath.
This procedure was repeated every 100 feet per gallon.

(2) The second exhaustion test was similar to the above, but
each set of solutions was exhausted to 200 feet per gallon before they
were changed.

(3) In the third exhaustion the solutions were changed every
200 feet per gallon, but the second bath was revived after 100 feet
per gallon, in a manner similar to that used with the single D-76 bath.
Figs. 11 and 12 show the changes in photographic characteristics
during the first two of these exhaustions.

When the second bath was discarded every hundred feet, very
uniform properties were obtained. The gamma obtained in 12
minutes varied between 0.70 and 0.64, while the greatest speed drop
was to a value equal to 85 per cent of the original. These values
should be compared with a gamma change from 0.68 to 0.55 and a
speed drop to 72 per cent for the single-bath D-76 developer with
revival.

As explained above, the quantity of chemicals added during
revival of the D-76 developer was equal to one-half the original
weight used except for the sulfite, so that the total quantity of
chemicals used to develop 200 feet of film by this method was one
and one-half times the quantity in the original formula.

Using the two-bath method, when the solutions were changed
every 100 feet per gallon, the quantity of chemicals required for each
200 feet was equal to twice the weight of the chemicals in the original
formula. Therefore, the two-bath method used only one-third more
chemicals than the single bath, and gave much more uniform proper-
ties.

When the second bath was discarded every 200 feet per gallon,
the drop in the rate of development and the loss in emulsion speed
were of the same order as those obtained with the single-bath D-76
developer revived, but the quantity of chemicals required to develop
200 feet of film by this method was equal to that in the original
formula, and thus was one-third less than that required by the D-76
developer revived.

When the second bath was revived after 100 feet of film had
been processed per gallon very little, if any, advantage was obtained
over the two-bath developer without revival, while the consumption
of chemicals was increased.

July, 1933] TWO-BATH DEVELOPERS 45

VI. PRACTICAL APPLICATIONS

One of the most important problems confronting motion picture
processing laboratories is that of securing a constant degree of
development, especially with sound-on-film records. Formerly it
was possible to compensate for variations in the gamma of the
negative by varying the gamma of the positive print, but present
practice requires that sound negatives be developed to a fixed gamma
ranging from 0.40 to 0.7 in the various laboratories. 11 This deter-
mines the gamma for the projection positives at values ranging from
1.8 to 2.3, which, in turn, determines the gamma of the picture nega-
tive. The absolute gamma aimed at varies in the different labo-
ratories, but this is immaterial so long as the product of the gamma
of the negative and that of the positive sound record is approxi-
mately 1.2. Any considerable departure from the standardized
gammas will cause a loss in quality of either the picture or the sound
or both, and for uniform results they must be maintained.

All developers undergo changes in their characteristics during
use as a result of the exhaustion of the developing agents and the
accumulation of reaction products such as sodium bromide, which
tends to restrain the development. This change is greatest for the
low energy developers, such as those of the borax type, because
they have much less power to resist the effect of bromide. Thess
changes in properties are evidenced in two ways: (a) the rate of
development is reduced, and (6) the effective emulsion speed is
lowered.

In practice, these changes may be partially offset either by lengthen-
ing the time of development so as to obtain the desired gamma,
or by adding booster solutions to revive the developer. The first
of these methods requires frequent tests to determine the necessary
time of development, while it does not compensate for the speed
loss. The second method could, theoretically, with the proper
control, maintain uniform results. In practice, however, it is difficult
to hold the ratio between the quantity of booster added and the
quantity of film processed at the proper value, with the result that
the processing is not always absolutely uniform.

A . Two-Bath Developers

In no case is it possible to obtain absolutely uniform development
properties throughout the life of a developer, but by the use of two-
bath developers this condition is approached much more nearly

CRABTREE, PARKER, AND RUSSELL [J. S. M. P. E.

than with single-bath developers which are suitably revived. The
two-bath developers give a practically constant gamma throughout
their life, and the emulsion speed loss is less than with single-bath
developers. Since the properties of the various types of two-bath
developers differ considerably, each type will be considered separately.

(1) Two-Bath Developers of Type I. The outstanding char-
acteristic of the two-bath developers of Type I is their ability to
give an almost constant gamma over a wide range of time of develop-
ment.

A satisfactory formula for a developer of this type designed to
give a gamma between 0.5 and 0.55 with Eastman motion picture
supersensitive panchromatic negative film is as follows:

Two-Bath Developer for Supersensitive Panchromatic Negative Film
(Formula SD-4)

Avoirdupois

Metric

Solution A First Bath
Elon

Hydroquinone

Sodium sulfite, desiccated (E. K. Co.)
Sugar

Sodium bisulfite (E. K. Co.)
Water to make

Solution B Second Bath

Sodium carbonate, desiccated (E. K. Co.)
Sodium sulfite, desiccated (E. K. Co.)
Potassium bromide
Potassium iodide
Water to make

Ibs.

5.0 grams

Ibs.

2.0 grams

100

Ibs.

100.0 grams

100

Ibs.

100.0 grams

Ibs.

5.0 grams

120 gallons 1.0 liter

10 Ibs.

100 Ibs.

V lb.

70 grains
120 gallons

10.0 grams

100.0 grams

0.5 gram

0.01 gram

1 . liter

The purpose of the sugar and the sodium bisulfite in the first bath
is to restrain development and prevent the appearance of an image,
thus greatly reducing the effect of exhaustion on the bath. If the
sugar is decreased, the time of appearance of the image is decreased,
while if the sodium bisulfite is decreased, the pH of the solution
increases and the time of appearance of the image is decreased.
It is not desirable that any image should appear during the period
of treatment in the- first bath as it would cause the more rapid ex-
haustion of the bath.

When the film is immersed in the first bath, the emulsion absorbs
a definite volume of the solution containing the developing agents;
and when the film is transferred to the second bath, the alkali acts on

July, 1933] TWO-BATH DEVELOPERS 47

the developing agents in the emulsion, causing a fairly rapid rate of
development. The developing agents held in the emulsion are soon
exhausted by the reaction or by diffusion into the surrounding
solution, so that after 3 or 4 minutes in the second bath no further
developing action occurs. This effect is shown by the time-gamma
curves in Fig. 2 at A. The curves also indicate that variation of
the time of treatment in the first bath has only a slight effect on the
gamma obtained.

Control of Gamma. Since it is not practicable with these developers
to vary the gamma by changing the time of treatment, the gamma

TABLE IX

Effect of Variation in the Composition of Type III Two-Bath Developers on the
Gamma with Constant Time of Development

Time of Treatment

First Second
Gamma First Bath Second Bath Bath Bath

0.5 A B 4 Min. 4 Min.

0.7 A with 5 grams per B 4 Min. 4 Min.

liter of hydro-
quinone.

0.7 A B with 50 grams per liter 4 Min. 4 Min.

of sodium carbonate and
double quantities of
bromide and iodide.

0.9 A with 5 grams per B with 50 grams per liter 4 Min. 4 Min.
liter of hydro- of sodium carbonate and
quinone. double quantities of

bromide and iodide.

must be varied by altering the composition of the baths. The gamma
obtained with a given formula will vary slightly with different
processing conditions, so that the exact composition for any desired
gamma must be determined by trial. Table IX indicates the results
obtained with various concentrations of developing agents and alkali.

The formulas in Table IX are only suggestions, since the same
gammas can be obtained with other combinations. An increase of
the elon and hydroquinone concentrations, separately or together,
increases the gamma. Increasing the alkalinity of the second bath
increases the gamma if there is hydroquinone in the first bath, but
this has little effect if elon is the only developing agent present.

Since the developing action is dependent on the developing agents
carried in the emulsion from the first bath, very heavily exposed
portions of the film tend to exhaust the developing agents before

48 CRABTREE, PARKER, AND RUSSELL [J. S. M. P. E.

the full density is reached. This effect is shown on the characteristic
H&D curve by the appearance of a slight shoulder in the over-
exposure region, as shown in Fig. 1. This effect is not of practical
importance, however, since it could not be detected in an ordinary
picture unless it were given at least eight times the normal exposure.

Graininess. Experiments have shown that the above two-bath
developer gives images, the graininess of which is of the same low
order of magnitude as that given by the borax type of developer,
which is as would be expected since the concentration of sulfite and
the times of development are approximately equal with both types
of developers.

Effect of Temperature. The effect of temperature on this developer
is not great, a 5-degree rise in temperature causing only a 10 per cent
increase in gamma and a 10 per cent increase in speed.

Life of Baths. Although this developer, by virtue of its constant
gamma characteristics, has advantages over the single-bath de-
velopers when the solutions are fresh, its greatest advantage is
realized on exhaustion. Since almost no development occurs in the
first bath, it is exhausted only very slowly. The second bath ac-
cumulates sodium bromide and other restraining agents, but it also
accumulates developing agents which are carried in by the film.
The restraining agents slow down the development during the first
part of the treatment in the second bath but the developing agents,
in turn, accelerate the development during the latter part of the
treatment. This effect is shown in Fig. 2 at B. If the quantity
of developing agents added is kept at a minimum by thoroughly
squeegeeing the film between the first and second baths, the gamma
obtained with 4 minutes' treatment in each bath remains practically
constant during exhaustion, and the emulsion speed loss is low.
The behavior on exhaustion is shown in Fig. 3. The first bath was
exhausted with 600 feet of film per gallon before it began to be
affected sufficiently to lower the gamma. The second bath was
discarded and replaced by a fresh bath after every 300 feet of film
per gallon had been processed. With more thorough squeegeeing, as
with an air squeegee, it should last even longer. Comparison of Fig. 3
with Fig. 5, which gives the characteristics of the single-bath borax
developer D-76, shows that the two-bath developer can process three
times as much film with an approximately equal speed loss and with
practically no change in gamma.

Use with Sound Recording Films. Two-bath developers of Type I

July, 1933] TWO-BATH DEVELOPERS 49

are equally satisfactory for use with sound recording films where low
gammas are required. The following formula is satisfactory :

Two-Bath Developer for Variable Density Sound Negatives (Formula SD-5)

Avoirdupois Metric

Solution A First Bath

Elon 71bs. 7.0 grams

Sodium sulfite, desiccated 10 Ibs. 10.0 grams

Sodium bisulfite 2 Ibs. 2.0 grams

Sugar 100 Ibs. 100.0 grams

Water to make 120 gallons 1 . liter

Solution B Second Bath

Sodium sulfite, desiccated 25 Ibs. 25.0 grams

Sodium carbonate, desiccated 10 Ibs. 10.0 grams

Water to make 120 gallons 1 . liter

Manipulative details have been given on p. 32.

(2} Two-Bath Developers of Type II. For those cases where
it is desirable to be able to vary the gamma by varying the time of
treatment, two-bath developers of Type II have several advantages.

With these developers the first bath acts as a booster solution for
the second bath, which has the same composition as an ordinary
single-bath borax developer. Since this booster solution is carried
into the second bath by the film itself, the quantity added is propor-
tional to the quantity of film processed. If the first bath is properly
compounded to fit the particular conditions of use, with this procedure
the rate of development will be maintained absolutely uniform.
The speed loss will not be eliminated but it will be diminished.
Fig. 6 shows the change in properties of a developer of this type
during exhaustion. It is seen that the gamma remained uniform,
while the emulsion speed gradually dropped.

(3) Advantage of Type III Developers. The two-bath developers
of Type III consist of two baths of an ordinary single -bath formula.
Their advantage lies in the fact that the film is treated for only
a short time in the first bath, which therefore remains comparatively
fresh. The short treatment in this fresh bath prevents the gamma
and emulsion speed from dropping as they would for total treatment
in an exhausted developer. Also, when the second bath becomes
completely exhausted, the first solution which is only slightly ex-
hausted may be used as the second bath with a fresh first bath.
Figs. 11 and 12 show the properties on exhaustion of two baths of the
D-76 formula.

50 CRABTREE, PARKER, AND RUSSELL [J. S. M. P. E.

B. Application of Two-Bath Developers to Machine and Rack and

Tank Systems

(1} Machine Systems. In machine development two systems are
in common use for handling the developer solutions: (a) the
recirculating system, and (b) the continuous flow system.

(a) Recirculating Systems. In the recirculating system, a large
volume of developer is held in a storage tank and circulated by means
of pumps from this tank through the tubes or tanks of the machine
and back to the storage tanks. If the developer is revived, the booster
solution may be added continuously at some point in the system or
it may be added at intervals to the storage tank.

With this system, the two-bath developers of Type I retain all
their advantages, giving constant gamma with variations in time,
constant gamma on exhaustion, and low speed loss on exhaustion.
The only precaution necessary is to see that the film is well squeegeed
between the first and second baths.

Developers of Type II and Type III retain their advantages of a
small gamma change and low speed loss on exhaustion. The de-
velopers of Type II give a constant gamma without increasing the
time of development, but the developers of Type III are somewhat
more economical of chemicals, since the first bath is not discarded
but is merely shifted to be used as the second bath. Also, there
is a less abrupt change in the developing power when the shift is
made to a fresh first bath. Neither of these types, however, has as
long an exhaustion life as Type I.

(6) Continuous Flow Systems. In the continuous flow system,
fresh developer is allowed to flow in continuously at one end of
the machine and to overflow continuously at the other end.

With this method, the developers of Type I retain all their ad-
vantages of constant gamma and long life, but the advantages of
Types II and III are of doubtful value. This system itself might
be considered as a multiple-bath developer with an infinite number
of changes. When the ratio between the quantity of developer used
and the length of film processed is held constant, an equilibrium is
reached with fresh developer at one point, exhausted developer at
another point, and various degress of exhaustion in between. The
degree of exhaustion at any point in the system remains con-
stant and therefore there is no change in the properties of the
developer.

July, 1933] TWO-BATH DEVELOPERS 51

Disadvantages of Two-Bath Developers with Machine Systems.
There are also certain disadvantages connected with the use of two-
bath developers in machines. These are briefly :

(a) Extra storage tanks, pipes, and pumping equipment are required for
handling the extra bath.

(b) It is necessary to compound two solutions rather than one, but this is
somewhat compensated for by the fact that the solutions last longer.

(d) With developers of Type III it is necessary to shift the slightly exhausted
solution from the first tank into the second tank. This shift could be accom-
plished quickly by means of valves if each storage tank were connected to both
sets of developing tubes or tanks.

(c) Rack and Tank Systems. With the rack and tank system,
when the rack is removed to the next bath, a considerable quantity
of solution is carried along, not only on the film but also on the
surface of the racks. This is a disadvantage with developers of Type I
because it causes the rapid exhaustion of the second bath. This
property is utilized by the developers of Type II, however, to pre-
vent the exhaustion of the second bath, and in this case a constant
rate of development can be maintained. With developers of this
type it is possible to use desensitizers.

Developers of Type III may be used and have the same processing
characteristics as when used in machines. The process of shifting
the first bath is relatively simple since it is merely necessary to change
the order of immersion of the racks.

The disadvantages of the two-bath method are practically the
same with rack and tank processing as they are with machine proc-
essing, namely, additional tank space is required, additional solu-
tions must be made up and handled and, in addition, any increased
amount of handling of the film on the racks subjects it to danger of
injury from careless manipulation and tends to reduce the output
of each operator.

VII. SUMMARY

(1) The photographic characteristics of two-bath developers
when used with Eastman supersensitive motion picture negative
film have been investigated. The developer solutions were placed
in two separate tanks, and the film immersed in the first bath for
a definite time and then transferred to the second bath where the
development was completed.

(2) Three types of two-bath developers were investigated, as

52 CRABTREE, PARKER, AND RUSSELL [J. S. M. P. E.

follows: Type I, in which all the developing agents were contained
in 'the first bath and all the alkali in the second bath; Type II,
class A, in which only developing agents and sulfite were contained
in the first bath, while both developing agents and alkali were in
the second bath; class B in which developing agents, sulfite, and
alkali were present in both baths but in different proportions; and
Type III, in which both baths were complete developers having
identical compositions.

(3) With developers of Type I, in which the first bath was re-
strained with sugar and sodium bisulfite, a practically constant
gamma was obtained with times of treatment varying from 3 to 5
minutes in the first bath and from 4 to 8 minutes in the second bath.
Control of gamma was obtained by varying the concentrations of
developing agents and alkali in the baths. Exhaustion tests showed
that the restrained first bath had a very long life, while the life
of the second bath was limited because "of the accumulation of
developing agents carried in by the film. With 4 minutes' treatment
in each bath, practically constant gammas were obtained throughout
the life of the baths, and the speed loss was low even after exhaustion
with 300 feet of film per gallon of developer.

(4) The developers of Type II, class A, were unsatisfactory
because of the rapid loss in emulsion speed on exhaustion.

(5) The developers of Type II, class B, in which the first bath
was compounded to act as a booster solution for the second bath,
had very satisfactory exhaustion properties with a constant rate of
development, and only a small speed loss resulted after exhaustion
with 200 feet of film per gallon of the first bath.

(6) When the film was treated even for only a small fraction
of the total development time in a fresh developer and for the re-
mainder of the development time in an exhausted developer, the
gamma and emulsion speed obtained were only slightly lower than
those for complete development in a fresh solution.

Tests have indicated that the use of two successive baths of the
same developer (Type III) has the following advantages:

(a) A more uniform degree of development can be maintained on exhaustion.

(b) The loss in emulsion speed during exhaustion is less than with a single
solution of the same volume as the combined volumes of the two-bath developer.

(7) The two-bath principle is applicable also to the development

July, 1933] TWO-BATH DEVELOPERS 53

of negative sound records, although a developer is required which
contains a lower concentration of sulfite than that for picture nega-
tives. In view of the extremely low gammas usually required, it
may be necessary to use borax in the second bath in place of carbon-
ate.

It has not been possible to compound a practical two-bath formula
which will give sufficiently high gammas for picture and sound
negatives.

REFERENCES

1 VON HUBL, A. F.: "Die Entwicklung der photographisches Bromsilver-
Gelatineplatte bei zweifelhaf t richtige Exposition," Phot. Rund., 11 (1897), p. 193.

2 VON JOANOVICH, P.: "Entwickler fur unter- oder iiberexponierten Flatten,"
Phot. Korr., 44 (1907), pp. 359 and 505.

3 KNAPP, A.: "A Developer for Underexposures," Brit. J. Almanac (1932),
p. 191. H. Greenwood & Son, Ltd., London.

4 DUNDON, M. L., BROWN, G. H., AND CAPSTAFF, J. G.: "A Quick Test for
Determining the Degree of Exhaustion of Developers," J. Sac. Mol. Pict. Eng.,
XIV (Apr., 1930), p. 389.

5 CRABTREE, J. I.: "Photographic Methods of Testing Developers," Brit. J.
Phot., 69 (1922), p. 153.

6 JONES, L. A., AND CRABTREE, J. I.: "Panchromatic Negative Film for
Motion Pictures," Trans. Soc. Mot. Pict. Eng., X (1926), No. 27, p. 131.

7 JONES, L. A.: "Photographic Sensitometry," /. Soc. Mot. Pict. Eng., Part I,
XVII (Oct., 1931), p. 491; ibid., Part II, (Nov., 1931), p. 695; ibid., Part III,
XVin (Jan., 1932), p. 54; ibid., Part IV, (Mar., 1932), p. 324.

8 TRIVELLI, A. P. H., AND JENSEN, E. C.: "Antifogging Agents in De-
velopers," J. Frank. Inst., 210 (1930), p. 287; ibid., 212 (1931), p. 155.

9 CARLTON, H. C., AND CRABTREE, J. I.: "Some Properties of Fine-Grain De-
velopers for Motion Picture Film," Trans. Soc. Mot. Pict. Eng., Xffl, (1929),
No. 38, D. 406.

10 NIETZ, A. H.: "Theory of Development," D. Van Nostrand Co., New York,
(1922).

11 MACKENZIE, D.: "Straight Line and Toe Records with the Light Valve,"
J. Soc. Mot. Pict. Eng., XVII (Aug., 1931), p. 172; also JONES, L. A., AND SANDVIK,
O.: "Photographic Characteristics of Sound Recording Film," J. Soc. Mot. Pict.
Eng., XIV, (Feb.. 1930), p. 180.

SENSITOMETRIC CONTROL IN THE PROCESSING OF
MOTION PICTURE FILM IN HOLLYWOOD*

EMERY HUSE**

Summary. It is the purpose of this paper to present a resume of the current
methods involved in the application of sensitometry to motion picture film processing
today in the major laboratories in Hollywood, Calif. During the past decade very
rapid progress has been made in the general technic of handling motion picture film
of all kinds in a variety of processes. This paper, however, discusses only that phase
of the handling that deals with the actual chemical development processes.

Ten years ago the development of motion picture film in Hollywood
was entirely a manual operation, being accomplished with the aid of
the well-known rack and tank system. During the last ten years
much work has been done, and a few papers written on the general
subject of machine development of motion picture film; with the
result that at the present time, with the exception of one or two
smaller laboratories, all film developed in Hollywood is handled by
machines. With machine development, one can feel assured that the
actual conditions under which film is developed are much more stable
than those conditions which existed under the now almost obsolete
rack and tank methods. In discussing sensitometric control in the
motion picture laboratories in Hollywood, therefore, it must be re-
membered that such laboratories as Paramount, Metro, Fox, Warn-
ers, Consolidated, Horsley, etc., use developing machines, operating
at speeds in the neighborhood of 90 feet per minute.

Prior to the actual use of the classical Hurter and Driffield method
of sensitometry in motion picture practice, certain less accurate con-
trol methods were in use. However, until recently no precise sensi-
tometric control instruments were available. All control was accom-
plished by visual judgment of the operator in actual charge of the de-
velopment processes. By the rack and tank method of development,
densities were matched and picture quality was obtained solely as a

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** West Coast Motion Picture Film Division, Eastman Kodak Co., Holly-
wood, Calif.
54

SENSITOMETRIC CONTROL 55

result of visual comparison. Naturally a procedure of control which
is dependent upon personal judgment lent for no standardization of
results. It was possible to establish a procedure, such as specifying
a certain number of shakes to the rack at certain stated intervals of
time during development but the developing solutions were not taken
care of from the replenisher standpoint with anywhere near the pre-
cision which now takes place with machine development and the
accompanying circulating and conditioning systems.

HURTER AND DRIFFIELD SENSITOMETRY

The first thing to be considered in discussing the Hurter and
Driffield system of sensitometry is the instrument with which a series

FIG. 1. A partial vertical section through the optical axis of the
Eastman Type lib sensitometer.

of exposures can be impressed upon a piece of film under known
conditions of light quality, light intensity, and exposure time. It
can readily be seen that an ordinary picture does not allow for a
complete technical analysis. It is not possible to determine the
absolute brightnesses which cause each of the different densities in
the negative ; and although the time of exposure given in a camera is
fairly well established it is still impossible to obtain a correct technical

E. HUSE

[J. S. M. P. E.

estimate of the total value of exposure as expressed by the simple
equation E = It. In a mechanical instrument designed for impressing
uniform exposures one is not confronted with a series of densities
which are distributed heterogeneously throughout the film. With
the aid of a properly designed instrument it is possible to obtain a
series of uniformly exposed areas differing as a result of a known ratio
of exposures. A strip of film containing a series of uniform areas of
density gives a means by which certain technical analyses may be
made of both the film and the developer. The first problem, therefore,
to be considered in setting up a sensitometric control is the establish-
ment of this exposure instrument, called a sensitpmeter, which must
be able to make exposures which can be definitely repeated. The
Eastman Kodak Company built and placed on the market for general
sale about two years ago an instrument which is called the Eastman
Type lib sensitometer. This instrument operates on the time scale

FIG. 2. A typical sensitometric record made with the Type lib sensitometer.

principle and makes use of precisely calibrated tungsten lamps as
the light source. This device has been adequately described by L. A.
Jones, 1 who designed and supervised the building of the instrument.

The Eastman Type lib sensitometer was designed especially to
meet the need of the modern motion picture film laboratory. It pro-
vides a precise and rapid means of making routine sensitometric tests
for the control of development processes. Fig. 1 shows a partial
vertical section through the optical axis of the sensitometer. This
instrument impresses on the film under test an accurately prede-
termined scale of exposures which may be maintained constant from
test to test over long periods of time. The exposure scale consists of
21 steps produced by exposures equal in illumination and ranging
from 1 to 1024 in relative times, each exposure being 1.414 (square
root of 2) times as long as the next shorter. This constant factorial
difference between steps permits the density readings to be spaced at

July, 1933] SENSITOMETRIC CONTROL 57

equal intervals along the log exposure axis in constructing an H&D
(density-log E) curve. Fig. 2 shows an actual sensitometric record
made with this instrument. Tables I and II herewith submitted
show the actual set-up qf the instrument for the exposure of positive
and negative films, respectively.

TABLE I

Positive Exposure

Lamp 72- watt, 6- volt locomotive headlight, calibrated for 2600 K.

Filter 78 B, correcting to 3000 K.

Intensity 27 meter-candles

t max. 4.99 seconds, at 50 cycles

log max. 2.13

TABLE II

Negative Exposure

Lamp 36-watt, 6-volt locomotive headlight, calibrated for 23(iOK.

Filter 79, correcting to 5400 K.

Intensity . 75 meter-candle

t max. 4.99 seconds, at 50 cycles

log E max. 0. 57

It is with this instrument that practically all sensitometric control
in the processing of motion picture film in Hollywood is accom-
plished. At the time of this writing there are nine of these instru-
ments in use there. It is interesting to note at this point that at the
Annual Awards Banquet of the Academy of Motion Picture Arts and
Sciences in November, 1932, this instrument was given official recogni-
tion by being awarded an honorable mention by the Committee on
Awards on Scientific and Technical Achievements.

After the development of the sensitometric strip made on the Type
lib sensitometer it is necessary, in order to attain the desired techni-
cal results, to find a means of measuring the densities of the various
deposits. A photometric instrument of one type or another is used
for this work. Such an instrument in common use in most of the
laboratories in Hollywood is the Eastman densitometer which has
been described by Capstaff and Purdy. 2 At the present time there
are approximately 25 of these instruments in use in Hollywood, both
in laboratories and in sound departments. Some of these depart-
ments make use of polarization photometers such as those made by
Schmidt and Haensch or by the Bausch.& Lomb Optical Company.

E. HUSE

[j. s. M. iv K.

The Eastman densitometer, which is shown in Fig. 3, is designed
to fulfill several conditions, namely : the ability to read densities from
0.00 to 3.00; to measure very small areas ( l /2 sq. mm.) utilizing the
same source of light for the illumination of the density to be measured
and furnishing the light for the comparison beam ; calibrated to read
direct diffused density; and designed to be portable, compact,
and inexpensive. It has been shown in actual practice that this in-
strument fulfills these requirements. The West Coast Laboratory of
the Eastman Kodak Company maintains a continual service in check-
ing the densitometers in the field for their physical condition as well

FIG. 3. Eastman densitometer.

as their photometric ability, calibrations being made against standard
densities originally calibrated in the Research Laboratories in
Rochester.

Inasmuch as the actual conditions of exposure of the sensitometric
strip are known, i. e., the time of exposure and the intensity and
quality of the exposing radiation, it is possible, with the values of
density available, to construct the characteristic density-log E cur\ c .
As the exposure increases so the density increases until upon compl
tion of the plotting of the curve a graph, such as is shown in Fig.
is obtained. There are three distinct portions to this curve, namely:

July, 1933]

SENSITOMETRIC CONTROL

the toe, which is that portion indicated between A and B; the straight
line, between B and C; and the shoulder, between C and D. It is
quite well known that these three portions of the characteristic curve
are referred to respectively as the regions of underexposure, correct
exposure, and overexposure. It is of value, therefore, in the applica-
tion of sensitometry to motion picture film processing to know the
characteristic curve resulting from the development of sensitometric
exposures in the negative, positive, and sound track developing
solutions. Quite naturally the negative developer is studied in terms

2 t

*^ J *

B/
fj

t C

.4 <

3 1

F .

3 1

6 1

9 I

2 K z

.5 i

A 3

LOG E.
FIG. 4. Typical H&D curve.

of exposures made on negative film, the positive developer i
of exposure made on positive film, etc.

in terms

GAMMA

There are in the Hurter and Driffield system of sensitometry
several constants regarding which data are desired. From the stand-
point of motion picture control, by far the most improtant character-
istic is the slope of the straight-line portion of the curve, which is
commonly referred to as "gamma." Gamma is defined as the tangent
of the angle formed by the straight-line portion of the H&D curve
and the log E axis, and is an indication of the degree of development.

E. HUSE

[J. S. M. P. E.

It is, of course, a constant of the emulsion itself; but for a given
emulsion developed in a given solution, gamma is a numerical specifi-
cation of the degree of development in that solution. It is important,
furthermore, to know that as development time increases, gamma in-
creases, meaning that the straight-line portion of the curve forms
increasingly greater angles with the log E axis as development time
progresses. This is an extremely important fact.

Probably the best way to determine for a given emulsion the reac-
tion which a developing solution will give is by a study of the rate at

FIG. 5. A series of H&D curves for identical
exposures but different times of development.

which gamma builds up with development time. Fig. 5 shows a
series of H&D curves, all of which had identical exposures in an
Eastman Type lib sensitometer, but each strip of which received a
different time of development under a constant developing condition.
After the determination of gamma for each of these curves, these
values were plotted as ordinates against the times of development as
the abscissa. A new curve is thus obtained which definitely shows the
relationship existing between gamma and development time and this
curve is referred to as the time-gamma curve. From the sensito-
metric standpoint this curve tells a great deal about the condition, or

July, 1933] SENSITOMETRIC CONTROL 61

rather the reaction, of that developer to that particular type of film.
It is now very easy to see that regardless of whether negative or posi-
tive film is to be developed, the determination of a time-gamma curve
for negative film in the negative solution and for positive film in the
positive solution is very essential. In actual practice complete series
of H&D curves for a variety of development times are not obtained
because with practice it is readily determined what the probable range
of development times will be and it is therefore only necessary to
construct time-gamma curves over the range which is within that
used practically. There are several other sensitometric constants
which, while important both from a technical and practical stand-
point, do not necessarily occupy the same important niche as that
filled by gamma. Reference is made to the constants speed, latitude,
and fog.

SPEED

Speed as a sensitometric constant has been subjected to many in-
terpretations. From the standpoint of the practical photographer,
speed and density mean much the same thing, for the reason that if
two samples of film are exposed simultaneously to the same quantity
and quality of light and then developed for the same time, that
sample showing the greatest density is considered the fastest. From
the standpoint of the Hurter and Driffield method of sensitometry,
speed has a little different meaning and it is arbitrarily defined as the
reciprocal of the inertia multiplied by a constant, i. e.,

S = 7 X k

The inertia i is defined as the exposure value at the point where on
the log E the straight-line portion of the H&D curve (Fig. 4), ex-
tended, cuts that axis. The value of the constant k should be so
chosen that it is sufficiently large so that values of speed for various
commercial emulsions are of convenient magnitude for practical use.
Speed determinations made by this method on the Eastman Type
lib sensitometer make use of the value 10 as the constant k. Speeds
as thus determined are not of any particular value to the practical
laboratory man. This method of speed determination has been dis-
cussed because it is technically important and, furthermore, it was de-
sirable to acquaint the reader with the fact that this constant exists.

62 E. HUSE [J. S. M. P. E.

LATITUDE

Latitude is a constant which has to do with the range of bright-
nesses which can be adequately rendered by a photographic emulsion.
The numercial specification of latitude is derived from the H&D
curve, such as is shown in Fig. 4. That portion of the curve with
which we are interested in the determination of latitude is the straight-
line portion. If from the limits of the straight-line perpendiculars are
dropped to the log E exposure axis, a simple determination can then
be made of the exposure value where each of these perpendiculars hits
the axis. The ratio between the two exposure values thus determined
gives a measure of the latitude. Quite a little consideration is given
to the latitude of an emulsion in the processing of sound records on
film. A thorough appreciation of the importance of latitude as it
affects negative and positive picture quality has not been attained.

FOG

Fog is an important constant in that it gives definite information
regarding the final results of the developed photographic images.
Fog may be considered as an actual density which has arisen from
sources other than intentional exposure to light. It may be consid-
ered under two general headings, inherent fog and development fog.
Inherent fog may be the result of certain of the silver grains being
made developable by the chemical processes involved during the
manufacture of the emulsion. It may also be due to slight exposure
to light during some stage of the handling, either prior or subsequent
to its final and intentional exposure.

Development fog arises from various causes such as the action of
fogging agents or reaction products in the developer, aerial oxidation,
etc. Fog is not detectable until after development. From a purely
practical standpoint no particular attention is paid to fog unless it
gets outside of accustomed grounds. For example, in the develop-
ment of positive film, a fog value of 0.03 to 0.05 is quite normal; and
unless fog builds up beyond this limit it is disregarded, other than to
record it. However, excess fog, which is readily detectable visually,
plays a detrimental part in both picture and sound quality. Pre-
caution is continually exercised to prevent fog of either the inherent or
the development type.

CONTRAST AS DISTINGUISHED FROM GAMMA

One very important item, which should be well understood in the

July, 1933] SENSITOMETRIC CONTROL 63

practical application of sensitometry to the control of the development
of motion picture negative film, is the fact that a sensitometric strip
which has been made under precisely controlled conditions of time
and intensity can only give data as to the degree of development ob-
tained. Thus, such a sensitometric strip developed with an exposed
picture negative will show a definite gamma value for the strip but
will not give any precise information as to the contrast of the picture
negative developed along with it. Contrast in the negative is not
only a function of gamma but also of the lighting balance in the set at
the time of exposure. If at all times the brightness balance would
be maintained constant in the various scenes to be photographed,
then the sensitometric strip would give a true indication of the con-
trast of those scenes after development. However, this is not pos-
sible. Each and every scene of different subjects has an inherent
brightness contrast characteristic of its own and upon controlled de-
velopment the resultant densities in each scene indicate the contrasts
which were in those scenes. The point which is being stressed is the
fact that, because picture negatives are developed in a solution which
gives a definite value of gamma from the sensitometric control strip, it
does not follow that these negatives have the same degree of contrast
as shown by the gamma value of that strip. They bear a very defi-
nite relationship to each other and under certain conditions can be
used synonymously; but this condition does not always hold. If we
refer to Fig. 4, it will be readily observed that the straight-line portion
of the H&D curve from which gamma is determined exists only be-
tween points B and C. This straight-line portion contains only part
of the various densities which go to make up the complete curve.
This being the case the densities which are beyond either extreme of
the straight line must play some part in the photographic rendering
of subject or scene. When one looks at a motion picture on the screen
and studies it for contrast, all thoughts of gamma disappear. What
the observer is looking at is the relationship existing between the
highlights and shadows. Sensitometrically this refers to the toe and
shoulder portions of the H&D curve. Therefore, it seems evident
in studying contrast sensitometrically that densities which lie on the
toe and shoulder must be considered. In evaluating contrast it is
necessary to determine the density at some fixed point on the toe and
shoulder and contrast can then be expressed in terms of a density
value. The difference between the density chosen on the toe and
that on the shoulder gives these data. It has been sensitometric

64 E. HUSE [j. s. M. P. E.

practice to consider the extremes of the curve as the density at those
points on the toe and shoulder where the slope is equal to 0.20.
From the curve as shown in Fig. 4, a 0.20 gradient on the toe and
shoulder would be slightly above point A and slightly below point D.
The word "gradient" signifies the slope of the curve at any given
point. However, along the straight-line portion of the curve the
gradient is constant and is equal to gamma.

As was previously indicated, there is one instance in practical sensi-
tometry where gamma and contrast can be used synonymously and
that is in the sensitometric control of sound track. Inasmuch as the
exposures on the track are based upon the straight-line portion of the
H&D curve, and for most recordings densities in the actual track
fall within the limits of the straight line, then in this instance where
the maximum and minimum densities recorded are completely in-
cluded in the straight line, it is quite simple to see that contrast and
gamma are identical.

Every major studio or commercial laboratory in Hollywood is
adequately equipped with, or has access to, the instruments which
have been described in the early part of this paper. Furthermore,
each laboratory has a man, or several men, taking care of the sensito-
metric routine. It can be stated rather strongly that with the advent
of sensitometric control motion picture film processing has attained
a degree of perfection which has not hitherto been possible. Of
course all of the current quality should not be laid to sensitometric
control because during the past several years much has been accom-
plished from the standpoint of improved machine development, photo-
graphic emulsions, developers, and processes which have aided in this
improvement. However, it is the candid opinion of the author that
sensitometric control has revealed deficiencies in the processing sys-
tems which have not heretofore been observed, or, if observed, were
not properly taken care of because there was no technical control
available to indicate the direction in which improvement should be
made.

In attempting to convey clearly a concise picture of the actual con-
trol methods in use it is felt that the subjects of control for negative,
positive, and sound track should be treated individually. Further-
more, that this paper might contain more than the opinion of the
author, data will be presented from actual production laboratories,
which data will show clearly the degree of consideration which is
given to sensitometric control.

July, 1933] SENSITOMETRIC CONTROL 65

Before discussing the actual laboratory data, consideration should
be given to the development of negative film. In Hollywood there
are two distinctly different methods by which negative film is de-
veloped. One method is that of a constant time of development.
The other method is that which is colloquially termed the test
method. According to the constant time method the developing
solution is maintained at a definite control gamma, as shown sensito-
metrically, and exposed negatives of all types, except very special
effect shots, are developed under this standard predetermined con-
dition. By the test method, it is necessary that the cameraman
photograph an extra portion of each scene to be used as a test for
development. These tests are developed for a constant time, which
time has been predetermined as normal for correct exposures. After
the development of these tests the responsible person in the labo-
ratory examines them and determines the time of development
which in his opinion would be best suited for each take. When one
considers a large production company with several individual com-
panies in production, the number of tests which go through by this
system is appreciable. Over a period of time it is found that many
of these scenes receive normal development. The remaining scenes
may vary from plus or minus one-half to several minutes from the
normal. In many instances, where scenes were shot under adverse
conditions, their photographic quality is materially improved by
this method of negative development. The time consumed is greater
and the work slightly more involved, but in view of the results ob-
tained the quality is very favorable.

The author after having observed over a five-year period the results
of the development of negatives by either of these two systems is con-
vinced that both have their merits and can be and are productive of
excellent results. From the standpoint of pure sensitometric control,
the developing solution itself is studied before any production work
goes through it. One of the first things done when a new developer
is put into the system is to run a sensitometric control, which consists
of the development of a group of sensitometric strips all made under
a set condition. It is determined from the results of these developed
strips just what time of development is necessary to give the desired
control gamma. Once this is established, then, regardless of which
of the two systems of negative development is to be employed, that
time becomes the standard for that machine operating at a definite
speed under a definite condition of developing solution and tempera-

66 E. HUSE [j. S. M. P. E.

ture. If the laboratory is operating purely on the time basis, then all
of the negative to be developed goes through that solution, for the
time which the tests indicate produces the desired control gamma.
The sensitometric control which is applied to this system consists of
periodic tests, such as at half -hour intervals, which give data showing
whether or not the degree of development was greater or less than
that determined by the original test. With knowledge as a result of
practice which now exists in the laboratories it is a simple procedure
to determine the rate at which a replenisher must be added to the
developing solution to maintain it at its predetermined controlled
developing power. Naturally, during the course of development by
either system the developing strength of the solution changes as in-
creased footage goes through it. It is necessary to find a means of
maintaining the fixed development condition, whether it be by alter-
ing the time of development or the chemical replenishment.

In some laboratories samples of a test negative are developed along
with their sensitometric strips. This test negative is usually a close-up
of a girl. Many laboratory men still feel that they can see more in
the picture than they can be told about that picture from the data
indicated by the sensitometric analysis. By developing both sensi-
tometric and practical tests a double check is made. It is remarkable
to observe the fine details of density and contrast which can be seen
by the experienced laboratory man in the examination of the test
picture negative. All laboratory men are becoming more thoroughly
educated in the art of sensitometry, and are becoming quite able in
determining from the sensitometric data the cause of any differences
which might occur and show themselves between two successive
tests. There are still in existence in Hollywood one or two smaller
laboratories operating by the rack and tank method. Their system
of sensitometric control is quite similar to that applied to machine
development. With very carefully laid down manipulative proce-
dures good results can be, and are being, obtained. One laboratory
in particular uses the rack and tank method for the development of
all film, including picture negative and positive, and sound track.
However, inasmuch as we are more particularly interested in the
modern methods of control, we shall not deal further with any system
except one which makes use of developing machines.

All of the foregoing, under the general heading of negative develop-
ment, has dealt with the procedure involved. Nothing has been
said quantitatively about the results obtained. During the past five

July, 1933] SENSITOMETRIC CONTROL 67

years there have been some rather definite changes in what is desired
from the standpoint of negative quality. In 1928 the author had
occasion to measure sensitometrically the control gammas of the
solutions used by the various major laboratories in Hollywood. This
was before any sensitometric methods were in use in these labora-
tories. The film which was used was panchromatic negative which
was then in style. The sensitometric exposures were made on a
time-scale sensitometer in the West Coast Laboratories of the Motion
Picture Film Division of the Eastman Kodak Company. This in-
strument was a duplicate of the time-scale sensitometer in use in the
Research Laboratories in Rochester. The gamma values resulting
from exposures made on this instrument were of the same general
order as those obtained currently with the Type lib sensitometer.
At the time of these tests, the average negative picture gamma was
very close to 0.80, although some laboratories used higher values.
A few years later, particularly after the introduction of the high-speed
supersensitive type of film, it was found that the trend in general
negative quality was toward a lower degree of contrast, which ex-
hibited itself from the sensitometric data made at the time. Measure-
ments of negative gamma from 0.55 to 0.65 were quite normal. At
the present time, 1933, the average negative gamma has increased
somewhat, and measurements show negative gammas between 0.60
and 0.75, with the average being very close to 0.67. It is not within
the scope of this paper to explain why this trend has taken place
even if it were possible to do so adequately, although it can be stated
in brief that as the details of the photographic method of recording
sound were improved, changes were necessitated in the entire proc-
essing system of both sound and picture films.

NEGATIVE CONTROL DATA

It was previously discussed in some detail that there are in existence
in Hollywood two different methods of developing negative film:
namely, the constant time method and the test system. Under this
general heading of negative control such data will be discussed as
applies to the constant time of development method. For that
purpose the writer has obtained from one of the major Hollywood
laboratories an exact copy of the sensitometric operations in the de-
velopment of their production negative during an entire night.
The procedure which will be described here is a nightly occurrence.

It is the procedure of this laboratory to begin production negative

68 E. HUSE [j. S. M. P. E.

development around 5 : 30 P.M. Early in the afternoon the circulat-
ting system of the negative developing machines is turned on. After
the developer has had time to recirculate thoroughly, which usually
takes about an hour, sensitometric control strips are developed,
consisting of exposures on negative film of the same kind and type
as used for picture work and exposed in the Type lib sensitometer
under the negative set-up conditions. These strips are put through to
determine the time of development necessary to produce the pre-
determined negative control gamma. If these first strips give values
which depart from the desired condition, another series of strips
is then developed under slightly altered conditions. Once the
desired gamma is attained the time of development which was
necessary to produce this gamma is considered the normal time of
development for that night's run. At intervals of approximately one-
half hour, together with the production work, further sensitometric
control strips are developed. This procedure is followed during the
course of the night. If at any stage during the development, gamma
increases or decreases to any marked degree, the time of develop-
ment is altered or replenisher is added to compensate for the change.

The laboratory in question submitted data obtained during the
production run on March 29, 1933. The first two tests which were
put through after the circulating system had been in operation for
a while were developed for the time which had been normal on the
previous night. In this instance, the time was 10 minutes, 37 seconds.
These first two tests went through the developing machine at 3:00
P.M. It so happened that these tests gave, upon plotting, a gamma
value of 0.62. If the gamma value obtained in such a test falls be-
tween the limits of 0.60 and 0.65, the time of development required
to do this is selected as normal. Inasmuch as these first two tests
at 3:00 P.M. gave a gamma of 0.62 and a further test at 5:26 P.M.
likewise gave the same gamma, production negative was immediately
fed into the machine, and the night's run was begun. From 5:26
P.M. until 12 :00 M. sensitometric strips were put through the system at
half-hour intervals, together with production and likewise with a
sample of test negative made under conditions as described earlier
in this paper. In Table III are given the exact data obtained sensi-
tometrically during the night, including the densities of each indi-
vidual strip together with the gamma obtained and also the amount
of footage which was developed between successive tests. These
data when plotted gave curves of the type indicated in Fig. 6. For

July, 1933]

SENSITOMETRIC CONTROL

-j05m^cotot-o<DOo-i05m^Goto>-'

Test

3:00

l-'tOtOGOt.Oia5O5'<lbo<>oi-'tOGOrfi.4^

-I bO O5 GO OOOOOO5COOOOOOi*>OOt^
OOOOOOOOOOOH-'h-i-'h-'h-'h-

Test

O5OO5tOtOOOCO<IOOOOCOO5tOtOOtf>-
OOOOOOOOOOOi-'i-'i-'h-'h-'i-'

3:00

H-i|-'tOGOGOt.CnO5V!OOCOOi-'tOtOGO*>.

5:26

1,400

6:05

3,500

i-'i-'tOtOGO*.OiO5^JOOCOOi-'tOtOGO.

6:20

^03005^^^GOGOtObOOt00^^5

OOOOOOOOOOOOn-'h-'H-'i-'H-*

4,600

Ko5to-to8oooo^05toioo

7:05

OOOOOOOOOOOi-'i-'H-'h-'t-'h-'

c H

6,700

t-'i-'tois3GO*>.aiO5<IOOCDOi-'lN3tOOOrfi.

7:35

o t^

GO O5 *"* 00 O5 M-* GO Go tO to tO CJJ C^ ^J O5 rf-^ O

S. w

"* M

OOOOOOOOOOOOii-'i-'h-'i-'

9,850

i-'i-'tOtOGO^CnOSMOOCOtOOi-'tOGOGO

8:20

^ 00 O O5 * to tO tO tO tO I 1 CO 00 O5 ^- O GO

OOOOOOOOOOOOt-'h-'i-'i-'i-'

12,300

BS8Skb^838iS98SKSS8

8:50

15,450

OOOOOOOOOOOOi-'i-'i-'i-'i-'

9:35

kb8"8S^8SiS83SB88

17,550

OOOOOOOOOOOOn-'h-'i-'HJh-'

10:05

h-'i-'tOtOGO^CnOS^IOOCDCDOH-'tOGOGO

OlOOH-0005^t000^0000505^000

20,700

OOOOOOOOOOOOv-'t-'h-'i-'i-'

10:50

i-'i-'tOtOGJ*CnO5<IOOCDcDOH-'tOGOGO

050oo-a^bototooooo50oo5tooco

OOOOOOOOOOOOi-'i-'i-'h-'h-.

23,150

l _i ( _i H -'tOGO t fi.CnO5-<I<IOCOOi-'tOGOGO

11:25

^05C005GoEo^tOOcD005^^tOOC!5

OOOOOOOOOOOOi-'t-'t-'i-'^-'

24,900

E S 8 S S fe S 8 S 2 g g S ^ 8 8

12:00

E. HUSE

[J. S. M. P. E.

the sake of brevity, the two curves shown represent the strips which
went through at 5 : 26 P.M. and 12 :00 M. These were the first and last
strips developed. It will be observed that there is a difference of
0.02 in gamma between these two tests, the 12:00 M. curve showing
the lower value. From the standpoint of density if one individual
step of the H&D strip is chosen, for example, step 11, it will be
observed that there is a maximum density change of 0.08. During
that night approximately 30,000 feet of negative were put through
the solution. These data are for one of the two negative developing
machines which are normally in operation each night at this labora-
tory.

From the standpoint of replenisher, an average of eight gallons per

i. Site p.m.
B. 12:00 Bit.

LOO XXFOSURI

5.57 1.17 1.77 1.37 1.9? O.W

FIG. 6. Characteristics of negative control strips showing the varia-
tion of gamma and the degrees of density with operating time.

hour of double-strength developer, minus bromide, was fed into the
system. If, at any stage during the development, gamma or density
had dropped appreciably, one of two things would have happened,
either the time of development would have been increased, or the rate
of replenishment increased. Inasmuch as the maximum density
change amounted to only 10 per cent, which is equivalent in speed to
approximately one Bell & Howell printer point, and furthermore,
inasmuch as gamma had changed less than 5 per cent, the same
time of development and the same rate of replenishment was main-
tained throughout the night.

This sample of data from the laboratory in question represents an
average condition. These data were not selected to represent either

July, 1933] SENSITOMETRIC CONTROL 71

a good or bad night's work. The slight development differences
which existed between negatives developed during this night required
very little unnecessary manipulation in timing or printing. Natu-
rally, all scenes of all cameramen did not print alike, but all negative
was well within the normal printing range.

Again it must be borne in mind that by this system of development
the solution was maintained constant within very narrow limits to
produce the same degree of development. The contrasts that were
exhibited by the various negatives were the result of the various
brightness contrasts in the scenes photographed. Inasmuch as it is
the practice of this laboratory to develop the bulk of its work for a
fixed time of development, the cameramen realize that any change
they make in their exposure conditions will be evidenced in their
negatives.

By the test system of development it is necessary first to construct
a short time-gamma curve, similar to* the one shown in Fig. 5. From
that curve it is possible to determine the time required to give the
normal gamma. In this system the time is varied, dependent upon
the judgment of the man in charge of the negative development; and
because each laboratory accurately records all sensitometric data it
is quite possible to determine just what control gamma is obtained
at any time of development other than normal. This is done, of
course, by referring to the time-gamma curve previously established.
The system of control of the solution over a period of time, when the
test system is used, is done in a manner identical to that which is
described above for the constant time of development method.

The developers in general use for the development of picture nega-
tives are modifications of the standard E. K. D-76 borax formula.
They are modified to fit the needs of the various types of developing
machines. Quite naturally, all laboratories do not use the same
formula, even though there is a similarity in the machines used.
Differences of opinion as to photographic quality and differences in
the recirculation and agitation of solutions are factors which enter
into the question of developer formula differences. In the final
analysis, there is not a great deal of difference between the various
negative formulas in use in the various laboratories in Hollywood.
It is desirable, however, to include in this paper a typical machine
negative developer formula. Developers of this general composi-
tion are in use today and produce excellent results. The formula
quoted in Table IV shows the chemical composition of such a developer.

E. HUSE

[J. S. M. P. E.

TABLE IV

Picture Negative Formula

Elon

Sodium sulfite, desiccated

Hydroquinone

Borax

Water to make

Avoirdupois

1 lb. 15 ozs.
96 Ib.

4 lb. 13 ozs.
1 lb. 12 ozs.
120 gals.

Metric
1 . 93 grams
96.0 grams
4 . 82 grams
1 . 75 grams
1.0 liter

POSITIVE CONTROL DATA

From the standpoint of positive control the procedure followed is
very similar to that described for negative film. The final positive
print contains both picture and sound track records. It is important

POSITIVE CONTROL STRIP

Dt : 3/26/33
TIM : 2:06 ..
UaohllM: No. 4
SpM4 : 100 ft. />ln

1.13

T.73

FIG. 7. A typical H&D curve having a gamma of approximately

2.00.

from the standpoint of the sound that positive gammas be specified
and maintained. For that purpose rigid control is applied to positive
film development so that the predetermined positive gamma may be
maintained consistent throughout. This is particularly true in
release print development.

From another major studio, laboratory control data have been
obtained for a period of one day. This laboratory submitted all of
their data on positive control for March 26, 1933. This consisted of
each individual sensitometric curve for every machine in operation
during that day, at intervals of about one hour. During the day's
work there were five developing machines in operation, each machine

July, 1933]

SENSITOMETRIC CONTROL

having in its system the same chemical formula. At this laboratory
a positive control gamma of 2.00 is desired. By preliminary sensi-
tometric tests, which consist of the development of Type lib positive
sensitometer exposures on the positive film emulsion which is in use,
gamma determinations are made before any production work goes
through. With these data it is possible to determine the time of
development to give this desired gamma value. This time is con-
trolled by the machine speed. In Fig. 7 is shown a typical positive
H&D curve giving an approximate gamma of 2.00, as obtained at
this laboratory in one of the machines on the day in question. From

Hourly Control Teats For Positive Film ^26/33

Nof

No. 5

No. 6

No. 7

No. 8

2.1

*__.

2.0

ifoflfD

2.1

- -S

tooott/

> t0
1.9

loppea

2.1

l.d

----

2.1
7 2.0
1.9

;--

...,.

- r ~-

-0--0--

.-.-.

....

^-S-

B -.-

2.1

a 2 -

1.9

o o

* o H

AM J

3 1

i t

f <

FIG. 8. Curve showing the variation of gamma from hour to
hour on several machines.

this graph the general density range covered by the sensitometric
strip, as well as the general shape of the curve, is readily seen. Once
the time is determined to give this desired gamma, production starts
and hourly thereafter a pair of sensitometric exposures are sent
through. Each strip is then read for density and plotted. Upon
the completion of the work, a final curve is obtained showing gamma
plotted against the time interval of development. Naturally there
are some tests which show values greater or less than the desired value.
Certain tolerances are, of course, allowed, usually a plus or minus
0.05 deviation from the normal. In the data which are presented in

E. HUSE

[J. S. M. P. E.

Fig. 8 the actual variations are readily observable for each machine.
The work during this day was not representative either of a good or a
bad day, but did show a perfectly average set of conditions. For any
given machine, for example, machine No. 8, there is only a 0.06
maximum variation between the highest and lowest gamma strips.
If all of these values of gamma were averaged, a numerical gamma of
1.995 would be obtained. The maximum deviation, considering all
machines, was only 0.07.

In the event that any general trend is shown by successive tests,
then slight alterations are made in the actual time of development to
compensate for that trend. It is the function of the foreman in
charge of the positive developing machines to maintain the desired
gamma of 2.00. From the data presented from this laboratory it
can be seen that the results are excellent.

Although the purpose of this paper is to describe the sensitometric
routine in vogue and to show the results obtained, it is only fair to
state that the results shown from this laboratory on their positive
film control is indicative of the type of work obtained by all of the
laboratories in Hollywood.

As was the case with negative developer formulas, differences
exist also among the positive formulas. To complete this section
on positive film control the tabulation of an average positive formula
is considered necessary. Table V gives such a formula.

TABLE v

Picture Positive Formula

Elon

Sodium sulfite, desiccated

Hydroquinone

Sodium carbonate, desiccated

Potassium metabisulfite

Potassium bromide

Potassium iodide

Water to make

Avoirdupois

12 ozs.

40 Ib. 8 ozs.
4 Ib. 14 ozs.
26 Ib.

1 Ib. 7 ozs.
10 ozs.
178 grains
120 gals.

Metric
. 75 gram
0.5

44.87

26.0
1.43

0.63

grams
grams
grams
grams
gram

0.025 gram
1 . liter

SOUND CONTROL DATA

A complete discussion of the sensitometric control for sound
film development would be extremely voluminous. In the discussion
of this subject in this paper no attempt will be made to give actual
data from any individual studio. The control methods are similar

July, 1933] SENSITOMETRIC CONTROL 75

to those described for positive and negative film, but are much more
detailed in many respects.

There are at the present time two major sound recording methods
in vogue, the RCA and the Western Electric systems. A brief dis-
cussion is necessary for each of these two methods. They differ
appreciably and should be discussed separately. The RCA system
makes use of a variable width sound track, while the track of the
Western Electric system is of the variable density type.

VARIABLE WIDTH SYSTEM

The particulars regarding this system of recording sound are very
well known and this paper will give only the necessary details to de-
scribe the sensitometric requirements which are submitted to their
licensees by the RCA Victor Co. They are quite simple, and are
readily maintained in the processing laboratory.

The specification laid down by RCA Victor engineers for negative
gamma is that the negative sound track should be developed to give
a control gamma of from 2.00 to 2.20. The unmodulated track
density, and in this instance reference is made to the symmetrical
track, should fall within the density range of 1.40 to 1.50. The sensi-
tometric set-up necessary to follow these specifications requires that
the laboratory determine the time of development on the film being
used for the recording to give a gamma within the limits just pre-
scribed. It is then necessary for the recording unit to provide an
exposure which will give a density upon development within the den-
sity limits cited above. The RCA Victor Company recommends for
positive control gammas between 2.00 and 2.20, similar to the nega-
tive, while the positive track densities should be from 0.15 to 0.20
less than the negative density. For example, if the negative un-
modulated track density at the proper gamma is 1 .45, then the posi-
tive track density should be within the limits of 1.25 to 1.30. Na-
turally, during the course of processing, solution control must be
maintained, and this is accomplished in a manner very similar to
that previously described under positive film. Once the sensito-
metric conditions are determined, both from the exposure and de-
velopment standpoints, then it is the function of the laboratory to
maintain those conditions in their developing procedure. The
developer normally used for this type of work is the regular picture
positive formula, an example of which was cited under the heading
of positive film.

76 E. HUSE [j. s. M. P. E.

VARIABLE DENSITY TRACK

In the consideration of the Western Electric system of recording,
a much more detailed sensitometric discussion is necessary. In
processing variable density sound records which utilize the straight-
line portion of the H&D curve, it is necessary that the over-all gamma
characteristic, as determined by plotting projection densities versus
the logarithm of light valve openings, be held to the ideal value of
unity. This might also be expressed by saying that the product of
the positive and negative gammas multiplied by the projection factor
should be 1.00.

Before specifying numerical values for positive and negative
gamma, several factors entering into this computation should be
explained. The gamma value obtained from a series of exposures on
a light valve recorder and plotted against the logarithm of the light
valve openings will be designated as the light valve gamma, LVy.
The gamma valve obtained on the control strip, which has been ex-
posed on the Type lib sensitometer and developed with the light
valve gamma strip, will be called the negative control gamma and
designated as NCy.

In the measurement of positive gamma it is necessary to measure
the apparent printer gamma, which is obtained by printing the
negative control strip on a printer and developing this print with the
positive sound track print. This value will be called APy. When
the positive control strip, which has been exposed on the Type
lib sensitometer is plotted, its gamma will be designated as PCy.

Furthermore, as the positive sound track will be scanned in
projection by a photoelectric cell, it is necessary to determine experi-
mentally the difference in gamma as determined by visual measure-
ment of the diffuse densities and the quasi-specular measurements of
the photo-cell. This factor has been measured for standard pro-
jection conditions and found to be 1.30.

The conditions for correct reproduction, as recommended by
representatives of Electrical Research Products, Inc., are given in
the following data:

over- all gamma = LVy X APy X projection factor =1 (1)

If LVy = a X NCy and (2)

APy = b X PCy, then (3)

over-all y = NCy X PCy X a X b X projection factor (4)

July, 1933 J SENSITOMETRIC CONTROL 77

It is customary to assume that the factor a = 1 ; i. e ., the difference
between the negative control gamma and the light valve gamma is
negligible. In practice, the light valve gamma is found to vary by
plus or minus 5 per cent from this. The printer factor, b, is usually
measured daily in most laboratories; at least such a procedure is
recommended. While the printer factor may be as much as 10 per
cent, it is found in practice that this value is approximately equal to
the factor a, and opposite in direction, so that the two tend to cancel
each other.

Omitting these factors, then, from equation (4) we have

over-all gamma = NCy X PCy X projection factor (5)
and substituting

1 = NCy X PCy X 1.30, or (6)

NCy X PCy = 0.76 (7)

In other words, any combination of negative and positive control
gammas which gives a product approximating 0.76 would be cor-
rect for straight-line recording.

Tests have been made in four Hollywood studios, listed below by
letter, of the Type lib sensitometer negative and positive gammas.
In Table VI these data are given.

TABLE VI

Studio Negative Gamma Positive Gamma Visual Product

A 0.42 2.10 0.88

B 0.35 2.15 0.75

C 0.36 2.00 0.72

D 0.38 2.40 0.91

It will be observed that the maximum variation from 0.76 is
+0.15 higher. One studio at the time these tests were made
operated slightly lower than the desired 0.76, their visual product
being 0.72. A mean exposure of the light valve that will permit 90
per cent modulation without going into the toe of the negative H&D
curve is recommended. This results in an average density numeri-
cally equal to the negative gamma plus the toe density. By toe den-
sity is meant that value of density at which the toe departs from the
straight-line characteristic of the sound negative H&D curve.

78 E. HUSE [J. s. M. P. E.

Thus if the Type lib negative gamma equals 0.40 and the toe density
equals 0.15, then the correct operating density would be 0.55.

In the case of positive density a value that will not permit much
operation into the toe region is recommended. This depends upon
the shape of the toe of the printer H&D curve. Experience has
shown that a visual print density in the neighborhood of 0.70, for a
gamma approximating 2.00, is usually satisfactory. Some studios
that fail to obtain a correct over-all gamma of unity have recourse to
lighter prints, varying in density from 0.50 to 0.60. Based on ob-
servations, the following specifications cover most processing of light
valve records.

Negative Gamma . 35 to . 40

Negative Density . 50 to . 60

Positive Gamma 1 . 80 to 2 . 20

Positive Density . 65 to . 75

Before any recommendations for processing are made to any studio
by the Western Electric Co., or its subsidiary, Electrical Research
Products, Inc., the entire sensitometric control set-up, from light
valve to photoelectric cell, is examined. All the above recom-
mendations regarding Western Electric track control are quoted
from data obtained from representatives of Electrical Research
Products, Inc., in Hollywood.

In considering developers for sound track work, it is necessary to
realize that recordings are made on film which contains an emulsion
of positive characteristics. As can be seen from the specifications
quoted above, the gammas desired are very low. It becomes neces-
sary, therefore, that a developer of low contrast characteristic be
used. Quite often, use is made of the picture negative formula.
However, slightly better results are obtained with a developer quite
similar to the picture negative formula but with a smaller quantity
of sodium sulfite. A typical formula used for the development
of sound negative of the variable density type is given in Table VII.

TABLE VII
Sound Negative Developer

Avoirdupois Metric

Elon 1 lb. 1.0 gram

Sodium sulfite, desiccated 46 lb. 8 ozs. 46 . 5 grams

Hydroquinone 2 lb. 3 ozs. 2. 2 grams

Borax 1 lb. 1.0 gram

Water to make 120 gals. 1.0 liter

July, 1933] SENSITOMETRIC CONTROL 79

It is quite obvious that the formula used for developing the posi-
tive sound track is identical with that for the positive picture, inas-
much as the final print contains both the picture and the sound track.

In closing the author would like to express his appreciation to the
individual representatives of practically every studio and laboratory
in Hollywood, as well as representatives of the major sound units,
for the assistance which they rendered in the compilation of the data
presented in this paper.

REFERENCES

1 JONES, L. A.: "A Motion Picture Laboratory Sensitometer," J. Soc. Mot.
Pict. Eng., XVH (Oct., 1931), No. 4, p. 536.

2 CAPSTAFF, J. G., AND PURDY, R. A.: "A Compact Motion Picture Densi-
tometer," Trans. Soc. Mot. Pict. Eng., XI (Sept., 1927), p. 607.

DISCUSSION

MR. L. A. JONES: The author has stated in this paper that the character-
istic curve completely specifies the contrast of the material. To avoid misunder-
standing on the part of the reader it might be mentioned that the word "contrast"
as used by most of us can refer to any of several quite different things. For in-
stance, when we speak of the "contrast of a negative" we refer to the extreme
density difference or to the highlight-shadow brightness ratio. Obviously the
characteristic curve of the negative material does not specify "contrast" in this
sense.

MR. J. COFFMAN: It would be interesting for Mr. Huse to explain exactly how
he reconciles the control of processes that are essentially intensity-scale processes,
such as printing and negative developing, with the use of a time-scale instrument
such as the lib sensitometer.

MR. HUSE: All our sensitometric tests are a means of solution control. The
sensitometer does not give us a picture of the entire photographic process, but of
solution control, primarily.

MR. COFFMAN: I should like my own attitude to be understood. The lib
sensitometer is well designed, if we accept its compromise principle. I also
agree that it is possible to establish an arbitrary control in which data obtained
with a time-scale instrument such as this are related to film printing character-
istics. But it would seem desirable, before proceeding too far in standardization,
to determine whether it is not possible to design an instrument that will be more
closely related to the processes that we wish to control. I do not want to put
obstacles in the way of standardizing, because standardization of sensitometric
technic and terminology is very desirable. But I do feel that it is too early to take
steps to standardize time-scale instruments. All tests that we have made have
tended to show that reciprocity and various other factors enter into the situation;
and while approximate control can be achieved with a time-scale sensitometer,
I do not believe that absolute control can be achieved with it. And while there
is no perfectly satisfactory intensity-scale sensitometer, yet I believe Mr. Huse
and the members of the Kodak Research Laboratory, if they will concentrate on

80 E. HUSE [J. S. M. P. E.

the idea, can probably turn out an intensity-scale instrument that will be more
satisfactory than the lib sensitometer.

MR. JONES: Mr. Coffman's statement is one with which we all must sympa-
thize from the purely academic point of view. The intensity-scale sensitometer
is ideal for use with an intensity-scale process. We have such an instrument
in the laboratory, and have described it to the Society. It is very expensive, and
is difficult to reproduce from specifications.

In our design of the lib sensitometer it was our purpose to lay down a structure
that could be accurately reproduced in any machine shop; an instrument that
would combine maximum precision with minimum cost. We think this instru-
ment is entirely adequate for the purpose for which it was built namely, the
control of processing. Now, in the control of variable density sound processes
there are factors built into the instrument that do take care of reciprocity failure
in a practical manner.

If sensitometry is to be applied to the theoretical study of a tone reproduction
problem, it is desirable to have an intensity-scale instrument. Moreover, this
instrument should reproduce the time and intensity and quality of radiation used
in the process.

Unfortunately, when one considers the multiplicity of practical problems
the picture problem, the sound recording problem, the printing problem, each
with its many variations in different organizations and even within the same
organization he is faced with the necessity for a multiplicity of sensitometers.
I do not see any way of building a single instrument of the intensity-scale type
that could meet all the requirements.

Therefore, it seems to me we should add confusion by attempting to meet those
conditions. Instead of one instrument at least four or five would be required,
and the result would be chaos.

The time-scale instrument results can for practical purposes be applied to other
problems by using arbitrary conversion factors. For the present I think this
type of instrument offers the best chance at standardization of sensitpmetric
procedure.

MR. T. E. SHEA: Is the work of the Sub-Committee on Sensitometry leading
toward standardization of any particular type of instrument for practical work?

MR. JONES: I do not feel that I can speak with the authority of the Com-
mittee on that point because we have not perhaps had sufficient discussion of the
subject to reach any conclusion. The proposals toward standardization in sensi-
tometry that can be made at the present time are rather limited. I think it would
be a mistake to set up too hastily and too arbitrarily sensitometric technics that
we all recognize are perhaps forcedly artificial. I do not believe it is the intent of
the Committee to propose that any particular instrument be adopted as standard.

However, there are certain rather basic things that can be established. For
instance, I think the Society should establish at once, or recognize at once, the
international unit of photographic intensity.

MR. COFFMAN: All I am asking for is a sense of values as related to the
industry as a whole. Everybody in the motion picture industry must pay a
tremendous amount of credit to the sound technicians who first forced sensi-
tometry on the industry. At the same time, sensitometry, as now practiced by
the industry, is not altogether a problem of sound and its control. As a matter

July, 1933] SENSITOMETRIC CONTROL 81

of fact, I believe that we realize today that we need sensitometric control for
pictorial quality far more than we need it for sound quality. And while the time-
scale sensitometer seems for most practical purposes well adapted to the control
of variable density sound negative processing, let us not forget that by far the
greater amount of footage used by the industry is represented by prints; and that
for every negative, either sound or picture, that is turned out, we are turning out
at least a hundred prints.

For the purposes of print control, all studies that we have made seem to indicate
that there are several variable factors that tend to upset systems of control based
upon time-scale sensitometric indications. And I believe that we should study
the matter quite carefully before we let the practice of the industry become crys-
tallized.

This is an industry that tends to follow the leader; sometimes even rushing
violently ahead of the leader. I do not believe that you would go as far as some
of your technicians have in approving this lib sensitometer for general use.
But if we simply continue to use the lib, trying other methods, and finally
determine what is the best system of control for each kind of process, then I
believe we shall be rendering a real service to the industry.

MR. JONES: I disagree somewhat with Mr. Coffman's apparent feeling that
the lib sensitometer can not be used satisfactorily for the control of pictorial
quality. I think it can. Any such instrument properly calibrated, when proper
conversion factors have been determined, can be applied to the control of pictorial
quality as well as it has been used for the control of sound quality. It is the old
story of learning the possibilities and limitations of a tool.

MR. R. F. MITCHELL: In checking some of the sensitometric controls in
connection with our new printer, we ran into a rather interesting situation.
Reference was made in Mr. Huse's paper to the girl test favored by many tech-
nicians. We, investigated that rather carefully and found a considerable lack of
agreement between different machines and between different laboratories in what
they got from the same girl test. We checked back, and instead of using the girl-
test negative, we used a clear film as a negative and made exposures at different
printer steps, say, every alternate step. A sensitometric exposure was made on a
piece of the same film and the two films developed together. The sensitometer
strip gave the gamma and that part of the H&D curve for which the sensitometer
was set. By taking the densities of the printer test and coordinating with the
H&D curve, we obtained a graphical representation of the relative exposures at
different printer steps. As far as the printer is concerned, specifically the Model
D, we found that the placement of the printer light in the printer itself affected
the difference in density obtained between different steps of the printer. There
was considerable variation between any two steps on different printers, which
seemed to be dependent on the placement of the light on the printer itself. That
means of checking back on the printers, and on the testing machines used for
making girl tests, provides a very desirable check for that type of control.

MR. COFFMAN: I believe that we have a sufficient volume of data available
to state positively that if we use a time-scale instrument we must have con-
version factors for every type of printer that we use. Mr. Mitchell has doubtless
found that he will get different results with the new printer from those obtained
with the old one.

82 E. HUSE

MR. MITCHELL: That is partly due to that factor that I mentioned. It is
also due to the fact that in the Model D we have more or less of an arithmetical
progression of the printer steps, whereas the new printer is designed to give a
geometric exposure progression.

MR. JONES: Mr. Coffman said we must have a number of conversion factors
if we use a time-scale instrument for various types of photographic procedure. I
personally feel that it is better to have a series of conversion factors with one
standardized instrument, than to have several different intensity-scale sensitome-
ters to represent each one of these procedures. I frankly think that the setting
up of intensity-scale sensitometers to match all these conditions is going to be very
much more confusing and much less economical than sticking to one sensitometer
with a series of conversion factors.

MR. COFFMAN: Not every laboratory has all these various kinds of work to do.
If we think of an experimental plant in which work of all kinds is done, that is one
situation. If we think of plants in which the work is strictly specialized, then
specialized instruments for those plants are obviously correct.

MR. JONES: I can not help feeling that the attempt to build sensitometers to
duplicate all the various practical conditions would mean that in the industry we
would have a very much larger number of sensitometers. Perhaps instead of nine
or ten in Hollywood, as we have now, we would have forty or fifty of these, of
several varieties. And I really believe that that would introduce more confusion
than the course we are now following.

SOCIETY OF MOTION PICTURE
ENGINEERS

OFFICERS
1933

President
A. N. GOLDSMITH, 444 Madison Ave., New York, N. Y.

Past-President
J. I. CRABTREE, Eastman Kodak Company, Rochester. N. Y.

Vice-Presidents

E. I. SPONABLE, Fox Film Corp., New York. N. Y.
W. C. KUNZMANN, National Carbon Co., Cleveland, Ohio.

Secretary
J. H. KURLANDER, Westinghouse Lamp Co., Bloomfield, N. J.

Treasurer
H. T. COWLING, 7510 N. Ashland Ave., Chicago, 111.

Board of Governors

H. T. COWLING, Burton Holmes Films, 7510 N. Ashland Ave., Chicago, 111.

J. I. CRABTREE, Eastman Kodak Co., Rochester, N. Y.

P. H. EVANS, Warner Bros. Pictures, Inc., 1277 E. 14th St., Brooklyn, N. Y.

R. E. FARNHAM, General Electric Co., Nela Park, Cleveland, Ohio.

O. M. GLUNT, Bell Telephone Laboratories, 463 West St., New York, N. Y.

A. N. GOLDSMITH, 444 Madison Ave., New York, N. Y.

H. GRIFFIN, International Projector Corp., 96 Gold St., New York, N. Y.

W. C. HUBBARD, General Electric Vapor Lamp Co., Hoboken, N. J.

E. HUSE, Eastman Kodak Co., 6706 Santa Monica Ave., Hollywood, Calif.

W. C. KUNZMANN, National Carbon Co., Cleveland, Ohio.

J. H. KURLANDER, Westinghouse Lamp Co., Bloomfield, N. J.

R. F. MITCHELL, Bell & Howell Co., 1801 Larchmont Ave., Chicago, 111.

E. I. SPONABLE, Fox Film Corp.. 850 Tenth Ave., New York, N. Y.

SOCIETY ANNOUNCEMENTS

PROJECTION PRACTICE COMMITTEE

At a meeting of the Projection Practice Committee, held on June 15 at the
Paramount Building, New York, N. Y., the final form of the targets, used in the
test reel developed by the Committee and presented at the convention last April,
was agreed upon. New targets were presented for detecting the presence of
travel-ghost and for testing for aberration of projection lenses.

Among the items included in the agenda of the Committee for the coming
season are the problems of monitoring the reproduction of sound in theaters, of
the increasing importance of maintaining the highest quality of projection, and
of exercising the greatest care in handling, shipping, storing, and cleaning the
"wide range" and "high fidelity" films now being used. Defects in these proc-
esses are glaringly evident with recordings of the newer types, and, if allowed to
persist, will nullify the benefits expected of the increased range of frequency of
the recordings.

STANDARDS COMMITTEE

At a meeting of the Standards Committee, held on June 2 at the General
Office of the Society, work was continued on the revision of the standards booklet,
which it is expected will be published in the Fall. The work has been completed
with the exception of bringing up to date the table of sprocket dimensions.

The Committee also investigated the question of standardizing the dimensions
of motion picture reels, and of adopting the recommendations of the International
Congress of Photography as regards the unit of photographic intensity for nega-
tive materials and the use of a non-intermittent exposure in making sensitometric
measurements. The following motion was passed:

"Resolved that the unit of photographic intensity adopted by the In-
ternational Congress of Photography for negative materials, and the
principle of non-intermittency in making sensitometric measurements,
be adopted as recommended practice."

As regards the standardization of the perforations used in 35-mm. film, the
following resolution was passed after considerable discussion:

"Resolved that a single perforation be adopted for all 35-mm. film,
and that this perforation be the present standard positive perforation,
to be known hereafter as the standard S. M. P. E. perforation."

BOARD OF GOVERNORS

The next meeting of the Board of Governors will be held on July 14 at New
York, N. Y. Consideration will be given to the financial problems of the Society,
the question of reducing the annual dues, and various administrative matters.
Officers for the year beginning October, 1933, will also be nominated.
84

SOCIETY ANNOUNCEMENTS 85

PACIFIC COAST SECTION

At a meeting of the Pacific Coast Section, held in the auditorium of the Bell
& Howell Co., Hollywood, Calif., on June 15, the following papers were presented:

"A New Development in Arcs for General Set Lighting," by Mr. E. C.
Richardson.

"The Lumenarc A Mazda Lamp-Gaseous Tube Unit for Daylight
Quality," by Mr. R. M. Maxwell.

"A New Development in Incandescent Lamps for Motion Picture Light-
ing," by Mr. R. E. Farnham.

SUSTAINING MEMBERS

Bausch & Lomb Optical Co.
Bell Telephone Laboratories
Burnett-Timken Laboratories

Eastman Kodak Co.

Electrical Research Products, Inc.

National Carbon Co.

RCA Victor Co., Inc.

HONOR ROLL
OF THE

SOCIETY OF MOTION PICTURE ENGINEERS

By action of the Board of Governors, October 4, 1931, this Honor Roll was estab-
lished for the purpose of perpetuating the names of distinguished pioneers who are
now deceased:

LOUIS AlME AUGUSTIN L,E PRINCE

WILLIAM FRIESE-GREENE

THOMAS ALVA EDISON

GEORGE EASTMAN
JEAN ACME LE ROY

PAMPHLETS, BOOKLETS, AND CATALOGUES

Manufacturers of motion picture equipment and supplies are requested to send
to the General Office of the Society copies of their descriptive pamphlets, book-
lets, and catalogues as issued. Notices of the issuance of this material will be
published in the JOURNAL, advising the readers that the material may be obtained
free of charge by addressing the manufacturers named. This editorial service
has been established in order to acquaint readers of the JOURNAL with the com-
mercial developments of the motion picture industry as quickly as they occur.

FILM RECORDING EQUIPMENT

Type 3S Com-
plete Record-
ing Equipment.

Single or Double System, Variable

Density or Variable Area, Studio

or Portable.

Write or cable for literature

JENKINS &ADAIR, INC.

3333 Belmont Ave.
Chicago, U. S. A.

Cable Address: JENKADAIR

ADVERTISE IN THE JOURNAL

OF THE

SOCIETY OF MOTION PICTURE ENGINEERS

Rates may be obtained by writing to the
General Office of the Society at

33 WEST 42nd ST., NEW YORK, N. Y.

EMPLOYMENT ADVERTISEMENTS

In order to assist members of the Society who desire to obtain positions, the
Board of Governors has authorized the establishment in the JOURNAL of an
"Employment Page." Advertisements will be available to members of the
S. M. P. E. who desire positions and to manufacturing concerns seeking trained
men. Material for publication is subject to editing, and must be sent to the
General Office of the Society by the 10th of the month prior to publication.

Each employment advertisement shall not exceed one-sixth page in length;
a charge of $2.00 will be made for a single insertion, or $5.00 for three con-
secutive insertions.

JOURNAL

OF THE SOCIETY OF

MOTION PICTURE ENGINEERS

Volume XXI AUGUST, 1933 Number 2

CONTENTS

Page

Report of the Projection Practice Committee 89

An Experimental Apparatus for the Projection of Motion

Pictures in Relief H. E. IVES 106

A New Alternating-Current Projection Arc

D. B. JOY AND A. C. DOWNES 1 16

Professional Motion Picture Photography with High-Intensity

Short-Life Incandescent Lamps

M. W. PALMER AND E. W. BEGGS 126

An Improved Potassium Alum Fixing Bath Containing Boric

Acid H. D. RUSSELL AND J. I. CRABTREE 137

A Practical Method and Photometer for Controlling Exposures

in Photography M. LASKY AND B. RUBIN 154

Sound Recording and Reproducing Using 16-Mm. Film

C. N. BATSEL AND J. O. BAKER 161

The Use of Mazda Lamps for Color Photography

R. E. Farnham 166

Officers 172

Committees 1^3

Society Announcements *"

JOURNAL

OF THE SOCIETY OF

MOTION PICTURE ENGINEERS

SYLVAN HARRIS, EDITOR

Board of Editors
J. I. CRABTREE, Chairman

O. M. GLUNT A. C. HARDY F. F. RENWICK

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers.

Publication Office, 20th & Northampton Sts., Easton, Pa.
General and Editorial Office, 33 West 42nd St., New York, N. Y.

Copyrighted, 1933, by the Society of Motion Picture Engineers, Inc.

Entered as second class matter January 15, 1930, at the Post Office at Easton,
Pa., under the Act of March 3, 1879.

REPORT OF THE PROJECTION PRACTICE COMMITTEE*

Projection provides the industry's closest contact with the public,
whose continued patronage is dependent largely upon the quality of
the projected picture and the reproduced sound the finished prod-
uct which embodies the work of all other branches of the industry.
Acceptance of the foregoing fact leads naturally to a consideration of
the means available for maintaining at all times a high standard of
quality. The Committee feels that every facility that aids, even re-
motely, in maintaining a high standard of projection should willingly
be provided.

TEST REEL

A serious deficiency in the projection field heretofore has been the
lack of an efficient test medium that would enable the speedy and con-
venient detection and correction of various defects common to both
visual and sound apparatus. To meet this requirement, the Com-
mittee, in collaboration with the RCA Victor Company, Inc., has de-
vised a test reel that serves two distinct purposes in that it provides
an accurate means of checking both the visual and the sound equip-
ment. This film is suitable for use on all makes of equipment ar-
ranged for projecting films that conform to the specifications of the
Standard Release Print.

The test reel is about 1000 feet long, of which about 500 feet are
devoted to various targets (test objects) to be used for detecting opti-
cal defects, the remaining 500 feet providing various means of testing
sound quality. The latter section has sound tracks on both margins
of the film, thus providing an effective test footage of approximately
1000 feet.

Optical Test Section

The "optical" section of the test film will be considered first. It
contains five test targets with appropriate descriptive legends in the
following order:

(1) Travel-ghost.

* Presented at the Spring, 1933, Meeting at New York, N. Y.

PROJECTION PRACTICE COMMITTEE [j. s. M. P. E.

(2) Picture-jump.

(3) Vertical lines for testing marginal and radial aberration of the projection
lens.

(4) Horizontal lines for testing the marginal and radial aberration of projec-
tion lens.

(5) Small squares for checking focus.

(1) Travel-Ghost Target (Fig. 1). Unless the shutter is properly ad-
justed and timed, "travel-ghost" will be evidenced by the blurring of
the bright portions of the screen into the dark portions, with a result-
ing loss of detail in the vertical direction. The target consists of

FIG. 1. Travel-ghost target.

white geometric forms on a black field. Travel-ghost may be con-
sidered to have been eliminated when the outlines of these white
forms are clearly and sharply defined on the screen against the black
field.

(2) Picture- Jump Target (Fig. 2). This target consists of two
rows of white rectangles placed corner to corner along the diagonals
of the screen image, against a black field. The amount of picture-
jump can be measured with a ruler held against the screen at the top
or bottom of any of these rectangles. By holding the ruler to the
vertical side of any of these rectangles, the amount of side-motion, or
weave, may be determined.

Aug., 1933] PROJECTION PRACTICE COMMITTEE 91

(3) Vertical Line Target (Fig. 3) . This target consists of a series of
white vertical lines against a black field. Projection of this image in
sharp focus over the entire area of the screen by a lens in one position, in
this and in the following test, stamps the lens as one that is commonly
referred to as having a "flat field." If, on the other hand, marginal
aberration is present, it will be indicated by a blurring of the lines at
the sides of the screen. Radial aberration will be indicated by a
blurring of the lines at the top and bottom of the screen.

(4) Horizontal Line Target (Fig. 4) . This target serves the same
purpose as the preceding target, except that it is partly intended to
compensate for visual deficiency (astigmatism) of the observer.

FIG. 2. Picture-jump target.

(5) Small Square Target for Checking Focus (Fig. 5). This target
consists of numerous white squares, arranged and numbered in verti-
cal and horizontal rows, against a black field. The position of the
lens for which the greatest possible number of squares are projected
in sharp focus is the most desirable lens position. The numbering of
the squares provides a means for making comparative tests of lenses.

Sound Test Section

The remaining portion of the test film, some 500 feet long, is re-
corded on both edges, as previously stated. On one side of the film
are recorded the following tests with suitable accompanying an-
nouncements :

(1) Buzz track for checking the position of the scanning light relative to the
sound track.

92 PROJECTION PRACTICE COMMITTEE [j. s. M. P. E.

(2) 6000-cycle and 9000-cycle constant frequency tracks for checking the
focus of the sound optical system.

(3) Selected frequencies for ascertaining the over-all characteristics, as fol-
lows: 50, 100, 200, 300, 500, 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000,
and 10,000.

(1) Buzz Track. This consists of a 300-cycle and a 1100-cycle fre-
quency recording, respectively, just outside the boundaries of the
standard sound track area, one on each side. When the 1100-cycle

FIG. 3. Vertical line target for testing aberration of projection lens.

note (the higher pitched note) is heard, it indicates that the film is
passing the light slit too closely to the sprocket hole margin. When
the 300-cycle note (the lower pitched note) is heard, it indicates that
the film is passing the light slit too closely to the picture margin.
When both notes have been eliminated by properly adjusting the
lateral guide rollers, or by adjusting the optical system, when such
adjustment is provided, correct film travel path may be assumed.
When one or both notes are intermittently heard, this indicates film
weave. If the weaving be due to warping of the film, its effect may
in some cases be reduced by adjusting the sound gate tension springs.

Aug., 1933] PROJECTION PRACTICE COMMITTEE 93

Incorrect relative alignment of the projector head and the sound head
also may cause weaving.

(2) 6000- Cycle and 9000 -Cycle Constant Frequency Tracks
When the maximum volume of sound is obtained for each of these fre-
quencies, it may be assumed that the optical system is correctly po-
sitioned for obtaining the best results; that is, that the scanning light
is sharply focused and perpendicular to the direction of travel of the

FIG. 4. Horizontal line target for testing aberration of projection lens.

film. The 6000-cycle note is used for making an approximate ad-
justment, and the 9000-cycle note, if audible, for still finer adjust-
ment. This positioning should not be undertaken unless the optical
system has been specifically designed to be adjusted and suitable
adjusting tools are available.

(3) Selected Frequencies. This track is so recorded that no voltage
calibration (on a volume indicator) is required. Assuming a perfect
scanning slit and a "flat" amplifier, the resulting over-all characteris-
tics, as determined by a volume indicator, would be flat. The ear will
naturally be more responsive to the 1000-cycle note, whereas the
higher and lower frequency notes will sound less loud. Of course, a

PROJECTION PRACTICE COMMITTEE [j. s. M. P. E.

volume indicator would admit of making precise measurements of the
sound level; this applies equally well to sections 2 and 3 above.

On the other edge of the sound test film are recordings of voice,
piano, and orchestral music. The vocal portion is to be used for test-
ing intelligibility of speech and theater reverberation; the piano re-
cording for detecting flutter ("wows"); and the orchestral recording
for noting the naturalness of reproduction, which is determined by
the range of frequencies reproduced by the equipment. This record-
ing contains notes ranging from the lowest notes of the tuba and
double bass to the very high overtones of the string and brass in-

FIG. 5. Small square target for checking focus.

struments. The metallic quality of special instruments, such as the
wire brushes, should be particularly noticeable.

OPTICAL ALIGNMENT TOOL

In conjunction with the test reel, the Committee recommends a tool
to be used in aligning the arc, condenser, projector aperture, and pro-
jection lens on the optical axis. The model, designed by the Com-
mittee, and referred to in the following description, conforms to the
standard 13.6-mm. carbons used in high-intensity lamps. To obtain
the maximum illumination and most uniform distribution of light on
the screen, it is of prime importance that the arc and all components
of the optical system be accurately centered. Fig. 6 shows the several
parts of this tool :

A is a disk having a hole through the center, which is placed in the

Aug., 1933]

PROJECTION PRACTICE COMMITTEE

condenser mount instead of the condenser. B is a cylinder having an
axial hole through the center, which is clamped into the projection
lens holder. Sis a short pointed bar, which is inserted through cylin-
der B and extended through to the aperture of the projector. The
bar should be in the center of the aperture. An aperture plug having
a centered circular hole with clearance for the bar can be provided as
an additional convenience.
L is a pointed bar, 36 inches long and approximately x /2 inch in

FIG. 6. An optical alignment tool in use.

diameter (the approximate diameter of the 13.6-mm. positive carbon),
which is inserted through the back of the lamp house, through the
positive carbon clamp and through the disk A, which occupies the
position of the condensers. This bar, which replaces the positive
carbon, should be aligned along its entire length with the center of the
condenser mount. The bar is then extended to the aperture where,
by manipulating the arc housing on the base of the projector, the
points of the two bars, L and S, may be aligned, as at RI.

Subsequently to this alignment, a confirmatory test is made by

96 PROJECTION PRACTICE COMMITTEE [j. s. M. P. E.

drawing the long bar back toward the disk A and extending the short
bar through the aperture so as to touch the point of the long bar as
indicated at R 2 . Further manipulation of the arc housing may be
required before the bars are exactly aligned. After this operation, it
should be possible to withdraw bar S, the smaller one, and extend bar
L through cylinder B without difficulty. (This device is a modifica-
tion of a similar piece of equipment used on a limited scale some years
ago. The Committee strongly recommends a much wider use of this
tool.)

POSITIVE PRINT DENSITY AND STUDIO PROJECTION SCREEN ILLUMINATION

It has been definitely established that the intensity of illumination
of the screens used in most studio projection rooms is greatly in excess
of the intensity that can be obtained in the theaters. The great dis-
parity that was found to exist was pointed out in a report of the
Theater Lighting Committee, published in the February, 1931, issue
of the JOURNAL. Complaints regarding dense prints still persist,
however, indicating the need of additional emphasis in the matter.

The Projection Practice Committee has conducted a partial survey
of typical theaters for the purpose of determining the values of screen
illumination obtaining in practice, and it is significant that its find-
ings, in a widely separated territory and after a lapse of two years,
check closely with the findings of the Theater Lighting Committee
referred to above.

The results of these two independent surveys indicate that the
average intensity of the projected light at the surface of theater
screens lies between 8 and 12 foot-candles, and that the average co-
efficient of reflection is about forty per cent, corresponding to intensi-
ties between 3.2 and 4.8 foot-candles, these figures being based on the
use of diffusive screens.

In the tests made by this Committee, each projector was equipped
with a two-blade, 90-degree shutter which caused a reduction of
approximately 50 per cent of the screen illumination as compared
with measurements made when the projector was at rest. In each
test, the auditorium illumination approximated that obtaining during
the presentation of pictures.

Invariably much smaller pictures are projected in the studio projec-
tion rooms than in the average theater, resulting in excessive screen
illumination. In addition, reflective screens are sometimes used in
studio rooms, adding further to the brightness of the projected pic-

Aug., 1933} PROJECTION PRACTICE COMMITTEE 97

ture. In the great majority of theaters, however, not only are much
larger pictures projected than are shown in the studio rooms, but, in
addition, diffusive screens are used. It is obvious that under such
circumstances it is extremely difficult to reconcile studio "screen
values" with theater "screen values," unless some compensating ad-
justment is made in the studio room. It seems highly desirable that
no reflective screens be used in studio rooms in which print density is
to be judged.

On the basis of these facts, therefore, and in view of the easy and
inexpensive manner in which the requisite change can be made, the
Committee suggests that in each studio projection room where the
screen illumination is excessive, each projector be provided with a
diaphragm or iris hi front of the lens to reduce the flux of projected
light while measurements of screen illumination are made. The
iris can be so adjusted that the screen reading approximates the
average value obtaining in the theaters, after which a diaphragm
mask having a fixed aperture of the proper size can be substituted
at will for the iris.

CHANGE-OVER MARKS AND THEIR LOCATION

It is obviously not good practice to attract the attention of the
audience to the marks used for start-motor and change-over cues.
Attention will, however, be drawn to the marks if important action
occurs near them.

Neglect of a corresponding precaution has resulted in giving undue
prominence to change-over marks in certain recent feature films.
Greater care in arranging the location of the indicating marks would
prevent such occurrences. The placing of indicating marks should be
in accord with desirable practice, as follows :

(1) No motion of objects directly toward or away from the mark is desirable,
and

(2) The marks should not be placed over moving objects or near significant

action.

The first requisite is that the indicating marks shall be properly
positioned; and next, that they shall be visible to the projectionist.
Unless both these requisites are emphasized, the crude punching and
marking of film by individual projectionists will undoubtedly con-
tinue, if not actually increase, and cause a pronounced loss of enter-
tainment value and waste of film stock.

The Committee proposes that the studios experiment with marks

98 PROJECTION PRACTICE COMMITTEE [J. S. M. P. E.

of different shape, in order to distinguish between start-motor and
change-over cues; e. g., a diamond for a motor-starting mark and a
circle for a change-over mark. This is an additional safeguard against
faulty change-overs.

In view of the fact that prints having marks that are not sufficiently
visible from the projection room are still being released, the Commit-
tee feels justified in referring again to the suggestion made in its pre-
ceding report; namely, to surround each black indicating mark with a
thin white circle that will be visible against a dark background. Two
of the major film producers have acknowledged the value of this sug-
gestion and have adopted it in practice.

HARRY RUBIN, Chairman
J. O. BAKER H. GRIFFIN

T. C. BARROWS J. J. HOPKINS

G. C. EDWARDS W. C. KUNZMANN

J. J. FINN R. H. McCuLLOucn

C. FLANNAGAN P. A. McGuiRE

S. GLAUBER R. MIEHLING

C. GREEN F. H. RICHARDSON

V. A. WELMAN

DISCUSSION

DR. GOLDSMITH: The company that has made the negative of the test reel,
the RCA Victor Company, has stated to the Society its willingness to make prints
and to furnish them to the Society or others upon the request of the Society.
That is, the Society can buy the prints and ship them, or the Society can act to
instruct the maker of the prints to ship them.

The Board of Governors has had data on that subject laid before it, but has
not yet fixed on a definite price for the prints, because there are shipping costs
and handling costs which tfie Society must consider. It is likely that as soon
as, and if, this reel is accepted and suitable prints become available, the Society
will undertake to announce to the theaters that this reel is available; and that
theaters, exchanges, laboratories, or others who wish such a reel for test purposes,
can address the headquarters of the Society. By sending a suitable remittance,
for the reel and for the shipping and handling charges, such a reel will be then sent
to that theater exchange, laboratory, or other organization. In other words, the
Society would become the clearing house through which the test reels will reach
the theaters, and the Society will instruct the shipper to whom to send them.

That procedure may not be the one that may be adopted. It is too early to
make a definite statement and it must be understood that I am speaking unof-
ficially. The Board of Governors has not as yet ruled on the matter.

MR. RICHARDSON: It seems to me that the Society itself might perhaps offer
to furnish these reels at a fair rental and to recommend that all exchanges carry
at least one copy of the film for rental to their customers.

Aug., 1933] PROJECTION PRACTICE COMMITTEE 99

DR. GOLDSMITH : The superior facilities of an exchange for arranging the rental
and for collecting the rental fees, would make the exchange a more logical agency
for distribution. We would have some hesitancy in saddling the office of the
Society with the task of arranging for the rental of these films, perhaps to hundreds
of theaters per month, and then arranging for the collection of rentals, the ship-
ping of the film, and the like. Our natural inclination would be to let an ex-
change, that was interested and that believed that the Society's recommendation
of this test reel was valid, buy such a film from the Society and rent it until the
reel no longer was usable, and then buy another one. In other words, the ex-
change would handle these films in much the same way that it would handle any
other theater film.

MR. RICHARDSON: In the legend accompanying the travel-ghost target, the
projectionist should be warned that he can not detect a faint travel-ghost from the
projection room, especially when the projection distance is long; not even with
an opera glass. It is necessary to approach within 25 feet of the screen, at least,
in order to be sure that faint travel-ghost is not present.

MR. TUTTLE: I do not believe that the definition test object is sufficiently
critical. Some sort of pie chart, or a series of pie charts, might be more critical.
When making the steadiness test, one notices not only the steadiness of projec-
tion, but the steadiness of the print as well. We assume that the print will be
accurately made, but I wonder whether an accurately punched film, the perfora-
tion of which would show on the screen, would not be a better kind of test object
to use. Also, I believe that the aberration test object would be somewhat better
if the horizontal and vertical lines were closer together.

MR. KURLANDER: Like Mr. Tuttle, I feel that some of the test objects used
in this film are too elementary to do what they are designed to do. Certainly
any method, to be an improvement over the means already at the disposal of the
projectionists, must be more precise than those described.

With respect to the first test object, as Mr. Richardson pointed out, he recently
had to go within 25 feet of the screen in order to detect travel-ghost. That wasn't
the projectionist's fault; the travel-ghost was present, but he couldn't see it.
It seems to me that an adjustment when once made, should be permanent, and
should be made carefully and very critically. If we give the projectionist the
proper kind of target, he will be able to make those critical adjustments pre-
cisely. Instead of the travel-ghost target used in this report, I recommend that
the travel-ghost figure be made up of alternate white and black horizontal bars of
equal width. A slight amount of travel-ghost would cause the white bars to in-
crease in width at the expense of the black bars. This would result in an altera-
tion of the entire pattern that would be easily detected.

As travel-ghost increased, the white bars would become broader and the black
bars narrower, until finally the black bars would disappear entirely and the pat-
tern would consist of alternate white and gray bars.

It is extremely difficult to detect an increase in the width of a white line; but
it is comparatively easy, with a regular pattern, to detect a change over the whole
pattern and I would suggest that horizontal bars of equal width be used. The
best width would have to be determined experimentally.

A test for lens definition was described some time ago in the Transactions which
consisted of a fine engraving of various characters in the form of an over-all pat-

100 PROJECTION PRACTICE COMMITTEE [j. s. M. P. E.

tern. Such an engraving would seem to be necessary in order to determine
whether or not the lens is a good one. The proposed bar pattern seems too crude
for the purpose.

As for the optical alignment tool, it is not new. It was used in the days of
Mazda lamp projection, and the greatest drawback to its extended use was its
cumbersomeness. In the first place, it is a little awkward to make. The average
projectionist probably would not go to the trouble of making such a tool, as it is
a machine job. I would suggest some optical method instead of the alignment
tool.

As regards the use of an iris diaphragm for framing, I would suggest that the
statement be made in the report that the iris diaphragm should be placed over
the mouth of the projection lens. It can be used also to diaphragm the condenser ;
or, in a similar manner, to diaphragm the mirror of the reflector arc. The most
logical way, however, is to diaphragm the projection lens.

With respect to screen intensity, I am not so sure that high intensities should
not be used in the screening rooms of studios. Screening is an inspection process,
and the observers ought to be very critical; certainly they should inspect films
under a much higher intensity than that to be used in the theaters. It seems to
me that the point to be stressed is the standardization of print density.

Now, with respect to the change-over marks, we have all the elements in a
modern projector for changing over automatically. There are present a photo-
electric cell, ample light, and electricity. Perhaps some photoelectric cell method
for automatic change-over can be devised that will eliminate the change-over
marks; or, at least, incorporate them in the film hi some way, so that the projec-
tionists won't have to watch for them.

The density of the film should be such as to suit the requirements of the average
theater, according to the illumination provided. We now receive many prints
that are satisfactory in the big theaters where the intensity of the light source is
sufficient; but they are far from being satisfactory hi the smaller houses of the
country. In order to determine the proper density to which the films should be
printed, they should be viewed in the screening room at about the same brightness
they will have in the theaters.

MR. FINN: The pattern of the travel-ghost target was designed with the defi-
nite idea in mind that it must be seen from the projection room. It would be
very difficult to see from the projection room a pattern like the one Mr. Kurlander
suggested. If travel-ghost is eliminated to the extent possible with this target,
there will be an enormous improvement in projection.

As for utilizing the photoelectric cell to actuate an automatic change-over, it
is true that we have available every necessary component. But it occurs to me
that whatever will actuate the photoelectric cell will have to be on the film itself.
Although an automatic change-over device is very desirable, I do not believe
that one rational idea has been proposed so far that would utilize a photoelectric
cell for a change-over.

As for the optical alignment tool, the Committee emphasized the fact that it
is not a new device; but that like so many other things that are proposed, it
finds its way into the library, and that is the end of it. Although this tool was
used in the days of the tungsten lamp, it is not being used now. True, the tool

Aug., 1933 j PROJECTION PRACTICE COMMITTEE 101

is a machine job; if it is to be used in the projection room, it will have to be
bought.

MR. EDWARDS : Anybody who has operated a projector for any length of time
has found out very early that it was impossible to eliminate travel-ghost while
looking at a picture. The old way of doing it was the way it is done today -to
wait for the title because the title contains horizontal lines, vertical lines, and
points, by means of which one can detect the travel-ghost. Unfortunately, in
sound pictures there is not very much in the way of titles.

Alternate bars of black and white will, it is true, show the travel-ghost fringe;
but that fringing will be three times as hard to detect on a bar pattern as on a
point pattern, because on the point pattern one sees the stream of light coming
right up to the points.

As for the change-over marks, I think every projectionist has had this experience
when starting a projector, that just at the time when his eyes should be glued to
the screen, something happens that diverts his attention for an instant, with the
result that the mark slips past. The other mark comes along a little later, and he
starts his machine at the change-over mark. I think the idea of having the start-
machine mark in the shape of a diamond, and a change-over mark in the shape of
a circle, will prevent many cases of that kind. The projectionist would know at
once what the mark was for when he saw it.

As for the automatic change-over, such schemes would be workable if every
theater always ran first-run film. The number of theaters that run first-run film
are a small minority. But what will happen in the theater that receives 30- or 60-
or 90-day films, and sometimes 200-day films, in which case one is particularly
lucky if he has the last scene in at all. How will the actuating device for the
photoelectric cell work then?

So many different kinds of marks appear on film that it would be almost im-
possible to place a mark that would not be duplicated several times throughout
the run of the film by accidental marks made by faulty equipment.

MR. KURLANDER : The value of the travel-ghost pattern rests on the fact that
when no travel-ghost is present, a pattern is seen ; and if travel-ghost is present
the pattern is thereby changed. It is true that the projection distance will deter-
mine what the width of those lines should be. That is a point the Committee
should investigate. The pattern, however, was advanced with the idea of detect-
ing a very small change or increase in travel-ghost.

With respect to photoelectric cells, Mr. O. H. Caldwell, in a paper presented on
the first day of this convention, certainly gave enough illustrations of the work
that can be done by small photoelectric cells.

MR. FINN: I am very familiar with Mr. Caldwell's paper. I still maintain,
however, that the fact that a photoelectric cell, or a light-sensitive cell, will operate
a relay and cause a garage door to open doesn't mean that it is suitable for change-
over purposes. A photoelectric cell must have some actuating means; and I in-
sist that the film travel controls the change-over process. A mark is required on
the film itself. What else is going to control the cell's operation?

MR. SCHLANGER : Some of the same films as, or films similar to, those usec
testing projection equipment, could be advantageously used by the exhibi
test the particular view of the screen that each seat in the theater may affo

A simple device could be used in conjunction with the test films. It would cot

102 PROJECTION PRACTICE COMMITTEE [J. S. M. P. E.

sist of a piece of glass or other transparent material, small enough to hold in the
hand before the eyes. On it would be printed a pattern, similar to the pattern
used in the test films, consisting of a series of horizontal lines in one case, and ver-
tical lines in another. By looking through this glass frame at the test film pro-
jected on the screen, the amount of distortion of the screen image from any seat
in a theater could be measured.

Such a testing method is intended chiefly for use in existing theaters having
seating arrangements in need of correction to obtain a proper view of the screen.
If such a device accompanied the test films distributed to exhibitors, and should
the exhibitors themselves witness the distortion of the screen image as seen from
the poor seats, they might show more interest in improving the sight lines of
their theaters.

MEMBER: The accuracy or value of the optical alignment tool depends en-
tirely upon the straightness of the rods. I wonder whether they might not become
bent during use without being noticed ; and whether a certain degree of springing
might not occur so that, when the test is made, it might be thought that the sys-
tem is faulty, when it isn't.

I would suggest the use of tubes instead of rods, with the idea that one might
sight through the tubes and correct the alignment that way. If the tubes were
bent, it would be noticeable. And I believe also that by that method it should be
possible to make the alignment with only one setting instead of two, as are neces-
sary in the case of a rod. A light would be placed at the end of the tube in the
projection head and one would sight through the tube in the lamp house.

MR. RICHARDSON: Suppose the tube is placed in the front end of the projec-
tion lens, and the lens itself is a bit out of alignment. Then the whole system
would have to be out of alignment to get the result sought.

MEMBER: Wouldn't that objection apply in the case of the rod?

MR. RICHARDSON: No, because the rod would extend clear through.

MEMBER: It would be necessary to place some object at the end of the front
tube.

MR. RICHARDSON: It seems to me that the rod is the logical thing, but I think
your objection is well put as to the possibility of bending.

MR. RUBIN: This tool is made of steel that can't be bent without great diffi-
culty.

MR. KROESEN: Many tests should be made before the rod is finally adopted,
because there are too many variances throughout the projector mechanism to war-
rant our saying it is accurate without testing it sufficiently. Many manufactur-
ing tolerances should be taken into consideration.

MR. JONES: The manufacturers of projector lenses certainly have during the
past years given the subject of testing those lenses a great deal of careful study.
I am quite sure that they have worked out tests that have proved satisfactory for
detecting the various kinds of aberration ; and I am wondering whether they have
been consulted with regard to the various test objects that are adapted to show
with the maximum of magnification these various defects that we wish to locate.
If not, I certainly think that before we decide to adopt any particular test film,
the matter should be discussed with them.

MR. RUBIN: The purpose of the Committee has been to design or suggest, a

Aug., 1933] PROJECTION PRACTICE COMMITTEE 103

set of tools for the projectionist that would be simple and easy to use, without re-
quiring the assistance of specially trained engineers. Up to the present time, the
projectionist has not been provided with any tools with which to test his equip-
ment. He has not been encouraged by any society, to provide himself with such
tools. The Projection Practice Committee has attempted to do so.

The tools the test film and the alignment tool have been so designed as to
require no special training on the part of those who use them, to require a mini-
mum of time in which to conduct the tests, and to require no additional equipment
whatever. Perhaps these tools can be improved; I am sure that the optical com-
panies have better testing equipment, but it is probably more elaborate.

For many years projectionists used titles for testing for travel-ghost. Since
the inception of sound and the absence of titles, he has had to substitute something
for the title, and has found that any vertical sharp point against a black back-
ground, or a white spot against a black background, is the best test for travel-
ghost.

In the present design the sharp points against the black background are easily
visible, and provide the best means for accurately correcting the travel-ghost.
Both fine lines and heavy lines are provided, and you will notice that the sharp
lines show the travel-ghost much better than the other lines; and furthermore,
until such time as means are developed that will make it unnecessary for the
projectionist to use spy glasses or to go within 20 feet of the screen, we shall have
to do the best we can with the system that we have.

MR. KURLANDER: The reason, I think, why titles were so effective in showing
up travel-ghost was simply that the titles approximate horizontal lines the height
of which was small in comparison with the extent of the travel-ghost. Not much
travel-ghost is needed to show in a title. But in the proposed test target, the
figure is large, and one can not see much travel-ghost in a figure' of that size.

With respect to the alignment tool, it is very difficult to induce projectionists
to use tools of any kind in aligning their equipment. A similar tool was tried
years ago and a few conscientious projectionists used it. But the larger number of
projectionists didn't care and they did not use it; that is why the tool went out
of use. I doubt whether such a tool will find extensive use.

MEMBER : Has any one looked into the possibility of designing some device for
adjusting the optical system?

MR. RUBIN: This Committee has been endeavoring to obtain the cooperation
of the manufacturers in designing such tools, but up to the present we have had
no such cooperation. The Committee would be very happy to have the coopera-
tion of every manufacturer. With the help of the RCA Victor Co. we were able
to prepare the test reel that you saw today. We have in mind a number of other
tools and have invited many manufacturers to cooperate in the work. I am sure
from this discussion here today that something will be done.

MEMBER: Why should the device for testing equipment exist separately from
the equipment itself? There is a possibility that the operator will pay no atten-
tion to a separate piece of equipment, whereas, if the equipment is made part of the
main apparatus, he may be induced to look at it occasionally.

MR. RUBIN: The lens manufacturer may have a little gadget for his lens, but
what about the mechanism, and the sound head, and the condensers, and t
lamp, and other elements that go with it?

104 PROJECTION PRACTICE COMMITTEE [J. S. M. P. E.

MEMBER: Are there no graduations on those things to show when they are
correctly aligned? Couldn't they be extended?

MR. RUBIN: Probably they could. But this Committee is working with
material already in the theaters ; we must obtain the best results with the equip-
ment that is in the field.

DR. GOLDSMITH: The point that has just been brought out is well taken.
There are two ways of determining the momentary condition of a piece of equip-
ment. One is by means of some indicator, which is an integral part of the device
itself, as, for example, a voltmeter or an ammeter or some mechanical indicator
which shows the instantaneous condition. The second method is by means of a
separate adjunct that can be brought by a service man to the location of the de-
vice.

The difficulty is that we face a condition where something over 10,000 theaters
in the United States have actual projection equipment in their projection rooms
and, economic conditions being what they are, it will be some time before most of
these projectors are replaced. When they are replaced, certainly the projector
manufacturers would be well advised to consider the inclusion ot self-contained
indicators wherever practicable.

Certainly the rod-type centering mechanism that has been described could
hardly be included in the projector, because it would occupy space to the ex-
clusion of the carbons, the film, and the lens. It is, therefore, a device that ap-
parently could not be permanently in place. But certain other things might well
be included in the projector, such as mechanical gauges or sighting devices. Even
if nothing else is accomplished by these suggestions except that we direct the at-
tention of the projector manufacturers to these points, the discussion will have
been of value.

Facing the conditions that it did, the best that the Projection Practice Com-
mittee could do was to devise external and separate indicators and devices
that could be carried into the projection room and used for tests, leaving until some
later date recommendations for the production of new projectors that would en-
able these tests to be carried out without any further test equipment.

MR. GOLDEN: With regard to the alignment tool, I believe that it should be
known that the entire Loew's circuit is now using such tools, which have been
found to be very effective and helpful in aligning equipment.

As to the possible chances of getting the projectionist to use the tool, after all,
the projectionist of today can not be compared with the projectionist of the early
days. With the advent of sound, he was forced to use tools that he had never
dreamed of. So, therefore, since he has been able to master and handle sound,
why can't he, if shown that this tool is practicable, be able to help put on a better
show. The projectionist, once shown that the tool will help him in his work, will
use it.

MR. BLIVEN: The entire procedure seems to me to represent good practice in
projection work ; but the only objection that I have to make is the possibility of
fatiguing the eye in checking the definitions. Why not use a very light straw
color instead of the pure whites?

DR. GOLDSMITH : It was suggested that a gray glass be slipped before the lens
but the difficulty is that as soon as the sharp edge contrast is lost, the sensitiveness
of the test goes down again.

Aug., 1933] PROJECTION PRACTICE COMMITTEE 105

MR. BLIVEN : May I suggest that the field of test film be limited, to cover spe-
cific points at a time. Probably a thousand-foot film would not be long enough for
that, or perhaps the cost of the films would be prohibitive.

DR. GOLDSMITH : The Committee had in mind using a vignette with one of the
test subjects so that only a portion of the field would be shown, the rest of the field
being dark; and letting the person watching the vignette travel around. But
difficult production problems would be involved, and more elaborate research
than was thought justified.

MEMBER : In the early days when the machine had to be taken apart and oiled
and cleaned, the obvious way for the operator to adjust the shutters was to place
a small flash-lamp in front of the projection lamp and project the beam on the
film; and then to set the shutters so that it covered the lens as the pull-down
started. In all the years that I used that scheme I never had to use a test object
on the screen to see whether I had travel-ghost. Travel-ghost is caused by the
shutters being out of time, and the lens must be closed while the film is being
pulled down. It is not difficult at all to set the shutter in the projection booth
without using any test object whatever on the screen. I think it is going to be
impracticable to use a long reel. If the tests were made on four-foot loops, each
one stored in a can by itself, the required test could be selected whenever desired.
There would be a number of such loops, one for definition, one for aberration, and
another for travel-ghost, or whatever the defects happen to be; and one would
select the particular four-foot loop needed, and run it for whatever time necessary
to adjust the projector.

AN EXPERIMENTAL APPARATUS FOR THE PROJECTION
OF MOTION PICTURES IN RELIEF*

HERBERT E. IVES**

Summary. An experimental demonstration apparatus is described for pro-
jecting motion pictures in relief by application of the principle of the parallax panor-
amagram. A series of thirty-two posed, still pictures is made by the use of a large
diameter concave mirror forming an image on a transparent concave ridged screen,
which in turn is imaged on lantern-slide plates. Positives from these negatives are
mounted on a slowly rotating disk in the slide plane of a projection lantern. A
flashing mercury lamp illuminates each picture as it comes into position. The
projected image is received upon the back of a translucent convex ridged screen.
When viewed from the front, the moving picture changes its appearance with the ob-
serving position, and exhibits stereoscopic relief.

While methods of projecting motion pictures in relief, depending
upon the use of apparatus at the eyes for separating the appropriate
views, have been successfully demonstrated, methods and means
by which motion pictures might be seen in relief without individual
viewing devices have not heretofore been developed, even in an
experimental form. Popular interest in the possibility has indeed
brought forth many attempts, but these have quite generally proved
to be scientifically unsound, and, in the vain hope of making the
apparatus simple and practicable, have failed to face squarely the
essential conditions.

Ideal requirements for a scientific solution of the problem (not
necessarily to be identified with commercial practicability) may be
listed as follows:

(1) No individual viewing apparatus for spectators.

(2) A single photographic exposure for each of the successive pictures out of
which the motion picture is built up.

(3) A single projection device.

Certain other goals, while undoubtedly desirable, may be dismissed
as being incompatible with the nature of the problem, such as the use

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Bell Telephone Laboratories, New York, N. Y.

106

APPARATUS FOR PICTURES IN RELIEF

107

108 HERBERT E. IVES [J. S. M. P. E.

of an "ordinary" motion picture camera and the use of an "ordinary"
projection screen.

These requirements may be met by applying the scheme of the
parallax panoramagram. As described in earlier papers, 1 multiple
strip pictures, produced by a single exposure with a large diameter

FIG. 2. Photograph of the taking apparatus from the rear.

concave mirror, in conjunction with a concave ridged screen, can
be projected upon a translucent convex ridged screen, and yield a
relief image. For this image to exhibit motion, it is necessary to
project a series of such pictures of a moving object taken in rapid
succession. It is this last step that has been achieved in the ap-
paratus here described.

Aug., 1933 j

APPARATUS FOR PICTURES IN RELIEF

109

The key problems to be faced in the transition from a single
projected relief image to a rapidly moving sequence are those of
accuracy of registration. Each multiple strip picture must be pro-
jected in turn upon the ridged screen with no relative distortion or
lateral shift with respect to the preceding and following ones, 'since
any varying distortion of position, even if exceedingly small, will
result in wavering or jumping of the image. In order to meet this

FIG. 4. One of the thirty-two parallax panoramagram pictures used
to build up the motion picture in relief.

requirement, since it was considered doubtful that pictures on
celluloid film would be likely to be sufficiently rigid or could easily
be guided with the requisite accuracy, the use of glass plates was
resorted to, and the series of pictures was mounted on a rotating
disk. This is, in effect, a return to the earliest form of motion
pictures, the magic disk of Plateau, one form of which, the zootrope,
was a common laboratory toy before the motion picture emerged
in its present form.

With this statement of the over-all form of the projection apparatus,
we may proceed to a description of the details of the actual system.

110

HERBERT E. IVES

[J. S. M. P. E.

The Camera. Fig. 1 shows diagrammatically the single exposure
camera used for making the individual "frames." At O is the
object; at C the 4-foot diameter (spherical) concave mirror, of
which a horizontal strip one inch wide is used. For simplicity of
fabrication, this mirror actually consists of sections of three mirrors,
carefully adjusted with their surfaces on the same arc. At Mi is a
half -silvered mirror set at 45 to the path of the light; at 5 a glass
plate on which are cut 180 concave grooves of semicircular cross-

FIG. 5. View of the projection apparatus from the rear, showing the
method of carrying the thirty-two parallax panoramagrams on a rotating
disk.

section; at M z is a front surface silvered mirror at 45, at L is a
high quality lens, and at P the sensitive plate. The distances and
dimensions are so chosen that the image at P falls conveniently inside
the dimensions of a plate the size of a lantern -slide.

In Fig. 2 is shown a photograph of the apparatus, and in Fig. 3
a sketch of the room built around it, constituting the "camera."

The essential features of this set-up have been previously described,
so that little need be added here. It will be recalled that the choice
of concave grooves on the screen S is occasioned by the decision to
use a convex rod transmission screen for projection ; an opaque line
grating or convex grooves could be substituted at 5 if projection

Aug., 1933] APPARATUS FOR PICTURES IN RELIEF 111

screens of other types were chosen. The single exposures made by
this apparatus can be instantaneous if sufficient light, such as pro-
vided by several flash-lamps, is available. In the actual experiment,
a group of incandescent lamps was used for illumination, calling for
exposures of several minutes with commercially available plates.
Because of the length of exposure required, the object was a
dummy head, mounted so as to be movable between exposures into
a series of positions, by two circular racks and pinions. Thirty-two
pictures were made, showing a complete cycle of motion of the head
from right to left and up and down. A print of one of the negatives
is shown in Fig. 4.

The Disk. Fig. 5 shows the projection apparatus from the rear
with details of the 30-inch diameter disk on which the 32 pictures
are mounted. Each picture is carried in a frame provided with
adjusting screws, permitting it to be accurately centered by ob-
serving the position of fixed points on the slide and on the projected
image. The whole disk is mounted on an accurate ball bearing,
true to 1 /2o,ooo inch. It may be emphasized that the success of the
experiment hinges on the accuracy that has been achieved in the modern
ball bearing. Since the panoramic strips are approximately Veo inch
wide, the positioning of the successive pictures is thus correct to
about 1 part in 300. The disk is driven at a speed of x /2 revolution
per second.

Any intermittent feed such as the Geneva movement used with
films is, of course, out of the question with such a large mass; so
the apparatus has been devised for continuous rotation, with inter-
mittent illumination. This is provided for by a vacuum mercury
arc, which is flashed by discharges from a condenser system, following
the general scheme described by Edgerton. 2 The housing of the
flashing lamp is shown at the right of Fig. 5. The flashes are timed
by a contact that slides over a ring attached to the disk, breaking the
current at narrow slots' cut in the ring.

The Lens. A highly corrected Tessar lens of 8V2-inch focus,
made by the Bausch & Lomb Optical Company for airplane mapping
work, is used to project the image of the slides, with a minimum of
distortion, to the screen.

The Screen. The screen, which is 3 feet by 3 feet in size, is built
up of 180 glass rods, 12 inches and 9 inches in length, arranged in a
stagger pattern so as to avoid prominent horizontal junction lines.
The curvatures of front (clear) and back (frosted) surfaces are such

112

HERBERT E. IVES

[J. S. M. P. E.

as to project a narrow parallel sheet of light into the observing space
from each linear image element on the rear surface. The disk,
lens, and screen are mounted on a rigid frame, with screw adjustments
for distance, tilt, and lateral displacement of the screen, as shown in
Fig. 6.

Performance. The apparatus when operated in the manner
permitted by its construction exhibits on the screen a moving picture

FIG. 6. View of the projection apparatus from the front, showing the
projecting lens and the rod screen.

that presents a different aspect for each direction of observation
(Fig. 7), constituting a true motion picture in relief, visible to a
small group of spectators by direct observation. It therefore con-
stitutes a scientific solution of the problem, to be ranked with the
exhibition of successive, posed photographs of many objects by a
disk in a projection lantern, which was the earliest form of pro-
jected moving picture. Just as the modern motion picture has
reached its present perfection by the contribution of practical details,
preeminent among which is the celluloid film, so this experimental

Aug., 1933] APPARATUS FOR PICTURES IN RELIEF 113

achievement may conceivably be followed by detailed refinements
that will put it into the realm of practical motion picture art.

Much, however, remains to be done to lift it to that state, and
the difficulties to be overcome are very great. Beginning with the
process of making the panoramic negatives, it is obvious that the
exposures required, although made singly and with a stationary
camera, are impracticably long. This difficulty may be overcome
in time by the development of far more sensitive photographic ma-
terials. An alternative taking device would be a battery of motion
picture cameras closely juxtaposed a complicated and costly outfit,
but required only in the studio. By projecting the negatives so
produced by a battery of projectors upon a convex rod screen and
rephotographing the back of the screen, prints similar to those here
described and suitable to be projected in the same way could be
made, thus circumventing the difficulty of over-long exposures.

Taking up next the disk, it is obvious that it would have to be
replaced by some device not limited in respect to the number of
pictures that it could carry. Roll film naturally suggests itself,
but such film would of necessity have to possess exceedingly fine
photographic grain, or be made much larger than regular film, if it
is to carry an adequate number of panoramic strips. Also, it would
have to be quite free from any tendency to warp out of shape, and
be so guided that the lateral hunting would be much less than the
ordinary perforations permit.

A characteristic limitation of the parallax panoramagram is the
falling off of definition in those parts of the scene that lie much in
front of or behind the plane of the picture, due to the finite size of
the elementary linear image elements, and their consequent over-
lapping. This defect is prominent in the projected relief pictures,
and can be overcome theoretically by attaining exquisite definition
of the panoramic images both in the negative making and in pro-
jection. This calls for optical work of excessive accuracy, both in
the camera and projection lenses, and in the screen elements, and
threatens to prove the most serious bar to any practical outcome of
this scientifically sound method of projecting motion pictures in
relief.

Acknowledgments. The author wishes to acknowledge the very
great assistance given by Mr. Howard Hall in the design and con-
struction of the mechanical features of the apparatus, and of Mr.
W. Knoop in the problems connected with the flashing mercury

114 HERBERT E. IVES [J. S. M. P. E.

lamp. The author also wishes to express appreciation of the co-
operation of Mr. W. B. Rayton and the Scientific Bureau of the
Bausch & Lomb Optical Co. in the fabrication of the glass rods used
in the projection screen.

REFERENCES

1 IVES, H. E.: "A Camera for Making Parallax Panoramagrams," /. Opt.
Soc. of Amer., 17 (Dec., 1928), No. 6, p. 435; "Motion Pictures in Relief,"
ibid., 18 (Feb., 1929), No. 2, p. 118; "Parallax Panoramagrams with a Large
Diameter Lens," ibid., 20 (June, 1930), No. 6, p 332; "Parallax Panoramagrams
for Viewing by Reflected Light," ibid., 20 (Oct., 1930), No. 10, p. 585; "Parallax
Panoramagrams with a Large Diameter Concave Mirror," ibid., 20 (Nov., 1930),
No. 11, p. 597; "Reflecting Screens for Relief Picture Projection," ibid., 21
(Feb., 1931), No. 2, p. 109; "Optical Properties of a Lippmann Lenticulated
Sheet," ibid., 21 (March, 1931), No. 3, p. 171; "The Projection of Parallax
Panoramagrams," ibid., 21 (July, 1931), No. 7, p. 397; "The Problem of Pro-
jecting Motion Pictures in Relief," /. Soc. Mot. Pict. Eng., XVIII (April, 1932),
No. 4, p. 417.

2 EDGERTON, H. E.: "The Mercury Arc as a Source of Intermittent Light,"
J. Soc. Mot. Pict. Eng., XVI (June, 1931), No. 6, p. 735; "Stroboscopic and Slow-
Motion Moving Pictures by Means of Intermittent Light," ibid., XVIII (March.
1932), No. 3, p. 356.

DISCUSSION

MR. P. A. McGuiRE: Why is it that when we photograph an approaching
object, the result conveys an illusion of depth?

DR. IVES: In motion picture photography there are a number of well-known
factors that contribute to the perception of depth. First, there is linear per-
spective the fact that lines converge to a distant point. If we see two men,
one occupying an angle of five degrees and another an angle of one degree, we
know they are both men and that their sizes are not in the ratio of those angles at
all. We immediately interpret the difference of angle subtended as meaning
that one man is farther away than the other.

Then we have aerial perspective the general desaturation, the blurring of
colors, due to intervening atmosphere. Distant objects exhibit less contrast
than nearer ones. Another and very important factor that assists in the per-
ception of relief is the relative motion of objects in different planes. As one moves
his head from side to side, for instance, distant objects do not appear to change
much as to their angular position, but nearer ones seem to move a great deal.
That is a signal, to our minds, that the objects are located in different planes.

But if the camera, or the object, or the observer, can not move in any way,
then we have to rel^ on other things for the perception of depth, of which the
most important is the binocular or stereoscopic relief dependent upon the use of two
eyes.

MR. J. I. CRABTREE: I don't quite understand the precise structure of the
parallax panoramagram. Of course, it consists of a series of strip images, the
number of images corresponding to the number of openings in the grating

Aug., 1933] APPARATUS FOR PICTURES IN RELIEF 115

through which one views it. But consider the left-hand strip: Is it a pano-
rama of the entire object; or, does it consist of a large number of narrow
strips of the left-hand portion of the object as viewed from different positions?

DR. IVES: The panorama developed in each strip is the panorama one obtains
on moving the eye laterally across the large lens or mirror, keeping it directed
always to the point on the scene that is focused on the grating strip. It is a
panorama in the sense of "looking around" the object from a series of points of
view, and thus differs from the pictures made in the usual panoramic camera,
when the lens "looks around" from a fixed point.

MR. CRABTREE: As one views the panoramagram through the grating he
sees a narrow strip of the first panorama, a narrow strip of the second panorama,
and so on, and these are integrated by the eye to constitute the picture. Is that
correct?

DR. IVES: Yes, that is correct.

MR. P. H. EVANS: If each successive strip in the photographs that were
shown on the screen is a panoramagram of the entire picture, why is there not
a greater similarity between strips as one's view progresses across the screen?
In the picture of the man's face, the center was white; there was no shadow
in it at all. And yet the strips to -the right had shadows in them, and the
strips to the left had shadows.

DR. IVES: The answer to your question lies in the fact pointed out to Mr.
Crabtree; namely, that the panorama in each strip is obtained by pivoting
around the point in the scene that is focused on the grating plane. A flat object
at the pivoting point will look the same no matter from what direction it is
viewed, and will result in a panoramic strip that is uniform across its width.
That is approximately the condition in portions of the man's face. In addition,
it is to be remembered that the angle embraced by the panorama is only the
angle subtended by the large mirror used for taking, namely, about 60 degrees.
If the entire 180-degree angle could be used there would be a greater variation
in the appearance of each strip.

MR. CRABTREE: How many lines are there to the inch on the viewing screen?

DR. IVES: The rods are about a quarter of an inch wide.

DR. H. ROSENBERGER: What is the significance of the horizontal lines on
the screen?

DR. IVES: The machines of the Bausch & Lomb Optical Company available
for grinding these cylindrical rods will accommodate only a twelve-inch length;
these rods were made, some nine and some twelve inches long, so as to stagger
them and avoid a bodily shift of one part of the picture in respect to another.

MR. CRABTREE: I believe you mentioned in your Swampscott paper that
before we could apply these principles to motion picture theaters we would need
emulsions that have a much finer grain and a tremendously higher speed than
what we now have.

DR. IVES: I grouped together all these questions of structure in my remarks
about our getting down to the wavelength of light. That applies also to the
grains of the photographic emulsion. They would have to be too small. And
the matter of speed I brought out indirectly by mentioning the exposure used in
taking these pictures, namely, about a minute. The photographic emulsion
speed required for making twenty exposures a second can be worked out arith-
metically on that basis.

A NEW ALTERNATING-CURRENT PROJECTION ARC*
D. B. JOY AND A. C. DOWNES**

Summary. A new type of arc operated on alternating current is a desirable
substitute for the low-intensity, direct-current arc used in the majority of motion
picture theaters. This arc is a modification of the well-known white-flame arc,
in which the light sources are concentrated at the electrodes by using heavy currents
and low arc voltages on specially designed carbons.

It is operated at current densities higher than those employed in direct-current
arcs, and produces a screen light of a blue-white color similar to that obtained with
the direct-current, high-intensity arc. The bluish white screen light is remarkably
steady and uniform. The power required to produce a given screen illumination
is considerably less than that required with the low-intensity, direct-current arc.

In projecting motion pictures, several grades or types of carbon
and several makes or designs of lamp are used. But all the grades
of carbons produce arcs of only three fundamental types ; and all the
lamps can be placed in one of three classes if the very few installations
still using lamps of the condenser type, burning either direct-current,
neutral-cored carbons or the white-flame, alternating-current special
carbons, be omitted.

The three carbon arc types are the high-intensity, rare-earth,
cored-carbon, direct-current arc, the plain neutral-cored, direct-
current arc, and the so-called special white-flame, alternating-current
projector arc.

The three lamp classes and the carbons burned in them are shown
in Table I.

Both Classes I and II use the direct-current, high-intensity arc,
which produces the brilliant light of blue-white sunlight quality
universally considered desirable in the theater. In the range of
current used in these two classes, there is no gap of any magnitude,
so that the theaters using these high-intensity arcs can easily arrive
at the level of screen illumination best suited to their particular
conditions.

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** National Carbon Co., Cleveland, Ohio.

116

ALTERNATING-CURRENT PROJECTION ARC

117

C 00 O I>

t-H i-l CO

fci.O

mt:

o
13

S <M

op ^

1C CO

CO i-l

<u D

%->

'55
o

? ?
x 7!

bo G

C 4J

a
S
a

1 1

" C

*"^

n~ C

I
H

C/)

i i

s 1

TABLE
rojector La

ondenser
carbon

bo bo

.s .s g

g .n "o .a

S 4J

<E 2 ^ o
<u u u w

' ^

rt rt

1 =

s g S S

"3 o

Tl VH

J -S

VH CS ^ f*

rt o '-C TS

SJ <

& (i fl

1
U

igh-intensity posit:

' Orotip negative c
-intensity positive
negative carbon
reflecting arc nega
" posii

White-Flame A-c.
Low-Intensity D-<
Condenser Lensi

ifl

< JS ft

> .SP '-g 6 E
o 8 E E

1 " E 5 S 2

o
u

rH CO

*< ?i^.i

.23

TH I-l

i-H ^ O5 "3 '^

1 1 I 1
1 1 t-H

118 D. B. JOY AND A. C. DOWNES [J. S. M. P. E.

There is a large drop in current value, however, between the high-
intensity arcs of Class II and the low-intensity arcs of Class III;
and the color of the light from the low-intensity installations, while
appearing a brilliant white viewed by itself, is yellowish white when
compared with the high-intensity sources of Classes I and II.

A large number of the theaters using the low-intensity light sources
desire the same blue-white light of the high-intensity sources, but
to obtain this desirable light color would require a change to more
expensive lamp equipment and higher power cost.

To bridge the gap between the high-intensity and low-intensity

FIG. 1. Direct-current, high-intensity arc.

arcs, and at the same time to give the small theater the advantage
of the blue-white light enjoyed by the larger theaters, has been the
object of researches in the laboratories of the National Carbon
Company for several years. The results of this long research show
that it is possible to fill the gap between the high- intensity and
low-intensity sources with an arc giving a light color very similar
to that of the high-intensity arcs, which will also provide a number
of other advantages to theater owners of Class III, who outnumber
the high-intensity classes by two to one.

The desired result has been accomplished by means of an alter-
nating-current arc burned on the secondary of a specially designed
transformer, without the ballast resistance, always necessary with
direct current, and also without the motor-generator set or rectifier
now required in the vast majority of theaters of all classes. This

Aug., 1933] ALTERNATING-CURRENT PROJECTION ARC

119

alternating-current arc is a modified white-flame arc with specially
designed carbons containing compounds of the cerium rare earth
group of elements. The accompanying Figs. 1 to 4 show the differ-
ences in the light sources of various carbon arcs and why this new
alternating-current arc should be of great value in projection.

Fig. 1 is a front and side view of a high-intensity arc, showing
very clearly the very brilliant, concentrated light source at the posi-
tive crater and why this arc is easily adaptable to an optical system
employing either a mirror or condenser lens. This type is used in
light source Classes I and II.

FIG. 2. Direct-current, low-intensity arc.

Fig. 2 is a direct-current, neutral-cored carbon arc (used in Class
III light sources), in which the light source is the brilliant crater on
the positive carbon. This figure shows the much less brilliant
negative tip and the very faint arc stream between the carbons.
This shows, as in the high-intensity arc in Fig. 1, that by far the
largest part of the light comes from the positive crater.

Fig. 3 is an alternating-current, white-flame arc, burning carbons
containing cerium group compounds. The light source is the brilliant
flame between the electrodes; but the electrodes themselves are
relatively dim, and emit only a very small fraction of the light.
Such an arc, while producing a great deal more total light than a
neutral-cored carbon arc using the same power on direct current, is

120

D. B. JOY AND A. C. DOWNES

[J. S. M. P. E.

obviously unsuitable for projection, as it is too large to be readily
focused through an optical system and its intrinsic brilliancy is too
low.

The light emitted by the white-flame arc at currents up to 30 or
35 amperes increases directly as the voltage, but as the square of
the current. The rate of increase in light, with increasing current,
decreases above 30 or 35 amperes until at very large currents the
rate of light increase becomes equal to the rate of current increase.

FIG. 3. Alternating-current, white-flame arc.

Increasing the current, which, if no other adjustments are made, will
also cause a small increase in arc voltage, will therefore very mate-
rially increase the light produced. If, instead of permitting the arc
voltage to increase with increasing current, the arc voltage and also
the arc length be decreased, the arc becomes steadier and the sources
of light are concentrated into smaller volumes near the electrode
tips, as shown in Fig. 4. This arc is quite different in character
from any other arc of which we know today. Either of these small,
brilliant light sources can be focused by means of a mirror, and
therefore can be used for projection.

In spite of the fact that practically all the useful projection light

Aug., 1933) ALTERNATING-CURRENT PROJECTION ARC

121

comes from only one of these sources, early experiments showed that
with this alternating-current arc at least 15 per cent more light could
be projected to a screen with a given optical system and aperture
plate than with a low-intensity, direct-current arc using the same line
power. The color of the screen light from the alternating-current
arc was blue-white, resembling that produced by the direct-current,
high-intensity arc.

In addition to these advantages is the fact that ballast resistances,
expensive switchboards, and motor-generators are unnecessary. The
motor-generator and ballast resistance are replaced by a relatively

FIG. 4.

Alternating-current projection
arc.

low-priced transformer, and the switchboard can be very simple and
cheap.

A number of years ago Mott found that if two alternating-current
arcs, one with resistance and the other with reactance ballast, were
operated with the same line power, the arc with reactance ballast
produced 33 per cent more light.

There is no reason why the new alternating-current arc can not
be controlled with a ballast resistance; but Table II, following,
shows that since the transformer will cost little or no more than a
good resistance unit, there should be no incentive to use the resis-
tance.

The carbons for this alternating-current arc service have been made
in 6-mm., 7-mm., and 8-mm. sizes for use at 40-45, 60-65, and 75-80
amperes, respectively. A large number of laboratory tests have
been made on these sizes. Since the current densities are about 800

122 D. B. JOY AND A. C. DOWNES [J. S. M. P. E.

to 1000 amperes per square inch of cross-section, it is necessary to
use metal-coated carbons.

The carbon consumption is approximately 4 to 4.5 inches per hour
for the 6-mm. carbons at 45 amperes, and 4.5 to 5.5 inches per hour
for the 7-mm. carbons at 65 amperes and the 8-mm. carbons at 80
amperes.

The uniformity of screen illumination is as good as that obtained
with direct-current arcs, as shown by Table III. The same optical

TABLE II

Power Required for Equal Screen Light Using Resistance and Transformer

Control

Line Arc Arc Power Consumption

Control Voltage Voltage Current in Watts

Resistance 115 30 80 9200

Transformer 115 30 80 2500*

system, consisting of a reflector, aperture plate, and objective lens,
was used in these tests. There was no rotating shutter in the system.
It should be noted that these data were obtained on a laboratory
set-up, and while they are perfectly comparable among themselves,

TABLE III

Comparative Screen Light from Alternating-Current and Direct-Current, Neutral-
Cored Carbon Arcs with the Regular Mirror Arc Optical System

Line Line Line Arc Arc Foot-Candles on Screen**

Carbons Cur. Volt. Watts Cur. Volt. Middle Left Right Top Bottom

7-mm. Pos. 9 d-c. 110 990 9 d-c. 50 21 20 19 20 21

5-mm. Neg.

Neutral Core

32832-2 8 d-c. 110 680 34 a-c. 19 27 25 25 26 27

they are not indicative of what might be obtained with any other
set-up of optical system and screen.

Table III shows that there is no question that the uniformity of
screen illumination with the new alternating-current arc is equal
to that of the low-intensity, direct-current arc, with the very distinct

* Assuming transformer efficiency of 95 per cent.
** Weston photronic cell.

Aug., 1933] ALTERNATING-CURRENT PROJECTION ARC

123

advantage that the blue-white color of the light resembles very
closely that obtained from the direct-current, high-intensity arc.

There are not yet available detailed comparisons of the arcs of
higher amperages, but a sufficient number of measurements have
been made to be certain that the following Table IV provides a general
idea of what may be expected from these alternating-current arcs
in terms of well-known direct-current projection systems.

TABLE IV

Performance of Alternating-Current Arcs with Large Currents

Carbons

Current

Screen Light Compared

Average Optical with SRA Carbons Line
Volts Line Watts* System at 35 Amp.. 55 Volt. Watts

6-mm. a-c.

40-45

22-25

945

Regular

Mirror

Arc

7-mm. a-c.

60-65

23-26

1580

8-mm. a-c.

75-80

24-29

2130

60-70%

85-95%
115-150%

3500

3500
3500

In arriving at the values for line watts an efficiency of 80 per cent
has been assumed for the motor-generator sets in the case of the
direct-current arcs and 95 per cent for the transformers for the
alternating-current arcs.

One practical test has been in progress since September, 1932, in a
theater with a 12 X 16V2-foot screen and a throw of about 100 feet.
There have been no complaints concerning the quality of the
screen light or the quantity, and no unusual operating troubles have
been encountered.

In order to furnish a general idea of the probable fields of use for
this arc there is given in Table V a general summary of carbons,
currents, voltages, wattages, and screen light in arbitrary units on
a screen of a given size. The optical systems were the conventional
ones employed with the different kinds of carbons and lamps in
actual use in the theaters.

* Transformer efficiency, 90 per cent.

124

D. B. JOY AND A. C. DOWNES

[J. S. M. P. E.

to
_c

i 1

3 2

o o
E E

M *O

3 .
S t

s
&j

j * *

Ji K B S

: E

!2 E

>ob

I||E|

CO CO ^J (N

J3 JS Ji JS i
_bC M bO to f

P3 P] Q PQ IH

E
1
o

en .S

a .T: .T: .t: .r: i> -- .t:

B g g g g *| g g ^

LS 8 S iic

_i T; * _ iir;

Aug., 1933] ALTERNATING-CURRENT PROJECTION ARC 125

DISCUSSION

MEMBER: Since the arc is supplied directly from the a-c. supply lines through
a transformer, every voltage variation of the outside lines will be transmitted
to the arc. In view of the fact that all power companies allow themselves a
variation of at least 5 per cent above and below the normal voltage (a total of
10 per cent between minimum and maximum), what would be the effect of such
variation on screen illumination and flickering?

MR. DOWNES: We have not found a 5 per cent variation in voltage in our
laboratories and factories except at two very definite times each day: in the
morning when the factory starts and in the evening when the greater part of the
power load is taken off the line. Such a condition may be found in an industrial
district where the loads are heavy, but we have thought the voltage changes in
our laboratories to be largely confined to our own lines on the low- voltage side
of the power transformers serving both factory and laboratory. If difficulty is
encountered with variable voltage, the trouble may be very easily corrected by
a small rheostat in the lamp circuit or by variable taps on the transformers.

MEMBER: How do the a-c. carbons compare in price with the present d-c.,
low-intensity carbons?

MR. E. R. GEIB: The operating cost of the new a-c. projector trim will be a
little over 50 per cent higher than it is with the present low-intensity, d-c. trim,
but this cost is just about one-half that of Hi-Low operating cost. The im-
provement in intensity and color of light with the new a-c. arc far more than
offsets the slight increase in operating expense.

MR. F. H. RICHARDSON: It was extremely difficult for projectionists to main-
tain steady screen illumination with the old style a-c. arc; the craters were
small and difficult to handle. Will this trouble occur in any degree with the
new a-c. carbons?

MR. DOWNES: I do not want to say that there will be no trouble with this
new arc system; but such difficulties as are encountered will be only those more
or less minor annoyances commonly found in all projection systems. You cer-
tainly will not have the difficulties encountered with the old a-c. arc, as the new
arc is entirely different.

MR. RICHARDSON: Another important consideration is the possible effect
upon light tone. The old-style a-c. arc furnished a rather harsh light tone;
and the light appeared to be very penetrating, much more so than light from a
d-c. arc.

MR. DOWNES: We have had many requests from the smaller theaters such as:
"Can't you give us the same blue-white light the larger theaters have?" There
seems to be a general consensus of opinion among the projectionists that the
blue-white color is far more desirable than the other.

MR. RICHARDSON: Are those carbons ready for the market now?

MR. GEIB: Yes.

MR. RICHARDSON: What is the amperage of the two lamps?

MR. DOWNES: The power consumption at the arc is the same; the one we
are demonstrating consumes between 40 and 45 amperes at 18 volts.

PROFESSIONAL MOTION PICTURE PHOTOGRAPHY

WITH HIGH-INTENSITY SHORT-LIFE

INCANDESCENT LAMPS*

M. W. PALMER** AND E. W. BEGGSf

Summary To increase the light output per watt, to improve the actinicity of
the light, to increase the proportion of blue light badly needed for color photography,
and to reduce the size and number and increase the output of the lighting units,
tungsten lamp filaments are heated to temperatures as near the melting point as is
practicable. Reference data are presented on lumen output, efficiency, color (energy)
distribution, and lamp life. Practicable safe operating limits are indicated. A
new "super photoflood" lamp is described and data on overvoltage lamp operation
for location lighting are presented for reference.

I. LAMP CHARACTERISTICS

Recent trends in the use of incandescent tungsten lamps for photog-
raphy indicate an increasing desirability of obtaining adequate
levels of illumination by operating lamps at higher efficiencies rather
than by increasing the wattage. A successful application of this
idea is the high-efficiency, short-life photoflood lamp, which has proved
to be a most convenient and satisfactory light source for indoor
amateur motion picture and still photography. In professional
motion picture photography, the operation of tungsten filaments at
high temperatures is especially appropriate where the available
power is limited and where undue heat on the set may result from
overcrowding the lighting units.

The advantages to be derived from operating tungsten filaments
at temperatures higher than those found in present practice are
pointed out in the illustrations, which show various relations be-
tween filament temperature, visible light output, photographic
effect, and lamp life.

Fig. 1 shows diagrammatically the tremendous increase Of visible
light emitted from a coil of tungsten wire at various temperatures up
to 3500K. The area beneath each curve represents the energy

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Motion Picture Lighting Co., Long Island City, N. Y.
t Westinghouse Lamp Co., Bloomfield, N. J.

126

PHOTOGRAPHY WITH INCANDESCENT LAMPS

127

resident in the visible light radiated at the temperature indicated.
These curves also bring out the fact that as the temperature of the
filament is increased, the increase of energy radiated at the shorter
wavelengths of light, i. e., violet, blue, and green, is much greater
than the increase at the longer wavelengths, i. e., yellow, orange, and
red. This is a fact that is of especial value in color photography.

3500 K

VISIBLE LIGHT

2900 K.

2840

40 50 .60 .70 .80

WAVE LENGTH IN MICRONS

FIG. 1. Distribution of visible energy radiated
from tungsten filament.

The change of photographic effectiveness due to this change of
spectral quality of the light emitted by a tungsten filament lamp at
various temperatures is illustrated in Fig. 2 for both panchromatic
and orthochromatic emulsions. The curves show distinctly that,
for a given intensity of illumination, measured visually, the higher
the filament temperature the greater the photographic effectiveness.
This, of course, is much more noticeable with orthochromatic than
with panchromatic film.

Increasing the temperature from 3000K. to 3500K. multiplies

128

M. W. PALMER AND E. W. BEGGS [J. S. M. P. E.

the lumens of visible light emitted from a certain tungsten filament
about five times. Actually, the power consumed is also increased;
but, due to the greater proportion of the energy radiated in the visible
range when the filament is heated to higher temperatures, the in-
crease of light far exceeds the increase of power required. These
relations between the power and the light output, for various filament
temperatures, are indicated in Fig. 3. The curve correlates the lumens
per watt, or efficiency of the tungsten filament, with the temperature
of the filament, and shows clearly how much more rapidly the light
output increases than does the power. For example, increasing the

24OO 260O 2 BOO 3OOO 32OO 3400 36OO

COLOR TEMPERATURE IN DEGREES KELVIN

FIG. 2. Relation of photographic effectiveness to color
temperature of incandescent tungsten.

filament temperature from 3000 K. to 3500 K. in a 1500- watt
lamp results in an increase of 20 to 35 lumens per watt, or a gain
in light output of 75 per cent.

Fig. 4 shows the relation between the temperature of the filament
and the life of the lamp. The molecules of tungsten that are
loosened from the incandescent filament vibrate more violently and
in greater numbers as the temperature increases, resulting in a grad-
ual loss of metal from the wire. Finally, by this process, called "sub-
limation," the filament wastes away until at some point in its length
it becomes too thin to carry the current; the filament melts at
that point, and the lamp burns out. The curve shows the life vs.
temperature characteristic of a tungsten filament lamp.

Aug., 1933] PHOTOGRAPHY WITH INCANDESCENT LAMPS

129

The data given in the curves of Figs. 2, 3, and 4 are the basic
data from which the best filament temperature for a given set of
conditions can be calculated. These data apply more to the studio
than to "location." They show that if the lamp manufacturers
were to make lamps especially for motion picture studio service they
might well allow the filaments to operate at higher temperatures
than they do now. Suppose, for instance, that the performance of
a 1500-watt, 115-volt, PS-52 bulb, 1000-hour general lighting lamp

tr
*28

2 20
& 16

lia

X
X

u. 1 2

Q. 8

- 1 4

2600 2800 3000 3200 3400 3600

FILAMENT TEMPERATURE. IN DEGREES KELVIN
FIG. 3. Efficiency vs. filament temperature; 115-volt,
gas-filled Mazda lamps.

were to be compared with that of a 1500-watt lamp similar to it
but designed to have a life of 10 hours. Such a comparison would
show some interesting facts.

The filament of a 1500-watt, 1000-hour lamp operates at a tempera-
ture of about 3000K., and produces an initial output of 32,100
lumens. However, if a lamp were designed to operate at 3500K.
and yet consume only 1500 watts, such a lamp would produce
about 55,900 lumens. This increase of 75 per cent is the increase
of lumens per watt at the higher temperature (see Fig. 3). At
the same time, the photographic effectiveness of the light is in-
creased (see Fig. 2), with the result that one such 1500-watt lamp

130

M. W. PALMER AND E. W. BEGGS

[J. S. M. P. E.

operating at a high temperature is photographically equivalent to
almost two 1000-hour, 1500- watt lamps.

This remarkable increase of light output practically doubles the
effectiveness of "broadside" lighting equipment and any other
fixtures now using the 1000-hour type of lamp. It also increases,
in the same proportion, the lighting capacity of the power generating
plant and the wiring system.

Similarly, the heat radiated to the set by these lamps is almost

100

20
10

3000 3200 3400 3600 3SOO

FILAMENT TEMPERATURE IN DEGREES KELVIN
FIG. 4. Per cent lamp life vs. filament temperature;
11 5- volt, gas-filled Mazda lamps.

halved, because the heat produced by the lamps is determined
directly by the amount of power consumed, and it is possible, owing
to the higher filament temperature, almost to halve the power.

The changes resulting from an increase of temperature from
3000K. to 3500K. apply only to lamps of those types the filaments
of which operate at the lower value. Such lamps in ordinary studio
work are the PS-52 bulb general lighting types, usually 1000-watt,
and the 1500- watt PS-52 bulb lamps.

The illumination in a motion picture set, according to present
practice, is obtained partly with general lighting PS-52 bulb lamps,

Aug., 1933] PHOTOGRAPHY WITH INCANDESCENT LAMPS

131

usually rated at 1500 watts, operating in broadside or overhead
units developing a diffused general illumination of high intensity
throughout the area of the set. The total required varies widely,
but in ordinary cases represents about half the total power used.
It is here that the greatest increases in photographic effectiveness of
the light can be achieved.

However, it is also possible to increase the filament temperature

00 $

100 120 140

PERCENT VOLTS

FlG. 5. Characteristic curves of
Mazda lamps.

of the higher powered lamps in a similar manner. The 2000-watt
G-48 bulb spotlight and the 5000-watt G-64 bulb lamp, both of
which are of the types generally used to create highlights, shadows,
and modeling effects, have filaments that operate at higher tem-
peratures than the 1500- watt, 1000-hour type discussed above. The
filaments of these lamps operate at approximately 3200K., so
that in such lamps, also, an increase of temperature is possible. This
would produce an appreciable improvement in their efficiency for

132

M. W. PALMER AND E. W. BEGGS [J. S. M. P. E.

9 10

FIG. G. Prints of frames photographed under different lighting conditions,
with the percentage of normal voltage of the lamps and the camera aperture
as follows: (1) 100%; (2) 115%; (3) 130%; (4) 145%; (5) 160%. For
higher voltages the lens was stopped down, as follows: (6) 160%, //5.6; (7)
160%, //8.0; (8) 160%, //ll; (9) 170%, //3.6; (10) 170%,//5.6.

Aug., 1933] PHOTOGRAPHY WITH INCANDESCENT LAMPS 133

black-and-white pictures, a very great increase for color photography,
and would reduce the total amount of heat radiated to the set for a
given lighting intensity.

As a result of these calculations and of the practical application of
the lamps in motion picture photography, lamp manufacturers have
designed PS-52 bulb short-life lamps (approximately 10 to 20 hours)
of 1500 and 2000 watts with their filaments operating at approxi-
mately 3450 K., which are now available for commercial use. The
standard higher-powered lamps for the present are being used in
actual studio practice with the impressed voltage set at about 110
to 120 per cent of the labeled voltage of the lamps. If the need
for special lamps in these sizes becomes apparent, they will be de-
veloped and supplied.

All the data presented in the foregoing discussion are concerned
with lamps especially designed for high filament temperatures. At
the outset, experiments were made with lamps having filaments
designed to operate at 3000 K., approximately, but operated at an
overvoltage in order to increase the temperature to higher values
on the scale. In Fig. 5 are four curves that show the result of over-
voltage operation on the performance of any given lamp. It will
be seen that an increase of voltage from 100 to about 175 per cent
of normal results in an increase of lumen output from 100 to 500
per cent. At the same time, the photographic effectiveness of the
light is increased not only five times, but with orthochromatic
film 6.25 times, and with panchromatic film 5.16 times. Of course,
such an enormous overload on the lamp may cause some part of its
structure to fail, even though the filament may actually be operating
below its melting point. This may result in early or immediate
burnouts. However, 1000-hour lamps have been operated at over-
voltages sufficiently high to reduce the operating life to less than
10 hours with surprisingly good results. Such operation of the
lamp is, of course, at the operator's risk. The lamp manufacturers
can not guarantee the operation of their product at such voltages,
since all the design work and practically all the tests are conducted
at the normal voltage, or at no more than 115 per cent of the normal
voltage.

II. PRACTICAL APPLICATION

In this section of the paper will be considered some practical appli-
cations of lamps burning at voltages greater than the normal in
professional motion picture photography. It becomes occasionally

134 M. W. PALMER AND E. W. BEGGS [J. S. M. P. E.

desirable on location to photograph by means of artificial light.
Very often in such cases the current that is available is insufficient
to provide the amount of light required. By using lamps having
a voltage rating less than that of the line voltage available, a con-
siderable increase of photographic light may be effected with an
increase of power.

The illustrations in Fig. 6 show the photographic results obtained
by raising the voltage in increments from 110 volts normal to 170
per cent of normal voltage. The lamps used were three 1000-watt,
110- volt PS-52 lamps mounted in porcelain enameled reflectors.
The position of the lighting units was not changed during the ex-
posures. The illustrations are prints of frames taken from a series
of exposures made with a standard motion picture camera operating
at a speed of 60 feet per minute and with the lens aperture set as
indicated in the caption. At 1 in Fig. 6 is shown an exposure made
with the lamps operating at the normal voltage, with which the
other exposures, made at voltages greater than normal, may be
compared. Examples 2 to 8, inclusive, are exposures made with
lamp voltages up to 160 per cent of the normal value. When 160
per cent of the normal voltage was attained, the lens was stopped
down.

Examples 9 and 10 show the results obtained with a still higher
voltage, 170 per cent of normal, with an aperture of //3.5.

As shown in the first section of this paper, there is a practical
limit to which the voltage may be increased, since with each increase
in voltage there is a corresponding decrease of the life of the lamps.
It is, however, possible to increase the photographic light by almost
100 per cent without increasing the wattage, a matter of great
importance in making it possible to conserve power to a considerable
extent in producing motion pictures.

DISCUSSION

MR. J. I. CRABTREE: Suppose the photoflood lamp is designed for a normal
line voltage of 110; at what voltage will the filament melt?

MR. BEGGS: There is naturally a limit, but the photoflood lamp, as now de-
signed, operates at normal voltage at a temperature that provides enough margin,
before reaching the melting point of tungsten, to accommodate all the published
voltages in the United States. These values also involve a secondary factor of safety.
The voltage in New York City is nominally 120. The power companies arrange
matters so that the average voltage at the lamp socket will be 120, which means
that some socket voltages will be somewhat greater and some will be somewhat

Aug., 1933] PHOTOGRAPHY WITH INCANDESCENT LAMPS 135

less than 120. These lamps have a factor in their design that accommodates all
such conditions. A surge of voltage would probably have to exceed 130 volts for
a period of over 5 seconds to "blow" the lamp.

MR. FARNHAM: The life of the photoflood lamp is usually given as two hours
at 115 volts, thus emphasizing the point that the life of the lamp is quite critical
as to voltage. A good general rule to remember is that a five per cent increase
of voltage halves the life, so that if the voltage on a photoflood lamp were in-
creased to 120, we should expect the lamp to last about one hour. However, if
the voltage is increased much beyond 120, the melting point of tungsten is soon
reached and the lamp fails immediately. Tests show that at 130 volts they
fail almost immediately, and that they last a short time at 125 volts; hence the
limiting voltage is undoubtedly between 125 and 130 volts.

In the course of a series of tests conducted by the General Electric Co., and the
Eastman Kodak Co., in connection with motion picture photography in color,
quite a large number of lamps of this type were employed. The process under
development requires an illuminant having at least twice the blue-violet radiation
as that obtained from the lamps generally used for black-and-white motion
picture photography, for an equal amount of red radiation.

The widely used 1000- and 1500-watt PS-52 bulb lamps operate at an efficiency
of approximately 21 lumens per watt, but if the efficiency is increased to 33
or 34 lumens per watt the percentage of blue-violet radiation is about doubled.
A 1500-watt lamp operating at that efficiency has the visual light output of a
2250-watt lamp and the approximate photographic effectiveness of a 4000-watt
lamp. The life of these lamps is roughly 12 to 15 hours.

An analysis of costs in connection with these tests shows that $80 worth of
lamps is sufficient to light a set of average size and permits the exposure of $2400
worth of film. This is considered a satisfactory ratio.

Because of the relatively short life of the lamps it is a good plan to operate
them at a reduced voltage, probably about 90 volts, when preparing for the
picture; then to increase the voltage to its full value just before photographing.
This procedure will greatly prolong the life of the lamp. It has been found that
the life of the lamp is greater when the voltage is increased to this intermediate
value first and the lamp allowed to "warm up" before applying the full voltage.
Where the work of photographing is interrupted by periods of preparation, the
voltage should be reduced to the intermediate value and then increased to the
full value again, rather than to turn off the lamps completely each time.

Lamps of this type blacken somewhat rapidly, hence a tungsten cleaning
powder, similar to that now employed in the 5- and 10-kw. lamps, is incorporated
in the bulb. Sliding this powder over the surface of the bulb from time to time
restores the light output to its initial value. A periodic cleaning schedule,
possibly once each day, is a good plan.

Because of the relatively great light output of these lamps as compared with
the more generally used types, the wattages employed on sets for color photog-
raphy are of the same order of magnitude as now employed for black-and-white
photography, and hence the increase of temperature is negligible.

MR. RICHARDSON: If I increase the voltage of the lamp and thus increase its
light-emitting power, will the same number of lumen hours result as would result
had the lamp been operated at normal voltage?

136 M. W. PALMER AND E. W. BEGGS

MR. MILI: A much smaller number of lumen hours, with a shorter lived
lamp. The life curve (Fig. 5) drops much more rapidly than the efficiency curve
rises, wherefore the total output of the lamp decreases throughout its life, so
to speak. But it is the intensity, possibly, that comes to be more important
in certain applications rather than the total amount of light emitted by the lamp
throughout its life.

MR. BEGGS: Regardless of the fact that the lumen hour output decreases
throughout the life of the lamp, it may be convenient to operate the lamps at
overvoltages in order to obtain the desired intensity, and to sacrifice the life
to some extent; and in projectors, only by using short-life, high-intensity fila-
ments can we obtain high intensities of screen illumination.

MR. CUTHBERTSON: What is being done to reduce the number of explosions?

MR. BEGGS: Every now and then a lamp does explode. We test thousands
of them, and a small fraction of one per cent do explode, and a very small fraction
of those explode violently. In most cases the glass simply cracks and gives way.
An explosion may occur if the filament should melt at a certain spot in the coil
of tungsten wire in such a way as to form an arc. If the arc is maintained for a
long period of time, the gas in the bulb becomes hot, creates an internal pressure,
and in some cases has been known to cause a bulb to explode.

However, explosions of Mazda lamps occur rarely, primarily because in lamp
making the bulb has been so annealed that it has no strains in it, and a link is
provided in the lead wire that fuses when an arc is formed. Two years ago we
used to put our monogram on the end of the bulb; in making the mark we used
silver metal and fired it in the glass, just as china is marked, thereby creating
strains. But for the past year and a half or two years that strain has been
eliminated, and every bulb is inspected by a polariscope for strains, so that arcs
and explosions are now very rare.

MR. KURLANDER: When the total wattage employed in illuminating a set
is very large, this method certainly will help to reduce lens focusing errors due
to atmospheric refraction.

MR. ENGLEKEN: To what extent are photoflood lamps being used in the
motion picture studios at present?

MR. FARNHAM: The small standard photoflood lamps are not used. The
high wattage lamps of the photoflood type, of 1500 watts' rating, were employed
in developing a color photographic process. Since they have proved to be satis-
factory, they will very likely be used where such processes are employed.

MR. BEGGS: I might say that the small photoflood lamp, in professional
studio work, is used occasionally in a lighting fixture a home lighting fixture,
for example that will appear in the set. But it is not used for general lighting,
because of the small size.

MR. ENGLEKEN: Is the photoflood lamp made at present in only the two
sizes?

MR. BEGGS: Yes.

MR. ENGLEKEN: Are the filament dimensions of the 1500-watt lamp the same
as those of the standard 1500-watt lamp?

MR. BEGGS: Slightly smaller. Physically, this lamp is interchangeable with
a standard 1500-watt, PS-52 lamp.

AN IMPROVED POTASSIUM ALUM FIXING BATH
CONTAINING BORIC ACID*

H. D. RUSSELL AND J. I. CRABTREE**

Summary. Most hardening fixing baths containing potassium alum tend to
lose their hardening properties and precipitate a sludge of aluminum sulfite long
before the fixing power of the sodium thiosulfate becomes exhausted. It is customary
in large-scale practice to offset this loss in hardening properties by adding acid to
the bath several times during its useful life. The addition of boric acid (5 grams
per liter) to a fixing bath containing potassium alum has been found to increase
the hardening life about four times and to minimize the sludging tendency to such
an extent that revival with acid is unnecessary during the useful life of the bath.

A detailed study has been made of the properties of fixing baths containing potassium
alum, acetic acid, sodium sulfite, and boric acid, and a suitable, formula is recommended
for motion picture work. Except for tropical use, it is considered that the formula
containing boric acid is to be preferred to most chrome alum fixing baths which
require frequent revival with acid during use in order to maintain uniform hardening
properties.

Fixing baths for motion picture film consist essentially of an
aqueous solution of sodium thiosulfate together with an acid harden-
ing agent, which consists of a mixture of sodium sulfite, a suitable
acid, and an alum. The alum may be either aluminum or chrome
alum and, according to the particular alum used, the fixing baths are
classified as (a) chrome alum baths, and (b) potassium alum baths.

Some of the properties of chrome alum baths have been described
previously, 1 ' 2 and although baths of this type are employed almost
exclusively in some of the larger film processing laboratories, they
require more precise control in order to maintain their hardening
properties during use as compared with potassium alum baths, and
for this reason the latter are to be preferred when facilities for such
control are not available.

The factors which determine the properties of potassium alum
fixing baths have also been investigated previously by Crabtree
and Hartt, 3 who have shown that the desirable properties of such

* Presented at the Spring, 1933, Meeting at New York, N. Y. Communica-
tion No. 519 from the Kodak Research Laboratories.

** Research Laboratories, Eastman Kodak Co., Rochester, N. Y.

137

138

H. D. RUSSELL AND J. I. CRABTREE [J. S. M. P. E.

baths exist only over a relatively narrow range of concentration of
the various constituents. The chief objection to the usual potassium
alum baths is that the hardening properties change with use and
ultimately tend to disappear as the bath is neutralized by the alkali
in the developer carried over by the films. Also, when the acidity
of the bath falls below a certain critical value, the bath deposits
a sludge of basic aluminum sulfite. This, in turn, tends to adhere
to the film surface, and appears as dirt or a scum on the dried film.
The quantity of developer which can be added before this sludge
appears is termed "the sludge life."

Attempts have been made to prevent this sludge precipitation by

o BORAX riXING. BATH

A ROCMCUUE.

" 6.0 /L.

10 ZO ao 40 bO 60 10 8O 9O IOO

FIG. 1. Effect of Rochelle salts on the hardening properties of
fixing baths.

the addition of relatively small quantities of organic acids, such as
oxalic, lactic, citric, tartaric, malic, and maleic or their alkali salts.
These acids have been shown by Sheppard, Elliott, and Sweet 4 to
prevent the precipitation of an alumina sludge in the fixing bath,
but they also lower the degree of hardening. These authors suggest
that possibly this behavior is due to the formation of complex ions
between the aluminum ions and those of the organic acid.

Mr. R. S. Becker, of the Eastman Kodak Company, has recently
devised the following fixing bath containing borax* for the super-
hardening of paper prints, and in comparison with existing potassium

* Patent applied for.

Aug., 1933] AN IMPROVED FIXING BATH 139

alum baths, this bath has an exceedingly long hardening life and
sludge life. It was considered that these properties were due largely
to the presence of the boron ion, so that further experiments on the
effect of the boron ion in fixing baths for film emulsions seemed worthy
of investigation.

Borax Fixing Bath

Hypo 250.0 grams
Sodium sulfite (desiccated) 15.0 grams

Acetic acid (glacial) 12.5 cc.

Potassium alum 30 . grams
Borax 7. 5 grams

Water to make 1 . Q liter

Tests have shown (Fig. 1) that the borax fixing bath has a harden-
ing life approximately four times longer than that of the following
F-14 bath (which has been used for many years for deep tank work),
while its sludge life is at least five times greater.

F-14 Fixing Bath

Hypo 300.0 grams
Sodium sulfite (desiccated) 7.5 grams

Acetic acid (glacial) 13.0cc.

Potassium alum 15.0 grams

Water to make 1 . liter

Tests have also indicated that the addition of 2.5 grams per liter
of Rochelle salts to the F-14 fixing bath does not appreciably affect
the hardening properties and doubles the sludge life. The addition
of as much as 5 grams of Rochelle salts per liter decreases the harden-
ing properties and greatly increases the sludge life. Preliminary
tests with the borax formula also indicated that the sludge life was
greatly dependent upon the rate of addition of the developer to the
bath. The bath did not sludge with the addition of 80 per cent of
a strongly alkaline developer if it was added drop by drop with
constant stirring, but if it was added in large quantities at one time,
the bath sludged with the addition of 30 per cent of the developer.
This sludge life was only slightly greater than that of the F-14
fixing bath. The addition of developer drop by drop simulates
practical conditions, because only a small quantity of the developer
is carried into the fixing bath at any one time by the film.

The various factors tested in the determination of the properties
of fixing baths containing boron compounds included the degree of

140 H. D. RUSSELL AND J. I. CRABTREE [J. S. M. P. E.

hardening, the hardening life, the sludge life, the sulfurization life,
and the pH value.

In the determination of the sludge life, a 100-cc. sample of a given
bath was titrated with rapid stirring at a rate of Vz cc. per minute
with increasing quantities of a given developer. The quantity of
developer which could be added to a given bath before the pre-
cipitation of an aluminum sludge occurred was considered as its
sludge life. The samples containing developer were used to de-
termine the effect of developer upon the degree of hardening, the
hardening life, and the pH. The degree of hardening was determined
in a manner similar to that described in previous publications. 2 ' 3
The pH of the baths was determined colorimetrically with La Motte
color standards.

Effect of Varying the Boric Acid Concentration. Although the
borax bath originally contained sodium borate, boric acid was used
in most of the tests because it did not materially change the acidity
of the baths. The effect of the addition of various quantities of
boric acid on the pH, sludge life, and hardening life of the following
F-l and F-2 formulas was determined.

F-l Fixing Bath

Water 1 liter

Hypo 240 grams

When dissolved add:

Water 80 cc.

Sodium sulfite (desiccated) 15 grams

Acetic acid, 28% 48 cc.

Potassium alum 15 grams

F-2 Fixing Bath

Water 1 liter

Hypo 240 grams

When dissolved add:

Water 50 cc.

Sodium sulfite (desiccated) 3 grams

Acetic acid, 28% 18 cc.

Potassium alum 6 grams

The results shown in Fig. 2 indicate that the addition of 0. 1 per cent
of boric acid to either formula does not appreciably change their
sludging and hardening properties. The addition of 0.5 per cent
increased the sludge life approximately four times, with a corre-
sponding increase in the hardening life, but this did not affect the

Aug., 1933]

AN IMPROVED FIXING BATH

141

degree of hardening. The addition of 1.5 per cent also increased the
hardening and sludge life, but decreased the degree of hardening.

The regular baths, F-l and F-2, sludged at a pH of approximately
4.8, when their usefulness as hardening solutions terminated, while
the baths containing boric acid sludged at a much higher pR and
maintained their hardening properties up to a pH of about 6.0.
These tests indicated that about 5.0 grams per liter is the minimum
concentration of boric acid which can be added to these baths if the
most desirable properties are to be maintained.

Effect of Varying the Alum Concentration. Since the composition

IO ZQ "iO 40 5O 6O IO ZO "iO 4-O &O 6O

FIG. 2. Effect of the concentration of boric acid on the properties
of the F-l and F-2 fixing baths.

of the original borax formula was analogous to that of the F-l, with
the exception of the alum concentration, and since the hardening
life of the former was superior, the effect of varying the concentration
of the alum on the properties of the borax fixing bath was investi-
gated. The results shown in Fig. 3 indicate that equal degrees of
hardening were obtained for equal pH values with concentrations of
alum between 6.0 and 30.0 grams per liter. The pH value and the
degree of hardening did not decrease as rapidly with the addition
of developer to the baths containing a high concentration of alum
as in the case of those with a low concentration, which indicates that
the alum increases the total acidity of the baths.

The hydrolysis of aluminum salts into basic aluminum compounds
possibly accounts for the increase in the total acidity of the baths

142

H. D. RUSSELL AND J. I. CRABTREE [j. S. M. P. E.

with an increase in the alum concentration. Basic aluminum com-
pounds have been reported in the literature and their formation
may be represented by the following equation:

y e oowng equaon:
2 (SO 4 ), + 2H 2 O ^ 2A1OHSO 4 + HjSC^

Aluminum , a r~t<. > Basic Aluminum Sulfuri
Sulfate Sulfate Acid

A1 2

ric

The most completely hydrolyzed salt which could be formed would

210

1-iO
90

90
210

no
1^0

90
210

IO GM.
I.O HTE.R,

6.6

4.6

10 ZO 30 40 5O 60 10 10 SO 40 50 CO

CC.OF D-l PER. OO CC. OP" riXlNG* BKTH

FIG. 3. Effect of the alum concentration on the
properties of borax fixing baths.

approach aluminum hydroxide (A1(OH) 3 ) as a limit and intermediate
compounds would have the general formula: AL^OH^SO^

Effect of the Concentration of the Acetate Ion. Preliminary tests
were made in an attempt to replace the acetic acid by other acids,
and the results indicated that the acetate ion was as essential to
the non-sludging properties of the baths as boric acid.

The effect on the hardening properties and sludge life of increasing
the acetate ion concentration by the addition of increasing quantities
of crystalline sodium acetate to the F-l fixing bath containing 5

Aug., 1933]

AN IMPROVED FIXING BATH

143

grams per liter of boric acid was determined. The results given in
Fig. 4 indicate that the addition of sodium acetate in quantities
greater than 25 grams per liter decreases the degree of hardening.

In measuring the sludging tendency the usual method of slowly
titrating the bath with developer was not employed, because under
these conditions none of the baths sludged. In testing the sludging
tendency, 100 cc. of the D-16 developer were added all at one time
to 100 cc. of the fixing bath. The results with this method indicated
that the sludging tendency was not decreased until after 25 grams
per liter of sodium acetate had been added. Since the acetate ions

FORMULA
MVPO

SODIUM suuriTE.

ACETIC ACID

BORIC. /VOID

SODIUM ACETATEI CRVSTAUS

AUUM

WATE.R TO

GM.
I. O LITER

SODIUM ACE1TATE. OO & /-
O " IO

CC. OF" D-l<i RER IOO CC. OF" FIXING

FIG. 4. Effect of sodium acetate on the hardening properties of a
potassium alum fixing bath containing boric acid.

decreased the sludging tendency, it was considered that if a relatively
high concentration of the acetate ion were used, a lower concentra-
tion of boric acid might be used. With the addition of 25 grams
per liter of sodium acetate to the F-l bath, the boric acid could be
decreased to 2.5 grams per liter. This was the minimum quantity
which could be used with any concentration of the sodium acetate
if the desired properties were to be maintained. If a quantity of
sodium acetate less than 25 grams per liter was added, the minimum
permissible quantity of boric acid was approximately 5 grams per
liter.

Effect of the Sodium Sulfite Concentration. An increase in the
sulfite concentration has been shown to increase the sulfurization

144 H. D. RUSSELL AND J. I. CRABTREE [j. S. M. p. E.

life of a given formula. 3 Most of the recommended formulas contain
between 3.0 grams and 15.0 grams per liter of this material.

A variation in the sulfite concentration between these limits had
no effect on the sludging properties of the F-l bath containing boric
acid. Higher quantities of sulfite decreased the developer capacity
of a given bath because it acts as a mild alkali and neutralizes some
of the acid in the fixing bath. The addition of boric acid to a given
bath did not appreciably change its sulfurization life.

Some fixing baths containing a relatively high concentration of
sodium sulfite and acetic acid evolve considerable quantities of sulfur
dioxide when fresh, especially when processing at high temperatures.
In rooms that are not well ventilated, the concentration of this gas
may become great enough to be very objectionable. Baths such as
the F-l containing boric acid can be compounded, in which this ten-
dency is greatly decreased by the substitution of sodium acetate for
a portion of the acetic acid. This produces a bath with a higher
initial pH and a shorter hardening life. However, if half the acetic
acid in the F-l formula were replaced with an equivalent quantity
of sodium acetate, the hardening life would be approximately equal
to that of the F-l formula without boric acid, with the added ad-
vantage that less sulfur dioxide would be evolved.

Substitutes for Boric Acid. Any boron compound which dissociates
in an acid solution to give an oxygen acid of boron may be sub-
stituted satisfactorily for boric acid. In substituting boron com-
pounds in a given formula with chemicals such as borax and the
alkali salts of oxygen acids of boron, it is necessary to add an addi-
tional quantity of acid sufficient to form boric acid or the life of the
bath will be impaired.

Compounds such as glycol borate, glycerol borate, phenyl borate,
and other esters of boric acid may be added to the fixing bath without
the addition of acid. Boron triacetate can be substituted in an
equivalent quantity for both the acetic acid and the boric acid in
the F-l formula.

Substitutes for Acetic Acid. Equivalent quantities of the following
acids were tested in comparison with acetic acid in the F-l formula
containing boric acid: Phenyl acetic acid, chloro-acetic acid, amino
acetic acid, propionic acid, lactic acid, furoic acid, tartaric acid, ma-
lic acid, maleic acid, glycollic acid, diglycollic acid, adipic acid,
succinic acid, butyric acid, fumaric acid, glycero-phosphoric acid,
sulfuric acid, and hydrochloric acid. The only acids which produced

Aug., 1933] AN IMPROVED FIXING BATH 145

baths similar to those containing acetic acid were propionic acid and
butyric acid. The other acids either did not affect the properties or
decreased the hardening properties in a manner similar to tartaric
acid.

A Satisfactory Fixing Bath Formula. The requirements for a
satisfactory alum fixing bath have been stated by Crabtree and
Hartt, 3 and these apply also to the compounding of an alum bath
containing boric acid. In general, 5 grams per liter of boric acid
added to a fixing bath containing hypo, alum, sulfite, and acetic
acid will increase its sludge and hardening life at least 3 to 4 times.
In case a bath with a longer life than this is desired, more acid should
be added. The total acidity of the bath can be increased by the
addition of further quantities of either acetic acid or potassium alum.
Although some alum is required to maintain satisfactory hardening
properties, it is more economical to increase the acidity by the
addition of acetic acid rather thart by the addition of alum.

The alum concentration has very little effect on the degree of
hardening, but in order to produce satisfactory hardening throughout
the useful life of the fixing bath, the concentration should be at least
10 grams per liter. The exact concentration of the alum is not critical
because even in high concentrations potassium alum does not harden
the gelatin to such an extent that there is danger of excessive brittle -
ness of the film.

Since the non-sludging properties are due to the presence of the
borate and acetate ions, it was considered that the minimum con-
centration of boric acid should be 5 grams per liter and that of the
glacial acetic acid, 10 cc. per liter. A lower concentration of either
of these decreases the hardening life and increases the sludging
tendency. A higher concentration of boric acid decreases the sludg-
ing tendency but also decreases the degree of hardening.

The acetate ion concentration may also be increased by the addi-
tion of sodium acetate, which decreases the sludging tendency and
does not affect the hardening life or the degree of hardening unless
an excess is added, such as a quantity greater than 25 grams per liter
of sodium acetate crystals.

The sulfite concentration should be sufficient to prevent sulfuriza-
tion of the bath over a period of at least one month at 70F. A
quantity which usually prevents this is equal to one-half the acetic
acid concentration. Baths containing equal weights of acetic acid
and sodium sulfite have a much longer sulfurization life and are

146 H. D. RUSSELL AND J. I. CRABTREE [J. S. M. P. E.

desirable when they are to be kept for long periods. An excess of
sulfite (a quantity greater than 20 grams per liter of desiccated
sodium sulfite) has no particular advantage, and may decrease the
hardening life because it acts as a mild alkali and neutralizes some
of the acid.

Baths for the processing of motion picture positive film should
produce a uniform degree of hardening without acid revival through-
out the period of processing 400 feet of film per gallon. With
negative films, a life of only 200 feet per gallon is necessary because
the fixing power of the hypo is usually inadequate at this stage of
the exhaustion.

Baths which meet these requirements should not precipitate a
sludge when a 100-cc. quantity is titrated, drop by drop, with 50 cc.
of the developer with which it is to be used. Developers such as
D-16 are moderately alkaline, and do not require a very acid fixing
bath. Baths which are to be used with very alkaline developers
may require as much as 20 cc. of acetic acid with an equal weight of
sodium acetate crystals in order to prevent sludging.

From the above discussion, it is evident that a safe rule to follow
in compounding a satisfactory bath is to add approximately equal
weights of potassium alum, acetic acid (glacial), and sodium sulfite
(desiccated) to a hypo solution with 5 grams per liter of boric acid.
The concentration of the alum, sulfite, and acid should be at least
10 grams per liter and not over 30. Less sodium sulfite may be used
if the bath is to be used within a short period after mixing, such as in
motion picture work. If less sludging tendency is desired, sodium
acetate may be added.

Fixing baths may also be compounded with boric acid which have
a relatively high pH value. Such baths, when fresh, do not evolve
as much sulfur dioxide as the regular baths without boric acid.

Baths such as the F-l and F-2 may be improved in this respect
by substituting one-half the quantity of acetic acid with an equiva-
lent quantity of sodium acetate. For each cubic centimeter of glacial
acetic acid, 2.5 grams of crystalline sodium acetate should be sub-
stituted. The hardening life and sludge life will be approximately
one-half that of the regular baths containing boric acid but will be
as long as that of the regular F-l and F-2 baths without boric acid.

The constituents other than the hypo can not be compounded
in a concentrated hardener such as the F-2a formula because of the
relatively low solubility of the boric acid, but more dilute hardeners

Aug., 1933]

AN IMPROVED FIXING BATH

147

may be prepared with a limiting concentration of 25 grams per liter
of boric acid. In this case, it would be necessary to add one part
of hardener to four parts of hypo solution in order to have 5 grams
per liter of boric acid in the final bath.

TABLE I

Effect of Boric Acid on the Properties of Chrome Alum Fixing Baths

Exp.
No.

Concentration per Liter in 30% Hypo

Per
cent
MQ-25

Melting
Point

Sludge Life

Chrome
Alum
[Grams)

| 0< Jj-" m Sulfuric

?Des> Acid
tJJes.; ,p x

(Grams) (CcJ

Acetic Boric
Acid Acid
(Cc.) (Grams)

Per
cent
MQ-25

Time to
Sludge at
70F.

15 5

110

100

1 Day

5 5

140

100

1 Day

200

110

Immediately

200

110

1 Day

200

110

2 Days

2.5

200

110

7 Days

2.5 5

200

110

10 Days

200

140

>60

10 5

200

140

>60

17.5 2.0

200

(F-23)

200

7 Days

110

Immediately

17.5 2.0

200

190

10 Days

110

1 Day

17.5 2.0

200

180

110

4 Days

Propionic

Acid

(Cc.)

20 10

5.0

190

110

5 Days

The Effect of Boric Acid on the Properties of Chrome Alum Fixing
Baths. It was considered that the addition of boric acid to a chrome
alum fixing bath might also have a beneficial effect on its hardening

148

H. D. RUSSELL AND J. I. CRABTREE [j. s. M. P. E.

and sludging properties. The effect of the addition of boric acid
to the F-23 formula and various experimental formulas is given in
Table I.

The results indicate that (1) the boric acid did not diminish the
degree of hardening (Expts. Nos. 5, 7, 9, 19, 22, and 25) ; (2) the
hardening life was not extended (Expts. Nos. 20, 21, 23, 24, 26, 27);

210

180
j 150
J 110

**'

2 180

IZO

180

-t PlMNG* BATH WIT.HOUT AC\O REVIVAL.
4.5
5.0
5.5

S=SLUDG,E1

-Z F-|XNC BATH WITH ACID REVIVAL.

R R R R R

R = REVIVAL. WIT
ACETIC ACID

5.O
R

6.0

REVIVAI_ WITH ACID

I _ I _ I _ I

BORIC ACID FIXING. BATH WITHOUT ACID REVIVAL.
4.5

-5.0 -

.0-

rEET PER GAL.L.ON
I I I I I

IZO

IOO ZOO aOO AOO IOO ZOO ^OO 100 &OO

& MIN. D-IQ> AND F-XEO O MIM.

FIG. 5. Effect of exhaustion on the hardening properties and H
values of various fixing baths. (D-16 developer.)

and (3) the addition did not increase the sludge life but prolonged the
time before sludging occurred (Expts. Nos. 6, 8, 10).

Since the addition of boric acid improves only those potassium
alum baths which contain acetic, propionic, or butyric acids, tests
were made to determine whether satisfactory chrome alum baths
could be compounded containing boric acid and one of these acids.
Previous tests 2 have indicated that acetic acid decreases the harden-
ing properties of chrome alum solutions and decreases the sludging
tendency.

Experiments Nos. 7 to 18, inclusive, indicate that the addition of

Aug., 1933] AN IMPROVED FIXING BATH 149

boric acid did not improve the properties of baths containing acetic
acid. However, acids may ultimately be found which will produce
a chrome alum bath having the desirable properties of a potassium
alum bath containing boric acid, and further work in this connection
is in progress.

Theoretical Considerations. A regular potassium alum fixing bath
such as the F-l will precipitate a sludge on the addition of a de-
veloper at a pH of approximately 4.8. The sludge consists of a
mixture of aluminum hydroxide and aluminum sulfite, the exact pro-
portion of each depending upon the conditions of precipitation.

Certain organic acids can be added to the bath and thereby increase
the pYL value at which sludging occurs. Acids such as citric and
tartaric behave in this manner, but they also decrease the degree of
hardening. Citric acid, undoubtedly, prevents the precipitation
of the sludge by forming complex ions with the aluminum ions,
which, in turn, are not hardening agents for gelatin.

The addition of boric acid to an alum fixing bath containing acetic
acid, also increases the sludge life, and may extend the hardening
life up to the sludging point. The addition of boric acid does not
produce an absolutely non-sludging bath; it only extends the pH
range over which sludging does not take place.

The hardening action of ordinary fixing baths usually decreases
rapidly at pH values greater than 5.0, but it is possible to compound
boric acid baths which have satisfactory hardening properties at a
pH value as high as 6.5 and which do not sludge at pH values less
than 8.0.

Any theory which explains this action must account for the fact
that boric acid produces the above properties only in a bath con-
taining potassium alum, sodium sulfite, acetic acid, and hypo, or
their equivalents. In this case, the aluminum sludging is probably
prevented by the formation of complex ions between the alum, the
acetic acid, and the boric acid.

Complex ion formation between hydroxy organic compounds and
boric acid has been investigated by Boeseken and Vermaas, 5 who
state that "when a diol or (a-oxy acid) is added to a solution of

boric acid in water, the compound R<T /BOH is primarily formed,

which, on account of the tendency of boric acid to assume the penta-
valent condition, can pass into the following mono-basic acids:

150 H. D. RUSSELL AND J. I. CRABTREE [J. s. M. P. E.

H R/ \B(OH)j or

R = any favorable diol or (a-oxy acid).

These acids are strong and are analogous to the acid HB(OH) 4
which has been shown to account for the acidity of a boric acid solu-
tion. 6

Aluminum hydroxide, in this case, may be considered as a favorable
diol and forms the following compound with acetic acid and boric
acid: 2A1(OH) 3 + 2CH 3 COOH + H 3 BO 3 ^

/Ov O

Al/

-CH S C

5H 2 O

The formation of this compound possibly does not take place
until aluminum hydroxide begins to form, which accounts for the
fact that a sludge can be obtained in the boric acid baths when
relatively large quantities of developer are added immediately.

In applying the mass law to the above equation, it is seen that an
increase in the concentration of borate ion or acetate ion would
increase the concentration of the compound and, therefore, produce
a bath with less tendency to sludge. This is in accord with experi-
mental facts.

SUMMARY

(1) The effect of varying the concentration of the various con-
stituents of a potassium alum fixing bath containing boric acid has
been investigated.

(2) The addition of boric acid to a fixing bath such as the F-l
increased the hardening life approximately four times by extending
the pH range over which hardening occurred and also by preventing
sludging.

(3) The sludging tendency of a given bath containing boric acid
can be decreased by increasing the concentration of the acetate ion
or borate ion. Five grams per liter of boric acid and 10 cc. per liter
of acetic acid was considered the minimum concentration of these
two constituents.

Aug., 1933] AN IMPROVED FIXING BATH 151

The sludge life of a given formula is determined largely by the
manner in which the developer is added. If a relatively large quan-
tity of developer is added rapidly, the bath will sludge immediately;
but if the developer is added drop by drop, the total quantity which
can be added before sludging occurs will be very much greater. This
latter condition simulates that existing in practice.

In formulas such as the F-l containing boric acid, the addition
of 25 grams per liter of sodium acetate decreases the sludging tendency
to such an extent that large quantities of D-16 may be added rapidly
without sludging. Concentrations of sodium acetate crystals
greater than 25 grams per liter decreased the hardening properties.

(4) An increase in the alum concentration increased the total
acidity of a given bath and, therefore, increased the hardening life
and sludge life.

(5) Any compound which dissociates in an acid solution to
produce an oxygen acid of borori can be substituted for boric acid.
Such compounds include borax, sodium metaborate, boron tri-
acetate, and esters of boric acid.

(6) Substitutes for acetic acid include boron triacetate, pro-
pionic acid, and butyric acid.

(7) Satisfactory fixing baths containing hypo and 5 grams per
liter of boric acid can be compounded by the addition of equal
parts of potassium alum, acetic acid (glacial), and sodium sulfite
(desiccated). For general use, the concentration of each of the
latter ingredients should be at least 10 grams per liter and usually
not more than 30. If a bath with a less sludging tendency is desired,
sodium acetate may be added. Also, if the bath is to be exhausted
within a short period after mixing, the quantity of sulfite can be
decreased one-half.

(8) The non-sludging properties of a potassium alum bath con-
taining boric acid have been ascribed to the formation of complex
ions between the alum, acetic acid, and boric acid.

(9) The addition of boric acid to chrome alum baths containing
either sulfuric or acetic acid or a mixture of the two does not affect
the hardening properties and does not increase the sludging life,
but merely 'retards the time required for sludging to occur.

PRACTICAL RECOMMENDATIONS

The following fixing bath formula is recommended for use with
motion picture positive and negative films.

152 H. D. RUSSELL AND J. I. CRABTREE [J. S. M. P. E.

Metric Avoirdupois

Hypo 300.0 grams 300 Ibs.

Sodium sulfite (desiccated)* 5.0 grams 5 Ibs.

Acetic acid (glacial) 10.0 cc. 1 gal. 26 oz.

Boric acid 5.0 grams 5 Ibs.

Potassium alum 10.0 grams 10 Ibs.

Water to make 1 . liter 120 gals.

Dissolve the hypo in one-half the required volume of water and then add the
remaining chemicals in the order given after dissolving in a small quantity of
water. Dilute with water to the required volume.

* This bath contains a minimum quantity of sulfite, which is such that the
bath will not sulfurize within a period of 3 to 4 weeks at 70 F. If the tem-
perature is apt to rise above 70 F., twice the quantity of sulfite should be used.

A comparison of the hardening properties and pH values obtained
during the exhaustion of the F-2 formula (with acid revival) and
the boric acid bath (without revival) is shown in Fig. 5. The baths
were exhausted with motion picture positive film which was de-
veloped in D-16 but not rinsed before placing in the fixing bath.
The F-2 formula was revived at intervals corresponding to 50 feet
of film per gallon with 7.5 cc. of glacial acetic acid.

During the exhaustion of the F-2 formula, a considerable change
in hardening properties and H value occurred when the bath was
revived with acid. The hardening life with acid revival was ap-
proximately 500 feet of film per gallon. Without acid revival, the
life was less than 100 feet per gallon. The degree of hardening of
the boric acid bath did not change during the processing of 400 feet
of film per gallon, but it decreased rapidly after 500 feet of film had
been processed.

A similar test was made by exhausting the boric acid bath with
motion picture panchromatic negative film developed in D-16.
The life of the bath in this case without rinsing the film was 400 feet
per gallon.

The degree of alkalinity of the fixing bath can be determined
readily by the addition of 1.0 cc. of a 0.04 per cent aqueous solution
of bromo-cresol-purple to a 10.0-cc. sample of the fixing bath. When
the bath is fresh, the color of the solution will be faintly yellow,
which will deepen during exhaustion, and finally have a distinct
reddish tint when the bath is exhausted. This occurs at an ap-
proximate pH value of 5.6. Although most of the boric acid baths
will harden the gelatin up to a pH value of 6.5, it is desirable to

Aug., 1933] AN IMPROVED FIXING BATH 153

discard the bath at this point because most of the acid has been
neutralized, and further exhaustion may produce developer stains.

The addition of boric acid to a potassium alum bath does not
interfere with the efficient recovery of silver from an exhausted
bath by the electrolytic process, and tests have indicated that the
wear and tear properties of film fixed in the above bath are not
inferior to those of film fixed in the F-2 formula.

Except for use under tropical conditions, it is considered that the
recommended boric acid formula which does not require revival
with acid during exhaustion, is to be preferred to most chrome alum
fixing baths. In order to maintain their hardening properties,
chrome alum baths require revival with sulfuric acid at intervals
during exhaustion, and for best results the precise quantity of acid
to be added must be determined by chemical analysis.

REFERENCES

1 CRABTREE, J. I., AND RUSSELL, H. D.: "Some Properties of Chrome Alum
Stop Baths and Fixing Baths (Part I), Chrome Alum Stop Baths," J. Soc. Mot.
Pict. Eng., XIV (May, 1930), No. 5, p. 483.

2 CRABTREE, J. I., AND RUSSELL, H. D.: "Some Properties of Chrome Alum
Stop Baths and Fixing Baths (Part II), Chrome Alum Fixing Baths," /. Soc.
Mot. Pict. Eng., XIV (June, 1930), No. 6, p. 667.

3 CRABTREE, J. I., AND HARTT, H. A.: "Some Properties of Fixing Baths,"
Trans. Soc. Mot. Pict. Eng., XIII (1929), No. 38, p. 364.

4 SHEPPARD, S. E., ELLIOT, F. A., AND SWEET, S. S. : "The Chemistry of the
Acid Fixing Baths," J. Frank. Inst., 196 (1923), p. 45.

6 BOESEKEN, J., AND VERMAAS, N. : "On the Composition of Acid Boric
Acid-Diol Compounds," J. Phys. Chem., 35 (1931), p. 1477.

6 HERMANS, P. H.: "Uber die Konstitution der Borsauren und einiger ihrer
Derivate" (On the Constitution of Boric Acid and Its Several Derivatives),
Zeit anorg. allgem. Chem., 142 (1925), p. 83. Also BOESEKEN, J., AND COOPS, J.:
"The Use of Boric Acid for Determining the Structure of Various Organic Com-
pounds; I The Dissociation Constants of Various Acids in the Presence of
Boric Acid," Rec. trav. chim., 45 (1925), p. 407.

A PRACTICAL METHOD AND PHOTOMETER FOR
CONTROLLING EXPOSURES IN PHOTOGRAPHY*

M. LASKY AND B. RUBIN**

Summary. The need of determining accurately the brightness of subjects to be
photographed so as to enable the cinematographer to determine the optimum exposure
of the film in motion picture cameras by instrumental means rather than by visual
estimate, is discussed at some length; and the errors inherent in the visual methods are
shown to be very considerable.

An exposure meter known as the Graphometer has been designed for the purpose of
enabling the cinematographer to determine the proper exposure of the film; it operates
on the photoelectric cell principle and may be used with a standard camera. A brief
discussion of the method of using the Graphometer is included.

Much progress has been made in the technical phases of the
motion picture industry, largely as a result of the introduction of
more precise methods of measurement and control in various opera-
tions relating to the handling of film. Camera exposures, however,
are still determined mainly by individual judgment because meters
for measuring exposure have not been adopted to any extent by the
majority of cameramen.

Motion picture technicians, however, are convinced of the necessity
of controlling exposure quite accurately. The rigid requirements
imposed by sound-on-film processing, together with the demand for
the utmost in pictorial quality and economy of operation, have
created a situation in which indefinite methods of determining photo-
graphic values and the attendant complications and waste can no
longer be tolerated.

In the laboratory, the inertia of technicians toward changes of
method was overcome by the advent of sound, which necessitated
a precise determination of the various factors involved: exposure
of the negative, development, timing of the print, development of
the print, etc., resulting in a fuller appreciation of the value of sensito-
metric methods and their general adoption by laboratory technicians.
In the case of exposure of the original picture negative, however,
the tendency has been, oddly enough, away from standardization.

* Presented at the Spring, 1932, Meeting at Washington, D. C.
** New York, N. Y.
154

PHOTOMETER FOR CONTROLLING EXPOSURES 155

Research has been in the reverse direction. Film emulsions are
produced with more latitude; negative developing formulas and
developing speeds are varied; printing machines are designed with
a greater number of light changes in order to compensate for errors
occurring in negative exposure because of human variability. More-
over, it is obvious that errors of judgment must occur. Instead of
assisting sensitometry in its general purpose, such measures, due to
their interdependence, conduce to a sacrifice of quality and complicate
laboratory operations, with resultant economic disadvantages.

In so far as the cinematographer is concerned, the predetermination
of light intensity is of great importance. A method by which the
camera operator could determine the intensity of the actinic light on
an object or scene would be extremely desirable, and would enable
him to confine the exposure of the film to the straight-line portion of
the characteristic curve of the emulsion, so as to maintain a constant
negative density, reproducible with fidelity from the highest light to
deepest shadow; it would also enable him to predetermine light
effects accurately, and have them reproduced with the light pro-
portioned as desired. In short, it' would provide a definite relation
between subject brightness and negative density.

Up to the present time the use of photometers has in general been
regarded apathetically, due mainly to a lack of appreciation of
sensitometric principles. In addition, the various kinds of photome-
ters or exposure meters available have been either impracticable
or inaccurate. Let us consider briefly some of these exposure meters
and the method employed to indicate the correct exposure.

The photoelectric type of instrument consists of a light-sensitive
cell, such as selenium, cuprous oxide, etc., and a meter calibrated
in / values, corresponding to lens diaphragm markings. In use, the
instrument with the photoactive surface is directed toward the
scene to be photographed, the light falling upon this surface causing
a physical change in the light-sensitive element which, in turn, actu-
ates the indicator of the meter. In order to appreciate the faults
inherent in such methods, the following illustration may be taken:

Two cards, A and B, Fig. 1, are equally illuminated. These cards
may represent light and dark sections of the scenes to be photographed.

As can be observed, the photographic brightness values of cc and
dd are equal. In other words, the photographic contrast in the two
cards is the same. Using the kind of exposure meter just mentioned,
a reading is made of card A. Since the white area of this card is

156

M. LASKY AND B. RUDEST

[J. S. M. P. E.

large, most of the light falling on it is reflected to the sensitive surface
of the photometer, and a reading is obtained indicating a lens dia-
phragm setting of //1 6 as the proper exposure. Since the small
white area of card B reflects proportionately less light than A, a
measurement would indicate a lens diaphragm setting of //2. Ac-
cordingly, one value of exposure would be indicated as 64 times that
of the other, yet both cards are equally illuminated and their photo-
graphic brightness contrasts are identical. Obviously, this method
of determining illumination for photographic purposes will not
suffice if precise measurements are required. Further, the possible

CARD A

CARD B

FIG. 1. The total brightness measurement of these two subjects
illuminated with the same intensity would indicate exposures in
the ratio of 64 to 1.

differences in light transmission of lenses having the same / rating,
and the decrease in relative aperture as the lens is focused at closer
ranges than infinity, precludes the possibility of obtaining accurate
results by any method which does not take into account these factors.

It can therefore be seen that any instrument designed to indicate
photographic exposure accurately demands consideration of the
following:

(1) Brightness of the Subject. For fidelity of tone, exposure
depends upon the brightness of the subject, not upon the quantity
of light impressed on the scene. Suppose it is planned to photograph
a black card; and then, using the same illumination and lens aperture,
to photograph a white card. In order to achieve a faithful repro-
duction of both cards, retaining the brightnesses in the same pro-

Aug., 1933] PHOTOMETER FOR CONTROLLING EXPOSURES 157

portions as in the original, both cards will require the same exposure.
The use of a highly reflective setting for these cards will simply
reduce the exposure needed for both. However, a knowledge of the
total amount of light coming from the same is of no use, as it does
not indicate the brightness of any one part; it is necessary to know
the brightness of definite parts of the subject.

(2) Control of Exposure through Lens to Be Used for Photographing
to Insure against Unknown Transmission Values. The light-passing
power of a lens may vary greatly from long shot to close-up because
of the change of relative aperture. Very frequently variations also
occur in the speed of lenses that are similarly marked, due to the
peculiarities in manufacture.

(3) Facility and Speed of Operation. In modern swift picture
production, it is desirable that measurements take but a few seconds
and that the instrument be easy to operate. This is of utmost im-
portance inasmuch as time is all-important from the economic stand-
point.

(4) Constancy and Dependability of the Instrument Used. Any
instrument to be used by cinematographers should be extremely
portable. In addition, frequent changes in picture making locale
necessitate that the instrument be unaffected by atmospheric condi-
tions or ambient temperatures. It must also retain its accuracy for
a reasonable period of time with normal usage.

The above requirements have been embodied in an instrument
known as the Graphometer. Its principles and dependability have
been proved under actual working conditions for more than a year,
during which time many suggestions for improvement in design and
operation of the instrument were offered by cinematographers and
technicians.

In its present form, the Graphometer is used in connection with
the Mitchell camera. It is possible, however, to make this unit
an integral part of the camera. Fig. 2 shows the design and place-
ment of the parts.

The meter may be calibrated for various types of emulsion in
terms of per cent of normal exposure or in light intensities, degrees,
etc. The power is obtained from a small 22V2-volt battery. Com-
pensation for depreciation of the battery is provided by means of
an adjustable rheostat.

In operation, the cinematographer proceeds in the usual manner
to set up, focus, and arrange the lights in their approximate positions

158

M. LASKY AND B. RUBIN

[J. S. M. P. E.

for the particular kind of lighting desired. Having focused the lens
to be used, the camera head is moved to the right to its farthest
position, and the Graphometer is placed directly behind the lens.
Guide rails on the instrument assure alignment of the focusing screen
in the same position behind the lens as the photographing aperture.
With the lens covered to exclude light from the photocell, the battery
is switched on and the meter indicator is set to read zero by adjusting

FIG. 2. Diagram of Graphometer; 1, focusing
screen ; 2, adjustable mask with an aperture ; 3, photo-
cell; 4, sighting tube; 5, photocell amplifier; 6, meter.

the rhoeostat. The light values are then measured as follows: a
white card (which is used as the standard of radiation or reflection),
20 inches square, is held by an assistant at any point in the scene
at which the light value is to be ascertained. Looking through the
sighting tube, the operator adjusts the moving mask so as to receive
only the image of the card in the 1 -millimeter opening. The photo-
cell is then brought into position, to receive the light admitted from
this opening, and a direct meter reading is obtained. Any deviation
from the normal may be compensated for by an increase or decrease
of illumination or adjustment of the lens diaphragm. Additional

Aug., 1933] PHOTOMETER FOR CONTROLLING EXPOSURES 159

readings can be taken at any other points in the scene and corrections
similarly made.

When photographing with a large and uniform light source, such
as daylight, only one reading is necessary to adjust the lens dia-
phragm properly. The occasion frequently arises, however, when
it is necessary to determine light values of sunlight and shade where
both are included in the scene. In such cases, the operator can favor
with accuracy either side, or choose an intermediate balance for the
best reproduction of both sides.

It has been found for all practical purposes that by using neutral
density filters, the same meter calibrations can be used for color
filters and different emulsions. For example, if the meter is cali-
brated for normal, using a supersensitive emulsion, and it is desired
to use a film with only half the speed, a neutral density filter having
a transmission value of 50 per cent may be slipped over the lens
when taking the reading, and the same normal point on the meter
would indicate the correct amount of light for that emulsion. Before
photographing, the filter must be removed from the lens. The light
can also be equalized when using color filters by the same procedure.
If it is desired to use a color filter having a factor of 10, a neutral
density filter transmitting only 10 per cent of the light is employed;
the lens diaphragm is adjusted so that normal is indicated on the
meter; then the neutral density filter is replaced by the color filter
before photographing. Corrections for differences in actinic values
of incandescent and sunlight can also be made by this method.

Since all measurements are made through the lens at the relative
aperture at which photographing is to take place, all the conditions
actually existing in the scene are considered.

Normal exposure as here referred to is meant to indicate the
exposure at which faithful reproduction is obtained for all the shades
and gradations, ranging from black to white, which were present in
the original scene. Variations of development factors in different
laboratories make necessary an initial test to determine the proper
normal dial setting for the laboratory that is to process the negative.
Practical determination of this point is best accomplished by making
a series of exposures of a light wedge or gradation chart, and noting
the meter readings for each exposure. After normal development,
the meter reading corresponding to the perfect exposure is set as the
normal. It will then be found possible to photograph the gradation
chart under different light conditions on the same film strip and at

160 M. LASKY AND B. RUBIN

varying camera distances, and maintain the negative density con-
stant throughout.

In some tests recently conducted over a period of months and under
light conditions that required lens diaphragm settings ranging from
f/2 tof/22, a negative was obtained of such consistency that all takes
were printed on one light intensity. These tests were made re-
peatedly with both incandescent lighting and sunlight with the same
results.

Unquestionably the application of sensitometric principles in
studio practice will effect much needed economies such as elimination
of light tests and the time consumed in making them; the main-
tenance of illumination in accordance with actual requirements with
consequent savings of current and equipment; elimination of retakes
because of under- or over-exposure; standardization of developer
compositions and speeds; elimination of waste and complications in
print processing; elimination of tests and retakes in trick and process
photography, etc. It is also practicable to combine picture and
sound track on the same negative without loss in quality of either.

Although these economies are vitally necessary, of equal importance
is the fact that only by sensitometric methods can full advantage
be taken of the photographic art; only by the definite knowledge
which these methods provide can the utmost in pictorial beauty
be attained.

SOUND RECORDING AND REPRODUCING USING
16-MM. FILM*

C. N. BATSEL AND J. O. BAKER**

Summary. This paper deals with the various problems involved in applying
sound to 16-mm. film and points out the advantages of the particular type of film
that was chosen. There are two general methods of producing 16-mm. positives,
the direct and the indirect. The direct method involves the recording of sound directly
on a 16-mm. negative, from which is printed a 16-mm. positive either by contact
or by the reversal process. The indirect method concerns various ways of obtaining
16-mm. positives from 35-mm. film. The problems involved in printing and repro-
ducing 16-mm. sound films are discussed.

The use of sub-standard film in the motion picture industry dates
back to 1897, but it did not gain much popularity until 1923. It
was at that time that the Eastman Kodak Company perfected the
reversal processing of their 16-mm. film, and announced it to the
trade together with the first hand-powered Cine- Kodaks and motor
driven Kodascopes. The Cine-Kodaks used the 16-mm. safety
film, and the daylight loading feature proved to be very popular
with the amateur.

Other, attempts have been made to popularize motion pictures
in the home; but none of them gained very much momentum,
lacking the desirable simplicity of operation of the Eastman camera
and the assurance of a good percentage of acceptable pictures with
the reversal process. There were no film libraries and, as a result,
the attempts to interest the amateur without the means of mak-
ing the films were soon forgotten. Some rapid changes took place
during the five years following 1923. In 1926, the Eastman Kodak
Company announced a spring-driven Cine-Kodak, and about the
same time the Bell & Howell Co. announced their 16-mm. line.
In 1927, film libraries began to spring up. Educational and indus-
trial pictures began to make their appearance along with 16-mm.

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Photophone and Applications Div., Engineering Dept., RCA Victor Co.,
Inc., Camden, N. J.

161

162 C. N. BATSEL AND J. O. BAKER [j. s. M. P. E.

versions of news items. In 1928 the Eastman Kodak Company
announced the Kodacolor process for 16-mm. film.

With these new developments, the public and the professions
quickly adopted the 16-mm. film, and its use became general in the
medical, educational, amusement, and business promotional
fields. Its popularity has continued to the present time, with only
little slackening due to the depression.

In 1928, it became apparent that the addition of sound to 16-mm.
movies would be the next great step; consequently, work was
begun by the Radio Corporation of America for the purpose of
accomplishing this. The efforts were restricted to sound-on-film
apparatus. At the same time a number of companies were an-
nouncing synchronized sound-on-disk for 16-mm. film. These
equipments consisted of 33 l / s rpm. turntables geared to the 16-mm.
projector, and employed the 33 l /z rpm. theater disks. The film
was reduced from the 35-mm. film. Some of the more elaborate
ones were mounted in cabinets. The equipment never became popu-
lar, as both the films and records were expensive and hard to obtain.

It is appropriate to mention here some of the problems that had
to be overcome to produce successful 16-mm. sound-on-film equip-
ment, inasmuch as after the adaptation of sound to 35-mm. film it
would seem that, by applying the same principles to 16-mm. film,
good results could likewise be achieved. Such is far from the case
In the first place, the linear speed of the narrow film had to be less
than that of the 35-mm. film if the usual 24 pictures per second
were to be projected. This was desirable in order not to use more
film than necessary and at the same time to maintain the same
proportions in the 16-mm. pictures as in the original 35-mm. pictures.
At this greatly reduced speed the fact that 16-mm. film has only
forty sprocket holes per foot, instead of 64, presented the very
difficult problem of providing a mechanical filter that would suc-
cessfully eliminate "wows," and at the same time eliminate the 24-
cycle sprocket tooth flutter. This problem was finally solved, and
16-mm. sound reproduction that is practically free from "wows"
and flutter is now attainable.

Not only did the low speed offer new filter problems, but it placed
a serious handicap on the frequency response. When one considers
that with a ratio of 2 l / z to 1 in speed the 4000-cycle wavelength on
the 16-mm. film is equal to the 10,000-cycle wavelength on the
35-mm. film, it can readily be seen with what the engineers had to

Aug., 1933] SOUND RECORDING AND REPRODUCING 163

contend. The film companies cooperated by improving the acetate
base and the emulsion, and in the latter part of 1932 perfected a
16-mm. sound negative that gave very good results. By suitably
equalizing the electrical circuits, it was possible to attain with this
film, on standard studio recorders and re-recorders, 16-mm. sound
track, the response of which is reasonably flat up to and even beyond
4000 cycles.

A third problem that had to be overcome before 16-mm. equip-
ment could be offered commercially was that of standardizing the
film. The RCA Victor Company in conjunction with several
others and one leading film company took the stand that the size
of the 16-mm. picture frame should not be reduced, as had been
done in the case of 35-mm. film when the sound track was added.
By eliminating one row of sprocket holes it was possible to place a
60-mil sound track on the film and retain the same picture size as
on the old 16-mm. silent film. Such a procedure presented certain
difficulties in printing, which will be mentioned later, but by re-
taining the old picture dimensions, the projectors could project
either the old silent films, many of which are valuable, or the new
sound films without having to change apertures in the projectors.
By thus using the sprocket hole space it was possible to record a
sound track of sufficient width to achieve a satisfactory ratio be-
tween the recorded sound and the noises inherent in the film and
amplifiers.

After producing satisfactory recordings and sound projectors the
next problem was to devise ways of making the 16-mm. prints
available to the purchasers of the projectors. There were two
alternatives: one of making all 16-mm. sound film directly, the
other of making 16-mm. prints from the existing 35-mm. film.
Both plans have been followed.

We shall now consider various methods of producing 16-mm.
sound film from the 35-mm. sound film. The first is to reduce
both sound and picture optically, by means of a continuous printer.
This is not easy to do with existing printers, due to a slight slipping
between the two films which causes the high frequencies in the sound
track to be lost. Also, no corrective compensation can be applied
between the 35-mm. and 16-mm. films. The difference between
the sizes of the pictures also prevents the simultaneous reduction of
sound and picture. If the proper kind of 35-mm. recording is
available we can now optically reduce the sound track satisfactorily.

164 C..N. BATSEL AND J. O. BAKER [J. S. M. P. E.

Another way that was considered was to use a combination step
reduction printer and re-recorder. While such a method would have
been satisfactory, the machines proved to be expensive and were
not adapted to the general layout of modern printing and processing
laboratories.

The most practicable plan that Has been developed up to the pres-
ent is to re-record the sound from a 25-mm. sound print on a 16-mm.
sound negative. The picture is reduced from the 35-mm. negative
to the 16-mm. print in a step printer, and the sound is printed by
contact on this print either before or after the picture is reduced.
The printing and processing then present no greater problems to
the laboratory than the printing and processing of any other sound
film. By restricting the re-recording to the recording studios, the
experience of the sound recordists in the 35-mm. sound* field can be
applied to the production of 16-mm. sound-film. When properly
compensated by an experienced recordist, good 16-mm. sound is
obtained from 35-mm. film which if optically reduced would be a
failure. After obtaining the 16-mm. negative, it may be printed in
any laboratory equipped to print and process 16-mm. sound film.

For making the 16-mm. recordings directly, the 35-mm. recording
channels have been duplicated, properly compensated for the differ-
ence in the frequency response of the 16-mm. and 35-mm. films.
These recorders can be synchronized with a 16-mm. camera, if
desired, or can be used to score existing 16-mm. films or add sound
to 16-mm. versions of the silent 35-mm. films. This method of
making 16-mm. film is now used extensively in making advertising,
educational, and other commercial film. It is fairly expensive,
however, and does not solve the amateur problems. In order to
popularize 16-mm. sound film among the vast army of amateurs and
others who might want to make 16-mm. sound film pictures, a com-
bination 16-mm. camera recorder has been developed.

Such a camera must be small and light in weight. The addition
of sound must not complicate the operation of the camera, and the
entire mechanism must be practically fool-proof. It must record
sound acceptably, and adhere to the accepted 16-mm. dimensions,
so that its product can be reproduced by standard reproducers.
In other words, the problem was to make a sound recording
camera very little larger than the existing silent cameras, and
to design it so that any one familiar with a silent camera could,
with a little instruction, make good sound motion pictures.

Aug., 1933] SOUND RECORDING AND REPRODUCING 165

As a result of a large amount of developmental work the RCA
Victor Company has produced a complete sound recording camera,
the over-all dimensions of which are 7 l /z inches high, 8 3 /ie inches long,
and 5 u /ie inches wide. Its weight is approximately nine pounds.
It is equipped with a spring motor, capable of running twenty-five
feet of film, which can be increased to 50 feet if desired, at one wind-
ing. It is equipped with a revolving turret head, telescope finder,
and a variable footage indicator. It has a one-piece removable door.
The film is threaded in the camera as simply and easily as in the
ordinary silent camera, over only one sprocket and through the
picture gate, and the only loops to be formed are those on either side
of the picture gate. It accommodates either 100- or 50-ft. rolls of
daylight loading reversal film. The entire recording optical system
has been reduced almost to the size of an average ink bottle, and
will record frequencies of a range sufficient for good reproduction
of speech.

There are to be two types of the camera: namely, the autophone
and the microphone. The autophone type is extremely portable
and simple to operate, and is designed to be used in difficult places,
and for making pictures to be described by the operator of the
camera. No amplifier is used, the recording mirror being actuated
by the sound waves from the speaker's voice. The current for the
recording light is supplied by three small flashlight batteries con-
tained within the camera.

The electrical type is identical in so far as the mechanical con-
struction is concerned, but contains an electrically driven galvanome-
ter and includes an amplifier and microphone as part of the equip-
ment. It is, of course, not as portable or as easy to operate as the
autophone type, but is much more flexible, in that in addition to
being able to do everything that the autophone can do, it can be
used for recording amateur plays, news items, and all events where
sounds emitted by the subjects are important to the picture.

Work has been done by the film companies, for the purpose of
producing a suitable film for use in this equipment. The earlier
reversal films were very poor for recording sound, so that a great
deal of experimentation was required of the film manufacturers in
order to produce emulsions that would provide good results with
sound and at the same time not impair the photographic qualities
now claimed for the reversal films. Sound films are now available,
perforated on one side, that meet these requirements.

THE USE OF MAZDA LAMPS FOR COLOR PHOTOGRAPHY*

R. E. FARNHAM**

Summary. By operating the filaments of high-efficiency tungsten lamps at higher
voltages than normal, not only are the operating characteristics of the lamps improved,
but a more desirable spectral energy distribution for color photography is attained,
the increase of the blue-violet radiation amounting to about 140 per cent of the increase
of the red-orange radiation. In view of the fixed voltages used on sets, the lamps are,
in effect, designed for lower voltage and wattage. Data on the candle-power distribu-
tion with suitable reflectors are presented.

The recent commercial availability of at least two satisfactory
systems of three-color motion picture photography will undoubtedly
result in a more extensive production of colored motion pictures than
has taken place in the past.

Since all the motion picture studios are equipped with incan-
descent lighting equipment, and the advantages of high-efficiency
tungsten lamps, such as reduced labor costs, inherent quietness,
accuracy of light control, etc., are well recognized, the studios are
naturally interested in the application of this light source to the color
photographic processes.

In the case of black-and-white photography using panchromatic
film we employ an emulsion that is highly sensitive at the blue- violet
end of the spectrum and somewhat less sensitive toward the red-orange
region. The Mazda lamps have their greatest radiation in that part
of the spectrum where most light is required and less output where
the film is most sensitive. This combination results in utilizing the
panchromatic film to its full advantage, and gives excellent rendering
of the faces and properties in the sets.

The color photographic processes require a light source having
more nearly equal proportions of all colors as well as approximately
fifty per cent greater illumination intensity than is necessary for black-
and-white photography. The simplest method of obtaining light
of the required quality is to employ a suitable filter, either at the lens

* Presented at a meeting of the Pacific Coast Section, June 15, 1933.
** Engineering Dept., General Electric Co., Cleveland, Ohio.

166

MAZDA LAMPS FOR COLOR PHOTOGRAPHY

167

or in front of the lighting equipment to absorb the excess red-orange-
yellow light and thus more nearly approximate a white light.

However, the amount of blue-violet radiation required is such as to
necessitate an excessively high wattage when lamps of the usual
studio type are employed, and the process becomes an extremely
wasteful one. Hence the desirability of having lamps which produce
not only a greater proportion of blue-violet light as compared to red-
orange, but a greater volume of light for a given wattage, since the
amount of heat on a set is largely determined by the wattage em-
ployed.

It is generally known by those familiar with the behavior of in-
candescent lamps that if the voltage applied to a lamp is increased,
the light output increases at a much faster rate than either the voltage ;

FIG. 1. Total spectral energy distribution of Mazda lamps.

or wattage ; in other words, the amount of light per watt is increased,
also the amount of blue-violet radiation increases at a greater rate
than does the red-orange light. For a more complete presentation
of the relationship of lamp voltage-life-light output and temperature,
the reader is referred to the paper on this subject by M. W. Palmer
and E. W. Beggs. 1

Referring to Fig. 1, showing the total radiation, and Fig. 2, showing
the visible output from two incandescent lamps, one operating at
2920K. (21 lumens per watt), typical of the lamps now used for
general lighting service in the studios, and the other at 3435K.
(33 lumens per watt), the photoflood type, it is evident that the

168

R. E. FARNHAM

[J. S. M. P. E.

increase in radiation in the violet region (4000-4500 A) is 270 per
cent while the increase in the red (6300-7400 A) is 55 per cent.
Based on an equal quantity of red radiation for the two lamps, the
increase in the violet is 140 per cent.

Thus by the simple process of operating the lamps on overvoltage,
we receive not only a greatly increased volume of light but light of an
improved color quality, applicable to color photography with minimum

^ - 1

70
00
i 50

35 ?<

jrtt-

ffCa

. ,"-

* t j'

Of'K

'/*

- I-'-

;~< : '-

' ''r'

_ 1

Grffn

v "

"f^*

TOO JCOO COOO 7<XW

s. 2. Visible spectral energy distribution of
Mazda lamps.

wattage. To be sure there is still some excess red-orange, which must
be filtered out, but the losses are very much reduced. Such filtering
can be accomplished by adjustment of the apertures in front of the red-
green-blue filter, by a color filter in front of the lens, by color screens
incorporated in the lighting equipment, or even the use of a colored
glass bulb for the lamp. The first two schemes mentioned permit
the more accurate control of the quality of light entering the camera.
The third and fourth methods remove the excess red and infra-red
radiation from the actors and sets. The use of colored glass for the

Aug., 1933] MAZDA LAMPS FOR COLOR PHOTOGRAPHY

169

lamp bulbs means that in effect the filter is thrown away with each
lamp renewal and, of course, lamp cost is increased.

Operating the lamp on overvoltage is not practicable from an operat-
ing standpoint since the voltages at the sets are fixed, but the lamp
can be designed for a high efficiency and accomplish the same result.
This permits operation on the standard 115-, 120- volt circuits. In
effect, the lamp is designed for lower voltage and wattage. This is
what has been done in the case of the small photoflood lamp and

60000

/MOO

FIG. 3. 1500-watt, 115- volt PS-52 bulb Mazda lamp
in M.R. rifle unit (21 L/W design).

what is being done in the case of the big 2000-watt photoflood or
color photographic lamps.

Another factor which aids very materially in placing sufficient
intensity of illumination on the sets with minimum wattage, is the
use of lamp equipment which directs a high percentage of light
within the angle where it can be used most effectively. The Mole-
Richardson "rifle" unit is a particularly good example of a unit of
this type. It is capable of directing 50 to 60 per cent of the light
output of the lamp within angles where it can be used. This is
several times more than is obtained with those equipments either not

170

R. E. FARNHAM

[J. S. M. P. E.

having good reflectors, or in which accurate control of the light has
not been attempted.

Fig. 3 illustrates the candle-power distribution obtained from a
standard 1500-watt PS-52 bulb lamp now employed for general
lighting, in an M. R. rifle unit, and Fig. 4 the distribution from a
2000-watt PS-52 bulb lamp of the color photographic type in the
same reflector. Analyzing these data we find that the 1500-watt

FIG. 4. 2000-watt 120-volt PS52 bulb
Mazda lamp in M.R. rifle unit (33 L/W
design).

lamp directed 17,200 lumens in the 60-degree useful angle. The
2000-watt lamp gave 34,400 lumens in this same 60-degree angle,
an increase of 100 per cent over the standard studio type.

Photographically, the increase is greater than that, since the photom-
eter employed has its greatest sensitivity in the orange-yellow,
where the increase is less pronounced. The use of the 2000-watt,
high-efficiency lamp, in place of the 1500-watt standard type, thus
doubles the illumination intensity on the set; and the general re-

Aug., 1933] MAZDA LAMPS FOR COLOR PHOTOGRAPHY 171

quirement of 50 per cent more illumination for color photography,
together with the filter absorption, means that satisfactory color
pictures can be made with approximately double the average wattage
now employed for black-and-white photography".

The 2000-PS-52 bulb color photographic lamps as designed for
studio service have a life of 15 to 18 hours, and it is therefore ad-
visable to "save" them as much as possible. A very good practice
is to operate the lamps at approximately 90 volts, obtained by either
field control at the generator, or by the use of grids outside the set,
when preparing, and bring them to full voltage only when actually
photographing.

With all the lamps operating at the same reduced voltage, lighting
balances, contrasts, and shadows are not altered when the lamps
are brought up to full voltage.

Lamps of this type darken sooner than the longer lived types, and
therefore contain a cleaning powder which allows the light output to
be restored to its initial value. A cleaning after each 5 to 6 hours
of full-voltage operation is sufficient.

The 2000-watt PS-52 bulb lamp adequately takes care of the
general lighting requirements. For modeling or spot-lighting
service, 5- and 10-kw. lamps, similar in every respect to the present
standard lamps but of 105-volt rating, should be used.

These lamps normally operate at 29 to 30 lumens per watt, and the
overvoltage to which they are submitted when operated on 120- volt
circuits not only increases the quantity of light emitted but makes
the color quality of the light the same as that of the color photo-
graphic lamps.

REFERENCE

1 PALMER, M. W., AND BEGGS, E. W.: "Professional Motion Picture Photog-
raphy with High-Intensity Short-Life Incandescent Lamps," /. Soc. Mot. Pict.
Eng., XXII (Aug., 1933), No. 2, p. 126.

SOCIETY OF MOTION PICTURE
ENGINEERS

OFFICERS
1933

President
A. N. GOLDSMITH, 444 Madison Ave., New York, N. Y.

Past-President
J. I. CRABTRBB, Eastman Kodak Company, Rochester. N. Y.

Vice-Presidents

E. I. SPONABLE. Fox Film Corp., New York. N. Y.
W. C. KUNZMANN, National Carbon Co., Cleveland, Ohio.

Secretary
J. H. KURLANDBR, Westinghouse Lamp Co., Bloomfield, N. J.

Treasurer
H. T. COWLING, 7510 N. Ashland Ave., Chicago, 111.

Board of Governors

H. T. COWLING, Burton Holmes Films, 7510 N. Ashland Ave., Chicago, 111.

J. I. CRABTRBE, Eastman Kodak Co., Rochester, N. Y.

P. H. EVANS, Warner Bros. Pictures, Inc., 1277 E. 14th St., Brooklyn, N. Y.

R. E. FARNHAM, General Electric Co., Nela Park, Cleveland, Ohio.

O. M. GLUNT, Bell Telephone Laboratories, 463 West St., New Yorfc, N. Y.

A. N. GOLDSMITH. 444 Madison Ave., New York, N. Y.

H. GRIFFIN, International Projector Corp., 96 Gold St., New York, N. Y.

E. HUSB, Eastman Kodak Co., 6706 Santa Monica Ave., Hollywood, Calif.

W. C. KUNZMANN, National Carbon Co., Cleveland, Ohio.

J. H. KURLANDBR, Westinghouse Lamp Co., Bloomfield, N. J.

R. F. MITCHBLL, Bell & Howell Co., 1801 Larchmont Ave., Chicago, 111.

E. I. SPONABLE, Fox Film Corp.. 850 Tenth Ave., New York, N. Y.

172

COMMITTEES

173

W. V. D. KELLEY

J. H. KURLANDER

J. CRABTREE
J. I. CRABTREE
A. S. DICKINSON
G. EDWARDS
R. M. EVANS
T. FAULKNER

G. A. CHAMBERS
W. CLARK
B. W. DEPUE
O. B. DEPUE

D. M. BALTIMORE
B. W. DEPUE

G. C. EDWARDS
R. EVANS

E. R. GEIB

A. A. COOK
W. B. COOK
H. A. DEVRY
E. GALE

Color

P. D. BREWSTER, Chairman
R. M. EVANS, Vice-Chairman

J. F. KlENNINGER

G. E. MATTHEWS

Convention
W. C. KUNZMANN, Chairman

Laboratory and Exchange Practice
R. F. NICHOLSON, Chairman

A. Hi ATT
E. HUSE-

D. E. HYNDMAN

E. D. LEISHMAN

C. L. LOOTENS
K. MAC!LVAIN

D. MACKENZIE

Historical and Museum
E. THEISEN, Chairman

C. F. JENKINS
W. V. D. KELLEY
G. E. MATTHEWS
O. NELSON

Membership and Subscription

E. R. GEIB, Chairman
W. H. CARSON, Vice-Chairman

]. G. T. GILMOUR
W. W. HENNESSY
W. C. KUNZMANN
E. E. LAMB

Non-Theatrical Equipment
R. E. FARNHAM, Chairman

E. R. GEIB
N. B. GREEN
H. GRIFFIN
L. A. JONES

H. B. TUTTLE

M. W. PALMER

J. S. MACLEOD
R. F. MITCHELL
H. RUBIN
W. SCHMIDT
V. B. SEASE
J. H. SPRAY

T. RAMSAYE
A. REEVES
F. H. RICHARDSON
A. F. VICTOR

J. E. MCAULEY
T. NAGASE

J. A. NORLING

N. F. OAKLEY
E. C. SCHMITZ

J. H. KURLANDER
R. P. MAY
R. F. MITCHELL
A. SHAPIRO

174

COMMITTEES

[J. S. M. P. E.

C. DREHER
P. H. EVANS
A . C. HARDY
E. HUSE

H. T. COWLING
J. I. CRABTREE

M. ABRIBAT

W. P. BlELICKE

L. BUSCH
G. A. CHAMBERS
A. A. COOK
J. A. DUBRAY

J. O. BAKER
T. C. BARROWS
G. C. EDWARDS
J J. FINN
C. FLANNAGAN

Papers

O. M. GLUNT, Chairman
G. A. CHAMBERS. Vice- Chairman

G. E. MATTHEWS
P. A. McGuiRE

D. McNicoL

Preservation of Film
W. H. CARSON, Chairman

A. S. DICKINSON
R. EVANS

Progress
J. G. FRAYNE, Chairman

R. E. FARNHAM
H. B. FRANKLIN
W. C. HARCUS
F. S. IRBY

E. E. LAMB

Projection Practice
H. RUBIN, Chairman

S. GLAUBER
C. GREENE
H. GRIFFIN
J. HOPKINS
W. C. KUNZMANN

W. C. MILLER
K. F. MORGAN
C. N. REIFSTECK
T. E. SHEA

T. RAMSAYE
V. B. SEASE

G. E. MATTHEWS
M. W. PALMER
G. F. RACKETT
P. SCHROTT
H. M. STOLLER
S. S. A. WATKINS

R. H. McCULLOUGH

P. A. McGuiRE

R. MlEHLING

F. H. RICHARDSON
V. A. WELMAN

E. R. GEIB
H. GRIFFIN

J. H. KURLANDER

R. E. FARNHAM
H. P. GAGE

F. C. BADGLEY
B. W. DEPUE

Projection Screens
S. K. WOLF, Chairman

W. F. LITTLE
A. L. RAVEN

Projection Theory
A. C. HARDY, Chairman

W. F. LITTLE

Publicity
W. WHITMORE, Chairman

D. E. HYNDMAN

F. S. IRBY

W. C. KUNZMANN

R. T. RASMUSSEN
H. RUBIN
C. TUTTLB

W. B. RAYTON
C. TUTTLB

G. E. MATTHEWS
D. McNicoL

Aug., 1933]

COMMITTEES

175

M. C. BATSEL
P. H. EVANS
E. W. KELLOGG

Sound
H. B. SANTEE, Chairman

C. L. LOOTENS

W. A. MACNAIR

W. C MILLER
H. C. SILENT
S. K. WOLF

W. H. CARSON
L. E. CLARK
J. A. DUBRAY
P. H. EVANS
R. M. EVANS
R. E. FARNHAM
C. L. FARRAND

Standards and Nomenclature
M. C. BATSEL, Chairman

H. GRIFFIN
A. C. HARDY
R. C. HUBBARD
L. A. JONES
D. MACKENZIE
G. F. RACKETT
W. B. RAYTON
C. N. REIFSTECK

H. RUBIN
H. B. SANTEE
V. B. SEASE
T. E. SHEA
J. L. SPENCE
E. I. SPONABLE
S. K. WOLF

L. J. BUTTOLPH

R. E. FARNHAM

Studio Lighting

P. MOLE, Chairman
C. W. HANDLEY

R. F. MITCHELL, Chairman
B. W. DEPUE, Sec.-Treas.

Chicago Section

New York Section

P. H. EVANS, Chairman
D. E. HYNDMAN, Sec.-Treas.

J. H. KURLANDER
W. J. QUINLAN

O. B. DEPUE, Manager
J. E. JENKINS, Manager

M. C. BATSEL, Manager
J. L. SPENCE, Manager

Pacific Coast Section

EMERY HUSE, Chairman C. DREHER, Manager

G. F. RACKETT, Sec.-Treas. J. A. DUBRAY, Manager

SOCIETY ANNOUNCEMENTS

BOARD OF GOVERNORS

A meeting of the Board of Governors Was held at the Hotel Pennsylvania on
July 14, at which time the various administrative and fiscal affairs of the Society
were reviewed with a view of making possible alterations conforming to the
changing conditions of the times. The problems incident to revising the rates
charged for dues and non-member subscriptions for the JOURNAL were also con-
sidered, and initial plans were made for the Fall convention.

FALL, 1933, CONVENTION

The next convention of the Society will be held in Chicago, early in October,
with headquarters at the Edgewater Beach Hotel. The exact dates will be an-
nounced later. An attractive program is being arranged by Mr. W. C. Kunz-
i nun ii, chairman of the Convention Arrangements Committee, and Mr. O. M.
Glunt, chairman of the Papers Committee.

The Chicago Section will collaborate with the Committees in order to make the
visit to Chicago a most interesting and profitable one. The members of the
Society are urged to make every effort to attend; an additional inducement is
afforded by the coincidence of the convention with the Century of Progress Fair,
which terminates at the end of October.

The Society regrets to announce the deaths of

Dr. Walter Akemann

April, 1933

and
William C. Hubbard

July 20, 1933

176

SOCIETY ANNOUNCEMENTS 177

SUSTAINING MEMBERS

Bausch & Lomb Optical Co.
Bell Telephone Laboratories
Burnett-Timken Laboratories

Eastman Kodak Co.

Electrical Research Products, Inc.

National Carbon Co.

RCA Victor Co., Inc.

HONOR ROLL
OF THE

SOCIETY OF MOTION PICTURE ENGINEERS

By action of the Board of Governors, October 4, 1931, this Honor Roll was estab-
lished for the purpose oj perpetuating the names of distinguished pioneers who are
now deceased:

Louis AIME AUGUSTIN LE PRINCE

WILLIAM FRIESE-GREENE

THOMAS ALVA EDISON

GEORGE EASTMAN

JEAN ACME LE ROY

Eastman Presents
A New Film

LXHIBITING extremely fine
grain combined with reasonably
high speed, Eastman Background
Negative admirably fulfills its
function as a negative medium
for composite shots. Both in
the camera and in the processing
laboratory it performs in a man-
ner that makes it an outstanding
film for this new era of the
motion picture . . . Make your
own tests of it as soon as possible.
Eastman Kodak Company (J. E.
Brulatour, Inc., Distributors,
New York, Chicago, Hollywood.)

EASTMAN

BACKGROUND NEGATIVE

To The

Members of The
SOCIETY OF MOTION PICTURE ENGINEERS

Announcement

of the
FALL, 1933, CONVENTION

CHICAGO, ILL., OCTOBER 16-18, INCL.
Headquarters, Edgewater Beach Hotel

Technical Sessions Film Programs Semi- Annual Banquet

Special Rates Garage Service

Entertainment

Visit The

CENTURY OF PROGRESS WORLD'S FAIR

Ending October 31st

(Complete information on page 256 of this issue oj the Journal)

JOURNAL

OF THE SOCIETY OF

MOTION PICTURE ENGINEERS

Volume XXI SEPTEMBER, 1933

Number 3

CONTENTS
Radio City Sound Equipment

Page
B. KRETTZER 181

Distortion in the Projection and Viewing of Motion Pictures
CLIFTON TUTTLE 198

Voice and Personality in the Motion Pictures. . . . I. L. BRADLEY 209

The Eastman Type lib Sensitometer as a Control Instrument
in the Processing of Motion Picture Film .

G. A. CHAMBERS AND

The Sound Film Program of the United States
Agriculture

I. D. WRATTEN 218

Department of
R. EVANS 224

The Preselection of Takes for Processing from
veloped Negative I

Exposed Unde-
). W. RIDGWAY 230

. . . G. M. BEST 236

Economies in Sound Film Processing

The History of the Animated Cartoon

. . . E. THEISEN 239

Officers

250

Committees

251

Obituary William C Hubbard

254

Society Announcements

255

JOURNAL

OF THE SOCIETY OF

MOTION PICTURE ENGINEERS

SYLVAN HARRIS, EDITOR

Board of Editors
J. I. CRABTREE, Chairman

O. M. GLUNT A. C. HARDY F. F. RENWICK

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers.

Publication Office, 20th & Northampton Sts., Easton, Pa.
General and Editorial Office, 33 West 42nd St., New York, N. Y.

Copyrighted, 1933, by the Society of Motion Picture Engineers, Inc.

Entered as second class matter January 15, 1930, at the Post Office at Easton,
Pa., under the Act of March 3, 1879.

RADIO CITY SOUND EQUIPMENT*
BARTON KREUZER**

Summary. The paper describes the equipment used in the Radio City Music
Hall, explains the mode of operating it, and the use of the component systems; it
refers to the engineering design of the various portions of the system, their per-
formance, and some of the installation problems involved.

The term Radio City has been somewhat loosely applied in the
past. Radio City is that part of Rockefeller Center occupied by the
Radio Corporation of America and its associated companies. Rocke-
feller Center lies between Forty-eighth and Fifty-first Streets and
Fifth and Sixth Avenues in New York City. The theatrical part of
this enterprise consists of two theaters, the "RKO Roxy Theater,"
seating 3700 persons, and the "Radio City Music Hall," seating 6200.
The remarks in this paper will be restricted to the latter theater,
since, from an equipment and engineering standpoint, the smaller
theater is a proportionately smaller copy.

A considerable number of systems were involved in equipping the
Radio City Music Hall. They may be summarized as follows:
I. Sound on film reproducing equipment.

A. Double channel equipment main projection booth.

B. Rear projection equipment.

C. Portable film phonograph.

D. Two complete preview rooms.

II. Public address and sound reinforcing equipment.

III. Rehearsal system.

IV. Sound effect system.

V. Stage manager's call system.
VI. Radio frequency distribution system.
VII. Custom built radio, phonograph, and monitoring system.
The sound-on-film reproducing equipment is of the latest "high
fidelity" type, employing the newly developed sound-heads shown in
Fig. 1. The main projection booth is equipped with four projectors

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** RCA Victor Co., Inc., Camden, N. J.

181

182

BARTON KREUZER

[J. S. M. P. E.

and a double-channel 80- watt amplifying system, which is completely
a-c. operated. This assures a complete "standby" equipment for
emergency use, which can be placed in immediate operation by
turning a single knob. A further guarantee against interruption is
the fact that the 80-watt output is obtained from two 40-watt
amplifiers coupled together so that all stage speakers will be supplied

FIG. 1. New sound-head.

with energy even if one of the two power amplifiers should become
inoperative.

Amplifier volume and fading from one projector to another are
controlled remotely by push buttons located adjacent to the pro-
jectors. Key-operated remote volume control stations have also
been placed throughout the theater.

Sept., 1933]

RADIO CITY SOUND EQUIPMENT

183

The stage loud speakers consist of three 10-foot directional baffles
and three 5-foot directional baffles. The larger speakers are placed
in a horizontal line so as to cover almost the entire house with the
exception of the front part of the orchestra, which is covered by the
three smaller speakers hung directly beneath the others. These

FIG. 2. Stage loud speaker.

r*i

L ! I

FIG. 3. Layout of stage loud speakers.

184

BARTON KREUZER

[J. S. M. P. E.

speakers are of the large throated, electrodynamic cone type. They
have been previously described in the JOURNAL. 1 Fig. 2 is a photo-
graph of one of the large directional baffles and Fig. 3 shows the
speaker positions.

These speakers are mounted in metal "cages" and are provided
with an electric motor system that folds the three large baffles,
removes all the speakers to the "wings" (four to the Opposite Prompt
side and two to the Prompt side) and "flies" all speakers in fifty-five
seconds. This achievement is due to Peter Clark, Inc.

FIG. 4. Film phonograph.

The rear projection equipment, which may be used to create
"panoramic" backgrounds or to project regular pictures, is similar
to that employed in Translux theaters. For the sound equipment
newly developed sound-heads are used and an a-c. operated booster
amplifier is provided for transmitting the signal to the main pro-
jection room where it may be fed into one of the several systems
mentioned.

The film phonograph is a portable table-mounted machine em-
ploying the magnetic drive which has been so successful on RCA
Photophone film recorders. This assures the reproduction of sound
of the highest quality from film, with absolute freedom from speed

Sept., 1933]

RADIO CITY SOUND EQUIPMENT

185

variations of any kind. This film phonograph is pictured in Fig. 4.
Two positions are provided for it, viz., in the rear projection room
and the mam projection room. Connections are quickly made to
wall receptacles.

The preview rooms, together with a broadcasting studio, re-
hearsal halls, etc., are located two stories above the projection room
level. These rooms are equipped with "high fidelity" film repro-
ducing equipment of the latest design, completely a-c. operated,
with push-button remote control. One push-button location is in

U LJ LJ U U U LJ U

FIG. 5. Microphone positions.

the auditorium, so that the listeners may regulate the volume to
their own satisfaction. This is particularly valuable in viewing
"rushes" of various types. In engineering this project, these two
preview rooms were treated as separate theaters and were so
equipped.

The public address and sound reinforcing system utilizes ribbon
microphones, 2 a-c. operated 80-watt amplifiers, and a mixer con-
sole and remote control panels, both of which are unusually flexible.
The system is used, of course, to reinforce sound emanating from
the stage or as a public address system that can be used, for instance,

186

BARTON KREUZER

[J. S. M. P. E.

by an unseen person making announcements to the audience, and
for other similar purposes. Six of the 3-foot directional baffles are
used for loud speakers. These are concealed behind lighting grilles
above and on either side of the proscenium arch.

There are 54 microphone positions
on or above the stage, including those
on the choral stairs. The most not-
able of these, perhaps, are eleven posi-
tions built into the disappearing foot-
lights and eight positions in metal
boxes whose tops are grilles made flush
with the floor of the orchestra elevator.
There are eight positions on the light
bridges above the stage. Fig. 5 shows
all the positions. Fig. 6 shows one of
the ribbon microphones.

Fig. 7 is a skeleton diagram of the
entire public address system. Many
liberties have been taken in making
this drawing for the sake of clarity
and typical circuits rather than actual
connections are shown. Audio circuits
are shown as solid lines and power
supply circuits as broken lines. Con-
trol circuits have been omitted, but
some of the functions of the control
racks have been marked adjacent to them. No interconnections
to other systems are shown.

The output of the microphones is carried on intermediate im-

FIG. 6. Ribbon microphones
(a) stand type; (b) hanging
type.

ro -I

HlBlN

h-0-- J

j-Q-- 4

r-l *

TtC L.MC

[*5532 W

FIG. 7. Schematic diagram of the public address system.

Sept., 1933]

RADIO CITY SOUND EQUIPMENT

187

pedance lines to the basement amplifier room where racks of micro-
phone amplifiers with non-microphonic vacuum tubes are located.
This type of amplifier is shown in Fig. 8. Filament current for these
amplifiers and field current for the ribbon microphones are supplied
by a 1000-ampere-hour glass cell storage battery located on the pro-
jection room level. This is a double-channel battery installation,
one set being charged while the other is in use. Other than a smaller
double-channel glass cell battery installation used for relay operation,
no other battery power is used in any of the systems.

FIG. 8. Microphone amplifier.

The microphone amplifiers derive their plate supply from mercury
vapor rectifier tubes in a power supply rack located on the projection
room level. These rectifiers are also in a double-channel arrange-
ment. The output of the microphone amplifiers is brought to the
"mixer" console where the lines terminate in jacks. All mixer
controls terminate in telephone plugs of the switchboard type with
flexible drop-cords. In this manner, any microphone position may
be connected to any mixer. There are four groups of these mixers,
each group colored differently. The output of each group is brought
to a correspondingly colored sub-master, which feeds into a single-
stage booster amplifier located in the console itself. The amplifier

188 BARTON KREUZER [J. S. M. P. E.

employs a non-microphonic tube. At the output of the booster,
the over-all master control is located. A volume indicator of the
copper oxide rectifier type is built into the console and may be patched
into any point in the circuit.

The console is shown in Fig. 9. It is constructed and placed so
that the operator may have an unobstructed view of the stage and a
portion of the auditorium proper. At the right may be seen the
cords and jacks referred to above, and to the left may be seen a jack
panel where tie lines to the other systems terminate.

FIG. 9. Microphone mixer console.

The output of the console is fed to the main amplifier or the
standby amplifier, both of which are 80-watt amplifiers similar to
those employed in the projection system. The output of this amplifier
passes through a control panel, where an individual volume level
may be established for each loud speaker, and thence to the loud
speakers themselves. This control panel is part of two control
racks located adjacent to the mixer console, shown in Fig. 10. Other
functions of these control racks are to furnish control of all power-
supply circuits by means of relays, selection of amplifier channels
both for the main amplifier and for seatphones, selection of pro-
jection or public address program for seatphones, etc. In addition
to this, a monitor loud speaker is provided on these racks fed by a
separate amplifier, which may be connected to any of the systems

Sept., 1933]

RADIO CITY SOUND EQUIPMENT

189

at the operator's discretion. This provides a rapid check on the
operation of all systems.

The rehearsal system, electrically, is a duplicate of the public
address system but employs three ribbon microphones, located in
the private box, the console box, and the twelfth row of the orchestra.
The last is built into a portable desk which is used only during re-
hearsals. In addition to these, one carbon microphone station is
built into the stage switchboard for the use of the chief electrician.
These microphone outputs pass
through a mixer panel located on
the control racks mentioned above
and thence to another 80-watt
amplifier. The output of this
amplifier supplies thirty- two 3 -ft.
directional baffles located at strate-
gic points off-stage, below stage, in
spotlight booths, projection room,
etc. In addition, magnetic speakers
of the wall type are used in a few
locations, such as in the orchestra
pit elevator shaft, choral stairways,
etc., where low volume is satis-
factory. The level of the speakers
is controlled in groups at the con-
trol racks.

Three-foot portable speakers are
used also to cover the front por-
tion of the stage when addressing
the cast, and are removed and
stored together with the twelfth
row microphone whenever there is
an audience present. This sys-
tem, which is most useful during rehearsals for talking to actors
and stage hands, is almost invaluable for cueing during stage pres-
entations when operated at a low level.

The sound effect system employs an 80-watt amplifier of the type
described previously, which in emergencies is used as standby for
the public address amplifier. Ordinarily, it is used to amplify
sound effects created either by microphone pick-up from backstage,
(}isk or film reproduction. Disk reproduction comes from a battery

FIG. 10. Control racks.

190

BARTON KREUZER

[J. S. M. p. E.

of four turntables capable of revolving at either 33 Vs or 78 rpm., and
allowing the pick-up to be pre-set at any desired point of the record.
The pick-up is suspended above the record while the latter revolves.
At the desired moment, a button is pressed, the oil-damped pick-up
is lowered gently to the record, and the desired "effect" is reproduced
without any delay in attaining the required speed or in locating the
correct groove. This arrangement allows very precise timing. The

FIG. 11. Public address, effect, rehearsal, and miscellaneous amplifier racks.

outputs of the turntables are controlled at a small mixer table placed
near the large mixer console.

The output of the "effect" amplifier supplies two gigantic di-
rectional baffles located in a separate room behind the cyclorama
at the rear of the stage. Each of these baffles has a mouth opening
of 19Vz by 12 feet, and is 20 feet long. The baffles terminate in four
throats, each of which is supplied with sound energy by an electro-
dynamic cone unit. These loud speakers reproduce frequencies as
low as 30 cycles per second.

Sept., 1933]

RADIO CITY SOUND EQUIPMENT

191

Each of the four systems described so far is provided with re-
motely operated volume controls, which can be manipulated by
means of push-buttons at the control racks adjacent to the mixer
console.

Fig. 11 shows the rack assembly comprising the public address,
effect, and rehearsal amplifier channels. In addition to these, a
miscellaneous equipment rack is shown, which contains a sensitive

(a)

FIG. 12. Eighty-watt amplifier
and seatphone amplifier with cover
plates removed (a) front; (6) rear.

radio receiver, a turntable, a tube tester, and patching panels. The
radio and turntable may be patched into any channel for testing or
for any special purpose. Fig. 12 is a front and rear view of one of
the eighty- watt amplifier channels with the cover plates removed.
Seatphone amplifiers may be seen at the bottom.

Fig. 13 (a) shows the two power-supply racks with the front covers
removed, exposing the mercury vapor rectifiers and the various

192

BARTON KREUZER

[J. S. M. P. E.

relays. Fig. 13(6) shows the rear of these racks, including the
main and standby channels for supplying field current to the loud
speakers. These consist of copper oxide rectifiers.

Fig. 14 shows the circuits employed in the voltage amplifiers,
Fig. 15 those used in the power amplifier and Fig. 16 shows the

if II I*

(a)

(*)

FIG. 13. (a) Power-supply racks with cover plates removed, front view;
(6) two guards removed, rear view.

over-all frequency response of the public address system from uniform
acoustic input to the microphone to acoustic output in the auditorium
from the loud speakers. The characteristics of sound reproduced
from film are equally excellent. The variations that do exist have
been definitely proved to be room characteristics, and no corrections

Sept., 1933]

RADIO CITY SOUND EQUIPMENT

193

have been made because the "weighting" of these factors is so often
open to question and the actual variations from uniformity are not
great.

The stage manager's call system consists of a carbon microphone
station, built into the stage switchboard, which feeds a 40-watt

FIG. 14. Voltage amplifier circuit.

FIG. 15. Power amplifier circuit

194

BARTON KREUZER

[J. S. M. P. E.

amplifier located in the basement amplifier room. This amplifier
supplies fifty-three magnetic loud speakers of the wall type located
in t"he various dressing rooms, offices, rehearsal rooms, cafeteria, etc.
By means of this system, announcements may be made and per-
formers called. When the microphone is not being used, the call
system "floats" on the public address system so that persons in any
of these rooms are aware of what is happening on the stage.

100 1000

Frequency in cycles per second.
FIG. 16. Over-all frequency response; public address system.

lOfXX)

FIG. 17. Plan view; projection room level.

An emergency key is also supplied with the microphone, which
enables the operator to connect the call system amplifier into the
circuit of the auditorium public address loud speakers for making
emergency announcements to the audience.

Lines are provided between all amplifier racks and between
racks and the mixer console so that a very flexible and efficient
system results. Any amplifier may be used as a standby for any
other. Microphone output, "effects," film phonograph, or pro-
jector output, may be patched into any of the systems at will; any
system may be coupled to the loud speakers normally operated by
another system, etc. A plan view of the main projection room level,
Fig. 17, shows the location of most of the equipment described so far.

Sept., 1933] RADIO CITY SOUND EQUIPMENT 195

The radio-frequency distribution system used to supply carefully
shielded broadcast signals to radio receivers consists of a single,
almost vertical, antenna wire feeding an a-c. operated radio-frequency
amplifier located in a fan room near the roof. This amplifier increases
the signal level and lowers the line impedance so that the signal may
be fed down a specially constructed, low capacity, shielded cable to
the various radio receiving channels in the building. At these
points, coupling transformers increase the impedance to a value
suitable for the input of a radio receiver. This system allows the
use of a single antenna for several receivers and assures a signal free
from "noise" pick-up within the building.

The custom built dial radio and monitoring system consists of a
basic amplifier and control cabinet, located on the projection room
level, controlled by telephone dials located at remote points, such
as an executive office and an executive dining room. At these points,
loud speakers are built into the walls behind ornamental grilles.
At one location, an automatic record-playing mechanism is recessed
into the wall. This machine will play a large number of records
without requiring any attention after the initial loading of the
magazine. If any selection is not liked, dialing a number instantly
rejects that record and the machine passes to the next record.

Instant selection of twenty radio stations is assured, and the stage
program or sound picture reproduction may also be heard. A high-
quality radio receiver is the heart of the system. It is operated by
the dial selectors of the same type as those used in automatic tele-
phone exchanges.

One difficulty peculiar to this particular theater which was over-
come is worth reviewing. As shown in Fig. 7, the stage consists
chiefly of elevators, three of which are rectangular elevators having a
rotatable section in the center made up of parts of the three elevators.
In addition, there is an orchestra pit elevator.

To connect the microphones to these elevators for public address
purposes was a relatively simple problem, which was solved by using
"swing" cables beneath the elevators. However, the musicians
comprising the orchestra sit in a motor- driven structure known as
the "bandwagon." This can travel anywhere from its "garage"
located under the first ten or twelve rows of seats to as far back as
the rear stage elevator (see Fig. 18). It can be moved below stage
out of sight if all the elevators are lowered (this necessitates raising
a steel "curtain") or it can be raised and moved across the footlights,

196

BARTON KREUZER

[J. S. M. P. E.

which sink into the floor, back to either the first or third elevators,
where it can be lowered and returned to the garage out of sight.
An alternative, sometimes made use of, is to stop the cycle and
reverse it. Maintaining connections to seven microphones on this
"bandwagon" was a difficult problem.

The problem was solved by using two cable receptacles on the
"bandwagon" connected in multiple. Outlets were provided on
the elevators and in the garage. Extension cables were used and
whenever the "bandwagon" moves from one position to another,

ELCVATIOM

FIG. 18. Stage elevators and "bandwagon."

a second extension cable from the new position is plugged into place
while the cable from the first location is still connected. Then the
latter cable is removed. This is performed behind the "bandwagon"
out of sight of the audience, while the "bandwagon" may be moving.
In this way, continuous contact is maintained, and the orchestra
may be picked up, amplified, and reproduced for the audience while
it is below stage on the elevators moving to a new position.

The author wishes to acknowledge the great amount of help
received in installing this equipment from Mr. G. A. Toepperwein

Sept., 1933] RADIO CITY SOUND EQUIPMENT 197

of the RCA Victor Engineering Department who completely engi-
neered and planned the project, from Messrs. Cochran and Lehman
of the same department, who are responsible for a great deal of the
apparatus design, and to Mr. Harry Braun who was responsible for
the installation of equipment at the RKO Roxy Theater and who
rendered valuable assistance in this installation.

REFERENCES

1 OLSON, H. F. : "Recent Developments in Theater Loud Speakers of the
Directional Baffle Type," J. Soc. Mot. Pict. Eng., XVHI (May, 1932), No. 5, p. 571.

2 OLSON, H. F.: "The Ribbon Microphone," /. Soc. Mot. Pict. Eng., XVI
(June, 1931), No. 6, p. 695.

DISCUSSION

MR. FRANK: The installation in the Radio City Music Hall cost something in
the neighborhood of $75,000, including supervision of the installation, but not
the electrical work connected with the installation.

MR. RICHARDSON: That, as I understand it, includes only the projection
equipment?

MR. FRANK: Only the sound equipment, not the projectors or arc lamps.

MR. KREUZER: It does, however, include the public address equipment and
the reinforcing.

MR. BLIVEN: Is there any correction from the center of the house?

MR. KREUZER: There has been no correction due to the fact that the house
is usually dead beyond the center.

MR. BLIVEN: I noticed that if one listens at the center he can hear the voices
from the stage as well as from the public address system. Is the microphone
set-up responsible for that?

MR. KREUZER: It is possible that the set-up of the next scene behind the
stage might have influenced the set acoustics at the time. I have occasionally
observed that if the stage is entirely bare and the performance is going on directly
in front of the large curtain, there is a rather large volume back stage, and it is
possible to get a small amount of vibration from the white tiled walls. I think
that could be prevented.

I can not verify your observation from my own experience, but I can say with
the sound coming from the actors on the stage, it might well happen. We have
made tests, and found that on most locations 100 per cent of the sound comes
from the public address system.

DISTORTION IN THE PROJECTION AND VIEWING OF
MOTION PICTURES*

CLIFTON TUTTLE**

Summary. At the instance of the Projection Screens Committee an investiga-
tion of the distortion in motion picture images was initiated. Although insufficient
data have been collected to j ustify generalization, it appears that the distortion that
may be tolerated by the average spectator is greater than one might at first suppose.
The paper discusses the subject particularly from the viewing and projecting angles.

The following discussion of distortion in the presentation of motion
pictures has been prepared at the suggestion of the Projection
Screens Committee. Although various phases of the subject have
frequently been brought before the Society, with the result that
some definite recommendations have been made, members of the
Committee felt that the subject is not entirely closed. It was sug-
gested in particular that an attempt be made to demonstrate the
distortion resulting from projection at an angle.

The whole subject of warped perspective and distortion can
conveniently be divided into three phases in accordance with three
causes which adversely affect true rendition of form on the screen
and in the eye of the observer. These are: (1) The discrepancy
between camera point of view and audience point of view involving
the relative values of camera and projector lens focal lengths. (2)
The vertical elongation of figures and the keystone effect resulting
from projection from a point above the screen. (3) The error in
perspective caused by the off-center view seen by many members of
the audience.

The three types of distortion are, of course, additive in their
effects, but for the sake of simplicity in this discussion it seems
best to treat them singly, leaving to the reader the task of evaluating
the combined effect for the set of conditions in which he is particu-
larly interested.

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Kodak Research Laboratories, Eastman Kodak Co., Rochester, N. Y.
198

DISTORTION IN MOTION PICTURES 199

It was pointed out by Hardy and Conant 1 that ideal conditions
of viewing appertain to only one point in the theater namely, a
point on the axis of projection, normal to the screen, and at a distance
D from the screen such that

focal length of camera lens

D = X projection distance

focal length of projector lens

At other axial points, either closer to or farther from the screen,
the observer sees an image distorted from its true perspective. If
the focal length of the projection objective is chosen as recommended
by Hardy and Conant to make the point D fall near the center of the
audience, the average distortion is reduced to a minimum. This
viewing distance defect in motion picture projection is obviously
of a nature which can not be overcome, although, fortunately, in
practice it does not seriously handicap the illusion for the average
person.

Projection at an angle and viewing at an angle also are causes of
distortion which can not be eliminated in practical presentation;
but since both these are controllable to some extent in the design of
theaters, it remains of practical interest to consider these matters
and perhaps to specify limiting values for the guidance of architects
and theater owners.

PROJECTION ANGLE DISTORTION

The question of projection angle has been frequently discussed
before this Society, the most thorough treatment of the problem
being that presented by the Projection Committee in 1929. The
Committee recommended, first, that the angle be kept as near zero
as possible, and then proceeded to analyze the amount of distortion
as a function of the angle of divergence of the projected beam and
of the angle which the axis of projection makes with the horizontal.
They expressed the distortion as the percentage increase in height
which results when the picture is projected at an angle to a vertical
screen.

To quote from this report: "Now, the maximum permissible
amount of distortion is a matter on which there seems to be con-
siderable divergence of opinion. The Committee feels that in recom-
mending 5 per cent as the maximum increase in picture height, it is
erring on the side of laxity rather than on that of rigidity." For
practical projection conditions a projection angle of 17 degrees re-

200 CLIFTON TUTTLE [J. S. M. P. E.

suits in an increase of image height of about 5 per cent. The Com-
mittee therefore fixed on this angle as the limiting value.

Through the kind offices of the chairman and members of the
present Projection Screens Committee, the author has been supplied
with data concerning the projection angles existing in one chain
of theaters. These data show that about 60 per cent of the theaters
in this group have projection angles equal to or greater than that
recommended by the Committee, and about eighty per cent have an
angle greater than the more rigorous specification of 12 degrees
recommended in the standards adopted by the Society as the limiting
value. Since the theaters of large seating capacity have the larger
projection angles, it follows that the vast majority of theater patrons
habitually see pictures which are projected at angles greater than 12
degrees. It would seem desirable either to bring the recommendation
of the Society into line with practice or to exert pressure to bring
future practice into line with our recommendation. Thus, it appears
that a continued discussion of the problem is not out of order.

From some points of view it seems indeed that the 5 per cent
increase in the ratio of height to width is a lax enough tolerance.
Consider for a moment the effect on the human figure. While the
pictures of some of our Hollywood actors and actresses might undergo
with aesthetic advantage a 5 per cent increase in vertical to hori-
zontal ratio, it hardly seems probable that the result would be highly
satisfactory in the majority of cases. The 17-degree projection
angle should in its effect be roughly equivalent to the once highly
advertised 18-day grapefruit diet. Greta Garbo, Ruby Keeler, Joan
Crawford, and others in the light-weight class apparently lose five
or six pounds by the treatment.

There is a strong tendency on the part of the general public to
accept what it sees on the motion picture screen as the last word in
fashion and beauty. Large projection angles may, therefore, be
largely responsible for the vogue for slender figures. It may thus be a
grave responsibility upon the motion picture engineer in the interests
of public health to prevent the motion picture screen from setting up
an ideal of dangerous emaciation.

In addition to elongation of figures there is, of course, a second
defect in an image projected at an angle which follows from the fact
that the bottom of the screen image is magnified more than the top.
What should be vertical lines in the image become convergent up-
ward. A rectangular screen shape is maintained by shaping the mask,

Sept., 1933] DISTORTION IN MOTION PICTURES 201

but nothing is done to rectify the convergence of lines within the
picture. It is well known that the eye is extremely sensitive to the
lack of parallelism between two lines. Mitchell 2 has recently thor-
oughly discussed this question for the edification of the cameraman,
pointing out several matters which should be observed in scene compo-
sition. In some actual cases in theater projection the convergence
amounts to five or six degrees, which is very apparent at the edge
of the screen.

Fortunately, the attention of the audience is seldom concentrated
on vertical lines in a picture. This is more true now than in the days
of the silent picture with its numerous titles. It seems, therefore,
that experimental demonstration of the practical effect of projection
angle should be confined principally to such subjects as make up the
greater bulk of motion picture presentation.

In preparing experiments to demonstrate and evaluate the limits
of allowable distortion, one is confronted by two diametrically
opposed points of view. Either 'one should seek to determine limits
which would prevent the audience from ever seeing an image in
which distortion could be recognized, or one should seek to find the
limits which in the majority of cases would not allow distortion to
destroy the illusion of naturalness.

In view of data which show that many successful theaters have
projection angles in excess of the arbitrary limit recommended by
the Society, it seems of interest to proceed on the latter basis; that
is, to determine the degree of distortion at which the illusion of
naturalness breaks down.

A number of still pictures of motion picture scenes were reproduced
as lantern slides. These were projected on a screen at vertical
angles which gave progressively 2.5, 5, 10, 15, and 20 per cent dis-
tortion. The screen picture was photographed at each position
and lantern slides were made of the results.

All the groups of slides thus obtained were thoroughly shuffled
together so that during projection no one of a series of slides would
follow another of the same subject. The slides were then shown
to a group of persons, each of whom was asked to select all pictures
which looked unnatural and to state the reason for the objection.

Results of this test are summarized in Table I. In column one of
this table, the subjects have been classified in a general way. The
terms "close-up," "semi-close-ups," and "full-length figures" apply
to human figures. Well-known inanimate subjects included pictures

202

CLIFTON TUTTLE

[J. S. M. P. E

of houses, doorways, wagons (showing wheels), etc. The remaining
columns headed by the value of distortion in per cent contain the
record of the relative number of observers who objected.

TABLE I

Summary of Data Showing Susceptibility of an Average Audience to Different

Degrees of Distortion

Subject

Relative number of observers who objected: in
per cent of total number of observers

Per Cent Distortion
2.5 5 10 15 20

Close-ups
Semi -close-ups
Full-length figures
Well-known inanimate subjects

0000 20%
0000 5% 25%
0000 15% 50%
000- 10% 40% 50%

Individuals participating in the test were not informed of its
object and they probably were neither less nor more critical than
members of the usual motion picture audience.

The conclusion from these data is that for scenes in which actors
or actresses form the principal interest, the image can be distorted
10 per cent before the illusion of naturalness is impaired. Stated
in another way, this means that most persons unacquainted with
what the real subject looks like are satisfied even though the picture
is decidedly distorted. In the case of very familiar objects of definite
shape as, for instance, a picture in which there was an axial view of a
wagon wheel, the tolerance is somewhat narrower some persons
objecting when the image is distorted as much as 10 per cent.

DEFINITION AND AREA LOSSES

This discussion thus far has been limited to true image distortion.
There are two other undesirable effects, however, which accompany
projection at an angle. The first is the loss of image area which
follows inevitably when the sides of the picture aperture are cut to
make the frame rectangular. The second is the effect upon top and
bottom image definition because of the path difference. Though
neither of these can properly be classified as distortion, an evaluation
of both has been included in Table II.* The third column shows

* Formulas sufficiently accurate for the calculation of these effects within a
few per cent for projection angle, 0, are as follows:

Distortion (per cent) increase of picture height = Secant 1.0. Image size,

Sept., 1933] DISTORTION IN MOTION PICTURES 203

approximately the area loss in per cent which results when a picture
is projected with a 6-inch lens to a rectangular vertical screen, and
the fourth column shows the diameter of the circle of confusion, or,
more accurately, the major axis of the ellipse of confusion at the
bottom and top of the field for a perfect lens working at an aperture
of//2.

TABLE II

Effect of Projection Angle upon Screen Definition and Loss of Area with Rec-
tangular Masking

Image Size
Projection Bottom and

Angle Per Cent Area Loss, Top of Screen,

(degrees) Distortion Per Cent Mm.

12.5 2.5 2 0.9

17.5 5.0 3 1.3

21.5 7.5 4 1.7

24.5 10.0 5 2.0

27.0 12.5 5 2.2

29.5 15.0 6 2.6

The area loss, that is, the amount which must be masked from
the lower corners of the picture, is not particularly serious at any
commonly used combination of projection distance and focal length
provided that the elongated picture is not masked off to maintain
the 3X4 picture.

The lack of definition as judged from a viewing distance of twenty
feet becomes noticeable if the image size exceeds about 1.5 mm.
Since the image size resulting from the projection angle is super-
posed upon the effect of the lens aberrations, it is probably reasonable
to state that with practical projection lenses of //2 aperture defi-
nition suffers noticeably at projection angles greater than 17 degrees.

SIDE VIEWING ANGLE DISTORTION

The second matter upon which some experimental data may
prove of advantage concerns the viewing angle forced upon all the

top and bottom of screen, for perfect focus at center =

H tan B sec 9

2 //number of projection lens

F - tan H/2
Area loss (per cent) = 100-100

In these expressions H is the film frame height, and F is the focal length of the
lens. Mitchell (loc. cit.} gives a method for determining the value of the area
loss which is accurate over a greater range of projection conditions.

204 CLIFTON TUTTLE [j. s. M. P. E.

members of a motion picture audience except those located opposite
the screen center. At first thought, this problem appears to be
closely related to the former one. Both projection at an angle
from the horizontal and viewing at an angle other than the normal
produce a similar kind of elongation of the screen image, and one
might expect that a distortion tolerance set up for the one case
might apply to the other. The conditions, however, are somewhat
different; a person viewing the screen from an angle is conscious
of his point of view, and instinctively makes a correction for some
distortion of the image. He is not conscious of the projection angle,
and therefore has no means of compensation which will aid him in
rectifying his concept of the picture.

It is common experience that motion pictures viewed from extreme
front and side seats in some theaters appear badly distorted. At
the same time, it is true that one's enjoyment is not adversely
affected until the angle becomes fairly large. The author has at-
tempted to determine the limiting angle experimentally by projecting
before a group of persons motion pictures of a screen image photo-
graphed at different angles.

There seems to be quite good agreement among the persons be-
fore whom these and other pictures of the same kind were shown
that any angle less than thirty degrees is not objectionable. Forty
degrees seems to be passable, but the opinion was unanimous that the
illusion is spoiled at angles greater than 40 degrees.

One can not say that these demonstrations adhere closely enough
to theater conditions to justify any general conclusions. One is
quite justified in questioning : How much does motion in the picture
affect the feeling that the illusion has failed? And how much does
the angular field of view change one's judgment?

A few trials were made in which a large black border was shown
around the rotated picture. This, it was thought, would supply a
comparison reference as to the amount of foreshortening to be
expected in the picture. The judgment of distortion did not seem
to be much changed. Pictures filling the screen appear better to
represent the view which a member of the audience has in a seat
close to the screen the only location in which the viewing angle
problem is serious. On the basis of a number of observations in
theaters during the projection of pictures, it appears that motion
in the picture does not affect the result to any great extent.

A seat which forces one to see any part of the picture at an angle

Sept., 1933] DISTORTION IN MOTION PICTURES 205

greater than 40 degrees is undesirable. For full-length figures the
judgment is more critical.

The larger the screen, of course, the worse is the distortion at the
farthest edge at a given viewing distance. In recommending practice
for the guidance of designers and architects, the specification for the
position of the extreme seat should be based upon the angle at
which this edge can be seen. Assuming that the distance from the
first row of seats to the screen is equal to 1.5 times the screen width,
the first row of seats should be not longer than 1.5 times the screen
width if the extreme viewing angle for the edge is not to exceed 40
degrees.' Data supplied to the author by the chairman of the Pro-
jection Screens Committee give the average maximum viewing angle
of a number of theaters at 34 degrees at the screen center an angle
which makes the extreme edge viewing angle somewhat in excess of
the distortion limit.

DISTORTION REMEDIES

Many attempts have been made in the past to cure the evil of
viewing angle distortion usually by the use of curved screens.
Anamorphotic lens systems also have been suggested. The fallacy
of such suggestions has been pointed out so frequently that ap-
parently no proposal of this kind has been made for several years.
There is, of course, no remedy except proper design of the seating
space. Any correction of the screen image for one position can be
made only at the expense of the perspective from other positions.

Correction of projection angle distortion is not so impossible
theoretically. Partial compensation can be effected by tilting
the screen. This means probably presents some mechanical diffi-
culties since it is not commonly used. Other partial remedies are at
least theoretically possible.

CONCLUSION

This subject was proposed to the author only a short time ago,
and it has not had the consideration which its importance warrants.
The principal purpose which this paper can serve is to stimulate the
gathering of further experimental data on the whole subject.

Whether distortion limits should be based upon the ideals of
truth or upon the ideals of showmanship is the most difficult question
to decide. Putting the results of the experimental data in the classic
form ascribed to Barnum, most of the people most of the time are

206 CLIFTON TUTTLE f j. S. M. P. E.

totally fooled, up to a projection angle distortion of 10 per cent.
All of the people all of the time are dissatisfied with a seat which
forces them to see part of the screen at an angle greater than 40
degrees.

REFERENCES

1 HARDY, A. C., AND CONANT, R. W.: "Perspective Considerations in Taking
and Projecting Motion Pictures," Trans. Soc. Mot. Pict. Eng., XII (1928), No.
33, p. 117.

1 MITCHELL, R. F.: "Projection Keystoning from the Cameraman's View-
point," Amer. Cinematographer , XIII (Jan., 1933), No. 9, p. 8.

DISCUSSION

MR. SCHLANGER: To evaluate image distortion, it is necessary to consider
the cumulative effect of its various causes. There are three distinct stages by
which an image or object may appear distorted to the spectator of the motion
picture.

The first occurs in photographing, and should be referred to as foreshortening,
rather than distortion. In this case, the foreshortened view of an image as
obtained by the camera, is an intentional and purposeful one. The cinema-
tographer chooses to tell his story more effectively by a combination of full,
unforeshortened and foreshortened views of the objects and images being photo-
graphed. Therefore, for best results, the images intended to be unforeshortened
or undistorted must remain so to the spectator; and images intentionally fore-
shortened must not appear any more distorted to the spectator than the cinema-
tographer originally intended.

Distortion in the second and third stages is the distortion due, first, to the
projection angle and, second, to the viewing angle in the theater. These two
forms of distortion destroy not only definition and beauty, but also, what is
more important, the effectiveness of the important foreshortened views, or dis-
cords, that are supposed to contrast vividly with the completely unforeshortened
or undistorted image. If all objects and images appear somewhat distorted to
the spectator, the effect of the value of the well-placed discords is completely
lost. The excessive projection angles and poor viewing positions in relation to
the position of the screen in motion picture theaters are serious causes of such
distortion.

MR. BLIVEN: How many architects, as a whole, take sufficient interest in
the work of picture projection, to familiarize themselves with sight lines and
other matters to get good viewing angles?

MR. CRABTREE: I think that many architects are very anxious for knowledge.
Mr. L. A. Jones and I were appointed by President Goldsmith as a committee
to discuss, with a committee of the Institute of Architects, this very problem of
educating the architects with regard to fundamentals of theater engineering from
the motion picture standpoint. The result was that the Institute proposed that
our Society should invite one of our members to prepare a paper on the subject
and to present it at one of our meetings, so that it could be discussed and officially
authorized by our Society. They promised that they would then re-present

Sept., 1933] DISTORTION IN MOTION PICTURES 207

the paper at one of their meetings, and see that it was subsequently published in
the architectural magazines. All that we have to do now is to get the paper.

MR. JONES: I have always agreed that distortion is a very serious matter.
However, in order to go to an architect or to an exhibitor, or to any one else,
and ask him to change his mode, we need some evidence to show that the present
method is wrong and should be changed.

From the study that Mr. Tuttle has made it does not appear that the geo-
metrical distortion is as objectionable to the observer as might be expected from
the computations previously made. He has collected considerable data from a
large number of observers, and in many cases those observers were entirely
unaware of the presence of this geometrical distortion in the picture. If the
observer is unaware of this distortion, and if it does not detract from the enjoy-
ment or appreciation of the picture, it certainly is unfair to say it is very objec-
tionable.

It is probable that more voluminous data must be collected before a final
conclusion can be drawn; but certainly if we wish to criticize the quality of pro-
jection when large projection angles are used, we should have some tangible
evidence that such a practice actually produces a picture of inferior quality;
and by that I mean a quality which interferes with the enjoyment of the picture
by the audience.

Mr. Tuttle has shown one very definite objectionable characteristic of a picture
projected at a large projection angle. I refer to the loss of definition which is
unavoidable when the projection angle exceeds 17 degrees. It is quite impossible
to obtain good definition from the top to the bottom of the screen if the projection
angle exceeds 17 degrees. That undoubtedly interferes with the perception of
detail in certain parts of the picture and is a strong argument against large pro-
jection angles.

MR. RICHARDSON: What Mr. Jones says is quite correct. Still, at the same
time, it seems rather far fetched for any one to assume that a thing that is dis-
torted in three different ways can be as beautiful as a thing that is undistorted.

MR. SCHLANGER: Mr. Jones seems to imply that we have nothing substantial
with which to go to the exhibitor or the architect, to inform him that he is
not exhibiting motion pictures under the best conditions. Sufficient data may
not as yet have been properly organized, but there should be no doubt about
there being sufficient reason for improvement in the exhibition of the motion
picture. The pictures that we saw today present only one phase of distortion.

As to the lack of complaint concerning the distortion, it is almost impossible
to find out how much the public will stand. But it is almost certain that the
public usually prefers the better and more beautiful effect when the difference
is made evident. Some mention was made that some spectators found that dis-
tortion of the image due to projection or viewing conditions is sometimes flattering
to the image. All the flattering and good effects of cinematography must be
achieved in the production of the master film, and not be subject to the pecu-
liarities of varying theater structures.

MR. TUTTLE: In doing this work I have attempted to find out how much
distortion people will tolerate without objecting. I have been surprised to
discover that not only are people uncritical of distortion, but that in some in-
stances they actually like it and believe that some faces are more pleasing to

208 CLIFTON TUTTLE

look at when they are distorted. It is my opinion that the limits which the
Society has attempted to set are a little stringent. That may be one of the
reasons why the recommendations of the Society have not been followed. If we
are recommending the abolition of something that doesn't turn people away
from box-offices, it is questionable whether we should make the recommendation
or not.

MR. JONES: I have taken, perhaps, an extreme position, one that was intended
to emphasize what we have to contend with. I believe firmly with Mr. Richard-
son that we should have a minimum of distortion, but we must try to evaluate
the distortion quantitatively, if we can, and find out just how much disagreeable
effect the several factors produce.

MR. RICHARDSON: There is only one way to compare the beauty of two
objects, and that is to observe them side by side. I am sure that if you had
projected the distorted picture on one side of the screen and the same picture
undistorted on the other side, the difference would have been painfully obvious.

MR. CRABTREE: I am inclined to agree with Mr. Richardson, because there
is a certain analogy in color photography. Suppose we photograph a subject
by a two-color process. The result is fairly pleasing. But if we compare it
with the subject photographed by a three-color process we are much less satisfied
with the two-color result.

Mr. Jones asks, is the annoyance sufficient to keep people away from the
theater? Personally, I don't think that it is. I don't think that this single
matter is very vital, but the integration of many such matters is important in
the design of theaters.

MR. RICHARDSON: If a theater patron finds a picture of real beauty on the
screen, is it not reasonable to suppose he will attend the theater more often
than he would were the picture less beautiful?

MR. CRABTREE: The fact is that most of us don't know how tall or how thin
the actors are in real life. As long as they look sufficiently thin and well pro-
portioned on the screen, that is satisfactory.

MEMBER: Many of these factors are subjective, and can not be evaluated
easily. I have yet to hear any one say that he went to one theater or another
because of better projection. However, it must be granted that a patron may
attend a theater as a matter of habit without knowing that he does so because
the projection is better.

MR. CRABTREE : When I have to sit in a side front row in a theater I certainly
make a rush for a center seat as soon as possible, because there is no question
that the beauty of the feminine face, at any rate, is greatly enhanced when viewed
from a position perpendicular to the screen. It may be that I am a little fussier
than some of the observers of Mr. Tuttle's pictures.

MR. TUTTLE: I do find that the extreme side viewing angles are very detri-
mental, but I have yet to find anybody who complains about the projection angle.

The fact that 75 per cent of the people who go to motion picture theaters today
are seeing projection at an angle which causes a distortion of about 10 per cent,
might be taken as an argument against the claim that the box-office receipts
would be greater if the present recommendations of the Society were followed.

VOICE AND PERSONALITY IN THE MOTION PICTURES*
IVAH L.BRADLEY**

Summary. The stylizing of many motion picture stars and the suppressing of
their natural selves has reduced them to mere "fashion comers" and robbed them of
their continued popular appeal. We have forgotten that a sincere emotional release
is more important. Instead of this training in conventionalized deportment the
artist needs instruction in coordinating the various bodily functions rhythmically for
vocal production and release of personality, for a close relation exists between physi-
cal obstructions to the free vocal production and psychological obstructions to the
release of the personality. The motion picture engineer and the director would find
it vastly to their advantage to know more of both the mechanics and psychology of
vocal production.

In the days of the silent motion pictures it seemed to us that the
personalities of the actors were more vivid. The acting was perhaps
not so smooth, but the actors conveyed to us more of themselves.
They had to, for they had only the visible expressions of face and
body with which to appeal to our eyes and emotions. We received
the impression and remembered it. In some respects Charlie Chaplin
is wise : you do not forget him ; you always see him in front of you,
and can recall his image immediately. If you stop to think of it,
you will find you can still easily recall certain personalities of the
silent films. They could not then rely upon the sound effects which
are today so important; they had to make an effort to convey them-
selves clearly and completely by visual means alone. The majority
of my friends seem about equally divided on the subject, but feel that
while pictures today have attained a greater dramatic value, the
personalities were more vivid in the silent films.

The cacophony that came from the screen between the days of
the silent film and those when sound effects became tolerable so
lacerated our nerves and taxed our endurance that in our relief we
have perhaps forgotten how we felt about the silent film. Then
came this immediate and imperative need to charm the ears as well
as the eyes of the public. Let us admit at once that the motion

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** New York, N. Y.

209

210 IVAH L. BRADLEY [J. s. M. P. E.

picture industry has met this need more quickly, in a more pro-
gressive and, let us hope, more scientific manner, than any other
group having a vital interest in the field of vocal education. How-
ever, in attempting to change so quickly their concept of the movie
actor's voice, the motion picture industry has lost sight of the most
important element for enhancing the beauty of vocal tone.

It is certain that the public, and perhaps the motion picture
producer, is aware that some subtle essence is slipping from some of
the artists that no one yet seems to have been able to discover, for
very many movie stars gradually slip down the same decline. One
actor after another enters Hollywood a vital human being, only to
become in a year or two a frozen creature who knows how to slither
across the floor as the world's most perfect mannikin, carrying her
clothes perfectly, setting the fashions, displaying emotion hi artificial,
cold movements, speaking with an artificial, cold voice they are
all alike, the same mannerisms, the same manner of speech, the same
kisses. The public is so tired of it, and I should think that the
engineers who have to look at it every day, year in and year out,
would explode in righteous wrath.

To begin with, it is criminal to suppress the natural release of
genuine human emotion and to cover it with a meaningless
artificiality. Have you forgotten that real, sincere emotion is so
much more beautiful? Ruth Chatterton used to thrill us, and we
rushed to see her; now she is little more than her clothes. "She
has lost something," the people say. Norma Shearer is undeniably
lovely, especially in Smilin' Through; but she is always the same,
and makes one feel that she is so conscious that every move is beauti-
ful, that she works day and night, constantly, to improve the per-
fection of her every line and gesture. She is a perfect example of
superficial beauty; nothing is ever disturbed, not even a hair. Her
facial expressions, even when she is supposed to be deeply moved,
are always calculated not to disturb her external shell of beauty.
Perhaps she works in a room lined with mirrors; but so did Isadora
Duncan, and she still retained her positive, creative vitality.

Now let us turn to Katherine Hepburn a vital creature, certainly.
In A Bill of Divorcement she was strong, dramatic, magnificent raw
material. I ask you, what are you going to do with her? Even in
her second picture she has already lost some of her spontaneity.
She has apparently two personalities: one masculine, with angular
movements and a hard masculine voice; the other feminine, with

Sept., 1933] VOICE AND PERSONALITY 211

a truly beautiful emotional feminine voice. The masculine ex-
pression is a self -developed protection for her emotional sensibilities.
I believe that if we could have an intimate talk with her we should
find the roots of that protection in some bitter hurt to her soul in
childhood, or in some childish desire to emulate the masculine virility
of her brother. To try to eliminate the hard exterior by developing
an artificial feminine svelteness will not solve the problem. Her
real personality is expressed by the rich, emotional feminine voice
that she uses only a few times in the picture.

I have been told that without a doubt the producers realize that
they have made a mistake in so quickly elevating her to stardom,
and that now she is making another father-and-daughter picture
with John Barrymore! I presume she will have to make father-and-
daughter pictures for the rest of her life! And the New Movie
magazine has already labelled her "The Fashion Comer." Could
anything reflect so dismally on the motion picture industry than that
it should have degraded the finest material it has had in years to the
level of a "fashion comer?"

In Katherine Hepburn the motion picture industry has the oppor-
tunity to develop an actress on a new basis. Instead of toning her
down, it should be the privilege of the producers to help her to become
aware of those two distinct expressions of herself, explain why they
exist, and help her to realize the fine emotional value of her real self
and show her how she might bring to greater fruition this undevel-
oped essence of her personality. She stands at the crossroads, a
strong, individual character; and if the cinema finally succeeds in
training her, and forcing her into just an empty shell of herself,
they will have killed a soul and impoverished themselves and the
public.

Clara Bow is another example of an emotional actress being forced
into an artificial mold. In Call Her Savage there was a disturbing
mixture of rather unconvincing rough behavior, artificial deportment,
and beautiful, deep, sincere emotion. I came away from the theater
somewhat saddened, saying to myself, "If they'd only leave that
girl alone!" You may smile, but my mind at once went back to
Eleanora Duse, remembering her perfect balance of rhythmic silence,
movement, and emotion ; and there were moments in Call Her Savage
when Clara Bow also had that. Her voice is not pleasant, but she
has all the material for a very emotional voice. I found that I
liked the fundamental Clara Bow, that she was fine and womanly,

212 IVAH L. BRADLEY [J. S. M. P. E.

and at heart a true actress. But will the movie producers insist on
this artificial stylization, or help her develop this very rare rhythmic
instinct of hers?

Now the motion picture industry seems not to have succeeded in
killing the soul of Greta Garbo she fought to keep it. And in
trying to make 250 imitations of her the movie people forgot that
it takes something more than a Garbo exterior to make another
Greta Garbo. She is what she is because she has kept her funda-
mental being inviolable.

Then we have those remarkably open and frank personalities:
Ida May Oliver beautiful technic, and yet a human being; Mae
West, the true vampire I never have been able to understand why
producers consider cold, artificial creatures vampires ; Helen Hayes
exquisite personality and fine technic ; Billie Burke a most gracious
lady in Christopher Strong; Leslie Howard supreme in every way.
And just look at what Gary Cooper is doing : he spoke so completely
from the depths of himself in Farewell to Arms that we almost forgot
it was a screen production. You will notice that all these actors and
actresses have a voice in keeping with and expressive of their personali-
ties. They may change it to suit different characters, but they them-
selves have attained a great degree of integration of their being.
They are artists sufficiently strong in themselves to resist an ex-
ternal artificial stylization; but others are weaker, and are swamped
by this peculiar style of deportment training and direction.

Professor Overstreet has pointed out that "the artist is first of
all a human being, and the quality of his art must be related to the
quality of his personality. Such a point of view saves the artist
from being a creature of pretty tricks and accords to him the high
dignity of being a revealer of life." So the problem of the actor
is not only to learn a physiological technic, but to develop and or-
ganize that subjective power that is the heart and core of the artist's
being. To keep alive this center of one's self, to expand his spiritual
understanding, should be the primary study of the artist. It is
the substitution of mechanical technic, or of "pretty tricks," that
has brought artists of all kinds to the present state of emotional
inarticulation.

Let me insert here a criticism of the last recital of John Charles
Thomas that appeared in the 'New York Times: "Yet despite all
this (i. e., the beauty of his voice), one sometimes felt a curious lack
of penetration into the inner significance of the lieder. The voice,

Sept., 1933] VOICE AND PERSONALITY 213

for all its extraordinary beauty, seemed to traverse the surfaces of
the music rather than translate into sound its more searching signifi-
cance." Anyone who understands the relation of posture to voice
would expect nothing more than just that, from the very manner in
which John Charles Thomas struts across the stage. It is one more
demonstration of the mechanical cultivation of a magnificent voice
without a simultaneous integration with the inner personality.

This brings us to the relation of bodily technic to artistic expression,
a relation that is highly important, for only through a rhythmic
coordination of physical and mental energies can one attain a har-
monious integration of one's being.

Let us return to Katherine Hepburn and her dual personality:
her masculine phase finds its physical expression in a depressed
forward larynx, a slightly protruding jaw, and a rigidity of breast-
bone, sides of the neck, and roof of the mouth. All this results in
a hard, clipped, dry masculine voice. When she becomes emotion-
ally feminine she releases all these contractions and permits the
breath to flow freely into the head so as to produce rich overtones.
The prevailing speech training does not change such conditions;
it devotes itself only to phonetics, which enable the actor to speak
clearly, with the proper enunciation, but allows him to retain his
peculiar vocal defects.

And again, take Carole Lombard. Several months ago, when this
paper was first written, she was accustomed, when speaking, to
throw her jaw forward so far as to be ugly; her larynx was pressed
so far forward, the fauces and palate pulled down so hard, the back
of the tongue so thick and low in the throat, her neck so distended,
as to be a serious menace to her appearance. Miss Lombard,
however, is a fine example of how we can change. Last week I
thought I had better go to see her again and check up on her. Im-
agine my surprise to find that the lady had corrected her former
faults to a very great extent, with the result that the voice was no
longer so husky, but clearer and more expressive of herself. People
imagine that a protruding jaw is something they were born with,
and that it can not be changed. That is incorrect it can be changed,
but in her case, as in most others, it was a habit of speech; and when
the jaw is brought back into the proper position the pressure on the
tongue and larynx is released and they fall into a position for flexi-
bility and consequent melody and beauty in the voice.

The vigor and power come into the voice from the floor of the

214 IVAH L. BRADLEY [j. s. M. P. E.

pelvis, which generates an internal rhythmic flow of breath for the
production of tone. In the motion picture industry this is under-
stood fairly well, because everyone keeps himself in such excellent
physical condition, which is naturally conducive to vocal power, but
it is made use of only in a very shallow manner. For in the majority
of actors the breath-power never penetrates to the head cavities,
where the overtones and real richness of voice are produced. It is
dammed back by the rigid palate and fauces, which choke the per-
sonality as well as the flow of breath. We might well think of a
securely corked champagne bottle, with all its bubbling effervescence
unable to find release for the stoppage in the neck. So it should be
clear that the physical coordination of all the parts of the body for
the production of rich, sincere emotion and beauty in the voice is
the physical correlate of the psychological coordination required for
the release of the personality. The finished artist is the unity of these
two.

The motion picture industry may feel that this coordination of
the physical, mental, and emotional phases of the actor's being
toward a unified expression of his personality is not their job. Per-
haps not; nevertheless they did undertake the responsibility of
repressing and eliminating the vital personality from their young
actors by forcing them into one uniform, standardized mold and
pattern of behavior. Now let them turn about and lead in the de-
velopment of the integrated personality of their rising artists.

I know that it has been publicly stated that it no longer makes
any difference how an actor uses his voice, because the engineers
have developed apparatus that enables them to mix vocal qualities
in any manner that is thought necessary in order to achieve what
are considered to be good vocal effects. Mr. Evans of Warner
Brothers generously undertook to educate me in this, and I should
like to explain to you what that machine does from my point of view.

To most of us the fundamentals of the voice mean the cloudy,
buzzy quality produced by most persons, who enunciate by pressing
the tongue down on the larynx. The apparatus of the engineer
clarifies the voice of that buzz, bringing the enunciation, as it were,
up into the mouth, resulting in clear, distinct enunciation but making
the voice what I should call white and metallic. When I use the
term overtones, I mean the quality produced by the whirling currents
of air above the palate. Scientific equipment can not yet produce
those overtones; it can reproduce overtones, where the breath is

Sept., 1933] VOICE AND PERSONALITY 215

allowed to flow through into the head cavities, but it does not yet
add vocal color to a white voice.

Last spring I was talking to a well known engineer; all his com-
parative vowel calculations were based on a very dark ah-color
almost aw the entire pronunciation was very low. I asked him
what would happen if I said ah. "Oh, you've changed the pitch,"
he said. Since then I have talked with others. They all used the
same aw sound to demonstrate vibrations. Of each I asked the
same question, and received the same reply. I said nothing, be-
cause none of them seemed to realize that I had changed, not the pitch,
but the vowel color and the tone color; but finally I got the courage
to say to one of them that I didn't think they quite understood those
things. He asked at once, "What difference would that make?"
I replied that I didn't quite know, but I thought it would certainly
change the calculations; for there is no such thing as a pure vowel:
they are all combinations of other vowels, which make the vowel
sound rich and musical. Ah is a combination of a, e, and the upper
dome o, and all other vowels have an ah-color under them. Can a
machine be developed that will be capable of so many colors and
emotional overtones or do we have to depend on the human being
after all?

I have felt lately that if engineers and producers knew more about
the subtleties of vowel color and tone color, they would know better
how to secure from the artist what they want. As an illustration
of this I might mention the extraordinary production by Maurice
Schwartz of Yoshe Kalb. There must have been some sixty persons
in the cast, all having the most gorgeous voices, each voice completely
expressive of the character. I was overwhelmed by the vocal and
rhythmic beauty of that production. I know nothing about Maurice
Schwartz, but he convinced me that he knows the voice, because he
drew from his players the most extraordinary variety of vocal color-
ing I have ever heard in a theatrical production. It was a veritable
symphony, and artistically most satisfying. What a difference it
would make if more directors, in the movies as well as in the theater,
were as sensitive to voices and to all that goes into vocal expres-
sion. If the director were to have such a thorough knowledge, not
only of the mechanics but of the psychology of vocal production, he
would find it so much easier to mold his artists into the precise
shade of characterization he has conceived in his imagination. He
knows what kind of character he wants he should also know the

216 IVAH L. BRADLEY [J. S. M. P. E.

psychology of that character on the voice. He should then be able
to work back and tell the actor precisely how to liberate his vocal
organs and his emotional self so as to produce the desired voice and
character. For just as in life the voice is the expression of the inner
personality, so in the theater and the movies is the voice the ex-
pression of an inner understanding of the character.

The motion picture industry is the largest, most widely organized,
and most progressive in the entertainment world. How magnifi-
cently it could lead in demonstrating to millions the value of the
unified expression of the body, voice, and personality. I feel that
some day education will begin with the voice not dry phonetics
or enunciation, but the voice as our most vital and beautiful ex-
pression. As Professor Overstreet has remarked, "If this psycho-
logical approach is widely followed, it will make a profound change
in the teaching of art, and consequently in the nature of our artists."

DISCUSSION

MR. EVANS: I should like to explain what I attempted to show Mrs. Bradley
when she visited our studio. I reproduced a dialog sound track, first at about
normal level, and then some 15 or 20 db. higher, so that the loud speaker fairly
shouted. In describing the effect, Mrs. Bradley and I apparently disagreed
on what had happened. We soon discovered, however, that the disagreement was
one of terminology and that actually we had observed the same effects.

I then equalized the higher level, reducing the level of the low frequencies and
the high frequencies, leaving the level of the middle frequencies undisturbed,
thereby attempting to show that at the higher sound level the quality more
nearly approximated the quality at the lower level than before equalization.
In other words, I merely attempted to demonstrate Dr. Fletcher's curves dealing
with the effect of loudness on the hearing at different frequencies.

I believe that it would be possible to do electrically what Mrs. Bradley wants
to do with the cavities of the head, if all speech were of the same fundamental
frequency. It would then be possible to make by electrical equalization the
same changes in the relative balance of overtones that can be made by cavity
resonance in the head -simply by emphasizing certain overtones. But when the
fundamental frequency of speech constantly changes it is impossible to do that,
because the equalizers would be constantly affecting different overtones. Due
to the nature of speech, the only way to accomplish the results for which Mrs.
Bradley is looking is to train the person who is speaking to bring out the over-
tones that improve the pleasantness of the voice.

MR. FARNHAM: From this paper I take a wider meaning than has been ex-
pressed so far. In general the criticism has been that our American movie
actors do not know how to use their voices. We have seen and heard pictures
this week, in some of which the cast was foreign and in others of which the actors
were American. In my opinion, the English actors in general demonstrated
that they know better how to use their voices than do our American actors.

Sept., 1933] VOICE AND PERSONALITY 217

MRS. BRADLEY: Yes. Their general manner of speaking is very much higher
than ours, and when we try to imitate what we believe they do, we become overly
English. The English do not speak that way; but we imagine it is that way,
and try to imitate them. The English really speak very much in the mouth.
They do not speak with depressed larynx and distended throat.

MR. EDWARDS: Might I ask Mrs. Bradley whether the quality of the tone of
the American voice does not depend somewhat on climatic conditions?

MRS. BRADLEY: I think we hide behind that beautiful excuse very much too
much. I have not found it to be so.

MR. EDWARDS: While traveling through Wales and England, I was struck
very much by the quality of the voice entirely different from what we hear
in this country. I am sure that the environment has a great deal more to do
with it than is generally credited.

MRS. BRADLEY: Yes. But that includes our psychological environment and
the influence of others on us; it is not simply climate, it is social structure.

MR. EDWARDS: Social structure may vary a great deal. Individuals who
come from a poor social district display the environmental influence in their
manner of u^^g the voice. But I am thinking of actors. Welsh singers, for
example, have much richer tones and enunciate much better than actors in this
country; I mean taking them as a whole.

MRS. BRADLEY: Yes, I think you are quite right.

MR. EDWARDS: I personally attribute the difference in the voice more to
climatic conditions than anything else.

MR. SHEA: To express what I believe Mr. Edwards has in mind, I should
like to ask whether you have any idea as to the reason for the Scotch burr or the
so-called London accent, or the Irish brogue. How do such things come about,
and what have they to do with good voice production?

MRS. BRADLEY: The action of the back of the tongue is the most important
thing in vocal production other than the general vigor of the body; all those
people use the back of the tongue much more than we do. Our actions are
extremely sluggish, and there is something about our present form of diction,
the American form of diction, in which the chin is used and the back of the tongue
is kept low, causing a peculiar kind of accent. It is the back of the tongue that is
important, and you will find that it is used in varying degrees. The flexibility
of the larynx, also, is related to the position of the back of the tongue.

MR. SHEA: It is simply a device that has come into use through long ex-
perience.

MRS. BRADLEY: and association. I did not mention the ear, although I
think the training of the ear is most important of all. We do not hear ourselves
speak we hear only an internal head resonance, which is completely different
from the tone color that the listener hears. We believe that the voice sounds the
same to him as it does to us. If this were true the cultivation of the voice would
be very simple. Not until we shall have learned to hear our own voices shall
we have any understanding of what to do with them. The ear will influence
very much what happens inside the throat. If we could hear ourselves speak,
and learn to hear our own voices, we would change them. This can be done.
It is merely the training of the ear to hear the voice in the room instead of inside
the head.

THE EASTMAN TYPE lib SENSITOMETER AS A CONTROL
INSTRUMENT IN THE PROCESSING OF MOTION PICTURE

FILM*

GORDON A. CHAMBERS AND IAN D. WRATTEN**

Summary. A description of the methods used in the periodic testing of the
Type lib sensitometers used in the laboratories in Hollywood. Data are presented
to show the order of agreement among the several instruments.

With the increased use of sensitometric methods in the control
of motion picture film processing that followed the introduction of
sound track development, various forms of sensitometers were built.
Some of these depended for their modulation of exposure upon some
form of step tablet that gave an exposure varying from step to step
in the intensity of exposure while the time remained the constant
factor. 1 Others were so constructed as to give a series of times of
exposure at a constant intensity. The adaptation of the well known
Cinex testing machine to sensitometric purposes is an illustration of
this latter type of instrument.

The difference between the results obtained with these different
types of instruments, 2 coupled with the necessity of some standardiza-
tion in order to promote the interchange of sensitometric data, led
to the development of the Eastman Type lib sensitometer. This
instrument has been described before the Society by L. A. Jones. 3 It
was made available to the industry in the Spring of 1931, and at the
present time nine of these instruments are in service in Hollywood.
The following list is given to show the location of the sensitometers
in order to indicate the usefulness which this instrument has found.
Type lib sensitometers are in use at the following: Agfa Raw Film
Corp., Roy Davidge Laboratory, Inc., Eastman Kodak Co., Fox
Film Corp. Laboratory, Metro-Goldwyn-Mayer Laboratory, Multi-
color Laboratory, Paramount Laboratory, Universal Pictures Labora-
tory and Warner Bros. -First National Sound Department. It might
be mentioned that the sensitometer at the Eastman Kodak Com-

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Motion Picture Film Dept., Eastman Kodak Co., Hollywood, Calif.
218

THE EASTMAN TYPE lib SENSITOMETER 219

pany's laboratory in Hollywood is regularly used by other laboratories
not having a sensitometer.

Thus it will be seen that this instrument has become the accepted
control sensitometer in Hollywood for making sensitometric strips.
This status was recognized by the Academy of Motion Picture Arts
and Sciences by the award of an Honorable Mention 4 by the Board of
Judges for the Scientific and Technical Award for the year 1931-32.
In addition, the Release Print Committee of the Academy has recom-
mended that the Type lib sensitometer be considered the standard to
be used for the compilation and specification of data on Release
Print Practice.

The Type lib sensitometer is regularly supplied for use on 110-
volt a-c. lines either 50 or 60 cycles in frequency. It so happens that
all instruments in the Hollywood area operate on 50-cycle lines so that
the speed of the motor, which controls the exposure time, is the
same for all instruments. With the exception of one sensitometer,
which has been adapted with a three-phase motor to suit the supply
mains of that laboratory, all instruments are functioning with the
single-phase autosynchronous motor supplied.

The light sources used are 6-volt locomotive headlight type lamps.
A transformer, either with or without a voltage control unit for regu-
lation to 1 /z per cent, is used to supply the lamp current. In the
case of two of the sensitometers in Hollywood this has been changed so
that the lamps are fed from a storage -battery supply. Of the remain-
ing seven instruments, five are equipped with voltage regulators.

The calibration of the lamps and filters for a sensitometer are
given on cards supplied with the instrument. These calibrations
are changed, of course, in the event of a lamp or filter replacement.
Spare lamps and filters are available in Hollywood in order that
such changes as become necessary may be easily made with no loss of
time.

The Motion Picture Film Department of the Eastman Kodak
Company conducts a photographic check test on each sensitometer
bi-weekly. At this time the instruments are checked physically and
any adjustments that may be necessary are made. Both the nega-
tive and positive exposing conditions are checked. Supersensitive
negative and positive film are used to test the respective conditions,
the sensitometric exposures being made successively on the rolls
with the direction of exposure always the same. When strips are
not made in this manner, difficulty is experienced because of the di-

220

G. A. CHAMBERS AND I. D. WRATTEN [J. S. M. p. E.

rectional effect 5 in machine development. It has become customary
in Hollywood to perforate that end of the strip which has had the
least exposure, and strips are put into the processing machine with
this end entering the developer first. Thus the effect of directional
development is discounted for practical purposes by being always in
the same direction. When measurements are to be made which
must be free from this error, two strips are exposed in opposite di-
rections and the readings from the two are averaged.

There is a phase of the Eberhard effect that may
enter here. One laboratory made a change in the
template in the exposure plane of its sensitometer so
that the steps were separated from each other by an
unexposed area. This was done in order to make the
positioning of the densities under the nosepiece of a
polarization densitometer more easy. It was found
that this changed the gamma value obtained from
that on a strip having a continuous series of steps.
Inasmuch as these results were then not comparable
with those obtained by other users of the Type lib
sensitometer and as the purpose of standardization was
defeated, this laboratory was persuaded to change the
template. The difficulty was solved by cutting small
notches along the template so that each step was
marked in the manner in which the center step is in-
dicated in Fig. 1 . This allowed the same ease as ob-
tained by the other method when the polarization
photometer was used.

The template furnished with the instruments was
so made as to allow the exposure of two strips side
by side longitudinally on the film. As this exposure
fell within the area normally used by the sound
track, most instruments now have had new templates
fitted, which made a single exposure down the center
of the film, thus allowing the exposure of a sensi-
tometric strip on the sound takes without interference.

After the test exposures have been made on all the sensitometers,
the two rolls of film are developed on continuous machines under
normal processing conditions for picture negative and positive films.
The resulting strips are then read and plotted.

In Table I are given the density readings and the gamma values

FIG. 1. Show-
ing method of
positioning
strip by
notched tem-
plate.

Sept., 1933] THE EASTMAN TYPE lib SENSITOMETER 221

obtained for each of the sensitometers, which are designated by
serial number, from a recent test. These values are for the negative
exposure condition. The last column gives the maximum variation
in density obtained for each step. These values illustrate the close
agreement among the different instruments.

TABLE I

Densities and Gammas of Sensitometers in Service in Hollywood

Negative

Serial No. 502

503

504

505

506

507

509

516

522

Maximum
Density
Difference

Step

1.66

1.69

1.66

1.67

1.66

0.03

1.62

1.59

1.62

1.60

1.61

1.59

1.60

1.61

0.03

1.52

1.51

1.49

1.51

1.50

1.52

0.03

1.42

1.39

1.41

1.39

1.42

0.03

1.32

1.34

1.32

1.31

1.32

1.33

0.03

1.22

1.26

1.22

1.23

0.04

1.15

1.14

1.15

1.13

1.14

1.12

1.15

1.14

0.04

1.04

1.02

1.03

1.02

1.03

1.02

1.04

1.03

0.02

0.91

0.92

0.90

0.89

0.91

0.89

0.03

0.79

0.80

0.79

0.80

0.77

0.79

0.80

0.03

0.69

0.67

0.69

0.68

0.70.

0.67

0.69

0.70

0.03

0.58

0.57

0.58

0.59

0.02

0.50

0.49

0.50

0.48

0.49

0.50

0.02

0.37

0.38

0.40

0.37

0.38

0.39

0.42

0.05

0.30

0.31

0.30

0.33

0.31

0.32

0.03

0.23

0.22

0.23

0.26

0.24

0.25

0.04

0.18

0.17

0.19

0.17

0.16

0.20

0.17

0.04

0.14

0.15

0.16

0.14

0.16

0.13

0.14

0.03

Average . 03

7 0.69 0.69 0.69 0.68 0.69 0.68 0.68 0.70 0.70

Table II contains comparable data from the positive test run on
the same date. This table illustrates very well the lack of agreement
of one instrument, in this case No. 505, with the others. It has been
found that the failure of a lamp or filter occurs gradually, as would be
expected, and for this reason the successive tests are watched closely
for any tendency on the part of one instrument to drift from the
mean. In order to prevent this, the Eastman Kodak sensitometer is
periodically checked against standard lamps and filters that are
carefully preserved and used for this purpose alone. When, as in the
example shown in Table II, one instrument is found not to agree,
careful tests are made to determine whether it is the lamp or the filter
that is at fault. When this has been determined, a report is made
to the laboratory, on a standard form, suggesting that the necessary

222 G. A. CHAMBERS AND I. D. WRATTEN [J. S. M. P. E.

replacement be made. Excellent cooperation has been obtained
from the laboratories in maintaining the instruments at standard
conditions by such replacements as have been necessary to date.

TABLE II

Densities and Gammas of Sensitometers in Service in Hollywood
Positive

509 516 522 ^DensUy

Difference*

0.04
0.04
0.05
0.06
0.05
0.05
0.05
0.04
0.06
0.05
0.03
0.04
0.03
0.02
0.04

Serial No. 502

503

504

505

506

507

509

516

522

Step
1
2

2.89

2.87

2.86

2.91

2.89

2.90

2.91

2.70

2.66

2.65

2.71

2.69

2.70

.52

2.51

2.49

2.46

2.53

2.51

2.54

.31

2.32

2.29

2.24

2.35

2.28

2.34

2.33

.12

2.14

2.09

2.02

2.14

2.10

2.14

1.93

1.88

1.83

1.92

1.88

1.91

1.88

1.66

1.62

1.56

1.67

1.62

1.67

1.66

1.42

1.38

1.31

1.42

1.39

1.42

1.13

1.08

1.01

1.12

1.09

1.12

1.14

0.83

0.81

0.74

0.85

0.82

0.84

0.82

0.59

0.57

0.49

0.60

0.57

0.60

0.59

0.38

0.37

0.31

0.39

0.38

0.40

0.38

0.21

0.19

0.14

0.19

0.21

0.20

0.21

0.08

0.09

0.06

0.07

0.08

Average

1.82

1.80

1.83

1.80

1.78

1.82

1.80

* Excluding No. 505.

Up to the present time seven positive lamps and two positive niters
have been replaced. As the sensitometers are used mostly for the
exposure of positive materials, it has been necessary to replace only
one negative filter and none of the lamps. It is obvious, therefore,
that the maintenance of the sensitometers has not entailed a great
deal of expense. From the mechanical standpoint, the sensitometers
have been very satisfactory, only two failures of this type having
been reported. In both cases the necessary repairs were made by
the mechanical departments of the respective laboratories.

It is felt by the writers that this testing service, which is done with-
out charge, is unique in the field of precision engineering instruments.
The service is maintained merely to assure agreement in results be-
tween the several sensitometers. The agreement among the instru-

Sept., 1933] THE EASTMAN TYPE lib SENSITOMETER 223

ments to date is believed to be excellent, and it is this order of result
that is to be desired.

REFERENCES

1 DORAN, A. B.: "Photographic Sensitometer," U. S. Patent No. 1,883,884,
October 25, 1932.

2 JONES, LOYD A.: "Photographic Sensitometry," Part I, /. Soc. Mot. Pict.
Eng., XV (Oct., 1931), No. 4, p. 491; Part II, Ibid. (Dec. 1931), No. 6, p. 695;
Part III, Ibid., XVI (Jan., 1932), No. 1, p. 54; Part IV, Ibid. (Mar., 1932), No.
3, p. 324.

3 JONES, L. A.: "A Motion Picture Laboratory Sensitometer," /. Soc.
Mot. Pict. Eng., XV (Oct., 1931), No. 4, p. 536.

4 Technical Bulletin, Academy of Motion Picture Arts and Sciences, Supple-
ment No. 18, Nov. 18, 1932.

5 CRABTREE, J.: "Directional Effects in Continuous Film Processing," /.
Soc. Mot. Pict. Eng., XVTH (Feb., 1932), No. 2, p. 207.

THE SOUND FILM PROGRAM OF THE UNITED STATES
DEPARTMENT OF AGRICULTURE*

RAYMOND EVANS**

Summary. The motion picture program of the U. S. Department of Agriculture
contemplates a gradual changeover from silent to sound films, largely of the lecture
type. A few sound films have been made, but realization of plans for distribution has
been slow for these reasons; (1) The depression has hit farmers harder than any other
class. Most of our 4000 County Agricultural Agents can not now buy sound equip-
ment at any price; (2) the cost of portable sound equipment has been outrageously
high; (3) uncertainty as to the ultimate development of 16-mm. sound-on-film equip-
ment has tended to delay decisions as to the purchase of equipment. The Department
feels that there can be no notable development in the field of educational sound films
until very cheap and efficient sound equipment is available.

The theme of this paper, as scheduled, is "The Sound Film Pro-
gram of the United States Department of Agriculture." After having
assembled material for presentation the conviction grows that
it should have been called "The Sound Film Problem of the United
States Department of Agriculture." Manifestly what we have to
present is more problem than program. This is largely owing to the
fact that the educational talking picture became a factor in visual
education just at the time when the farmers of this country were
slipping into the slough of the worst era of hard times that they have
experienced since 1873, if not the worst in our history. Because corn
sells in Iowa at ten cents a bushel and won't pay taxes on the land on
which it grows, and because a bushel of barley in the Dakotas has at
times been quoted so low that it would barely buy a three-cent
postage stamp, our sound picture program languishes. We have
made a few talking pictures, and the few copies of these that we have
been able to put into circulation are in good demand; but the vast,
potential field for agricultural educational talking pictures has barely
been touched.

To make this situation clear, we must first outline briefly the pur-

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Office of Motion Pictures, Extension Service, U. S. Department of Agri-
culture, Washington, D. C.
224

SOUND FILM PROGRAM

225

pose and scope of the work carried on by our Office of Motion Pic-
tures. This office is a unit of the Extension Service of the Depart-
ment, and its function is to make and circulate films designed pri-
marily for the use of the county agents who represent the Department
and the State Extension agencies in farming communities throughout
the country. Altogether there are over four thousand of these,
counting both agricultural agents and home demonstration agents.
Fig. 1 shows the distribution of these local representatives of our De-
partment. The areas they serve are shown in black.

For the past fifteen years these agents have used motion pictures
more or less in their work. Fig. 2 shows the extent to which our

COUNTIES HAVING COUNTY AGRICULTURAL AGENT
JUNE 30, 1931

FIG. 1. Distribution of local representatives of the Department
of Agriculture.

films are used by the county agents in an average year. Comparing
this with Fig. 3, showing the total use of United States Department
of Agriculture films for an average year, we see that the county agents
are responsible for a large percentage of our distribution. Thirty-
five-millimeter silent projectors are owned in a considerable number
by the county organizations that contribute to the support of these
agents, largely DeVry, Holmes, and Acme portables. Practically
all these equipments were bought before the beginning of the era of
sound. Since sound came in, the purchasing of projectors for this
purpose has apparently fallen off to almost nothing.

226

RAYMOND EVANS

[J. S. M. P. E.

This situation, I have indicated, is in large measure attributable
to the slump in farm income throughout the country. The county
agents, for the most part, are financed locally, and when farm incomes
shrink the budgets of the county agents shrink too. For the past
two or three years the county agent's very job has been in jeopardy.
Naturally, there has not been much money available for motion
picture equipment.

This, however, is not the only factor to be taken into consideration.
The excessive cost of sound equipment has served to keep sound
equipment out of reach of the few county agents who might have

AVERAGE ANNUAL USE Of UNITED STATES

DEPARTMENT Of A6KICULTUKE MOTION PICTURES
\Y__COUHTY AGENTS ,

etc.* OOT

FIG. 2. Chart showing extent of use of films of Department of Agriculture

by county agents.

bought sound equipment had prices been comparable to the prices
hitherto asked for silent equipment.

When the first sound-on-film portables, so-called, began to come on
the market, three or four years ago, the representatives of the manu-
facturers were doubtless vexed, and certainly incredulous, when we
told them that an outfit requiring a half-ton truck to carry it would
never be accepted by our clientele as a portable projector. And
when we told them that a price of four thousand dollars a pair, or
more, was utterly out of the question, that sound projectors would
have to sell for five hundred dollars or less, complete, and at the
same time be truly portable, before they could hope to do business

Sept., 1933]

SOUND FILM PROGRAM

227

with those who use our films they were aghast. In passing, I
should like to point out that good portable sound projectors, com-
plete, are now selling for about five hundred dollars. If times were
normal, I believe there would be a good market for equipment of this
type.

The manufacturer who is accustomed to thinking in terms of ex-
pensive professional equipment is not easily convinced that there is
not some way in which comparatively expensive equipment can be
sold for use in visual education. The fact that funds for equipment
for visual education are meager, even in the best of times, can not be
too strongly stressed. In this connection may be cited the phe-

AYERAGE ANNUAL DISTRIBUTION OF UNITED STATES

DEPARTMENT OF AGRICULTURE MOTION PICTURES

DAYS' USE,

FIG. 3.

Chart showing total use of films of Department of Agriculture
in an average year.

nomenal success of the film strip in supplanting the glass slide for
still projection. The film strip is admittedly a poor and grainy sub-
stitute for a glass slide, but a full set can be bought for fifty cents, and
the projector can be bought for thirty dollars, more or less. Whether
we like it or not, the film strip is going to take the field. It has al-
ready taken the field in the work of the United States Department of
Agriculture. There is nothing left to do but to hope for finer-grained
emulsions and a change to a standard in which the picture runs
lengthwise instead of across the film. Similarly, it is folly to hope
that the highest grade and most expensive sound equipment is going

228 RAYMOND EVANS [J. S. M. p. E.

to find a place in the educational field. It is reasonable, however, to
hope that truly portable projectors, with amplifiers and speakers
giving pleasing response, will find a good market in the educational
field when times get better.

Another consideration that has retarded our progress is the un-
certain status of 16-mm. sound-on-film. Coincidently with the
marked improvement that was made in 16-mm. illumination two
years ago, we began to issue nearly all our new silent films on 16-mm.
as well as on 35-mm. stock, and it is our belief that eventually a large
proportion of our distribution will be in the 16-mm. field. We have
made no 16-mm. sound films, thus far, however, for reasons that are
rather obvious.

In the first place there is no demand. The people who use our
films have no 16-mm. sound projectors. Furthermore, we would not
know what to advise if these people should come to us for advice as
to the purchase of 16-mm. sound equipment. We have listened
to good 16-mm. sound-on-film produced by re-recording and by both
of two systems of optical reduction, and we have heard rumors and
representations as to successful mechanical sound track on 16-mm.
film, but we have not yet seen on the market any 16-mm. sound sys-
tem that could be recommended to our clients as representing a
standard that could be accepted without question or argument.

Naturally, we have hoped that some system of producing prints at
a cost comparable to the cost of silent prints would eventually find
general acceptance. Though the quality of sound obtained by re-
recording has been satisfactory, we feel that the cost of re-recording,
thus far, has been prohibitive, as far as the educational field is con-
cerned. We realize, to some extent, we hope, the engineering and
merchandising difficulties to be overcome in making available for
visual education 16-mm. sound prints at a reasonable price. We
believe, however, that the game is worth more candles than have al-
ready been burned, and we hope that we have not yet heard the last
word on 16-mm. sound standards.

From what I have said it will be clear, I trust, that our Department,
in changing over its motion picture activities to a talking picture
basis, has a rather difficult problem to solve.

Our plans then, so far as we can envisage them now, contemplate
the continued production of 35-mm. sound pictures on a modest
scale, pending development along the lines we have discussed. We
feel that the efficiency of the 16-mm. projectors, as far as illumination

Sept., 1933] SOUND FILM PROGRAM 229

is concerned, is now such that we can recommend them as suitable
for the type of educational work done by our county agents. When
and if 16-mm. sound projectors settle down to a generally accepted
standard that promises to be fairly permanent, we shall begin issuing
our sound films in that size. We trust that we shall find a cheap and
satisfactory way of making such sound prints from our 35-mm. nega-
tives. We do not anticipate, however, a quick and general change
from 35-mm. to 16-mm. film on the part of our county agents. There
are over 1200 silent 35-mm. portable projectors in the hands of county
agents, and a fairly large percentage of these agents seem to favor
the standard film for serious educational work. It is safe to say that
many of the old projectors will be rebuilt for sound or turned in on
new 35-mm. portable sound projectors. Thus, it must not be as-
sumed that the 16-mm. projector is destined wholly to displace the
35-mm. portable in this field.

In this connection we must bear in mind the fact that portable
16-mm. sound equipment, irrespective of the projector itself, is as
heavy as, or heavier than, equivalent 35-mm. sound equipment.
It takes transformers to achieve great amplification, and transformers
are made of iron and copper. The 16-mm. projector loses its great
advantage over the 35-mm. projector as to portability when it an-
nexes a heavy amplifier and associated speaker. Thus, the relative
cheapness and portability of the film remains the only great advan-
tage of the 16-mm. system when it goes over to sound. This is a
factor that those who have not actually lifted and carried about a
complete 16-mm. sound-on-film outfit may not have taken into con-
sideration.

Our plans, therefore, do not contemplate a one-hundred per cent
change-over from silent to sound pictures. The demand for our
silent films has fallen off but slightly since the advent of sound, and
we expect to be circulating many silent educational pictures for some
years to come.

DISCUSSION

MR. RICHARDSON: I notice that Massachusetts, Rhode Island, Pennsylvania,
and Ohio are covered by the Department very thoroughly, whereas states like
Missouri, Michigan, Wisconsin, and others that are essentially agricultural, are
not so thoroughly covered. Why is that?

MR. EVANS: Due to local state extension service: in some states they believe
in motion pictures and in some they do not; some have projectors and some do
not. For instance, in Ohio every accredited high school has a motion picture
projector, which furnishes an outlet. That is not true in Michigan.

THE PRESELECTION OF TAKES FOR PROCESSING FROM
EXPOSED UNDEVELOPED NEGATIVE*

DAVID W. RIDGWAY**

Summary. Desired "takes" are preselected from rolls of sound film shot on
production, only takes required for printing being developed. Undeveloped "out"
takes are reversed, spliced together, and later used for printing "dailies." The
methods of "breaking down" the predeveloped negative, necessary precautions, and
the economies effected are described in detail.

In recording sound on film for motion pictures, many takes are
usually made on a single roll of negative, f Some of these takes are
good and are chosen to be printed; others are unsatisfactory. Pre-
viously to the adoption of the method described in this paper, all
takes, good and bad alike, were developed, whether prints were to
be made of them or not. The only exception occurred when no
satisfactory take was made on a roll of film, in which case the entire
roll was held undeveloped. After development, each roll was
broken down and the takes to be printed were separated from the
rejected, or "out," takes. The film that had been exposed on
unsatisfactory takes and the cost of developing it were wasted.

By separating the good from the bad takes before development,
two savings are made possible. In the first place, the cost of de-
veloping film that is not to be used in the finished product is avoided.
In the second place, owing to the fact that the sound track occupies
only a very narrow area near one edge of the film, the film may be
reversed and spliced together, and prints for rushes and editing
purposes may be made on the opposite edge.

Method. The method described was suggested by the writer and is

* Presented at the Spring, 1933, Meeting at New York, N. Y.

** RKO Studios, Hollywood, Calif.

t Sound is recorded on positive stock (1) because it costs only one cent per
foot as contrasted with the usual cost of four cents per foot for negative; (2)
because it has a fine granular characteristic. The positive stock upon which the
original sound record is made, is, of course, the sound negative from which, in
turn, prints are made. Prints are also made on positive stock.
230

PRESELECTION OF TAKES FOR PROCESSING 231

now being used on sound film at RKO Studios in Hollywood.* The
application of the method to picture negatives and the additional
problems involved will be discussed later in this paper.

It has always been customary to expose a ten-foot strip of film to be
used by the laboratory for determining development time at the begin-
ning of each roll of sound negative. When takes are to be selected
before development, a test strip twice the usual length is exposed.
Half this test strip is sent to the laboratory with the takes selected for
development, and the other half is retained with the "out" takes,
to be used in case any of the latter are later ordered to be printed.

Between takes a notch is punched at the edge of the film. Each
take is identified by hand-punched marks designating scene and
take numbers. The operator on production makes two sets of re-
ports, one enumerating all the takes that were shot and the other
only those takes that are chosen to be printed.

Equipment. In the breaking-down operation, a small amount of
inexpensive equipment is required. The room in which the film
is broken down is the usual positive darkroom. It is lighted, how-
ever, by a Wratten series OA yellow safe-light, which affords sufficient
illumination to render easily legible not only the pencil writing on
the reports attached to the film magazines by the recorder operator,
but the punch marks that identify the individual takes, as well.

A special film rewind, which accommodates simultaneously two
separate rolls of film, is used. On one roll are wound the takes
selected to be printed, and on the other the rejected or "out" takes.
One-half the exposure test strip is separately wrapped and put into
the can with the film that is to go to the laboratory. The remain-
ing part of -the exposure test strip is put into cans with the "out"
takes.

The operator who breaks down the film, wearing gloves so
as to avoid making finger-prints, allows the film to pass be-
tween his fingers, stopping the rewinds whenever he comes to a
notch between takes. The film is broken between the accepted and
rejected takes, and each take is wound on its proper spool. No
attempt is made to splice takes together in the break-down room.
All film to be developed is spliced at the laboratory, where film to

* Experiments were undertaken and the system was placed in practical opera-
tion under the direction of the RKO Technical Committee, composed of Messrs.
C. Dreher, J. V. Maresca, W. Eglinton, J. Wilkinson, J. Swain, F. Garbutt,
L. E. Clark, and John Cass.

232 DAVID W. RIDGWAY [j. s. M. p. E.

be processed is checked for any weakness or breaks and where any
film weakened by notches is strengthened.

Risks. In breaking down the predeveloped negative, there are
the usual hazards incident to handling and processing film: (l)
scratches and rubs, (2) static, (3) breaks in developing tanks due to
faulty splices, (4) abrasions of sprocket holes and edges of film,
(5) fog, and (6) dirt.

Precautions. The danger of scratching and rubbing may be
minimized by carefully handling the film and by taking care not to
wind it too tightly. Static may be largely avoided by rewinding the
film slowly. Little difficulty has been encountered in the form of
splices breaking in tanks. The possibility of abrading the film at
the sprocket holes or at the edges has not proved to be great. If
the darkroom is properly equipped, the danger of fogging is insig-
nificant. Proper ventilation of the darkroom and careful handling
of the film are necessary to avoid getting dirt on the film.

Special care must always be taken in all phases of handling un-
developed film. A conscientious and painstaking operator, who
will guard not only against the various hazards, but who will also be
meticulous about checking and rechecking film and reports, in order
to send the proper takes to the laboratories, is, therefore, essential.

Added Safety Factor. Although certain risks are involved, an
important factor of safety is gained as the result of breaking down
the negative before development. As was mentioned before, pre-
viously to the adoption of this system, all takes of scenes were sent
to the laboratory at the same time for development. If a failure
should occur in the laboratory, for example, and the negative were
stopped in the developer, all takes of a given scene might be damaged.
If only one take of a scene is at the laboratory, only that one take
can be damaged. Since usually two or more satisfactory takes are
made of a scene, the good takes which have been held at the studio
can be developed and printed, and retakes avoided. This addi-
tional factor of safety ordinarily makes it unnecessary to use two
recorder heads and to make duplicate sound records on location or
at other places where the danger of damage during processing has
to be specially guarded against.

Storage of "Out" Takes. Storage space for the undeveloped film
must be provided and the film in the cans must be indexed so as to
be readily available. The undeveloped film is ordinarily kept until
the picture has been completely edited and shipped.

Sept., 1933] PRESELECTION OF TAKES FOR PROCESSING 233

Reversal and Use of Printing Stock. When it has been determined
that there will be no further call for the rejected, undeveloped takes,
they are removed to a splicing room, which is lighted in the same
manner as the positive darkroom. Here the notched portions, the
identification marks, and the five feet of film preceding and follow-
ing, that might possibly have been fogged, are cut out. In
practice, no take of less than forty feet in length is used. All the
other takes are spliced together into continuous, 1000-foot rolls.
The rolls are put into cans labeled according to the emulsion numbers
on the film. The laboratory, using the portion of the film opposite
the part that was originally exposed in recording, makes prints of
daily rushes and prints for editing purposes on the reclaimed
stock. This procedure was suggested by Mr. J. Wilkinson of RKO
Studios.

Splicing Precautions. Special precautions must be taken in the
splicing operations. Here again, the skill and conscientiousness of
the film splicer are all-important. He must wear gloves at all times
to prevent finger-prints and grease from getting on the film. The
gloves must be changed frequently, as they become soiled or wear
through from constant contact with the film. A carefully ground
emulsion scraping blade must be used on the splicing machine. It
must be set exactly in the scraper so that the blade thoroughly
scrapes the emulsion from the portion of the film to be spliced and
yet does not cut into and weaken the celluloid base. It must scrape
cleanly and evenly. To do this effectively, it has been found that
if the emulsion is scraped from the center outward in both directions,
a better splice is made possible than when the emulsion is scraped
off in complete sweeps from side to side. Well prepared contact
surfaces and a film cement of high quality assure the greatest possible
strength in the splices. With these precautions, practically no
difficulty has been encountered from broken splices.

It has been necessary to make reprints occasionally when spliced
stock has been used, because of fogged spots that have been un-
noticed by the splicer. However, less than four per cent of the
footage of the prints made on reclaimed spliced stock have to be
reprinted for any reason. This percentage is being decreased as
various minor difficulties are remedied and eliminated.

Economies Effected. At RKO Studios, laboratory developing
charges for sound film have been reduced somewhat more than fifty
per cent as a result of processing only selected takes. About ten

234 DAVID W. RIDGWAY [j. s. M. P. E.

per cent of the total studio outlay for positive raw stock is saved as
a result of using salvaged spliced stock.

Use of System on Predeveloped Camera Negative. Warner Bros.-
First National Studios have been breaking down predeveloped
picture negative for several years. The system was adopted there,
not primarily as an economy measure, but at a time when the
laboratory facilities were not adequate to develop all negative. The
situation was met by developing only the good and usable takes. The
procedure followed there, with certain exceptions, is similar to that
of the RKO Studios on predeveloped sound negative.

Exposure test strips are made each time photographic conditions
are changed. While in the camera, the negative is notched at the
end of each exposure strip and between takes. Reports are made
listing the "selected," the "hold" and the "NG" takes. "Selected"
takes are those from which prints are to be made immediately.
"Hold" takes are those that are deemed usable, but not ordered to
be printed. The "NG" takes are those that probably will not be
used.

Camera takes are identified by photographing a slate bearing the
scene and take numbers instead of by punching these numbers on
the film, as is done with sound negative. The photographed identi-
fication is invisible, of course, until after the film has been developed.

The person separating the undeveloped takes into their proper
classes must work in almost complete darkness. His only source
of illumination is a small flashlight, which makes it possible for him
to read the tickets classifying the various takes. Knowing the
number of takes and tests on a roll of film, the operator, by counting
notches, can separate the takes and the tests that go with them into
their proper classes.

The "selected" and "hold" takes are developed. The "NG"
takes are put into cans and held until the picture has been edited
and shipped.

The hazards that are present while working with undeveloped
sound track are greater when working with undeveloped picture
negative because the risk of fogging is greater and because the area
covered by picture, and consequently subject to damage, is larger
than the area covered by sound track. The precautions previously
suggested in connection with sound negative should, accordingly,
be more painstakingly observed when working with picture negative.

Mr. Fred Gage, in charge of Warner Bros.-First National Labora-

Sept., 1933] PRESELECTION OF TAKES FOR PROCESSING 235

tories, who furnished this information on the breakdown of pre-
developed camera negative, reports that delays and difficulties
encountered are negligible and that the savings effected justify the
use of the system.

Since the picture covers the greater part of the film, camera nega-
tive, unlike sound negative, once exposed, can not be used again.
As a consequence, the only saving resulting from breaking down the
camera negative before development is the cost of the laboratory
processing of the film that is not developed.

At the RKO Studios, a semi-automatic film notching device that
may be attached directly to the camera or to the sound recording
head is being developed by Mr. T. Winchester. This device will
simplify and make possible faster notching of film.

Conclusion. Faced with the necessity of decreasing the costs of
production, the studios have sought economies in every phase of
their work. The methods described in this paper are helping to
reduce costs without detriment to the quality of the finished product.

DISCUSSION

MR. PALMER: I did not quite understand the purpose for which this sound
record was intended. If for prints, would the prints be silent or sound prints?

MR. OELLER: They would be sound prints; the originally recorded negative
would show on one side of the picture. For editing that is immaterial, as the
prints are not sent to the theater; they are used only for rushes and cutting
purposes.

MR. PALMER: I don't see anything of value in the silent print. If there is
no sound on it, it is no good anyway.

MR. OELLER : It is not a silent print. The negative take that is good is printed
on that stock. The "out" stock, which has been "canned" for use in printing
dailies, has the sound print of the good take on one side and the picture on the
other, as usual; and then, in addition, it has the originally recorded negative
which was "NG." In other words, the originally recorded negative is shown
with the picture after the print is made.

MR. McNAiR: Who is charged with the responsibility of selecting the takes
to be developed?

MR. OELLER: That is the director's responsibility. A continuity girl tabu-
lates the takes to be held, those that are "NG," and those that are satisfac-
tory.

MR. J. I. CRABTREE: How would one know that nothing went wrong with
the recording?

MR. OELLER: The recordist makes a report at the end of the take, and the
girl asks whether the picture and the sound are "O.K." It is the responsibility
of the cameraman or the sound man to state at the time whether he believes the
take was good or not.

ECONOMIES IN SOUND FILM PROCESSING*
GERALD M. BEST**

Summary. The preselection method, which permits sound track negative that
)ias been exposed and rejected to be salvaged and used again, is discussed with par-
ticular emphasis on the economies achieved at Warner Bros. Studios by applying the
method. Large savings in film and laboratory costs have been made possible without
investing additional funds in equipment or changing the ordinary recording, process-
ing, or editing routines in the slightest degree.

It has been customary at the Warner Bros. Studios to process
only the "choice" scenes of picture film, retaining the "NG" and
"hold" scenes in storage at the laboratory until it is determined
that they will not be required, after which the film is disposed of
as scrap, or fogged and used for leader. Until recently, the sound
track negative, however, was processed in its entirety, the "NG"
and "hold" scenes being set aside after processing, and stored in
vaults until the end of the production season. It was not considered
advisable to break into the rolls of undeveloped sound track to
remove the selected scenes, due to the possibility of damaging the
negative in handling.

Hence, at the end of a production season there would be several
million feet of sound track negative stored in vaults, for which there
was no use. In order to avoid this waste, a system of preselecting
the takes, first introduced at the RKO Studios 1 in Hollywood, was
adopted at Warner Bros. Studios, and a most gratifying saving has
been achieved with no delay in film processing, no impairment of
sound quality, and no expenditure for new equipment or modifica-
tion of existing equipment.

The "NG" scenes are placed in containers marked with the numbers
of the rolls from which they were removed, and such other informa-
tion as is needed to identify them should they be required for ref-
erence or emergency use. The "hold" scenes are placed in other
containers marked and stored in vaults under the name of the picture

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Warner Bros. West Coast Studios, Hollywood, Calif.
236

ECONOMIES IN SOUND FILM PROCESSING 237

for which the film was recorded. If at the end of 72 hours the
"NG" scenes are not called for, they are removed from their con-
tainers, spliced into 1000-foot rolls and used for printing dailies, etc.
As three feet of film ahead of each scene is fogged when operating
the recording machine, this length of film is torn off and discarded.

The splicer has been so adjusted that the scraper does not remove
the emulsion over the full width of the patch, but leaves a very small
margin on each edge so that there will be no white lines of clear film
after the patch has been made. This avoids annoying "pops" in
reproduction, and unless the reproduced volume is very great, the
patches are not ordinarily audible.

Due to the fact that different emulsions are used in recording,
during any period of several months or more, the emulsion number
is marked on each roll of film sent to the laboratory, so that each
roll of spliced stock can be made up of sections of the same emulsion.
If different emulsions are spliced into the same roll, serious variations
in print transmission are likely to occur due to differences of speed
of the several emulsions. By printing all the dailies with the picture
aperture open, the negative track is practically entirely fogged out,
so that there is no danger of confusing it with the positive track on
the opposite side. Even if the negative track is not fogged out by
this means, it can still be readily distinguished, as it becomes ex-
tremely dense due to the high gamma to which it is developed.

The "hold" scenes are stored until the picture has been released,
as occasionally in the editing of the picture, the action or sound in a
choice scene is not suitable; in which case the "hold" scene is ordered
processed for inclusion in the picture. When the picture is released,
the "hold" scenes not called for are removed from the vault, spliced
into rolls as are the "NG" scenes, and used for printing, thereby
using all the film that was not originally processed.

The saving represented by the preselection method has been quite
appreciable. A total of 4,250,000 feet of positive raw stock was
purchased, and of this, 3,100,000 feet were used on actual produc-
tion, 940,000 feet for music score, dubbing, sound effects, voice tests,
etc., and 90,000 feet for recording tests and miscellaneous require-
ments of studio production routine.

Of the 3,100,000 feet of production recording, 1,487,000 feet were
used in choice takes, processed and printed. "NG" or "hold" takes
comprised 1,622,000 feet, the" NG" scenes becoming immediately
available for splicing and printing dailies, and the "hold" scenes

238 GERALD M. BEST

becoming eventually available. Thus a surplus of 135,000 feet
over the amount of film required for printing the dailies was created,
to which should be added the "NG" takes of music score, dubbing,
etc. Allowing a 3 per cent waste in splicing, and deducting a small
number of "hold" takes later ordered processed after being held for
some time, a substantial quantity of spliced stock is thus left over
for other uses, such as dupe picture prints occasionally required in
dubbing and music scoring, or for conditioning the developer in the
laboratory. The only new print stock purchased during the season
was for printing music score, sound effects, and other material used
in dubbing, where spliced stock would not be suitable due to the
possibility that the splices might interfere with the music or the
effects.

On this basis, at a cost of 1 cent per foot, it is obvious that an
actual saving of more than 50 per cent of the purchase price of print
stock was effected; and to this must be added the saving in proc-
essing costs, due to the fact that less than half the negative sound
track formerly processed is now put through the negative developing
machines. It was found that one batch of developer lasted at least
50 per cent longer than before, on account of the smaller quantity
of negative film processed, and the saving in chemicals, power,
and general laboratory overhead amounted to several thousand
dollars during that time. The new method did not reduce labor
costs in the processing, as the personnel released from the developing
machine crew due to the smaller quantity of film processed was put
to work breaking down the film before processing. It did, however,
eliminate the large amount of labor required in sorting out, storing,
and finally throwing out the millions of feet of "NG" and "hold"
scenes that had been processed and stored throughout the season;
and incidentally the problem of film storage space has been ma-
terially simplified.

The preselection method thus permits large savings in film and
laboratory costs without the investment of additional funds in
equipment, or changing the ordinary recording, laboratory, and
editorial routines in the slightest.

REFERENCE

1 Ridgway, D. W.: "The Preselection of Takes for Processing from Exposed
Undeveloped Negative," /. Soc. Mot. Pict. Eng., XXI (September, 1933), No. 3,
p. 230.

THE HISTORY OF THE ANIMATED CARTOON*

EARL THEISEN**

Summary. The history of the animated cartoon is traced from the earliest de-
vices used to depict motion, before the introduction of photographic processes, to the
realistic and artistic colored cartoons of the present day. The various innovations
developed for reducing the labor and cost of producing the thousands of different
photographs in seriatim for motion picture cartoons, is. described briefly in relation
to their chronology and application.

The history of the animated cartoon goes back farther than that of
the motion picture; in fact, motion pictures had their beginning as
hand drawn pictures. Long before photography had become prac-
ticable, many devices were introduced that portrayed motion by a
series of cartoon pictures. These early devices were nothing more
than toys, and were impracticable for depicting a story; however,
they were popular and did much to crystallize the demand for motion
pictures.

Five years after the discovery of the "persistence of vision" by
Peter Mark Roget, in 1826, the first attempts were made to show
motion pictorially by a series of drawings. With a device, called
the Phenakistoscope, invented by Joseph Antoine Plateau, motion
was depicted by a sequence of drawings, fourteen in number, each
drawing blending with the next in the series to show some simple
bit of action. The device was composed of two disks mounted on a
shaft, the front disk having a series of slits around its outer edge,
while the rear disk carried the drawings. The drawings were aligned
with the slits, and on peering through the slits as the two disks
revolved, the illusion of motion was created.

This was followed by the Daedaleum, or Wheel of the Devil (Fig. 1),
invented by William George Horner, in England, in 1834, which
consisted of a shallow cylinder, mounted on a stand, having slits

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Honorary Curator, Motion Picture Division, Los Angeles Museum.

239

240

EARL THEISEN

[J. S. M. P. E.

around the top. The drawings, made on strips of paper about 2*/2
feet long, were inserted on the inside of the cylinder. In these
drawings, the chief character was the devil, waving his trident.
The Daedaleum was later re-invented in France as the Zoetrope by
Desvignes, in 1860. It came to be known as the Wheel of Life be-
cause it showed action, and portrayed little every-day happenings,
such as a child jumping rope, or a man pumping water, or a cast of
actors, including an erring husband, his wife, and her rolling pin.
The rolling pin here used may be said to be one of the forerunners
of the assorted "props" that are now so valuable to the motion picture.
Many other events were faithfully recorded by hand drawings for the
Zoetrope, which were motion pictures 2 l /2 feet in length.

The Wheel of Life was first introduced in the United States by

FIG. 1. The Wheel of Life, one of
the early devices for showing animated
pictures.

William Lincoln, in 1867, and was patented on April 23, 1867, which
can be said to be the date of introduction of the animated cartoon
into this country.

The most notable of the pre -photographic inventions was the
Praxinoscope, devised by Emile Reynaud, in 1877, in France. To
Reynaud goes the credit of drawing short bits of dramatic action in
the form of plays, which he projected on a screen in the Reynaud
Optical Theater (Fig. 2). His most notable picture was Pauvre
Pierrot, drawn on a thirty-foot length of a transparent medium which
he termed "crystaloid." It should be noted that this was twelve
years before either the first Eastman raw stock with celluloid base, or
the Edison motion picture apparatus was demonstrated.

Space will permit mention only of the first few pioneers who

Sept., 1933]

HISTORY OF THE ANIMATED CARTOON

241

struggled to make pictures move before photography was available
to them. There were a great number of others, some of whom spent
a lifetime at the work. One man continued grimly to peer into his
devices until he sacrificed his eyesight. Another lost his sight,
yet continued his work with the assistance of his wife. To such men
as these our gratitude must be extended. The revaluation that
followed the perfection of the screen cartoon should not be allowed
to discredit those early movie devices; in perspective they may seem

FIG. 2.

Reynaud's Optical Theater, showing an audience viewing the play
Pauvre Pierrot.

crude. But they were received in their day with all the acclaim
accorded to Mickey Mouse today.

It was not until 1906 that the first cartoon was made on motion
picture film. It was a picture made by J. Stuart Blackton, for Vita-
graph, and was entitled Humorous Phases of Funny Faces (Figs.
3-6). A recent screening of this picture, with its 1906 copyright,
and Vitagraph trademark, showed it to consist of such cartoon bits
as a man rolling his eyes and blowing smoke at his sweetheart, a
large-nosed Semite, a dog jumping over a hoop, and ended by showing
Blackton doing a chalk-talk type of drawing, in which apparently
the cartoon starts as one thing and ends as another. This first

242 EARL THIESEN

cartoon picture required about three thousand drawings and its
running for an early-day audience, which was largely composed
of the more solid citizenry, was a signal for great mirth. The man
blowing the smoke at his sweetheart was the highlight of the picture.
Since cartoon technic at that time permitted the girl to show her
displeasure only by suitable eye movements, the picture had its
elements of humor.

The next man to make animated pictures was Winsor McCay.
The first was completed early in January, 1911, and was known as
Little Nemo. It was photographed in one-reel length by Walter
Arthur, directed by J. Stuart Blackton and released by Vitagraph as
Winsor McCay Makes His Cartoons Move. It contained over 4000
drawings, each complete with a background and was considered a
mammoth undertaking at that time, despite the fact that the present
cartoon requires as many as 12,000 drawings, which are run for a
screen time of only six minutes. McCay's second picture, How a
Mosquito Operates, was made in December, 1911, in 600 feet and was
sold to Carl Laemmle. The third, Gertie, a Trained Dinosaur, was
sold to William Fox. These pictures were used also as a vaudeville
act by McCay, who toured with them and explained their making
and technic.

To McCay goes the credit of making the first serious attempt at a
dramatic cartoon. The Sinking of the Lusitania, released on August
15, 1918, by the Jewel Productions, was a cartoon of feature length.
According to the Motion Picture News, of August 18, 1918 [it was],
"made from 25,000 drawings on gelatin by the famous artist, Winsor
McCay, requiring 22 months of work." The picture attracted
attention at this time by virtue of its length and because it was a
propaganda picture for the war. To date it has been the longest
cartoon ever made.

John R. Bray, during the period 1914-16, was granted several
patents on making animated cartoons. The first, filed January 9,
1914, and granted the same year as No. 811,165, describes a method
of registration so as to hold each picture in correct relation to every
other for photographing. The most important claim in this patent,
however, relates to the use of a translucent background over the
character drawing. It will be noted that this is a departure from the
tedious process of drawing each cartoon complete with a background
and character, as was heretofore done. With this patent, Bray
introduced the idea of making one background serve for all the

I HUMOROUS '".'

^PHASES I

FtlWY FACtS

14 15 17

FIGS. 3 TO 6. Copies from specimens of cartoon films made by Blackton in
1906 (FiG. 3, picture of Blackton doing a chalk talk in his 1906 cartoon).
FIGS. 7 and 8, the first Bray cartoon, The Artist's Dream, by J. R. Bray.
FIG. 9, Pathe newsreel cartoon of about 1912 (note the Patents Company
stencil on the edge of the film). FIG. 10, Gaumont cartoon of about 1912.
FIG. 11, Packer cartoon of 1916. FIG. 12, the first Disney series made in
St. Louis, 1921. FIG. 13, one of the Alice Cartoons made by Disney, 1923;
a combination of real-life characters and cartoon. FIG. 14, first Mickey
Mouse, Steamboat Willie. FIG. 15, first Silly Symphony, Skeleton Dance.
FIG. 16, first method of synchronizing sound and picture; originated by
Disney. The bouncing ball kept time for the musicians when scoring.
FIG. 17, a recent Mickey Mouse, showing the sound track.

244 EARL THEISEN [j. s. M. P. E.

action occurring in that scene (Fig. 7). He printed his background
from a zinc plate on thin translucent sheets of paper, which he laid
over the character drawing for photographing. This system did
not permit good quality and was used only for the first few pictures,
after which he adopted a system similar to the Earl Hurd method.
Bray further eliminated unnecessary drawing by introducing the
"stationary" drawing, which comprises the use of separate sheets of
celluloid when a part of the character is motionless while other parts
are moving. One drawing is made for the motionless part, while the
action of only the moving part of the character is drawn out.

Cartoon history may be said to date from the announcement, on
June 12, 1913, of the first Bray cartoon, The Artist's Dream (Fig. 8).
This, while not the first of the animated cartoons, was the forerunner
of the cartoon vogue. Previously to that, cartoons were largely
considered a novelty, or photographic trick. They had been used at
the end of newsreels, or vaudeville acts, and were shown more or
less apologetically. Now, audiences seeing The Artist's Dream were
left in a mood bordering on the hysterical, from laughter, and de-
manded more cartoons. The central character of this cartoon was a
dachshund, with the long "wheel base" and the short legs. This
dog, which resembles so much an animated sausage, experienced
difficulties with a flea, which interrupted the dog hi obtaining his
sausages for dinner.

Another Bray cartoon, Col. Heeza Liar, which was the "Mickey
Mouse" of that day, was by far the most popular of the early
cartoons. The first of the series, Col. Heeza Liar in Africa, was
released by Path6 in December, 1913. They were discontinued
after about five years, and again resumed in 1922 as an Out of the
Inkwell combination. Walt Lantz, who draws the Universal Oswald,
drew the later series, which consisted of a combination of the con-
ventional motion picture into which was introduced Col. Heeza Liar.

Bray has the distinction of having made a hand-colored cartoon in
1917; it attracted much attention, but was impracticable because of
the high cost of coloring each frame.

Earl Hurd introduced the modern technic of making cartoons.
On December 19, 1914, he filed an application for a patent which,
on June 15, 1915, was granted as No. 878,091. In this patent,
he claimed the use of a transparent medium bearing the moving
parts of the cartoon over an opaque background. Hurd was the first
to use celluloid for his action drawings, which he laid over a back-

Sept., 1933] HISTORY OF THE ANIMATED CARTOON 245

ground, as is done today. It will be remembered that Bray, in
his first efforts, drew his backgrounds on a translucent medium,
which he laid over his characters for photographing, and wherever
the background interfered or covered the character, that part of the
background was removed. Earl Kurd's first cartoons were the
Bobby Bump series. Bray and Hurd combined their patents and
formed the Bray-Hurd Company early in 1917.

Another early worker was Sidney Smith, who made Old Doc Yak
for the Selig Polyscope Company. The first of this series was released
on July 8, 1913. Wallace Carlson, who made Dreamy Dubb, and
later the Caminated News, which was an Out of the Inkwell
combination, was prominent at that time. Paul Terry drew Farmer
Al Falfa. Leslie Fenton drew the Hodge Podge series which were
released at the end of the Pathe newsreels.

Max Fleischer was the first to make the Out of the Inkwell
type of drawing. This is a photographed picture to which is added a
cartoon character by photographing a series of opaque cartoons
drawn on celluloid placed over previously photographed conventional
motion pictures. Fleischer's first series was Koko, the Clown, re-
leased by Paramount in 1917.

During this period Leon Searle made what was known as "cut-
outs." They were jointed characters cut out of paper and animated
across a background. Their animation was rather jerky, as were the
marionettes made about the same time by Tony Sarg. The Sarg
marionettes were figures illuminated from the rear, thus producing a
silhouette effect.

Raoul Barre, who began making the Edison cartoons, introduced
the "slash" system, whereby the motionless parts of the characters
were drawn once, and the animated parts of the characters torn
away. These moving parts were then drawn on another sheet so as to
coincide with the stationary parts. The two were then photographed
simultaneously, one over the other, thus saving unnecessary drawing.
Barre also originated the use of registering pegs and punch holes
in the drawings for holding them in place during the photographing.

Bill Nolan, working with Barre, was the first to use a panorama
background. The panorama, with the characters moving past the
background, was an innovation and made a decided improvement
in the action. That was in 1916.

Most of the cartoons shown on the screen in 1917 were greatly
inferior to those of today. They were crude and the characters im-

246 EARL THEISEK (J. S. M. i>. .

perfectly synchronized. They would walk either too fast or not fast
enough; the leg movements seemed to create the illusion that the
feet were being dragged somewhat in the manner of a skater's sliding
over the ice. Another characteristic was the "bubble" type of title.
This title was similar to the present press cartoon title, in which the
wording appears in a balloon with a line leading down to the character.
When the title appeared on the screen, the character would come to a
pause, face the audience, and "yap" or rapidly open and close its
mouth to represent talking. This, of course, greatly interfered with
the continuity of the story.

Into this period entered the International Feature Syndicate,
formed by William R. Hearst. He placed Gregory La Cava in charge,
who immediately set about improving the cartoons. He increased
the number of drawings from the 2000 of the average cartoon of the
time to 3500, resulting in smoother animation. Further, he changed
the animation of the characters from the stiff, angular movements of
the legs and arms to a smooth "rubbery" animation, such as is used
at present. La Cava also discontinued the "bubble" title for the
conventional title of the silent days.

Beginning in 1917, the International Syndicate released such
cartoons in series as Jerry on the Job (Fig. 18), drawn by Walt
Lantz, Katzenjammer Kids, by John Foster, Tad's Indoor Sports,
drawn by Bill Nolan and released at the end of the International
Newsreel. Happy Hooligan, drawn by Jack King, Bringing Up
Father, by Bert Green, Krazy Kat (Fig. 19), drawn also by Bill Nolan
and Leon Searle, and the best of the Internationals, Silk Hat Harry,
were the principal cartoons released at this time by that company.
This last-named was drawn by Walt Lantz and La Cava, and was
first released in 1918.

The first International cartoons were made somewhat after the
principle of the first Bray cartoons, in which the background was
drawn on a translucent medium and the characters on an opaque
sheet. The background was then laid on top of the character draw-
ings. Where any part of the background interfered with the char-
acter animation, that part of the background was drawn on the same
sheet with the character. This system was discontinued after the
first few cartoons in favor of the now conventional "celluloid over the
background" method.

Other famous cartoons during the 1917-20 period were the Mutt
and Jeff series, made by Budd Fisher. The Kay Company released

S>ept., 1933] HISTORY OF THE ANIMATED CARTOON

the Terry Cartoon Burlesque, and Sterling Pictures the Zippy series.

Skipping over the years to the sound era, we come to Walt Disney
and his Mickey Mouse series, which were the first cartoons with sound.
Steamboat Willie (Fig. 14), was the first of this series and had its
premiere on September 19, 1928, at the Colony Theater in New York.
An earlier Mickey Mouse had been made but it was released later as
Plane Crazy. Mickey Mouse is probably the most popular of any
screen character, whether in real life or cartoon. He is certainly
the acme of all that the screen has to offer as entertainment.

Disney started cartoon making in St. Louis, in 1921, when he made
the Laugh-O-Gram (Fig. 12) series. In October, 1923, he and his
brother, Roy, went to Hollywood and produced the Alice Cartoons
(Fig. 13), which were a combination of real life characters and
cartoons.

(a) (b) (c)

FIG. 18. Jerry on the Job, animated by Walt Lantz in 1917; the drawing
(a) is placed beneath the drawing made on celluloid (b), and the two photo-
graphed together, producing the effect (c).

Disney's first Silly Symphony, The Skeleton Dance (Fig. 15), was
released at the Carthay Circle, in Los Angeles, in July, 1929. It was
later shown at the Roxy in New York. It was the first cartoon
picture to be rebooked for a second showing at the Roxy.

The method of synchronizing the first Mickey Mouse was by the
"bouncing ball" method (Fig. 16), in which a ball was made to
bounce in time with the music as a guide for the musicians, who
watched the picture and the ball as they appeared together on the
screen. This ball was photographed along the edge of the film,
which space was later occupied by the sound track in the release
print. Disney next used a wavy line, and finally adopted an aural
method. The last method, employing head-phones, is still in use.
Disney controls many of the sound cartoon synchronization patents.

The first Silly Symphony in color was Flowers and Trees, first
shown at Grauman's Chinese Theater, Hollywood, on July 15,

248

EARL THEISEN

[J. S. M. P. E.

1932. This was the first cartoon to employ the Technicolor Cartoon
Process, a three-color imbibition process. Judging from today's
standards it seems that it will be impossible to improve upon the
beauty of these Disney cartoons colored by this process.

Many will remember the cartoon sequence that served as an intro-
duction to the Universal picture, King of Jazz, released on March 30,
1930. It was colored by the Technicolor two-color process and was
the first cartoon on record to be mechanically colored.

Another cartoon to follow this was Ted Eshbaugh's Goofy Goat,
made by Multicolor and released at the Loew's State Theater, Los
Angeles, on March 2, 1932. It had been previewed earlier at the
Warner's Alhambra Theater, on July 6, 1931. Many will credit
this cartoon with being the first in color, since it was the first com-

(a) (b) ()

FIG. 19. Felix the Cat, animated by Wm. Nolan for International in 1917.
showing the "slash" system used by Raoul Barre in the early Edison cartoons;
the drawing of the cat (a) is cut as indicated by the dots and is superimposed
on the background drawing (b). The photograph of the combination is shown
at (c).

plete cartoon story done in color, whereas the earlier Lantz cartoon
was only an introduction for a real-life picture.

The current cartoon characters besides those named are Oswald
and Pootch-the-Pnp, drawn by Walt Lantz and Bill Nolan for Uni-
versal ; Krazy Kat and Scrappy made by the Mintz Studio ; Looney
Tunes and Merrie Melodies made by Leon Schlesinger; Flip-the-Frog,
by U. B. Iwerks for M. G. M. release; Aesop's Fables and Tom and
Jerry by the Van Buren Corporation ; Betty Boop by Max Fleischer ;
Bosko by Harman and Ising, released by Warners as a "Looney Tune ;"
Magazine of the Screen by Bray; and Terry Tunes by Paul Terry.
The Wizard of Oz will shortly be released as a series, in Technicolor.

Such is the history of cartoons. It is interesting to observe their
popularity today and then to recall their reception back in 1911-14,
when they were always coupled with real-life characters in order to
give a reason for their existence. The Artist's Dream had as an intro-

Sept., 1933] HISTORY OF THE ANIMATED CARTOON 249

duction an artist who drew a picture, and fell asleep ; then the drawing
came to life. The McCay cartoon had, as an introduction, a bet that
he could not draw motion. He was pictured making the bet, and
then the cartoon followed. Pathe, more or less hesitantly, ran a
few short, terse bits of action in cartoons on the ends of their news-
reels during 1911.

Today cartoons are a source of wonder. Those making animated
cartoons lift out and re-shape human experiences in their more lovable
form. They instill into the screen a gaiety and glow that depicts
human traits in their more desirable form. They re-create again lost
childhood. It is a form of entertainment that the screen must never
lose.

SOCIETY OF MOTION PICTURE
ENGINEERS

OFFICERS
1933

President
A N. GOLDSMITH, 444 Madison Ave., New York, N. Y.

Past-President
J. I. CRABTREE, Eastman Kodak Company, Rochester, N. Y.

Vice-Presidents

E. I. SPONABLE. Fox Film Corp., New York. N. Y.
W. C. KUNZMANN, National Carbon Co., Cleveland, Ohio.

Secretary
J. H. KURLANDER, Westinghouse Lamp Co., Bloomfield, N. J.

Treasurer
H. T. COWLING, 7510 N. Ashland Ave., Chicago, 111.

Board of Governors

H. T. COWLING, Burton Holmes Films, 7510 N. Ashland Ave., Chicago, 111.

J. I. CRABTRBE, Eastman Kodak Co., Rochester, N. Y.

P. H. EVANS, Warner Bros. Pictures, Inc., 1277 E. 14th St, Brooklyn, N. Y.

R. E. FARNHAM, General Electric Co., Nela Park, Cleveland, Ohio.

O. M. GLUNT, Bell Telephone Laboratories, 463 West St., New York, N. Y.

A. N. GOLDSMITH, 444 Madison Ave., New York, N. Y.

H. GRIFFIN, International Projector Corp., 96 Gold St., New York. N. Y.

E. HUSE, Eastman Kodak Co., 6706 Santa Monica Ave., Hollywood, Calif.

W. C. KUNZMANN, National Carbon Co., Cleveland, Ohio.

J. H. KURLANDER, Westinghouse Lamp Co., Bloomfield, N. J.

R. F. MITCHELL, Bell & Howell Co., 1801 Larchmont Ave., Chicago, 111

E. I. SPONABLE, Fox Film Corp., 850 Tenth Ave., New York, N. Y.

250

COMMITTEES

251

W. V. D. KELLEY

J. H. KURLANDER

J. CRABTREE
J. I. CRABTREE
A. S. DICKINSON
G. EDWARDS
R. M. EVANS
T. FAULKNER

G. A. CHAMBERS
W. CLARK
B. W. DEPUE
O. B. DEPUE

D. M. BALTIMORE
B. W. DEPUE

G. C. EDWARDS
R. EVANS

E. R. GEIB

A. A. COOK
W. B. COOK
H. A. DEVRY
E. GALE

COMMITTEES
1933

Color

P. D. BREWSTER, Chairman
R. M. EVANS, Vice- Chair man

J. F. KlENNINGER

G. E. MATTHEWS

Convention
W. C. KUNZMANN, Chairman

Laboratory and Exchange Practice
R. F. NICHOLSON, Chairman

A. Hi ATT
E. HUSE

D. E. HYNDMAN

E. D. LEISHMAN

C. L. LOOTENS

K. MAClLVAIN

D. MACKENZIE

Historical and Museum
E. THEISEN, Chairman

C. F. JENKINS
W. V. D. KELLEY
G. E. MATTHEWS
O. NELSON

Membership and Subscription

E. R. GEIB, Chairman
W. H. CARSON, Vice-Chairman

J. G. T. GILMOUR
W. W. HENNESSY
W. C. KUNZMANN
E. E. LAMB

Non-Theatrical Equipment
R. E. FARNHAM, Chairman

E. R. GEIB
N. B. GREEN
H. GRIFFIN
L. A. JONES

H. B. TUTTLE

M. W. PALMER

J. S. MACLEOD
R. F. MITCHELL
H. RUBIN
W. SCHMIDT
V. B. SEASE
J. H. SPRAY

T. RAMSAYB
A. REEVES
F. H. RICHARDSON
A. F. VICTOR

J. E. MCAULEY
T. NAGASE
J. A. NORLING
N. F. OAKLEY
E. C. SCHMITZ

J. H. KURLANDER
R. P. MAY
R. F. MITCHELL
A. SHAPIRO

252

COMMITTEES

[J. S. M. P, E.

C. DREHER
P. H. EVANS
A . C. HARDY
E. HUSE

H. T. COWLING
J. I. CRABTREE

M. A B RIB AT

W. P. BlELICKB

L. BUSCH
G. A. CHAMBERS
A. A. COOK
J. A. DUBRAY

J. O. BAKER
T. C. BARROWS
G. C. EDWARDS
JJ. FINN
C. FLANNAGAN

Papers

O. M. GLUNT, Chairman
G A. CHAMBERS. Vice-Chairman

G. E. MATTHEWS
P. A. McGuiRE

D. McNicoL

Preservation of Film
W. H. CARSON, Chairman

A. S. DICKINSON
R. EVANS

Progress
J. G. FRAYNE, Chairman

R. E. FARNHAM
H. B. FRANKLIN
W. C. MARCUS
F. S. IRBY

E. E. LAMB

Projection Practice
H. RUBIN, Chairman

S. GLAUBER
C. GREENE
H. GRIFFIN
J. HOPKINS
W. C. KUNZMANN

W. C. MILLER
K. F. MORGAN
C. N. REIFSTECK
T. E. SHEA

T. RAMSAYB
V. B. SEASE

G. E. MATTHEWS
M. W. PALMER
G. F. RACKETT

P. SCHROTT

H. M. STOLLER
S. S. A. WATKINS

R. H. McCULLOUGH

P. A. McGuiRB

R. MlEHLING

F. H. RICHARDSON
V. A. WELMAN

E. R. GEIB
H. GRIFFIN

J. H. KURLANDER

R. E. FARNHAM
H. P. GAGE

F. C. BADGLBY
B. W. DBPUB

Projection Screens
S. K. WOLF, Chairman

W. F. LITTLE
A. L. RAVEN

Projection Theory
A. C. HARDY, Chairman

W. F. LITTLE

Publicity
W. WHITMORE, Chairman

D. E. HYNDMAN

F. S. IRBY

W. C. KUNZMANN

R. T. RASMUSSEN
H. RUBIN
C. TUTTLB

W. B. RAYTON
C. TUTTLE

G. E. MATTHEWS
D. McNicoL

Sept., 1933]

COMMITTEES

253

M. C. BATSEL
P. H. EVANS
E. W. KELLOGG

Sound
H. B. SANTEE, Chairman

C. L. LOOTENS
W. A. MACNAIR

W. C MILLER
H. C. SILENT
S. K. WOLF

W. H. CARSON
L. E. CLARK
J. A. DUBRAY
P. H. EVANS
R. M. EVANS
R. E. FARNHAM
C. L. FARRAND

Standards and Nomenclature
M. C. BATSEL, Chairman

H. GRIFFIN
A. C. HARDY
R. C. HUBBARD
L. A. JONES
D. MACKENZIE
G. F. RACKETT
W. B. RAYTON
C. N. REIFSTECK

H. RUBIN
H. B. SANTEE
V. B. SBASE
T. E. SHEA
J. L. SPENCE
E. I. SPONABLE
S. K. WOLF

L. J. BUTTOLPH

R. E. FARNHAM

Studio Lighting

P. MOLE, Chairman
C. W. HANDLEY

R. F. MITCHELL, Chairman
B. W. DEPUE, Sec.-Treas.

Chicago Section

New York Section

P. H. EVANS, Chairman
D. E. HYNDMAN, Sec.-Treas.

Pacific Coast Section

EMERY HUSE, Chairman
G. F. RACKETT, Sec.-Treas.

J. H. KURLANDER
W. J. QUINLAN

O. B. DEPUE, Manager
J. E. JENKINS, Manager

M. C. BATSEL, Manager
J. L. SPENCE, Manager

C. DREHER, Manager
J. A. DUBRAY, Manager

WILLIAM C. HUBBARD

Mr. William C. Hubbard, member of the Board of Governors of the Society
of Motion Picture Engineers, died at his summer home at Long Lake, N. Y.,
on July 20, 1933. He had been associated with the Society since 1918, and had
been very active in administrating its affairs in the capacity of Treasurer from
1925 to 1930, as Vice-President from 1930 to 1932, and as an elected member of
the Board of Governors from 1932 until his death. Mr. Hubbard was born in
Plainfield, N. J., on October 15, 1866. He attended the local schools of that city;
and, after graduating from Plainfield High School in 1884. attended Alfred Uni-

WILLIAM C. HUBBARD

versity and was later graduated from Rutgers in 1890, with the degree of Electrical
Engineer. In his early business life he was connected with the Westinghouse
Electric & Manufacturing Company, but in 1905 became associated with the
Cooper Hewitt Company, of Hoboken, N. J., now known as the General Electric
Vapor Lamp Company.

Mr. Hubbard was also very active in numerous civic, religious, and fraternal
organizations. In addition to having been closely affiliated with the Society of
Motion Picture Engineers, he was a member of the American Society of Me-
chanical Engineers, American Institute of Electrical Engineers, and the Illuminat-
ing Engineering Society. Mr. Hubbard is survived by his wife, Mabel Potter
Hubbard, and a daughter, Dorothy.

The Society regrets very deeply the loss of one of its most esteemed members.
254

SOCIETY ANNOUNCEMENTS

FALL MEETING OF THE SOCIETY, EDGEWATER BEACH
HOTEL, CHICAGO, ILL.

OCTOBER 16-18, INCLUSIVE

CONVENTION ARRANGEMENTS COMMITTEE
W. C. KUNZMANN, Chairman

J. H. KURLANDER M. W. PALMER

LOCAL ARRANGEMENTS COMMITTEE
R. F. MITCHELL, Chairman

J. E. JENKINS
H. T. COWLING
H. H. DEVRY
A. WARMISH AM

R. P. BURNS
G. W. BAKER
O. B. DEPUE
B. W. DEPUE

O. HOLMES
J. E. McAuLEY
J. H. GOLDBERG
O. F. SPAHR

MRS. R. F. MITCHELL
MRS. H. T. COWLING

LADIES COMMITTEE

MRS. O. B. DEPUE, Hostess

assisted by

MRS. B. W. DEPUE MRS. J. E. JENKINS
MRS. A. WARMISH AM

OPENING OF CONVENTION

The Convention will convene at 12:30 P.M., Monday, October 16, at the Edge-
water Beach Hotel, opening with an informal luncheon, during which the members
of the Society will be addressed by several prominent speakers. The morning
preceding the luncheon will be devoted to registration, committee meetings, and
various organization matters.

SESSIONS

All technical sessions and film programs will be held in the East Lounge of the
hotel, where also will be located the registration headquarters. The Berwyn
room will be provided for the Board of Governors and the technical committees.
Technical sessions will be held on Monday, Tuesday, and Wednesday afternoons,
and on Tuesday and Wednesday mornings. The film program will be held on
Monday evening, and will include several recent outstanding productions, in ad-
dition to several reels of the Century of Progress World's Fair, photographed by
Mr. H, T, Cowling.

255

256 SOCIETY ANNOUNCEMENTS [J. S. M. p. E.

BANQUET AND DANCE

The S. M. P. E. Semi-annual Banquet and Dance will be held in the Ball Room
of the Edgewater Beach Hotel, Tuesday evening, October 17, at 7:30 P.M. an
evening of dancing and entertainment, with no banquet speeches. Banquet tickets
should be obtained at the registration headquarters; tables reserved for eight or
ten persons.

SPECIAL EXTENDED RATES

Excellent accommodations are assured by the management of the Edgewater
Beach Hotel, and minimum rates are guaranteed. Room reservation cards
should be returned immediately to the Edgewater Beach Hotel in order to assure
satisfactory reservations, particularly on account of the heavy advance registra-
tion incident to the large attendance at the Century of Progress Fair. Special
rates have been arranged by the hotel, which will be effective during the stay of
S. M. P. E. delegates and their guests in Chicago, should they wish to visit the
Fair before or after the Convention.

CENTURY OF PROGRESS WORLD'S FAIR

Members of the Society who attend the Convention will be able to take advan-
tage of the opportunity of visiting the Century of Progress World's Fair, now
being held in Chicago, and which will close October 31. They will also be able
to benefit by the special rates provided by the Edgewater Beach Hotel and the
reduced railroad fares now in effect because of the Fair.

LADIES' HEADQUARTERS

A reception suite will be provided for the use of the ladies attending the Con-
vention, and an attractive program for their entertainment is being arranged by
the Ladies' Committee.

TENTATIVE PROGRAM
MONDAY, OCTOBER 16

9:30 A.M. Committee Meetings, Organization, Registration.
12:30 P.M. North Room:

Informal luncheon for members, their families, and friends; speakers.
2:30 P.M. East Lounge:

Report of the Secretary, J. H. Kurlander.

Report of the Treasurer, H. T. Cowling.

Convention Announcements, W. C. Kunzmann.

Society Business and Election of Officers.
8:00 P.M. East Lounge:

Interesting program of recent motion pictures and views of the
Century of Progress World's Fair, produced by H. T. Cowling.

TUESDAY, OCTOBER 17
9:30 A.M. East Lounge:

Program of technical papers.
2:30 P.M. East Lounge:

Program of technical papers.
7:30 P.M. Ball Room;

Semi-annual Banquet and Dance.

Sept., 1933] SOCIETY ANNOUNCEMENTS 257

WEDNESDAY, OCTOBER 18

9:30 A.M. East Lounge:

Program of technical papers.
2:30 P.M. East Lounge:

Program of technical papers.
Convention adjournment.

Papers Committee Convention Committee

O. M. GLUNT, Chairman W. C. KUNZMANN, Chairman

SUSTAINING MEMBERS

Bausch & Lomb Optical Co.
Bell Telephone Laboratories
Burnett-Timken Laboratories

Eastman Kodak Co.

Electrical Research Products, Inc.

National Carbon Co.

RCA Victor Co., Inc.

HONOR ROLL
OF THE

SOCIETY OF MOTION PICTURE ENGINEERS

By action of the Board of Governors, October 4, 1931, this Honor Roll was estab-
lished for the purpose of perpetuating the names of distinguished pioneers who are
now deceased:

LOUIS AlME AUGUSTIN L,E PRINCE

WILLIAM FRIESE-GREENE

THOMAS ALVA EDISON

GEORGE EASTMAN

JEAN ACME LE ROY

A New Film for
Composite Shots

I ROM the general standpoint of
fineness of grain, speed, and proc-
essing characteristics, Eastman
Background Negative is very
definitely superior to every film
hitherto available for composite
shots. Tests in the laboratory
and on the lot prove this. They
indicate that this new Eastman
film will go far to enhance the
beauty and effectiveness of to-
day's motion pictures. Eastman
Kodak Company (J. E. Brulatour,
Inc., Distributors, New York,
Chicago, Hollywood).

EASTMAN

BACKGROUND NEGATIVE

Society of Motion Picture Engineers

33 WEST 42ND STREET

NEW YORK, N. Y.

EXHIBIT OF NEW APPARATUS

AT THE
FALL CONVENTION

Chicago, III., October 16-18, 1933, inclusive
Headquarters Edgewater Beach Hotel

Arrangements are being -made to hold an exhibit of newly
developed motion picture apparatus, in order better to acquaint
the motion picture engineer with the newly devised tools of the
industry that may be of value to him.

It will not be of the same nature as the usual trade exhibit.
There will be no booths, although each exhibit will be allotted
definite space and all exhibits will be arranged in one large room.
The following regulations will apply:

1. The apparatus to be exhibited should be new or have
been developed or improved within the past 12 months.

2. Each exhibitor will be permitted to display a card giving
the name of the manufacturing concern, and each piece of equip-
ment shall be labeled with a plain label free from the name of the
manufacturer.

3. A technical expert capable of explaining the features of
the apparatus exhibited should be present during the period of the
exhibition.

4. A charge for the exhibit will be made in accordance with
the space occupied, as follows: up to 20 sq. ft., $10.00; 20 to 30 sq.
ft., $15.00; 30 to 40 sq. ft., $20.00; 40 to 50 sq. ft., $25.00.

Please make requests for space to Mr. Sylvan Harris, Editor-
Manager of the Society, 33 West 42nd St., New York, N. Y.,
stating the number and nature of the items to be exhibited.

Checks should be mailed to the editor-manager at the General
Office of the Society.

JOURNAL

OF THE SOCIETY OF

MOTION PICTURE ENGINEERS

Volume XXI OCTOBER, 1933 Number 4

CONTENTS

Page

National Standardization in America P. G. AGNEW 261

A Historical Summary of Standardization in the Society of Mo-
tion Picture Engineers L. A. JONES 280

Sound Film Printing J. CRABTREE 294

Wave Form Analysis of Variable Width Sound Records

O. SANDVIK, V. C. HALL, AND J. G. STREIFFERT 323

Military Training and Historical Films F. W. HOORN 337

Book Review 342

Society Announcements 343

JOURNAL

OF THE SOCIETY OF

MOTION PICTURE ENGINEERS

SYLVAN HARRIS, EDITOR

Board of Editors
J. I. CRABTREE, Chairman

O. M. GLUNT A. C. HARDY F. F. RBNWICK

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers.

Publication Office, 20th & Northampton Sts., Easton, Pa.
General and Editorial Office, 33 West 42nd St., New York, N. Y.
Entered as second class matter January 15. 1930, at the Post Office at Easton,
Pa., under the Act of March 3, 1879. Copyrighted. 1933, by the Society of
Motion Picture Engineers, Inc.

Officers of the Society

President: A. N. GOLDSMITH, 444 Madison Ave., New York, N. Y.
Past President: J. I. CRABTREE, Kodak Park, Rochester, N. Y.
Vice-President: E. I. SPONABLE, 277 Park Ave., New York, N. Y.
Vice President: W. C. KUNZMANN, Box 400, Cleveland, Ohio.
Secretary: J. H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J.
Treasurer: H. T. COWLING, 7510 N. Ashland Ave., Chicago, 111.

Governors

P. H. EVANS, 1277 East 14th St., Brooklyn, N. Y.
R. E. FARNHAM, Nela Park, Cleveland, Ohio.
O. M. GLUNT, 463 West St., New York, N. Y.
H. GRIFFIN, 90 Gold St., New York, N. Y.
E. HUSE, 6706 Santa Monica Blvd., Hollywood, Calif.
R. F. MITCHELL, 1801 Larchmont Ave., Chicago, 111.

NATIONAL STANDARDIZATION IN AMERICA*

P. G. AGNEW**

Summary. Standardization in America is discussed in relation to the classi-
fication of standards as company, association, national, and international standards.
The function of the American Standards Association and the standardizing activities
of the United States Government are described. Considerable attention is given to
the factors that should be considered in deciding whether to standardize, and as to the
methods to be followed in standardizing. The paper closes with a discussion of the
human factor in standardization projects, and the responsibilities of participating
groups and individuals engaged in standardizing.

The scope of the standardization activities in which technical
societies and trade associations are engaged is entirely too great to be
covered in a single paper. I shall, therefore, omit from this discussion
many important activities in such fields as the standardization of
accounting methods, bills-of-lading, trade practices, codes of ethics,
methods of arbitrating disputes, etc., and deal primarily with the
standardization of things.

In order to be explicit on fundamentals we may say that this type
of industrial standardization means to single out specific products
and materials, to settle upon their properties and dimensions, and to
concentrate upon them in production and in use all to the end of
bringing about the greatest over-all industrial efficiency.

CLASSIFICATION OF STANDARDS

Within this seemingly limited field of industrial standardization
there are actually a great many different types of standards which
must be included in any complete standardization framework. It
will be helpful to classify these various types. It is almost always
necessary to start with the standardization of nomenclature, if our
subsequent standards are not to be wasted because their terminology
is confused. We have, then :

(1) Nomenclature:

(a) Definitions of technical terms used in specifications and in contracts,
and in technical literature.

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Secretary, American Standards Association, New York, N. Y.

261

262 P. G. AGNEW [j. s. M. P. E.

(ft) Abbreviations.

(d) Graphical symbols (ideographs or pictographs) used on drawings,
schematic diagrams, and the like.

(2) Uniformity in dimensions necessary to secure interchangeability of parts
and supplies, and the interworking of apparatus.

(3) Specifications for quality of materials and products.

(4) Methods of test for materials and products.

(5) Ratings of machinery and apparatus which establish test limits under
specified conditions as a basis of purchase specifications, or which establish re-
quirements as to performance, durability, safety, etc., under operation.

(6) Provisions for safety of workers engaged in production or use of machinery
and equipment.

(7) Standard processes and operations for industrial establishments.

(8) Standards providing for concentration upon the optimum number of
types, sizes, and grades of manufactured products.

This is a simple classification for practical purposes. Other classifi-
cations may be and have been worked out on other bases, such as the
following, developed by the U. S. Bureau of Standards :

(1) Standards of measurement.

(2) Standards of constants.

(3) Standards of quality.

(4) Standards of performance.

(5) Standards of practice.

Another important classification of standards relates to the scale
upon which they are planned and carried out. The process of in-
dustrial standardization may be classified roughly into four stages,
namely :

(1) By individual companies.

(2) By societies or associations.

(3) On a national scale.

(4) On an international scale.

Every industrial plant is carrying on standardization of its own
products and processes, and its competitive success largely depends
upon the cleverness and thoroughness with which it has studied and
solved these problems. Standardization within the plant has been an
essential factor in the development of mass production; and mass
production, in turn, has been the chief contribution of the United
States to industrial development.

Standardization within individual companies, which had its
greatest rate of growth during the last half of the nineteenth century,
gave rise to collective standardization for entire industries. Such

Oct., 1933] STANDARDIZATION IN AMERICA 263

standardization by industries, carried on by technical societies
and trade associations, has for the most part been a product of this
century. The present extensive use of electric motors and lamps
would have been impossible had it not been for the collective stand-
ardization carried on in the electrical industry, not only in funda-
mentals such as voltages and frequencies, but also in such details as
the interchangeability of lamp bases and sockets. To cite another
example, the adoption of a standardized track gauge and of a standard
system of interchangeable brakes and couplings was a necessary step
in the development of our railroad system. Had these standards
not been adopted early in the period of railway construction, it would
have seriously retarded not only the growth of transportation fa-
cilities, but also our entire national industrial development which
required these facilities.

Just as standardization by individual companies led to standardi-
zation by industries, so standardization by industries has in turn
been found insufficient because so many problems affect numerous
diverse industries and require common solution. This has led to
inter-industry or national standardization in behalf of industry as a
whole, technical and trade associations here playing the same role
as individual companies in group standardization.

That standardization can not reach its greatest effectiveness until
it is treated as a national problem has been recognized in all the
leading industrial countries. At present there are national standardi-
zing bodies in 22 countries, all but one of which were organized
during or since the War. The American Standards Association, the
national standardizing body in this country, was organized in 1918.
Just why national standardization is in so many cases necessary
simple industry standards not being sufficient can best be judged
from an actual case.

The establishment of specifications for railroad ties would seem
to be a simple and straightforward matter. Yet even this very spe-
cialized subject is a striking illustration of the complexity of group
interests. Twelve national organizations were officially represented
on the technical committee on railroad ties organized under ASA
procedure. Representation of three organizations was required to
cover completely the steam and electrical railways, five organizations
to cover producers, and an additional one for the wood preservers.

Naturally much less has been accomplished in international in-
dustrial standardization than in national work. Yet beginnings have

264 P. G. AGNEW [j. s. M. p. E.

been made in several fields, and in some lines there has been sub-
stantial progress.

GROUP STANDARDS

Much of the pioneering hi group standardization activities has
been done by technical societies. Nearly all the group standardiza-
tion activities have developed during the present century. The
interests and activities of associations have, however, been rapidly
increasing and the trade association has already become a dominant
factor in the movement.

In a trade association, corporate management as such is directly
represented. In most professional societies this is not the case, since
the basis of organization is the individual. This is, I believe, the basic
reason for the steadily increasing role which the trade association is
playing in the standardization movement. Decision and authority
can not be permanently separated. It is fundamental in human
nature that management should wish to be in a position to control
decisions affecting its own policies. The normal agency for bringing
this about in cooperative work affecting competing companies is the
trade association.

These fundamental differences in points of view and in types of
organization have, in a number of instances, led to jurisdictional
struggles between technical societies and trade associations as to
which organization should be in charge of standardization programs.
In extreme cases these jurisdictional disputes have been based on
radically different philosophies of what the nature of a standard
should be ; whether it should be an ideal placed before industry as an
inspiration and a goal, or whether it should be a day-by-day working
tool of industry, forged through the give-and-take necessary in the
work-a-day industrial world, and therefore necessarily bearing the
marks of the grime of the shop. I once heard this difference in point
of view put in this extreme form: Should a standardization process
partake of the nature of a gentlemen's tea-party, or must it needs
be a business decision worked out under the stress of various, and
often conflicting, industrial forces?

It is unnecessary to add that people generally are not much
interested in "paper standards," but only in such standards as are
useful as working industrial tools.

Nearly all the national engineering societies are carrying on
standardization activities, some of them on an extensive scale.
While by no means all trade associations carry on standardization

Oct., 1933]

STANDARDIZATION IN AMERICA

265

work, several hundred of them are active. Of the 474 national
organizations participating in the work of the ASA, the great
majority are trade associations and technical societies. Four hundred
and eleven technical societies and trade associations carry on stand-
ardization activities of sufficient importance to have separate sum-
maries of their work in the Standards Year Book issued by the U. S.
Bureau of Standards.

AMERICAN STANDARDS ASSOCIATION

The central clearing house of the national industrial standardiza-
tion movement in this country is the American Standards Associa-
tion, which provides the forum by which industry carries its stand-
ardization work to the national stage.

The ultimate authority for the policies and affairs of the ASA,
which is essentially a federation of national organizations, rests hi

OMMIUTNM Or

IDA&D

1.1935

FIG. 1. Organization of American Standards Association.

the hands of the 37 constituent organizations known as Member-
Bodies. These include eleven technical societies, eighteen national
industrial associations, and eight departments of the Federal Govern-
ment. Dues for Member-Bodies are $500 a year for each representa-
tive on the Standards Council, no Member-Body being permitted
to have more than three representatives. Company members pay
annual subscriptions based upon a minute percentage of their total
annual business.

266

P. G. AGNEW

[J. S. M. P. E.

General direction of the affairs of the ASA is in the hands of a
Board of Directors of sixteen industrial executives.

Method of Work. In its work as an industrial forum or legislature,
the ASA brings together all those directly concerned with a project
to formulate a workable and acceptable standard, and submit it for
approval to the authorized committee representing all the groups
having a major interest in the problem in hand including producers,

Standardization proposed by respon-
sible body.

General conference called by ASA to
consider proposal, and to decide
whether the work shall be done, and
if so, how.

Administrative Sectional
Committee

Sponsorship and scope acted upon by
ASA.

Sectional committee organized through
coSperation between sponsor and ASA.

Make-up and personnel of sectional
committee approved by ASA.

Standard drafted by sectional com-
mittee.

Draft standard published or circulated
for criticism.

Proposed standard approved by sec-
tional committee by letter ballot.

Proposed standard approved by spon-
sor.

Proposed standard submitted for ap-
proval to ASA by sponsor.

Standard approved by ASA.

Autonomous Sectional
Committee

Sectional committee organized
ASA with advice of conference.

Standard drafted by sectional com-
mittee.

Draft standard published or circulated
for criticism.

Proposed standard approved by sec-
tional committee by letter ballot.

Proposed standard submitted for
approval to ASA by sectional com-
mittee with endorsement of at least
one cooperating body.

Standard approved by ASA.

Standard published by sponsor.

Standard published by ASA.

FIG. 2. Development of a standard under sectional committee procedure of
the American Standards Association.

consumers, and general interests. When the great preponderance of
committee opinion is favorable to the standard in question, it is sub-
mitted to the ASA for approval as an American Standard, with the
full knowledge that it represents a real national consensus.

Any responsible body, whether a member of the ASA or not, may
request the initiation of a standardization project. Opportunity to
cooperate in technical work is in no wise contingent upon financial
support. A technical staff carries on the central office work in con-
nection with the various projects. The final approval of Standards

Oct., 1933] STANDARDIZATION IN AMERICA 267

rests with the Standards Council made up of representatives of the 37
Member-Bodies of the Association.

Some 474 national organizations are officially cooperating in the
establishment of American national standards through 2700 technical
experts appointed as their representatives.

A Typical Undertaking. A fine example of the cooperative
methods used in national standardization projects relates to specifi-
cations for wood poles for power and communication lines, which
were drawn up by a technical committee consisting of 40 men repre-
senting 23 national bodies. These include representatives of the
power, telephone, telegraph, municipal, and steam and electric
railway groups, and also such diverse groups as the American Wood
Preservers' Association and the Federal Government all of which

An existing or a proprietary standard submitted to ASA for approval by a respon-
sible body.

Investigation of facts regarding consensus through ASA agencies.
Standard approved by ASA.
Standard published.

NOTB: Revisions of approved "proprietary" standards are at the discretion of the sponsor.
Revisions of approved "existing" standards are made by the sponsor or by the sectional com-
mittee to whose charge the standard has been consigned.

FIG. 3. Approval of "existing" and of "proprietary" standards by the American
Standards Association.

have a substantial interest in the project. The work was completed
on a thoroughly sound basis. Confusing elements intended merely
for sales effect were eliminated. Also a false basis of competition
between consumer groups was done away with; some of these con-
sumer groups had attempted to rig their private specifications so as to
"cut the heart out of the pie," leaving the crust for other groups.
The new national specifications have already brought about a much
freer national market than ever before existed. The product can now
flow from the forest, through the treating plant and into use without
having to be earmarked from the beginning for particular consumers.
The far-reaching influence on the $60,000,000 industry is indicated
by the sale of 30,000 copies of the specifications in a year's time.

Late in 1932 a draft report was issued which contained definitions
of some 3500 terms used in all branches of electrical science and

268 P. G. AGNEW [J. s. M. P. E.

industry. On the technical committee in charge of the project, which
was initiated hi 1928, there are about forty persons representing
thirty-three organizations, including national engineering, scientific
and professional societies, trade associations, government depart-
ments, and miscellaneous groups. This committee has seventeen
subcommittees with a total personnel of 120 at work on different
phases of the subject, and these hi turn have called freely upon non-
member experts, so that in all over 300 men in all the various fields
concerned are aiding in the task. An organization chart of the ASA
is shown in Fig. 1, and flow charts of the principal methods used in
the development of standards hi Figs. 2, 3, and 4.

GOVERNMENT ACTIVITY

Various standardization activities are being carried on by the
Federal Government which is interested in industrial standardiza-
tion in two ways : first, as a purchaser it is interested in an extraor-
dinarily wide range of specifications for materials and apparatus;
second, through its great research and service bureaus it is interested
in innumerable standardization questions.

The Bureau of Standards is the most important of the govern-
mental standardizing agencies. Its activities cover the necessary
fundamental scientific research preparatory to the setting up of
master standards for units of weights and measures as well as the
maintenance of these standards and the derivation of working stand-
ards therefrom. Much pure scientific research is involved as well as
research of immediate and practical value to industry, such as
methods of technical analysis, testing of materials, etc. The Bureau
acts as technical and scientific advisor on specifications to other
departments and to state and municipal governments, and carries on
formal liaison work with trade associations and technical societies.
Through its Division of Simplified Practice, recommendations leading
to the elimination of excessive varieties of products are made, while
through its Commerical Standards Unit industry is aided in estab-
lishing standards primarily useful in marketing commodities.*

The Bureau of Agricultural Economics of the Department of Agri-
culture maintains extensive activities in setting up standard grades,
inspection, etc., for agricultural products.

The U. S. Food and Drug Administration is charged with the
administration of federal acts concerning food, drugs, naval stores,
and the like. Fundamental hi this work are the U. S. Pharmacopoeia

* See Addendum on p. 279.

Oct., 1933] STANDARDIZATION IN AMERICA 269

and the National Formulary, which are essentially technical specifi-
cations for a large number of the more important drugs and medi-
cines. Any drug which is sold by a name or a synonym recognized
in the Pharmacopoeia or in the Formulary must conform to these
specifications.

The Federal Specifications Board is composed of representatives
of the departments and independent establishments of the Federal
Government purchasing materials in accordance with specifications.
The specifications prepared by the Board are binding upon all govern-
ment departments, bureaus, agencies, and offices of the government.
The extent of this work may be realized from the total of 780 specifi-
cations which have been promulgated by the Board.

Standardization proposed by responsible body.

Survey of field made and draft standard produced by working committee.

Draft standard considered by general conference.

Proposed standard presented to all groups concerned for their written acceptance.

Standard approved by ASA.

Standard published.

FIG. 4. Development of a standard by the American Standards Association
through "general acceptance."

The extent to which departments and agencies of the government
are interested in standardization is shown by the fact that in the 1933
issue of the Standards Year Book published by the Bureau of Stand-
ards more than one page is necessary in the table of contents to list
the various bureaus and departments active in one phase or another
of standardization.

The Federal Government is cooperating with industry and the
various trade associations and technical societies in their stand-
ardization work, and is also tying in with the national movement
through the ASA. Cooperative relations between the Bureau of
Standards and the ASA are both intimate and active. The Bureau
is officially represented on 83 ASA projects, hi 14 of which it is taking
official leadership. By special cooperative arrangement between the
Bureau and the ASA, certain Commerical Standards of the Bureau,

270 P. G. AGNEW [J. S. M. p. E.

at the specific request of the industries concerned, are being sub-
mitted to the American Standards Association for approval as
American Standards.

INTERNATIONAL COOPERATION

An important aspect of the national standardization movement is
that of international cooperation, which constitutes the fourth and
most difficult stage in the standardization process. Naturally, much
less has as yet been accomplished internationally than nationally,
yet beginnings have been made in several fields and in some of these
there has been substantial progress.

Several international bodies are undertaking standardization work
in particular fields, and the national bodies are in touch with each
other and are interchanging information on many projects in process
of development in their respective countries. Through these and
other means considerable international uniformity is being brought
about in a number of specific subjects.

The International Standards Association is a federation of the
national standardizing bodies of eighteen countries, of which the
American Standards Association is one. The main object of the
international body is to promote uniformity between the national
standards of the different countries. Through it an important
service has been rendered in the systematic interchange of information
between the various national bodies. Some thirty international
technical committees have so far been organized.

One of the oldest and most experienced of the international bodies
whose primary purpose is standardization is the International
Electrotechnical Commission, which was organized in 1906. The
Commission consists of national committees in each of the twenty-
nine countries represented. Agreements have been reached and
published on a considerable number of electrical subjects.

The International Commission on Illumination, which was mod-
eled somewhat after the lines of the International Electrotechnical
Commission, was organized in 1913.

The international organization which has the most direct bearing
on motion picture work is the International Congress of Scientific
and Applied Photography. The S. M. P. E. is cooperating with this
organization, which has an American committee.

There are many and serious obstacles in the way of international
standards, such as differences in language and industrial background.

Oct., 1933] STANDARDIZATION IN AMERICA 271

On the other hand, there are other important considerations tending
to further international cooperation, and a great deal of essential
work has been accomplished in the use of fundamental units and
methods of measurement, which are prerequisite to nearly all phases
of industrial standardization.

Whatever the ultimate outcome may be, it seems that the next
step in any case is the same to develop as full and as free an inter-
change of information as conditions will permit.

FACTORS FOR CONSIDERATION IN DECIDING WHETHER TO STANDARDIZE

Let us turn now to some of the factors which should be
considered in deciding as a matter of policy the attitude of a
society or an association toward a particular standardization pro-
posal. Some of these factors are positive and some are negative.

(i) We should never standardize anything merely for the sake of
standardizing.

What conceivable merit can there be in uniformity for the mere
sake of uniformity? It takes variety and spice from life, it tends to
make us walk in lock-step, and is probably one of the chief reasons
for the charge that is sometimes made that standardization or even
the machine itself is tending to make robots of us all.

(2} We should standardize only when a preliminary analysis of the
facts makes it seem fairly clear that doing so may be expected

(a) to result in important economies;

(b) to simplify and clarify operations;

Stated in this way we may seem to be drastically restricting the
field for standardization, but a little reflection will show that these
three categories cover an enormous field.

(Note: Let us turn aside a moment to speculate on a phase of social philosophy.
These first two principles seem to leave no place for those who are intent on
"standardizing our minds." If we accept this easy generalization we are at once
swept on to condemn fashions and styles in general completely and out of
hand.

Even more, these principles would condemn not only styles, which play and
always have played an unbelievably important role in civilization, but they
would as well condemn all other social standards which are also based solely on
herd-mindedness even those folkways which constitute our very strongest
social controls, such, for example, as codes of etiquette and moral codes in so far
as these standards are based on herd-mindedness.

While such speculations would carry us to depths which have not been ex-

272 P. G. AGNEW []. S. M. P. E.

plored from this point of view, one wonders whether such an exploration would
not validate the method even in such fields.)

(5) Standards should as far as possible be planned as a structural
part of a coordinated industry, not stuck on as mere patches.

Unfortunately most industrial standards are as yet of the patch
type. It is certain, however, that any fundamental economic plan-
ning, whether national, regional, or by industries, must embrace
standardization as a basic part of its structure.

(4) Standards should be developed wherever possible in the logical
order, the basic ones and those of most general application coming first,
and the more specialized and detailed ones coming later and being based
upon the fundamental ones.

Here again commerical considerations and other realities of the
situation often prevent what would otherwise be the most economical
and logical order of development.

(5) Very early consideration should be given to the policy to be
followed in the introduction of a new standard into practice. This
should be done not only during the process of its formulation, but it should
be considered even in arriving at a decision on whether to standardize.

For example, a company or an industry will frequently find it
greatly to its advantage to cooperate fully in the development of a
standard, even though it may not be feasible for the company to
put the standard into operation immediately. The actual introduc-
tion can await an opportune time when other changes make it possible
to introduce it economically. Numerous illustrations of this point
could be given. For example, some groups are actively cooperating
in the new specifications for manhole frames and covers, although
they do not expect to adopt them at once, but will rather work toward
their introduction at such time as developments make this eco-
nomical and convenient.

(6) A standard based on performance requirements is nearly always
preferable to one based on construction details.

The reason for this is that it gives greater freedom in methods of
production and frequently stimulates new developments. This is a
point which should normally be considered in arriving at the initial
decision as to whether a particular piece of standardization work
should be undertaken.

(7) The old idea of making things "special" in order to get the busi-
ness of supplying parts and making repairs is now pretty thoroughly
discredited.

Oct., 1933] STANDARDIZATION IN AMERICA 273

Under modern conditions such a course usually keeps a firm out
of more business than it retains for it. The evolution of industry
under mass production methods has outmoded such narrow policies.

(8) Standards dealing with dimensions, specifications, rating, etc.,
should be sufficiently clear and definite to serve as criteria in determining
whether material, work, or products comply with the standard.

That is to say, the requirements of the standard should be suffi-
ciently sharp to serve as a "go" and a "not go" gauge to accept or
reject material, work, or products sold as complying with the standard.

AMERICAN NATIONAL STANDARDS AND ASSOCIATION STANDARDS

"A national standard implies a consensus of those substantially
concerned with its scope and provisions .... The basic test to be
applied in all cases is the fact of the assent, affirmatively expressed,
of the groups having substantial interest in the standard."

This quotation from the first, section of the formal procedure of
the ASA lays down the fundamental requirements for an American
Standard. Approval of a standard by the ASA means that all
organizations concerned have had an opportunity to participate in
the work, that the work has been carried out under a procedure that
has been regular, open, and above board, and that the standard
represents a real national consensus on what is best in American
industrial practice, and hence that it either already does or may
reasonably be expected to play a significant, if not a control-
ling, role in regard to the materials and processes involved in the
standard.

There is a widespread but erroneous opinion that the primary
difference between a standard suitable for approval as an American
standard and one that should remain a society or association stand-
ard is the length of time which is liable to elapse before revision be-
comes desirable. As a matter of fact, this is a wholly secondary
matter. When the ASA was first organized a provision was inserted
in the procedure preventing the revision of an American Standard
more frequently than once in three years. Experience at once
showed that any such a provision was wholly unnecessary and un-
workable. The real question is whether industry finds the standard
a workable and useful tool. Whenever additional information, new
developments, or changing conditions in the industry make it de-
sirable, a revision should be promptly carried through in order that
the standard as a tool shall at all times have a good cutting edge.

274 P. G. AGNEW [j. s. M. P. E.

From the point of view of utility (which is the only reason for setting
up a standard) it is immaterial whether a standard be revised in one
year or hi ten years. Revision should be made when, and only when,
conditions make it desirable to resharpen the tool. The National
Electrical Code, which is one of the most widely used American
standards, is regularly revised once every second year in order to
keep it abreast of developments.

It is the business of a gyro-stabilizer to keep the ship on an even
keel when it is being buffeted by waves, not to stop the ship. Just
so it is the business of a successful standard to help industry to main-
tain itself in dynamic equilibrium, not hi a static condition. The
ASA has for years carried the following statement on the title page
of its Year Book:

Standardization is dynamic, not static. It means not to stand still,
but to move forward together.

The number of group or association standards is, of course, enor-
mously greater than the number of American national standards, and
doubtless will always remain very much greater. This must needs
be so, since a vast number of standards are required for highly
specialized fields.

As the whole national movement develops, these association
standards should be brought into consistency with such national
standards as apply to the particular field. Similarly, the standards
of individual companies, which in total will be far greater than the
number of group or association standards, should in most part be
brought into conformity both with the national and with the group
standards which apply.

In a great number of cases a standard may be of interest only to a
particular consumer interest and to the producer of the product
covered. In such cases the standard is likely to remain in the group
or association stage. For example, if an American Railway Asso-
ciation specification for locomotive whistles assures a product which
is satisfactory to the railways and the manufacturers, there would
be little to gain by advancing this specification to the status of Ameri-
can Standard.

It frequently happens that a society or a trade association may
wish to assist in carrying out a policy or in bringing about a result
by developing a provisional standard which blazes a trail into new
territory. This can frequently, but by no means always, best be
done by handling the work as an association standard, even though

Oct., 1933 ] STANDARDIZATION IN AMERICA 275

the nature of the work is such that it should be used as a step in the
development of a national standard.

There are often situations in which controversies or differences
of opinion make it impossible, for the time being at least, to secure
sufficient support for a standard for it to attain the status of an
American standard. Yet from the point of view of the group, main-
tenance of such a standard may be desirable, even though lack of
support or even opposition makes it impossible to get national
recognition.

THE HUMAN FACTOR

The human factor is far more important and far more difficult
to handle in standardization work than are the purely technical
and industrial sides of the problem. This is widely recognized by
those who have had any considerable experience in such work.
This applies not only to the negotiations often involving "give and
take," necessary in the development of a standard, but to the stage
of deciding whether a standard shall be undertaken. The latter is
closely connected with the instinctive conservatism, not- to say
suspicion, of the great majority of men to new developments and
ideas.

Standardization is essentially a cooperative undertaking, whether
the undertaking be in the company, in the group, or in the national
stage. And, whether it be recognized or not, the human factor is
preeminently the factor which requires most attention in the ad-
ministration of any cooperative undertaking. For example, the
question of the prestige of an individual, or of a company, or of a
society, often overshadows the engineering or even the economic
difficulties encountered in the development of a simple engineering
standard.

It is unfortunately a fact that as human beings most of our acts
are determined by our emotions, when we think that we are making
decisions and doing things on a purely intellectual basis.

It seems that the technic of influencing human behavior in such
undertakings, which necessarily have to be carried out by conference
and committee methods, deserves an almost unlimited amount of
study. Perhaps through such studies it may prove possible to trans-
form it from an art into the beginnings of a science. For the present
it is essential to remember that true cooperation can only be won;
it can not be commandeered.

276 P. G. AGNEW [J. S. M. P. E.

RESPONSIBILITIES OF PARTICIPATING GROUPS AND INDIVIDUALS

One of the weak points in the whole standardization movement
is that the participating organizations and their representatives
frequently fail to live up to their responsibilities. Unfortunately,
most persons do not fully understand the meaning of representation
and the responsibilities which it entails.

It is the duty of each participating organization to carry out with
administrative orderliness, competence, and with reasonable prompt-
ness, that part of the work for which it has assumed responsibility.

Likewise, each participating organization should assume responsi-
bility before the world for the consequences of the acts of its auth-
orized committees and representatives, scrupulous care being taken
that no effort shall be made to shift this responsibility to others or
in any sense to "hide behind the skirts" of any other organization or
individual.

Furthermore, it is the duty of every representative: (1) to keep
sufficiently in touch with his organization, so that he can correctly
interpret its attitude in the development of the work and can par-
ticipate in decisions in committees; (2) to keep his organization
informed of developments; (3) to act as a leader in the formulation
of the policies of his organization in regard to the matters with which
he is dealing; and (4) to refer back to his organization questions
upon which he feels unauthorized to speak for it.

As a result of many sad experiences the ASA has included the
above principles in a formal statement of the principles under which
it operates. Judged by these criteria, the majority of the numerous
associations and societies with which it has had contact can not
justly claim a consistent record for responsibility.

SUMMARY

Perhaps it may be of advantage to set forth in categorical form
what seem to be essential functions in the movement which should be
performed by the company, by technical and trade associations,
and by the American Standards Association as the national stand-
ardizing body.

The Company

(1) Standardization work should be specifically provided for and systemati-
cally organized, each department concerned taking an active part.

(2) The company should cooperate actively in standardization work of the
trade association and, through it, in the development of national standards.

(3) Conversely, the set-up should be such as to permit an immediate start

Oct., 1933] STANDARDIZATION IN AMERICA 277

in the introduction of each new national or association standard which concerns
the work of the company.

(4) The head or heads of the standards organization should act as authorita-
tive spokesmen of the firm in standardization work in their trade association,
and should cooperate in the standardization activities of technical societies.

Societies and Associations

(5) Technical and trade associations should have effective machinery for
promptly getting a real consensus of all members concerned with a particular
subject.

(6) An educational function of the technical and trade associations should be
to bring home to their members and to the executives of their member companies
the economic importance of standardization, and its significance as a managerial
tool.

(7) In consultation with other groups and with the national body, each
organization should decide what part of its standardization work is to be handled
purely as an association matter on account of the limitation of its scope and in-
fluence to one narrow field without reflexes upon other industries, and what part
needs cooperation with other groups, -from the point of view of national standardi-
zation.

(8) The organization should play a thoroughly responsible part, and should
require responsibility on the part of its members and committees.

The National Body

(9) In the present state of industrial development a national standardizing
body should occupy a pivotal position in the whole industrial standardization
movement As more and more of our industries are becoming so integrated as to
function on a national scale, the standardization activities of the company and
of the association have arrived at the point where they must head up in a true
national movement, centered about a federated clearing-house organization, but
one whose functions should extend far beyond mere clearing-house work a
body which should not only strengthen, but which should give a new and broad-
ened impetus and direction to the whole movement, and in particular to company
and to association standardization activities.

(10) The national body should provide simple, systematic methods for bring-
ing about acceptable national standards. Its organization and methods should
be sufficiently flexible to utilize existing machinery and yet leave free the use of
other simple, cooperative methods.

(11) Its methods should make provision so that the organizations within
any major division of industry may carry on their standards work more or less as a
unit.

(12) In general, each problem should be in the hands of a joint technical
committee made up of representatives accredited for the purpose by the various
industrial, technical, and governmental organizations concerned.

(13) One of its most important immediate undertakings should be to broaden
and unify the great number of existing standards into a consistent system of
national industrial standards.

278 P. G. AGNEW [J. S. M. P. E.

DISCUSSION

MR. FARNHAM: When is the proper time to standardize? The question has
often come before the Standards Committee of this Society as to whether it
should hasten to standardize a certain dimension, when a number of new manu-
facturers desired to enter the field, or to wait and let the "best man" win. In a
particular case, if we standardize a diameter too early, developments may require
a very quick change. If we wait for economic forces to bring about unification,
it may fail to happen; or if it does, then official standardization on what has
already happened will have no meaning.

DR. AGNEW: That depends largely on the state of organization of the in-
dustry and the degree of cooperation which is possible. If you have free and
active cooperation of all the elements of the industry, official standardization
can take place very early. Experience has shown that you should do all the
development you can in advance, and then set up a standard with the idea that
revision may be necessary next year, or the year after, as development takes
place. The assured cooperation will make revision easy. On the other hand,
if you do not have real cooperation, you may have to fall back on the competitive
method, letting the play of economic forces wreak their way until such time as
one or more standards may be agreed upon. That is, of course, a much more
costly method, but sometimes the only one possible.

MR. BEGGS: In the lamp socket we have a device that holds many lamps
in the course of its life. Is it better to standardize upon the device that is re-
newable and is consumed, or upon the receptacle into which it fits? Both films
and lamps are, for example, consumable products.

DR. AGNEW: That is a case of dimensional standards for interworking parts.
Both should be standardized at the same time if they are going to fit together.
There may be economies in requiring better workmanship and smaller tolerances
on one part than on the other. If a relatively high accuracy, i. e., a smaller
tolerance, is provided for the lamp socket, then less accuracy is necessary in the
lamp base; and consequently a greater tolerance can be allowed than if the
precision of both parts were the same. I think it is clear that the economical
way is to require the greater precision in the permanent rather than in the
replaceable part; i. e., place the greater burden on the permanent part, and the
lesser burden on the replaceable part.

MR. BEGGS: I think I have stressed the lamp socket and the dimensions
more than I should. Suppose it is a housing in which the lamp is to be used.
It has been argued that if the manufacturers would design the lamp houses to
accommodate the standard line of lamps, the users would profit by years of lamp
consumption. That is, the burden should be upon the man who makes the
lamp house rather than demand that the lamp maker produce a special size of
lamp to fit every sort of lamp house.

DR. AGNEW: That is an example of the principle that the more basic standard
should be the first to be developed. If we were starting anew, I think the first
thing would be to work out a thoroughly satisfactory form of lamp, since it is
renewable and is used in much larger numbers; and all housings would then be
designed to accommodate that standard lamp. However, we are often faced
by the historical fact that it has been done the other way round. This being

Oct., 1933] STANDARDIZATION IN AMERICA 279

the case here, I would not attempt to answer off-hand. If the job has gone too
far, it will often be necessary to reverse what would otherwise be the normal
procedure, since to do so may cost less in the end than to obliterate all that has
been done and start afresh, even though that might be the strictly logical course.

MR. SHEA: Is it not almost inevitable that contradiction should come about
in many such cases? After all, the lamp was the fundamental invention and
not the housing; and the first job of the housing maker was merely to make
something to accommodate the lamp. In this case the lamp came first, then the
housing.

DR. AGNEW: Just as in England the bayonet instead of the screw base has
been adopted. We are here facing the realities of a commercial situation.

ADDENDUM

Since this paper was prepared, the Secretary of Commerce, on July 19, 1933,
made the following announcement concerning the standardization activities of
the Bureau of Standards:

"Actuated by economy requirements and the desire to place the responsibility
for a unified national industrial standardization program in a single national or-
ganization representative of both governmental and private interests, arrange-
ments have been completed to transfer certain of the commercial standardizing
activities of the Bureau of Standards to the American Standards Association in
New York.

"The task of turning over this work of the Divisions of Simplified Practice,
Specifications, and Trade Standards, and the Sections of Safety Standards and
Building and Plumbing Codes, will be effected gradually under the direction of
the Secretary of Commerce and the Director of the Bureau of Standards."

A HISTORICAL SUMMARY OF STANDARDIZATION
IN THE SOCIETY OF MOTION PICTURE ENGINEERS*

LOYD A. JONES**

Summary. The history of the Society in respect to its standardizing activities is
traced from the date of organization, July, 1916, to October 5, 1932. One of the
principal reasons for the formation of the Society was "the standardization of mecha-
nisms and practices" employed in the motion picture art, the constant application
of which principle has resulted in a large group of standards formulated by the Society
for the motion picture industry during the sixteen years of its existence. The stand-
ards so formulated have become recognized internationally, and undoubtedly have
contributed enormously to the rapid advancement of the art.

The Society of Motion Picture Engineers was organized at a meet-
ing held in Washington, D. C., in July, 1916, which had been called
by Mr. C. Francis Jenkins. The meeting was attended by a group
of some ten or twelve men vitally interested in the then relatively
young art of motion picture engineering. At this meeting, which
may be regarded as the organization meeting, a constitution and by-
laws were drawn and adopted, in which the aim of the Society was
set forth as "the advancement in the theory and practice of motion
picture engineering and the allied arts and sciences, the standardiza-
tion of the mechanisms and practices employed therein, and the main-
tenance of a high professional standing among its members."

Those responsible for the drafting of this paragraph consciously
and wisely made the aims and the purposes of the Society rather
broad, providing for a development of activities into many and
diverse fields of an industry destined to grow to enormous size and
to cover in its activities a wide range of scientific and artistic en-
deavor. There is little doubt, however, that the one thing upper-
most in the minds of those men who formed this Society, the one
thing which they considered paramount, of prime importance, and of
immediate necessity for the welfare of the industry, was the standardi-
zation of materials, mechanisms, and practices. Although the phrase

* Presented at the Spring, 1933, Meeting at New York, N. Yt
** Eastman Kodak Co., Rochester, N. Y.
280

STANDARDIZATION IN THE S. M. P. E. 281

in the above quotation relating to standardization does not occupy
the first place in the sentence, but is placed secondary in order of
mention, all the evidence as gleaned from the early activities of the
Society shows that the founders of the organization in mentioning
the advancement of theory and practice were really thinking that
this advancement was primarily to result in standardization. The
history of standardization by the Society of Motion Picture Engineers
is therefore of itself the history of the development of the Society.

At the first meeting, so far as the author can learn, only one paper
was presented, one by Henry D. Hubbard, which consisted entirely
of a discussion of standardization and of the necessity for standardi-
zation in the motion picture industry.

At the second meeting, which was held in New York, N. Y., on
October, 1916, the officers and directors were elected, Mr. C. Francis
Jenkins being chosen as president. At that time the membership
numbered twenty-five. The president's address was a plea for
standardization, and two out of three papers on the program dealt
directly with the standardization of motion picture film and mecha-
nisms. One of these, Motion Picture Film Perforation, by Donald J.
Bell, of Chicago, 111., discussed in considerable detail the dimensional
characteristics of 35-mm. film. As part of the paper a dimensional
drawing was included, which represents the first step taken by the
Society of Motion Picture Engineers in the direction of standardiza-
tion. This drawing is reproduced in Fig. 1. It is interesting to note
that the dimensions as proposed by Mr. Bell are those which at the
present time remain as the dimensional standards adopted by our
Society and, moreover, adopted by the entire world.

Immediately upon the completion of the organization, four com-
mittees were appointed by the president. These were not called
at that time "standards committees," but, in fact, their chief function
was to study their various fields from the standpoint of developing
standards. The four committees in question were those dealing with
"Cameras and Perforations," "Motion Picture Electrical Devices,"
"Picture Theater Equipment," and "Optics." No "standards com-
mittee" as such existed.

So far as the author is able to ascertain from records available,
there was no clause in the original constitution and by-laws, nor has
there been in any of the subsequent revisions any clause specifying
the mode or method by which the Society shall adopt standards.
In the very early days, of course, the membership was small, num-

282

LOYD A. JONES

[J. S. M. P. E.

bering approximately twenty-five members in October, 1916. It
was a simple matter, therefore, for the entire membership to function
as a perfect democracy in such important matters as the adoption
of standards, and, in fact, such was the custom. The procedure to be
followed for the adoption of standards by the Society has at all times
been specified by the Board of Governors.

The third meeting of the Society was held at Atlantic City, N. J.,
April, 1917. Again the president hi his annual address reiterated

078%

-S5

133-

FIG. 1. Reproduced from Motion Picture Film Perforation, by D. J. Bell,
presented at the second meeting of the Society, October, 1916, at New York,
N. Y.

his previous statements as to the prime objective of the Society
that of standardization. For instance, we read, "Now, the prime
purpose of the organization of this Society was to standardize our
industry. The way to standardize is to standardize. We will not
accomplish it all this session, but without leaving this room there is
plenty that can be done to help the industry wonderfully.

"There are many looking to us to establish these standards.
Most of those who have written to me say they do not care so much
what the standards are, being ready to accept any standards so
far as they can be assured that it is the consensus of opinion of the

Oct., 1933] STANDARDIZATION IN THE S. M. P. E. 283

engineers of this Society. In fact, they are anxious to do so as they
have no other authoritative body to consult.

"So let's get busy now and tabulate the majority opinion of
our members on the several subjects to come up for consideration
at this session."

In spite of Mr. Jenkins' high hopes that something definite would
be accomplished at the Atlantic City meeting toward the establish-
ment of dimensional standards, the records do not show that any
standards were actually agreed upon. However, the nomenclature
list, which in later years was to assume considerable importance,
was initiated; and a list of definitions of some thirty-six words and
phrases was "adopted in committee of the whole Society."

The fourth meeting of the Society was held in Chicago, 111., July,
1917. The president, again, in his opening address reiterated em-
phatically the pressing necessity for the adoption of standards, and
at this meeting the first action by the Society in the adoption of
dimensional standards was taken. This first list of standards includes
twelve items, as follows: film speed, frame line, projection angle,
projection lens foci, projection lens mounting, projection lens height,
picture aperture, film perforation, standard picture film, lantern
slide mat opening, thumb mark, and lantern strip. The introductory
paragraph to this group of standards is interesting; it reads as
follows :

"The following have been adopted as standards by the Society of Motion
Picture Engineers, and are promulgated to encourage uniformity and standard
practice throughout the industry as a whole. Their early universal adoption
will save the industry a great deal of present annoyance and monetary loss."

The various items included in this list were proposed by the standing
committees as previously mentioned. Each item was discussed at
length by the whole Society and each standard was voted on
separately by the entire Society, the list as published being prefaced
by the remark "adopted in committee of the whole Society."

Fig. 2 shows the drawing which was adopted as a part of this
group of standards. This conforms precisely to the dimensional
drawing presented as a part of Mr. Bell's paper at the October,
1916, meeting, and all dimensions shown thereon are in agreement
with the present standard.

The fifth meeting of the Society was held in New York, N. Y.,
October, 1917, and at that time a few additional items were stand-
ardized, again the adoption being "by committee of the whole society."

284

LOYD A. JONES

[J. S. M. P. E.

In about 1917 or 1918, the demand for portable motion picture
equipment became insistent. Due to the inflammable nature of
the nitrate film which was universally used for standard 35-mm.
film, rather stringent restrictions were placed on the use of this film
outside of fire-proofed "booths." Cellulose acetate base, at that
time referred to as non-inflammable film, was available, and it was
proposed that only this should be used in portable equipment.

JJ3*

J33

.078

C
C

D C

-MO' ooe .. .//Or
*? ooo ^

D
D
D

"\ . _ f

^.187"
9* /)??* r

j
-j

D *f ^C
y^7^"

1.3 fd

D C
DZ^KT C

y*' c

D
D

FIG. 2. Reproduced from report on standards
approved by the Society at Chicago, 111., July, 1917.

At the sixth meeting of the Society, held in Rochester, N. Y., in
April, 1918, Mr. A. F. Victor presented a paper on The Portable
Projector; Its Present Status and Needs, in which he made a strong
plea that the Society establish a substandard film width and that
this be made only on non-inflammable base. In Fig. 3 is shown the
original dimensional diagram proposed by Mr. Victor in the com-
munication mentioned. The Transactions of the Society do not
contain a definite statement that this dimensional standard was

Oct., 1933]

STANDARDIZATION IN THE S. M. P. E.

285

accepted at the April, 1918, meeting, but, according to a statement
by Mr. W. B. Cook, in a paper presented at the Cleveland meeting
(November, 1918) and published in Transactions No. 7 (p. 86) this
action was actually taken, "At the last meeting in April at Rochester
your Society adopted a new size of narrow width, slow burning film
as the standard for all portable projectors." In the Transactions
No. 10, 1920, p. 5, the standards "adopted in committee of the whole

,0*7

.*<7

FIG. 3. Original dimensional diagram for
sub-standard film, reproduced from T&e
Portable Projector; Its Present Status and
Needs, by A. F. Victor, presented at the
sixth meeting of the Society, at Rochester,
N. Y., April, 1918.

Society" were again published and this list contains for the first time a
diagram for the 28-mm. "safety standard" film as an officially
adopted standard.

Until this time no committee to deal specifically with standards
had been appointed but, in fact, practically all the committees were
working upon matters of standardization. Moreover, each standard
was adopted after discussion by the whole Society. Such procedure
worked very satisfactorily as long as the membership of the Society
was fairly small and practically all members attended all meetings

286 LOYD A. JONES [j. s. M. p. E.

at which the standards were decided upon. By 1921 the membership
had grown to about eighty, and this method of adopting standards
began to be cumbersome and unsatisfactory.

In May, 1921, the Board of Governors passed a resolution creating
a standardization committee, and at about the same time the board
also passed a resolution that negotiations be opened with the Ameri-
can Engineering Standards Committee and an effort be made to have
that body recognize the standards adopted by the Society of Motion
Picture Engineers. Shortly after the May, 1921, meeting, the first
standards committee was appointed; the members of this committee
were W. E. Story, Jr., Chairman, A. R. Dennington, A. C. Roebuck,
and A. F. Victor.

At about the same time it was found expedient to alter the pro-
cedure by which standards were adopted, and the Board of Gover-
nors passed a resolution providing that recommended standards
should be presented to the Society and receive tentative approval
at a regular stated meeting of the Society, and that after a lapse of
six months these standards should again be presented to the Society
and receive final validation. This procedure was adopted with the
idea that it would prevent the too hasty adoption of proposed
standards which might possibly be ill advised, and that it would
give all members an opportunity to know just what was going on in
the standardization work.

The first report submitted by the standards committee, published
in Transactions No. 13, 1921, p. 160, dealt largely with the question
of submitting the 35-mm. standard and safety standard film di-
mensions for approval to the American Engineering Standards
Committee.

Following its organization, the first standards committee became
very active, and during the subsequent year or two the previously
adopted standards were revised to some extent, this revision in most
cases consisting not of actual changes but merely changes in the form
of expressing the standards. Due to the procedure prescribed for the
adoption of standards, progress was slow, and some confusion exists
in the records as to the exact date on which some of these standards
were finally and officially adopted.

Late in 1923, the newly elected president appointed a new stand-
ards committee consisting of L. C. Porter, Chairman, J. G. Jones,
H. Kellner, F. F. Renwick, and F. H. Richardson. The same com-
mittee was given responsibility for nomenclature, the committee

Oct., 1933] STANDARDIZATION IN THE S. M. P. E. 287

being called the "Committee on Nomenclature and Standards."
In spite of the great care that had been exercised throughout the
Society's life in the adoption of standards, certain minor incon-
sistencies had been perpetuated and, in order to clear up the matter,
the new standards committee, under Mr. Porter's direction, made a
complete study of the entire work beginning with the formation of
the Society.

In Transactions No. 18, May, 1924, p. 236, a detailed report was
made on the status of the standards at that time and, in order to
clarify the whole situation, all standards with regard to which any
doubt existed as to their having been officially adopted by the So-
ciety and all those requiring modification in any detail were re-
drafted and presented to the Society at the May, 1924, meeting.
The dimensional standards for newly cut and perforated film were
prepared in the form of dimensional diagrams, including standard
35-mm. positive film, standard 35-mm. negative film, safety standard,
28-mm. positive and negative film, and standard 16-mm. positive
and negative film. The 16-mm. film for amateur use had been intro-
duced to the trade shortly prior to this date, and this represents
the first recognition by the Society of the dimensional standards for
this material.

For some little time prior to this date various proposals had been
made to modify the form of the perforation, especially for use on
standard 35-mm. positive film. It was maintained by some that the
use of perforations with rounded corners afforded better wearing
quality in the case of positive film. Evidence supporting this
contention seemed very strong and the Standards Committee, on the
date mentioned, therefore proposed the adoption of the rectangular
perforation with rounded corners as standard for 35-mm. positive
film. At the May, 1924, meeting, however, the Society was not
prepared to give this proposed change its initial approval and the
question was referred back to the standards committee.

The other proposed dimensional standards for film were given
initial approval. The dimensional diagrams in question will be found
in Transactions No. 18, May, 1924, p. 238.

At the following meeting, October, 1924, the Society gave its final
approval to all of the standards which had been accepted at the
previous meeting. The question of positive film was again taken up
and the Society gave initial approval to the proposed dimensions,
with the exception of perforation size and shape. Agreement could

288 LOYD A. JONES [J. S. M. P. E.

not be reached on this and the matter was referred back to the
committee. One other point which had been under discussion for
some time was the maximum cutting width for 35-mm. film. At
this meeting (October, 1924) the Society gave its initial approval
to 1.378 inches instead of the 1.375 inches, thus making this width
correspond as precisely as possible to 35-mm. rather than to l*/8
niches.

The committee had made a careful study of camera and printer
apertures, the projector aperture having already been standardized
as 0.725 inch high by 0.950 inch wide. At the next meeting, May,

1925, the question of camera and printer apertures was again raised,
but the Society refused to give final sanction and it was referred
back to the committee.

In the latter part of 1925, the incoming administration appointed
a new standards committee consisting of J. G. Jones, chairman,
H. P. Gage, H. Griffin, L. C. Porter, F. H. Richardson, C. M. William-
son, and C. A. Ziebarth. It is interesting to note that as the Society
grew in membership, the number of individuals on the standards
committee was steadily increased. This was done in an effort to
make the standards committee represent as many as possible of the
diverse interests and special fields.

At the meeting held in October, 1925 (Transactions No. 24, p. 5),
the question of shape and size of perforations for positive film was
again raised. The committee had consulted with the English,
German, and French interests, and as a result of this correspondence
it was decided to recommend the adoption of two forms of perfora-
tion for positive film, one known as the Kodak standard and the other
the Path6 standard. The Society at that time gave its first approval
to these two standards.

The committee also had given exhaustive study to the problem
of standardizing sprockets for film handling mechanisms but was
not prepared at that time to make a final recommendation. The
same subject was discussed again at some length at the October,

1926, meeting (Transactions No. 27, p. 20). Tentative dimensions
for sprockets of various types had been formulated which were based
largely on an exhaustive study of the subject by Mr. J. G. Jones
(Transactions No. 17, October, 1923, p. 55). No definite recommenda-
tions were made, however, by the committee at that meeting and
hence no formal action was taken.

At the following meeting, April, 1927 (Transactions No. 30,

Oct.. 1933] STANDARDIZATION IN THE S. M. P. E. 289

p. 402), the second and final approval was given to the Kodak-Pathe
perforations for positive film. The proposals for standardization
of sprockets were put into final form and the Society voted to give
these proposed dimensions the initial approval. This proposal
involved somewhat different dimensions for feed and take-up sprock-
ets, since it was pointed out by the committee that these sprockets
function under somewhat different conditions and most perfect
handling of film could be achieved only by using sprockets of slightly
different dimensions. Camera and projector apertures were again
discussed but the dimensions proposed by the committee were not
accepted.

At the September, 1927, meeting (Transactions No. 31, p. 443),
the Society gave its second and final approval to the sprocket di-
mensions previously accepted. At this time the camera and projector
speeds for use in connection with sound film were first discussed,
but no definite recommendations were made by the committee.

At the meeting held in Hollywood, April, 1928 (Transactions
No. 34, p. 258), a complete summarized report of standards adopted
up to that time was prepared and published in the Transactions,
and, in addition, dimensions for the three varieties of 16-mm. sprock-
ets (feed, take-up, and combination) were presented and received
the first approval of the Society. The committee prior to this
meeting had given considerable study to the standardizing problems
arising as a result of the introduction of sound into the motion
picture industry, but opinion had not sufficiently crystallized to
permit the formulation of definite proposals for dimensional stand-
ards.

Some months prior to the April, 1928, meeting, contact had
again been made with the American Engineering Standards Com-
mittee. All the dimensional standards and recommended practice
which had been formally adopted by the Society had been collected
and published in a booklet which was submitted to the American
Engineering Standards Committee and received that body's approval
April 9, 1928. This booklet represents the first publication by the
Society, outside of its Transactions, of its dimensional standards
and recommended practice.

In September, 1928 (Transactions No. 36, p. 899), dimensional
standards relating to 16-mm. sprockets, camera, and projector
apertures, and 16-mm. splices, which had received the first approval
at the preceding meeting, were again presented to the Society and

290 LOYD A. JONES [J. S. M. P. E.

received the second necessary approval. In the report presented
at that tune, tentative dimensional standards for 35-mm. sound-
film positive were discussed but no definite recommendation was
made by the standards committee for adoption.

At the May, 1929, meeting (Transactions No. 37, p. 29), the
committee presented to the Society proposals for standardization
of taking and projecting speed for sound pictures, a dimensional
drawing indicating the position of the sound track, and a definition
of the term safety film. All these were given the first approval by
the Society.

The procedure as set forth by resolution of the Board of Governors
for the adoption of standards had proved to be very cumbersome
and a source of great delay in the prompt adoption of standards.
On the recommendation of the chairman of the standards committee,
the Board of Governors on July 26, 1929, passed a resolution modi-
fying this procedure. It reads as follows :

"Resolved that the Standards and Nomenclature Committee shall prepare
and present their report and recommendations at a convention. After this report
shall have been thoroughly discussed, the report with discussions shall be pub-
lished in the first issue of the Transactions following the convention. Approxi-
mately two weeks after the distribution of these Transactions, a letter ballot
shall be sent to the active membership."

The adoption of any proposed standard depended upon the receipt
of a majority affirmative vote, and no second approval was required.
It was thought that this method would appreciably accelerate the
action of the Society in the adoption of proposed standards and,
at the same time, would also give an opportunity to members unable
to attend the convention to vote and express an opinion as to the
wisdom of the proposed standards.

During 1930, the official standards booklet was brought up to
date by the addition of those standards which had been approved
by the Society subsequently to the publication of the standards
booklet mentioned previously. This booklet, the contents of which
were published in the JOURNAL of May, 1930, p. 545, was submitted
to the American Standards Association, which was the new name of
the organization previously operating under the title of American
Engineering Standards Committee. The standards contained
therein were approved by the American Standards Association on
September 28, 1930.

During 1930, the problem of standardizing a film wider than the

Oct., 1933] STANDARDIZATION IN THE S. M. P. E. 291

standard 35-mm. material was given lengthy study by the committee,
but no definite recommendations were made or action taken.

The procedure for the validation of standards as set up by the
Board of Governors in the resolution passed July 26, 1929, continued
in force until May 8, 1932. On that date a letter addressed to
President Goldsmith, by M. C. Batsel, chairman of the standards
committee, which had been formulated as a result of a lengthy
discussion by the standards committee, pointed out that the pro-
cedure was still too complicated and involved unnecessary delay in
the validation of standards. It was the opinion of the standards
committee that the necessity of presenting a report on standards at a
meeting of the Society might on occasion cause undue delay in the
adoption of standards. As a result of this letter, the Board of Gover-
nors, on the date mentioned, rescinded the action of the board taken
on July 26, 1929, which prescribed the mode of adopting standards.
The Board of Governors then passed a motion as follows:

"The Standards Committee shall prepare a formal resolution, proposing in
detail the method of validating and approving standards, to be presented at the
next convention; but that the present report of the committee, to be presented
at this convention shall be presented for validation in the manner prescribed in
the Minutes of the Board of Governors' meeting of July 26, 1929, on page 4."

The standards committee, acting in accordance with the above
motion, presented to the Board of Governors on July 7, 1932, a
proposal for improvement in the technic of validating standards.
After a somewhat lengthy discussion by the Board, the following
motion was made and passed :

"The Board of Governors will consider proposals of the Standards Committee
that are supported by a three-fourths recorded affirmative vote of the ballots
received from members of the committee within thirty days after the date of
mailing those ballots to the entire membership of the committee. The Board
may then accept or reject the report. If accepted, the report will be published
in the JOURNAL, accompanied by an invitation to members of the Society to sub-
mit their comments. At the next meeting of the Board of Governors, not sooner
than thirty days after the appearance of the report in the JOURNAL, the Board
may consider the comments received from the membership and take final action
upon the validation of the proposals for standardization."

The action of the Board of Governors was communicated to the
standards committee and evoked criticism from some members
of that body. These members felt that the procedure still was
somewhat cumbersome and that the likelihood of holding up the
adoption of standards still existed. It was felt by them that to de-

292 LOYD A. JONES [J. S. M. p. E.

mand a three-fourths vote of the members of the standards committee
voting, in order to obtain publication of a proposal favored by a
majority of the committee, might, in some cases, result in unfortunate
delay. Moreover, the procedure required an initial and a final
approval by the Board of Governors. It seemed doubtful whether
this would be much more expeditious than requiring an initial and
final approval by the Society. The standards committee, through its
chairman, M. C. Batsel, therefore, again communicated with the
Board of Governors, setting forth its objections to the resolution as
passed and suggesting certain changes. This letter was considered
by the Board in its meeting October 5, 1932, and as a result a motion
was made and passed that the method of validating and approving
standards be reconsidered. A motion was then made and passed
which reads as follows :

"That a majority affirmative vote of letter ballots received within thirty days
on specifically stated recommendations of the Standards Committee shall be
authorization of early publication of corresponding material in the JOURNAL,
together with an adjoined request for comments from all those interested. It
should be the duty of the office of the Society to bring to the attention of the
Board at its earliest meeting, all communications relative to the published
recommendations. Such recommendations, if later approved by the Board,
shall be re-published as S. M. P. E. Recommended Practice."

This paper, as indicated by its title, is intended to be a historical
summary of the standardization work of the Society. An attempt
has been made to fix, as nearly as possible, the dates of the adoption
of the more important dimensional standards. Other items falling
within the category of dimensional standards and many other actions
in the general category of standardization, such as recommended
practice, nomenclature, etc., have been taken by the Society from
time to time. The last issue of the standards booklet, which has been
mentioned previously, published in 1930, and which has received the
approval of the American Standards Association, may be regarded
as an authentic collection of all the standards which had been offi-
cially adopted up to the time of its publication. The author regards
the publication of the 1930 Standards booklet as terminating the
period with which this report is intended to deal. Standardization
occurring subsequently to the publication of the booklet is regarded
as current rather than historical, and something, therefore, to be
dealt with more appropriately by the current standards committee
rather than in a historical summary.

Oct., 1933] STANDARDIZATION IN THE S. M. P. E. 293

DISCUSSION

MR. CARVER: Has there been any close cooperation between this Society
and the European societies in international standardization?

MR. JONES: We have been in contact with them from time to time, but the
contact has not been close and the cooperation has not been good. The problem
of bringing standardization to the international stage is probably much more
difficult than bringing it to the national stage. The differences of language and
the difficulties of communication, the lack of actual meetings at which the men
can talk things over, make it much more- difficult to reconcile differences of
opinion which in many cases are very slight and immaterial.

The only organization through which we have been able to operate at all in
the direction of international standardization is the International Congress of
Photography, which meets at intervals of several years, depending upon condi-
tions. The organization of the International Congress is rather loose, and for
some reason or other we have riot succeeded in actually setting up a smoothly
working arrangement to bring forward international agreement.

SOUND FILM PRINTING*
J. CRABTREE**

Summary. The production of sound-film prints from variable density negatives
by the Model D Bell & Howell printer has been studied from the point of view of high-
frequency response and uniformity of product. Differences in frequency response
are noted between prints made in different commercial laboratories. Individual
prints in the majority of cases show variations in unmodulated track density and in the
amplitude of wave envelope at high frequencies. The former are due chiefly to irregu-
larities in the film drive and possibly to some extent to variations inherent in film.

The effect of simple mechanical filtering by the addition of a flywheel is discussed.
High-frequency wave envelope irregularities result from momentary loss of contact
at the printing aperture. The influence of threading, gate adjustment, printing
speed, air pressure, type of illumination, degree of shrinkage, and variations in
aperture height are discussed. The changes in high-frequency response produced by
modifications of mechanical parts do not cover the range of difference observed among
commercial prints. A part of this difference in loss characteristic is likely to be
f ound in the developing operation.

It is known that a sound print may show a loss of photographic
definition as compared with the definition of the negative from which
it is made. This condition will result in a loss of modulation in-
creasing in amount with the frequency.

The extent of the loss appears to be indefinite, and few statements
relative to it can be found in the literature. Lewin* states "the loss
at 6000 cycles is at least 4 db." Verbal expressions of opinion by
other observers in commercial studios confirm such a figure and imply
that the loss is variable, the variation not being ascribable to any
known cause. In a series of frequency prints made from negatives
exposed in the Sound Picture Laboratory of Bell Telephone Labora-
tories, Inc., but developed and printed in several commercial labo-
ratories, the loss at 7000 cycles, derived from a comparison of the
frequency response of negatives and prints, varied from zero to 8 db.
This loss of amplitude is usually attributed to the printing process;

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Bell Telephone Laboratories, Inc., New York, N. Y.

t Electronics, January, 1931.
294

SOUND FILM PRINTING

295

but it can not be assigned wholly to that cause since some loss of
definition in the print must occur because of the limit of resolving
power inherent in positive emulsion.

The greater proportion of sound prints produced in the motion
picture industry are made on the Bell & Howell Model D continuous
printer; therefore, in experiments that were conducted to determine
the extent and cause of printing loss, and the possibility of its being
reduced, our attention was confined to that make of printer with the

FIG. 1. Schematic arrangement of Bell & Howell Model D
continuous motion picture printer.

idea that the findings would be directly applicable to commercial
laboratory practice and with the expectation that conclusions could
be drawn with regard to the fundamental problems involved.

The B & H printer consists essentially of a split sprocket 1 (Fig. 1),
a portion of the periphery of which is illuminated by light from the
diffuse light source 2. The effective area of 2 is controlled by a dia-
phragm and is illuminated by the source at 3.

In addition to the above are the sprockets 4 and 5, the idlers 6 and
7, and the weighted idlers 8 and 9. The area of film illuminated is
restricted by means of plates 10 and 11 and an aperture at 12 con-

296

J. CRABTREE

[J. S. M. P. E.

sisting of a pair of jaws having the same curvature as the sprocket 1 .
The mechanism is so housed that light reaches the negative and posi-
tive films only at the aperture 12. A polished steel shoe 13, the face
of which has the same radius of curvature as the printing sprocket, is
held by appropriate means against the rear surface of the positive
film and at a distance from the sprocket that can be varied by an
adjusting screw. The sprocket 1 is driven by means shown in Fig. 2.
The negative and positive films are propelled past the light aper-
ture by the sprocket 1, the leading edges of the perforations being
held against the faces of the driving teeth by the tension effected by

FEED

SPROCKET
GEAR

FIG. 2. Mechanical drive system; Bell & Howell Model D printer.

the weight of the idlers 8 and 9. The films are also held in contact
by this tension and presumably by the contact of the shoe 13. The
paths of negative and positive films are shown on Fig. 1.

In the experiments to be described two stock B & H Model D
printers were employed; and, except where indicated later, no changes
were made in the mechanism. In certain motion picture laboratories
printers are dismantled and the mechanical parts checked and cor-
rected, when found necessary, before being placed into service. For
the present purpose, however, it was considered preferable to study
the machine as received from the manufacturer.

In choosing the type of negative image to be used in making the
experimental prints it is obvious that a modulated film with a per-

Oct., 1933] SOUND FILM PRINTING 297

fectly defined image would be best. It would be difficult to produce
such a film and to translate the results of such tests into quantitative
terms applicable to sound recording. In order to make the results
of more practical value it was decided to use constant-frequency
records of the variable density type for the negative image and to
consider the high-frequency response of the positive made therefrom
as a measure of the effectiveness attained at the printer.

Negatives of different types were used, varying from those made
with the standard recording system at constant input to equalized
negatives made with a recently developed recording objective of im-
proved design. The range of frequency covered was 500 to 9000
cycles. The prints were made on Eastman positive film and de-
veloped in Eastman D16 developer in an Erbograph machine, all
operations being conducted in the Sound Picture Laboratory of
Bell Telephone Laboratories.

Frequency measurements were made in a re-recording machine, the
output being measured both by vacuum tube voltmeter and by ther-
mocouple and milliammeter.

The first requirement in a contact printer is that perfect contact
between the negative and positive films be effected to ensure that no
spreading of the printed image takes place, as will occur when space
exists between the two emulsion surfaces.

It is known that contact between two pieces of film is best attained
at the printer gate by bending the films together longitudinally at
the region of the printing aperture. By reason of the properties of
sheets of materials, curvature in the transverse direction is thus pre-
vented and under, the correct conditions, contact of the two films may
be assured.

This principle is embodied in the B & H printer; but the question
remains, in view of variations among commercial prints, whether
perfect contact is attained and whether imperfections in the surfaces
over which the films pass do not militate against the desired condition.

When the printers under observation were threaded in the usual
way, the gate assembly removed, and the film at the aperture viewed
with a low-power microscope while the machine was in operation,
it was seen to flutter irregularly. The same fluttering could be seen
with the gate in place a window being cut in the side of the shoe to
permit observation except when the gate clearance was quite small.

Mechanical Adjustment. Loss of contact during operation being
thus suggested, prints were made (a) under varied conditions of loop

298 J. CRABTREE [J. S. M. P. E.

settings, such as might be expected to result from faulty threading,
and (b) with different gate clearances and adjustments. Measure-
ments of the frequency characteristics of the prints showed :

(a) Loop setting. That no difference in high-frequency response could be de-
tected between prints made with different settings of upper and lower loops ex-
cept in one case, wherein no tension existed in the upper loop with a shoe setting
of 85 mils' clearance, when a loss of 2.5 db. at 7000 and 9000 cycles resulted. The
same tension condition with 37 mils' clearance showed no loss from the normal
conditions.

(b) Shoe setting. That no difference in high-frequency response could be de-
tected between prints whether made with a tight or slack gate (up to 85 mils'
clearance), or with a wrongly adjusted shoe that made contact with the film on
one side only.

It would appear from these results that, in general, losses of high
frequencies are not likely to occur in printing through maladjustment
of the printer gate or errors in threading, and that contact is main-
tained under quite widely varying conditions. That loss of contact
between negative and positive does cause a loss of high frequencies
in this printer was established by making prints definitely separated
from the negative by interposing different thicknesses of film base.
Separations of 5, 10, and 15 mils were used in printing a negative
recorded on the standard Western Electric System, resulting in losses
at 9000 cycles of 2.5, 4.0, and 5.0 db., respectively, greater than those
from the regular print.

Air Pressure. Though these results suggest that loss of contact is
not easily attained in the normal operation of the Bell & Howell
printer, they do not demonstrate that the contact is perfect and can
not be improved. A series of experiments was therefore conducted
in which the positive and negative were forced against one another
under air pressure in an endeavor to improve the contact.

In certain commercial laboratories, air pressure is applied to the
base side of the positive films to ensure contact with the negative.
Where this is done, some modification of the gate or aperture has
usually been made, so that the results of the following experiments do
not necessarily apply to such cases.

A brass shoe was constructed, identical in contour with the stock
shoe, but perforated with three Y^-inch holes having Y^-inch separa-
tions immediately opposite the gate aperture at the location of the
sound track. These three holes were suitably connected to an air
pressure line. On the negative side, an air jet with a Ys-inch orifice
was arranged so as to direct a stream of air on the negative at a point

Oct., 1933] SOUND FILM PRINTING 299

immediately opposite the jets from the shoe. Both air lines were
provided with valves and gauges. In this manner air pressure could
be applied in any desired amount to either the positive or the nega-
tive or to both, at the point where the printing was being effected.
Prints were made from a high-quality negative at a film speed of 60
feet per minute. In the following table the pressures shown are in
pounds per square inch in the pipe supplying the jet in question.

AIR PRESSURE
(in Pounds)

Negative Side Positive Side

200
220
370
4 16

502
607

7 16

8 '2 2
977

10 16 16

Two series of prints were made: one with a gate clearance of 13
mils and one with a clearance of 37 mils. In no case was any measur-
able difference of frequency response noted, leading to the conclusion
that forcing the film together by air under the pressures used does not
improve the contact in a stock model Bell & Howell printer.

Static Prints. In all the above tests no method has been indicated
by which the contact may be improved. It does not follow, however,
that the normal print produced by this machine is a "perfect" print
in so far as the contact is concerned. A comparison was made,
therefore, between prints made in the printer and so-called "perfect"
prints made statically under conditions of contact and illumination
that probably could not be surpassed in a continuous printer.

Two methods were employed, both utilizing a printing frame in
which short lengths of negative and positive were held in static con-
tact under pressure. Exposure was made to a point source at a dis-
tance ensuring nearly parallel light.

In the first method 6-foot lengths of constant frequency negative
were used, carefully aligned with corresponding lengths of positive
and exposed, in a Cirkut printing frame, to light from a 250-watt
projection bulb at a distance of 60 feet. These exposures were
spliced together and developed in the developing machine along with

300

J. CRABTREE

[J. S. M. P. E.

prints from the same sections of negative made in the Bell & Howell
printer. The resulting positives were run through the re-recording
machine at one-third the normal speed, and the output measured by
a volume indicator in the usual way.

In the second method one-inch sections of the various frequencies
of the negative were cemented to the glass of an 8- X 10-in. printing
frame in correct alignment (Fig. 3). Positive film was placed in
contact with the negative assembly, the back of the frame plated in
position, and exposure made to the light of a 3-cp. automobile lamp
at a distance of 8 feet in a black- velvet-lined tunnel to ensure specular
illumination.

FREQUENCY IN CYCLES PER SECOND
1000 3000 5000 7000 9000 UNMODULATED

D D D

q q p q o 00000 o p o o o 00000 00000

Q D

FIG. 3. Method of mounting negatives for production of static

prints.

A number of exposures were made, spliced together, and developed
along with the corresponding prints made in the Bell & Howell
printer. These latter were from 12-inch sections of the negatives
immediately adjacent to the portions selected for the printing frame
exposures. These sections were spliced together, printed with a
close gate setting, and developed along with the printing frame ex-
posures. The frequency response of the resulting prints was derived
from their microdensitometric traces, the response being computed
from the amplitudes of the wave envelopes. A specimen record made
in this manner is shown in Fig. 4.

Three different types of negatives were examined in the above

Oct., 1933] SOUND FILM PRINTING 301

manner. In each case the printing frame print gave a slightly higher
response than the print obtained from the Bell & Howell printer, the
maximum difference shown being 1.5 db. at 9000 cycles. This differ-
ence, however, is not greater than the loss that would be expected
to result from slippage in the printer due to the pitch dimensions of
the negative, positive, and printing sprocket (negative shrinkage 0.21
to 0.27 per cent, positive raw stock 0.16 per cent) and from the type of
illumination employed.

The conclusion is therefore indicated that the printing losses ob-
served are not due to loss of contact in the printer.

It was noted, however, that when prints were made with greater
clearance between the shoe and the gate, the envelopes of the wave
of the high frequencies, as recorded by the microdensitometer,
exhibited occasional constrictions. More extended records were,

FIG. 4. Specimen microdensitometric record for measurement of
frequency response.

therefore, made to determine the nature of these constrictions.
From the prints made in the printer it was observed that the envelopes
of the higher frequencies were quite irregular, the extent of the ir-
regularities increasing with the frequency. Scannings of sections,
three frames in length, of such a print are shown in Fig. 5 for the fre-
quencies shown. The constrictions are obviously not due to density
variations but result from a loss in modulation, i. e., a blurring of the
positive image, since microdensograms of the negative from which
the print was made showed no such losses (Fig. 6). This blurring
can not be a result of slippage due to shrinkage since such an effect
would be periodic with a frequency equal to that of the perforations
and of very short duration, whereas the losses in question extend over
as much as half a frame in length. They are, therefore, probably
due to momentary loss of contact between negative and positive at
the printer aperture.

302

J. CRABTREE

[J. S. M. P. E.

An effort was made to determine whether it would not be possible
to prevent the loss of contact by mechanically forcing the films to-

500 CYCLES PER SEC

1000 CYCLES PER SEC

^^^^ "**k^** ,-^**A-*--* 1 V^

t^^^-'-'"

"""-

7000 CYCLES PER SEC

9000 CYCLES PER SEC

FIG. 5. Relation of print irregularities to frequency; each frequency
to be developed as shown on diagram, *. e., 500, 1000, 3000, 5000, 7000,
9000 cycles, in order from top to bottom.

gether. From a number of tests it was found that operation of the
printer with the shoe set for no clearance (i. e., riding on the base of
the positive) gave an improved envelope. Equally good was the

Oct., 1933]

SOUND FILM PRINTING

303

case wherein the shoe was removed entirely and pressure applied by
the finger to the base side of the positive at the aperture over the

FIG. 6. Negative; 9000 cycles.

sound track region. Fig. 7 shows representative records of these
conditions; also of normal operation, and of operation without shoe
or pressure of any kind.

It is important to determine how prevalent are the above-described

FIG. 7. 9000-cycle prints: (A) gate clearance, 18 mils; (5) no gate
clearance; (C) no shoe, pressure by finger at aperture; (D) no shoe,
no pressure.

defects in prints made in commercial laboratories. A series of prints
was available, made in the Fall of 1931 by twelve commercial labora-

304

J. CRABTREE

[J. S. M. P. E.

tones on the east and the west coasts from a frequency negative
recorded in the Sound Picture Laboratory. Unfortunately, the
highest frequency usable for the purpose of scanning was 5000 cycles,
a much less sensitive index than the 9000-cycle records in Figs. 5, 6,
and 7. Microdensitometer graphs of 3 -frame sections of these
prints are given in Fig. 8.

.-'' >.->^--',-- ~ .-*'.'""\

' " ..-.-. . - " '

FIG. 8. Envelopes of 5000-cycle prints from industrial laboratories.

Oct., 1933 J SOUND FILM PRINTING 305

The majority of these prints show irregularities in high-frequency
output, some of which are of the type discussed; that is, they result

FIG. 8 (Continued). Envelopes of 5000-cycle prints from industrial
laboratories.

306 J. CRABTREE [j. s. M. P. E.

from the loss of contact between the negative and the positive.
Others show variations in output due to changes in average density
(e. g., no. 1), a defect which will be referred to later. Fig. 8 suggests
that variations in high-frequency amplitude of prints occur widely
and merit consideration.

It should be noted that an apparent discrepancy exists between the
results shown in Fig. 7 and the observation on page 303 that no loss
in high-frequency response results from operation of the printer under
different conditions of gate adjustment, etc. The 9000-cycle prints
shown in Fig. 7 vary obviously in average output, yet volume indica-
tor measurements of these and similar prints failed to disclose differ-
ences between the several conditions. A range of 1 db. difference in
output at 9000 cycles was noted, but the direction of the differences
bore no relation to the quality of the print as shown by micro-
densitometric traces. Measurements by thermocouple and milli-
ammeter showed the same order of difference but here the direction
was such as would be predicted from the envelopes. It is evident,
when using the vacuum tube voltmeter, that so long as the duration
of the constriction in the wave envelope is less than the period of the
meter needle, the movement of the latter will be little affected, and
discrimination between uniform and non-uniform prints, up to a cer-
tain point, will not be shown. If the print is projected at half
speed, or less, then the irregularities of the wave become apparent
in the needle deflections. It is probable that where a measure of
average response is desired, use of the thermocouple and milliammeter
is to be preferred.

Illumination. The respective merits of specular and diffuse illu-
mination for printing purposes are apparently debated points, al-
though there seems to us to be a definite objection to the use of diffuse
illumination for the following reason :

Consider, for the sake of simplicity, the case of an image consisting
of alternate, uniformly dark and clear spaces simulating a frequency of
10,000 cycles. These spaces would, at 18 inches per second, have a
finite length of 0.0018 inch per cycle, or approximately 1 mil for each
dark or light space. The approximate thickness of the emulsion
layer is 1 mil, and, assuming that the image occupies the whole
depth of this layer, we have, on printing, the condition shown in
Fig. 9, where the blackened areas represent the darkened areas of the
negative. Where the printing illumination is normal to the surface,
as at S, no image spread will occur from this cause. With diffuse

Oct., 1933 j

SOUND FILM PRINTING .

307

illumination, those rays not normal to the surface, as at D, will cause
a spread of the image. This, of course, is a consequence of the fact
that the thickness of the emulsion is of the same order as the width
of the density area under consideration. With perfect contact, there-
fore, we should expect a print from negative images of this order of
frequency (10,000 cycles per second) to suffer appreciable loss if
printed with diffuse as compared with specular illumination and this
loss will be accentuated by any lack of contact between films.

As stated earlier, the light source that illuminates the printing

FIG. 9. Specular vs. diffuse illumination in printing.

aperture of the Bell & Howell printer is an illuminated area of ground
glass. The area of this diffusing screen exposed by the diaphragm
at the highest "printing point" depends to a certain extent on the
ratio required between the lowest and highest printing points and is,
in our case, 3 / 4 inch high by 15 /ie inch wide. At the lowest printer
point it is l /& inch high. The angles subtended at the center of the
printing aperture by the height of the ground glass at printer points
22 and 1 are 12 degrees and 2 degrees, respectively. The illumina-
tion received at the printing aperture is, therefore, reasonably specu-
lar; but reflections from the inside surfaces of the printing sprocket
may modify this condition.

308 . J. CRABTREE [j. S. M. P. E.

In order to determine whether any improvement could be effected
in the Bell & Howell printer by the use of more nearly specular illu-
mination, and to check the above conclusions respecting diffuse illumi-
nation, two high-quality negatives were printed under conditions of
(a) specular illumination; (6) normal printer illumination at full
shutter opening; (c) diffuse illumination as follows:

(a) Specular illumination was provided by a 3-cp. automobile lamp at the ex-
treme rear of the printer lamp house. This latter was lined with black coffin
paper, and the inside surfaces of the sprocket were painted with optical flat
black. A practical "point source" of illumination was, therefore, attained; and
since the angle subtended by the aperture at the source was only 1.5 per cent, the
illumination beam may be considered as essentially parallel. The ground glass
in the light change shutter was removed. Control of the intensity of the source
was effected by voltage change.

(6) Full shutter opening (step 22) was used; light change was effected by volt-
age change.

(c) Diffuse illumination was attained by mounting a thin piece of pot opal
glass at the printing aperture, the glass being secured to the lower jaw of the gate
by a small piece of modeling clay.

The results showed an average gain in high-frequency response at
9000 cycles of 1 db. by using the specular illumination. Diffuse
illumination showed a loss at the same frequency of 3.5 db. as com-
pared with specular illumination, confirming the deduction made
above.

Printing Speed. The Bell & Howell printer was originally equipped
to print at a linear speed of 60 feet per minute. The speed of print-
ing is generally increased beyond this value in industrial laboratories
for production reasons. To determine whether the higher speeds
might increase the printing losses, comparative tests were made at
normal speed (60 feet per minute), 90 feet, 120 feet, and 180 feet per
minute, with gate shoe clearances of both 20 and 35 mils, using high-
quality negatives.

No differences in output could be measured except at treble speed
(180 feet per minute) with gate clearance of 35 mils. In this case
the prints showed at 9000 cycles a loss of 1.5 db. by vacuum tube
voltmeter and 2.0 db. by thermocouple over normal operation.
Envelopes of the 9000-cycle print made at 180 feet per minute with
both the close and wide gate settings are reproduced in Fig. 10.
They show that apparently the increased centrifugal force at the
higher speeds aggravates the momentary losses of contact unless

Oct., 1933] SOUND FILM PRINTING 309

checked by a close gate setting that prevents the film from bulging
away from the aperture.

Since a speed of 180 feet per minute on this type of printer is ex-
cessive from the standpoint of film breakage, it is not likely that such
speeds are in actual use. It would be reasonable, from the stand-
point of frequency response, to print at such a speed if the gate
setting were sufficiently close.

Reflections from Shoe. The standard printer gate shoe is con-
structed of polished steel and thus affords a light-reflecting surface.
This surface being in contact with the base surface of the positive
film, it follows that a portion of the light passing through the positive
emulsion will be reflected back to the positive film. The photo-
graphic light transmission of positive film is about 20 per cent, so

FIG. 10. 9000 cycles printed at 180 feet per minute: (A) gate clearance
18 mils; (B) gate clearance 35 mils.

that this proportion of the light transmitted by the negative will be
incident on the shoe. The reflection factor of the shoe being about
50 per cent, we see that 10 per cent of the total amount of light
falling upon the positive film will be reflected back to the emulsion
layer of the film, where it will reenforce the primary image if the re-
flected rays are normal to the film surface ; or degrade the definition
if the reflected rays are scattered, as they will be to some extent
by passage through the diffusely transmitting positive emulsion layer.
Assuming a reflected ray normal to the surface, we can readily
determine from the H&D characteristic of the positive material
whether the reflected component is of sufficient intensity to be of im-
portance. Calculations made in this way show that under normal
conditions for "straight-line recording," an increase in signal of about
0.2 db. will result from the use of a reflecting shoe. This is a negligible

310

J. CRABTREE

(J. S. M. P. E.

factor, and the resulting change in wave-shape may be assumed to
be negligible also.

In the same manner it can be shown that no appreciable volume
distortion of noiseless recording can result from the difference in the
reflected components of the light during printing of the biased and
unbiased portions of the negative.

To verify the above conclusions, prints were made on several
occasions, using (7) the normal polished steel shoe; (2) a brass shoe
having an inlay of hard rubber opposite the aperture (Fig. 11-4),
the contour of the shoe being the same as in (7); (5) a stock shoe in
which a slight recess (5 mils at the deepest point, and 3 /8 inch in
height) was ground opposite the aperture (Fig. 1LB). This recess

HARD RUBBER
INSERT

FIG. 11. Printer gate shoes (elevation).

was painted with flat black lacquer. Shoes (2) and (3} have a very low
reflected light component due to the black material opposite the aper-
ture in each case.

Prints having unmodulated track densities from 0.50 to 1.10 were
made with each shoe. No differences in frequency characteristic,
measured by vacuum tube voltmeter or thermocouple, could be
detected between the prints made with the various gates either at
low or high densities.

The polished shoe is, however, undesirable in one important re-
spect. When making a combined picture and sound print bearing
constant frequencies, the levels of the latter were affected by the
character of the picture scene used and by the printing light required.
This was found to be partially due to the spreading of reflected light

Oct., 1933] SOUND FILM PRINTING 311

incident at the position occupied by the black separation line between
the picture and the sound track. This line is printed through clear
film, and, the light intensity on the positive being high, the reflected
light spreads over to the adjacent portion of the sound track. The
density thereby added to that of the sound track proper has been
found sufficient to affect the level of the latter in the case of thin
sound prints by as much as 2 db. between picture scenes which re-
quired exposures that differed by several printer points.

Shrinkage.- It is well known that motion picture film shrinks
slowly in storage after being perforated. Measurements of Eastman
positive film made at the Sound Picture Laboratory have shown
shrinkage varying from to 0.16 per cent, averaging about 0.1 per
cent. When film is processed, some of the residual solvents or plas-
ticizers in the film base are dissolved or evaporated in the drying
cabinets, resulting in further shrinkage. The negative is therefore,
in general, shrunk to a greater degree than the positive on which it is
later printed; and, in consequence, the perforations of the two will
not match. If, however, negative and positive are curved over a
circular form, with the negative on the inside, that side of the nega-
tive that is in contact with the positive will be stretched by virtue of
its curvature, so that if the radius of curvature be properly chosen,
the stretching can be made such as to bring about a match in pitch
of the contacting emulsion surfaces.

This principle is utilized in the design of the printing sprocket and
gate of the Bell & Howell printer. The curvature has been so
chosen, according to the manufacturers, that with a positive film
having a shrinkage of 0.079 per cent and a negative film of 0.368 per
cent, the perforations will match perfectly in the plane of contact.
With any other value of negative shrinkage one film will slide over the
other as each successive sprocket tooth assumes the drive, until the
leading edge of each perforation of both films makes contact with
the driving face of the tooth. Since printing is continuous, some im-
age blurring must take place as a consequence of such slipping.

The value of negative shrinkage assumed by the printer manufac-
turer seems to be excessive for modern sound negatives, which in this
laboratory show an average shrinkage of only 0.10 per cent from
standard pitch. When printed on the Bell & Howell printer, de-
signed for approximately 0.3 per cent shrinkage, slippage during
printing is unavoidable.

The amount of slip occurring each time the film drive is shifted

312 J. CRABTREE [j. s. M. P. E.

from one driving tooth to the next is the difference between the
negative and the positive pitches at the surface of contact of the two
films; e. g., if we print a negative of zero shrinkage (pitch 0.187 inch)
on a positive of zero shrinkage (0.187 inch) then, due to the curvature
at the printing plane, the pitch of the negative at the surface of con-
tact with the positive will be 0.187 inch + 0.3% of 0.187 = 0.1875
inch. Every time that a perforation passes the aperture the positive
must slide back along the negative by 0.0005 inch.

With an infinitely narrow aperture, demodulation of the print
would occur over this distance only; but with apertures of finite
size, the new space relationship of negative to positive after slippage
will affect the images already formed, and the manner in which the
latter is affected is a function of the height of the printing aperture.
With a very narrow aperture only very short distances will be af-
fected by the slippage; the demodulation over these distances will be
abrupt and appreciable, but the average signal will be little affected.
As the aperture is increased, the average signal will decrease, but the
amount of variation will also decrease.

The Bell & Howell Company adopted Vie inch as the standard
height of the printing aperture height of the Model D printer. To
determine whether any measurable increase of high-frequency re-
sponse could be gained in prints made from negatives of various
degrees of shrinkage by the use of smaller apertures than this, the
following tests were made.

The printer aperture was reduced in height by the insertion of a
small piece of sheet metal of the necessary size attached to the lower
jaw of the aperture by modeling clay and arranged just to clear the
film plane. Aperture heights chosen were 3 / 32 inch, 3 /is inch, and the
full opening of 6 /ie inch. These values correspond approximately
to */2 perforation pitch, 1 perforation pitch, and ! 2 /3 perforation
pitch. The negatives used were a high-quality equalized negative
having a shrinkage of 0.21 per cent from standard pitch, and a nega-
tive made by the standard recording system having a shrinkage of
0.16 per cent. Sections of the latter were taken and shrunk to the
following values: 0.43, 1.13, and 1.81 per cent, respectively. The
positive film used had a shrinkage of 0.16 per cent. By reason of
printer design mentioned earlier, the negative that would print on
this positive without slippage would have a shrinkage of 0.16 +
0.30 = 0.46 per cent. The 0.21 per cent shrunk negative then,
printed on the 0.16 per cent shrunk positive, would cause an effective

Oct., 1933] SOUND FILM PRINTING 313

slippage of 0.16 + 0.30 0.21 = 0.25 per cent; and so on, in like
manner, for the other negatives.

Prints were made from these various negatives in the printer
using the three aperture heights mentioned above. The frequency
loss values of the resulting prints, referred to 1000 cycles as zero,
were as given in Table I.

TABLE I

Variation of High- Frequency Loss with Size of Aperture

Aperture 1000 3000 5000 6000 7000 8000 9000
(Inches)

Experimental Negative*

(Shrinkage, 0.21%. Effectiie Slippage, 0.25%)

Vie -1.0 - 1.0 +1.0 - 0.5 -1.0 - 3.0

Vie -1.0 - 0.5 +1.5 +0.5 - 1.5

3 /32 -1.0 - 0.5 +2.0 +0.5 - 2.0

Standard Negative

(Shrinkage, 0.16%. Effective slippage, 0.30%)

Vie -1.0 - 5.5 ... -10.5 . -16.5

Vie -1.0 - 5.0 ... - 9.5 . -15.5

3 /32 -1.0 - 5.0 ... - 9.5 . -15.5

(Shrinkage, 0.43%. Effective slippage, 0.03%)

Vie -1.0 - 5.0 ... - 9.5 . -16.0

Vie -1.0 - 5.0 ... - 9.5 . -15.5

3 /32 -1.0 - 5.0 ... - 9.5 . -15.5

(Shrinkage, 1.13%. Effective slippage, 0.67%)

Vie -1.0 - 6.5 ... -15.0 . -22.5

Vie -1.0 - 5.5 ... -12.0 . -20.0

Vsz -1.0 - 5.5 ... -12.0 . -18.0

(Shrinkage, 1.81%. Effective slippage, 1.35%)

Vie -1.5 -10.0 ... -19.0 . -25.0

Vie -1.5 - 6.0 ... -14.0 . -23.0

V2 -1.0 - 5.5 ... -12.0 . -19.0

* An experimental negative not representative of commercial recording.

The results show that with negatives having shrinkages such as are
ordinarily encountered (0.10 to 0.20 per cent), a slight improvement
in high-frequency response (about 1 db. at 9000 cycles) is obtained
by reducing the aperture height from 6 /ie inch to 3 /ie inch, but no
advantage is to be gained by reducing to a dimension below this
figure. With negatives of high shrinkage, large gains in high-fre-
quency response are obtained by further reducing the aperture size
(6 db. at 7000 cycles with slippage 1.35 per cent and aperture 3 /32
inch) . Such negatives, however, will but seldom enter into considera-
tion in sound printing.

314 J. CRABTREE [j. s. M. P. E.

It should be noted here that a print made with a very narrow aper-
ture is viewed favorably by the volume indicator, since here the
period of the demodulated portion is so small a fraction of the whole
that the volume indicator needle is not able to follow the change.

Factors other than high-frequency response, however, enter into
determining the optimum aperture height to be used. Irregularities
in the rate of movement of the film past the aperture will obviously
affect the degree of exposure it receives from point to point, and this
will be the more apparent the shorter the time occupied in traversing
the aperture relative to the period of the irregularities.

Such irregularities may be expected to enter from errors in the
gearing that serves to drive the printer sprocket, as well as from any
errors of tooth spacing or of tooth contour in the drive sprocket.
That these exist is readily seen by observing the movement of gears
and sprocket teeth with the aid of a stroboscope.

FIG. 12. Film flashed in printer using an aperture 5 mils high.

Density Variations in Prints. As mentioned above, the film moves
past the aperture of the Bell & Howell printer in an irregular manner,
which may be demonstrated stroboscopically or by making "flash"
exposure of film for which the printing aperture has been reduced.
Fig. 12 shows such a piece of film exposed through an aperture 5
mils high. Variations in transmission of a film flashed in the printer
to a low density may be seen visually even when the standard aper-
ture of 8 /i6 inch is used. These variations, as measured on film ex-
posed to produce a density of 1 .0 in the printers used in this labora-
tory, show a maximum variation of density of approximately 0.05,
*. e., == 1 db. in transmission. Greater variations have been observed
in "flash" exposures made in printers in commercial laboratories,
amounting in one case to a maximum of 3 db.

Variations in density of the print cause a volume variation of the
signal in the reproducer and a flutter in the sound from the horn.
Such a flutter, though not always obvious in a first print, can usually
be detected in the print made from a duplicate negative. The varia-
tions aggravate the unsteadiness of the output meter needle when

Oct., 1933]

SOUND FILM PRINTING

315

measuring constant-frequency records with the vacuum tube volt-
meter, particularly at the higher frequencies, when the amplifier
gain must be raised to bring the needle to the zero reading.

It was felt that the production of such irregularities is a distinctly
undesirable feature in a sound printer and remedial measures were
therefore sought.

Variations in density of the print (assuming a perfectly uniform
negative) may result from :

(1) Variation in exposure from irregularities in the rate of motion of the film
past the exposing aperture due to mechanical defects in the driving mechanism.

(2) Unevenness in development.

(3) Irregularities in the light-sensitive layer of the photographic film on which
the print is made.

Mechanical Defects. As might be expected, a certain amount of
flutter results from the train of commercially cut gears used in the

B WITH FLYWHEEL

C COUNTERSHAFT AND FLYWHEEL DRIVE

FIG. 13. Densograms of printer flashes made through normal
aperture; length, 4 feet (lines of zero transmission located 2.2 cm.
below line of mean transmission).

speed-reducing mechanism. Some commercial laboratories attempt
to improve this condition by "lapping" the gears with a very fine
abrasive. Although this may polish off fine irregularities on the
face of the teeth, it was felt that such were but minor causes of flutter
and that any improvement gained would be more than offset by the
change of contour of the gear teeth caused by the wear thus brought
about.

The substitution of expensive, specially cut gearing was not re-
garded as a "commercial" solution, so an attempt was made to
smooth out the film movement at the printing sprocket by placing a
heavy, carefully balanced flywheel of 30 pounds in weight and 20
inches in diameter, on the shaft of the printing sprocket. The drive
was otherwise unchanged, except that several types of belt were com-

316 J. CRABTREE [j. s. M. P. E.

pared with the standard leather belt. Examination by the strobo-
scope revealed no improvement from the use of the flywheel or from
the different types of belts. "Flash" exposures on positive film
scanned in a recording densitometer revealed no differences that
could be ascribed to the flywheel, the type of belt, or to reductions in
the load imposed by the feed and take-up mechanism.

The effects produced by gear inaccuracies were then eliminated by
driving the printing sprocket directly by means of a fabric belt
around the periphery of the flywheel.

The necessary speed reduction was attained through a counter-
shaft driven from the driving motor by means of a pulley and fabric
belt. The take-up mechanism was driven by a separate belt.
Stroboscopic examination showed a change hi character of the
motion of the sprocket with some improvement, but "densograms"*
of flash exposures indicated this to be but slight in the print (Fig. 13).
Prints made from constant-frequency negatives showed little im-
provement in irregularities, as shown by the degree of output meter
needle unsteadiness.

It seemed, then, that in the printers under test, little is to be gained
by the addition of a flywheel to the printing sprocket when the latter
is driven through the normal gearing; nor is much more to be gained
when the flywheel is driven directly.

This may not be the case where the sprocket drive irregularities
have become accentuated by gear wear, such as must occur in many
of the machines in commercial use. Fig. 14 shows ' 'flash' ' exposures of
density 1.0 from twelve different printers representative of a total
of 28 examined from commercial laboratories on both the east and
west coasts, and illustrates the wide variation in performance. Many of
these could doubtless be improved by the addition of a suitable flywheel.

The "flutter" at the sprocket teeth observed stroboscopically was
irregular and of low period (about 1 /g to 1 second), and with an
aperture of B /ie inch seemed of too small an amplitude to account for
the variation in density shown hi the density records of the flashed
exposures.

It had been observed that volume indicator readings of prints made

* The "densograms" in Figs. 13 and 14 show the variations in transmission of
the particular record from point to point longitudinally over a distance of 4 feet.
The distance from the datum line of zero transmission to the line of mean trans-
mission is shown on each chart. The graph thus shows directly the comparative
degrees of modulation of transmission of the samples regardless of their density.

Oct., 1933]

SOUND FILM PRINTING

317

FIG. 14. Densograms of flashed exposures: density, 1.0, length, 4 feet;
made in printing machines in twelve different motion picture laboratories
(line of zero transmission located 2.2 cm. below line of average trans-
mission in each case).

318 J. CRABTREE [j. s. M. P. E.

with or without filtered drive from constant-frequency negatives
showed increasing unsteadiness with increase in density, being satis-
factory from this standpoint below a visual diffuse unmodulated
density of 0.40 and increasingly unsatisfactory above this point.
"Flash" exposures to produce a range of density were therefore
made in a recorder known to be in good condition from the standpoint
of mechanical flutter. These were scanned in the recording den-
sitometer and also in a standard reproducer using a modulated light
source. In the latter method the variation hi signal resulting from
density irregularities was observed in the output meter. Com-
parisons were made with similar exposures made hi the printer and
developed under different conditions. The variations in signal level
were found to increase from 0.25 db. at a density of 0.10 to 1.0
db. at a density of 1.10, and the order was the same whether the
tracks were exposed in the recorder, stock printer, or "filtered"
printer, and whether developed under conditions of poor or of good
circulation of the developing bath. This progressive increase in
irregularity of needle response with increase in density is borne out by
the densograms (Fig. 15.4, B) as is also the lack of definite superiority
of one exposing device over the other.

If these variations result from modulation of the record by uneven
motion of the film past the exposing aperture, then the smaller varia-
tions at low densities may be explained by the smaller slope of the
H&D curve of the material for values up to a density of 0.4 or 0.5.
Above this point, however, the variation should not increase with
density, as was usually found to be the case. The particular re-
corder used in making these tracks was stroboscopically observed to
be free from low-frequency flutter of the period of variation of the
volume indicator needle or of the irregularities shown in the "denso-
grams." These considerations suggest the presence of a factor of
variation approximating the order of the exposure irregularities
existing in any of the machines used in the experiments.

To obtain a record in which any possibility of variation of exposure
from point to point was excluded, positive film was exposed statically
in a 6-foot printing frame to a point source of light. No glass was
interposed between the light source and the film. A number of such
exposures was made which would give on development a series of
tracks of various densities. These tracks were developed by a ma-
chine having a very efficient developer circulating system and then
scanned in the recording densitometer.

Oct., 1933]

SOUND FILM PRINTING

319

^DENSITY
0.10

FIG. 15. Densograms of positive film (length, 4 feet), uniformly
exposed in (A ) recording machine; (B) normal printer; (C) mechanically
filtered printer (line of zero transmission located 2.2 cm. below line of
average transmission in each case).

320

J. CRABTREE

[J. S. M. P. E.

The "densograms" of a typical series of such static exposures are
shown in Fig. 16. They reveal the fact that irregularities occur
even when the light exposure is uniform, and that the irregularities
usually increase with density. These variations are, therefore, prob-
ably inherent in the photographic emulsion itself. It should be

FIG. 16. Densograms of static uniform exposures on positive film; length,
4 feet (line of zero transmission located 2.5 cm. below line of average trans-
mission in each case).

mentioned here that a number of such exposures were scanned and
that the order of variation in respect to density was not always the
same. Occasionally records were obtained in which an increase in
density did not show increased variation. The average, however,
showed a decided trend in the direction indicated.

The reason for the density variations in the film lies possibly in the
variation in thickness of the sensitized emulsion layer. That the

FIG. 17. Densogram of positive raw stock; length, 4 feet (line of zero
transmission located 2.5 cm. below line of average transmission).

thickness does vary in a manner similar to that of the photographic
density resulting from uniform exposure is seen from Fig. 17, which
shows the variations in transmission of a 4-foot length of positive raw
stock. It is not unreasonable to assume that where the image result-
ing from exposure and development of such a film is of a low order of

Oct., 1933] SOUND FILM PRINTING 321

density, it will be little affected by the thickness of the emulsion
coating, since, as is well known, such an image is largely confined to
the upper surface of the sensitive layer. With increasing density,
the image penetrates more deeply into the emulsion layer, and hence
will be increasingly affected by variations in thickness of the latter.
The variations in coating thickness doubtless vary from sample to
sample, as does the order of variations in the image density.

These facts suggest that little would be gained by further refine-
ments in the mechanical filtering with the 8 /i 6 -inch aperture. Of
course, with narrower apertures irregularities in exposure will be
accentuated. With any given aperture and filtering device, only
operating tests would reveal whether the limit imposed by the film
irregularities had been reached.

Frequency Characteristic. Although the preceding discussion covers
many phases of the printing operation, it will be noted that no satis-
factory explanation has been found that would account for the
widely varying print losses (0 to 8 db. at 7000 cycles), noted on page
304 as measured on prints of the same constant-frequency record,
developed, and printed in different commercial laboratories.

The prints made in the Sound Picture Laboratory from the same
series of negatives showed less spread in 7000-cycle loss, suggesting a
real difference hi printing efficiency among laboratories. However, a
decided tendency was noted in our prints toward the production of
a more or less constant high-frequency output. Negatives having a
higher 7000-cycle level showed a greater printing loss than did nega-
tives of lower 7000-cycle level. This fact suggests the presence of a
limiting factor in print response level in this and probably other
laboratories, the most likely factor being the developing operation.

DISCUSSION

MR. MITCHELL: We have done considerable work along somewhat similar
lines, using a different method, in connection with the new fully automatic
printer. We found the very minute differences in the pitch and shape of the teeth
of the main sprocket to be so important that we installed a special machine ex-
clusively for making the printing sprockets. We now can control the pitch of
the teeth, and the shape of the teeth, to within 0.0002 inch ; so that in the new
printer, with synchronous drive, we can reduce the picture printing aperture to
almost 3 /ie inch now and obtain a very satisfactory response, without the "dips"
mentioned by Mr. Crabtree. We use 3 /32 inch on the sound printing aperture.

I should like to ask Mr. Crabtree whether he has the new sprockets in the
printers in his laboratory.

MR. J. CRABTREE: We have the new sprocket.

322 J. CRABTREE

MR. J. I. CRABTREE: Since when have they been available?

MR. MITCHELL: They have been available commercially for about nine
months or a year. They are now made of stainless steel, as we find that in con-
stant use the chemicals in the film tend to erode ordinary steel. The use of
stainless steel seems to afford the possibility of maintaining efficiency over a long
period of time. Stainless steel sprockets have been available only within the
last week or so.

MR. J. I. CRABTREE: I am glad you have done this. We had to work several
weeks on the sprocket of our printer to put it into condition. Are sprockets
available calculated for shrinkages other than 0.3 per cent?

MR. MITCHELL : As far as I know 0.3 is still the standard. We find the present
sprocket is almost ideally suited to the film shrinkages normally met with at the
present time.

MR. J. CRABTREE: I did not make it clear that the data presented are all on
the basis of the old sprocket.

MR. SHEA: Have you data on the improvement that the new sprocket
represents?

MR. J. CRABTREE: Yes, but not here.

MR. SHEA: What is the general order of improvement?

MR. J. CRABTREE : A flatter envelope.

MR. MITCHELL: Some idea of the improvement may be gathered from prints
made in Hollywood with the new printer in which are incorporated these vari-
ous improvements. The difference was so great that it was found impossible
to print one reel on the new printer and the other reels on the ordinary printers.
They all had to be printed on one type of machine or the other. So in practice
there is a very decided and definite difference.

MR. O. SANDVIK: The transmission of cine-positive film varies with the wave-
length, being practically opaque to the ultra-violet, and increasing as the wave-
length increases. In transmitting a wavelength of 400 millimicrons then, we
have a transmission of the order of one part in 10,000 to one part in 100,000; that
is, from a hundredth of one per cent to a thousandth of one per cent, depending
on the emulsion.

Regarding the vacuum tube volume indicator, it can be made to indicate ap-
proximately r.m.s. values, average values, or peak values. I believe Mr. Crabtree
said that the vacuum tube volume indicator, as used in these measurements,
indicates peak values. Is that correct?

MR. J. CRABTREE: Both vacuum tube voltmeter and thermocouple with
milliammeter were used as volume indicators. The former was a peak voltage
device. The thermocouple integrates the output better, but we could not always
detect the differences due to constriction of the envelope.

MR. SHEA: Is it your feeling that experimentally it is within the 2 db. that
you mentioned?

MR. J. CRABTREE: Yes.

WAVE FORM ANALYSIS OF VARIABLE WIDTH SOUND

RECORDS*

O. SANDVIK, V. C. HALL, AND J. G. STREIFFERT**

Summary. The results of a study of the relation between sensitometric condi-
tions and the harmonic content introduced in the photographic process are described.
Sinusoidal wave-forms of very low initial harmonic content were recorded, varying
the negative exposure, negative development time, and positive exposure so as to
cover a range of negative densities, negative gammas, and print densities. Sets of
such records were made at several levels of modulation and several frequencies from
100 to 4000 cycles per second.

The records were analyzed on a recording microdensitometer which automatically
recorded the transmission of the sound track. These traces, after being enlarged to
the appropriate size in order to determine the harmonic content, were analyzed with
a harmonic analyzer. The results show that sensitometric conditions for good quality
are much more critical at the higher frequencies and that in general the harmonic
content does not increase at those frequencies.

The two general types of photographic sound records and the
methods and processes used in making them are so well known that
they need not be described here. This paper describes the results
obtained during a study of the relation between the sensitometric
conditions, harmonic distortions, modulation and frequency char-
acteristics of variable width sound records.

The problem was first considered by Hardy, 1 who concluded that
when the width of the image of the recording aperture is small com-
pared with the wavelength of the record, the quality of the sound
reproduced does not depend on the conditions of exposure or de-
velopment of either the negative or the positive. Cook 2 has investi-
gated the effect of a finite aperture on the quality of the reproduced
sound. More recently this problem has been treated by Foster. 3
He concludes from his theoretical investigation that by proper
choice of the conditions of exposure and development, sound can be
recorded and reproduced which is practically free from non-linear
distortion. When the image of the recording aperture is rectangular
and uniformly illuminated, the following conditions are necessary:

* Presented at the Spring, 1933, Meeting at New York, N. Y. Communica-
tion No. 518 from the Kodak Research Laboratories.
** Eastman Kodak Co., Rochester, N. Y.

323

324 SANDVIK, HALL, AND STREIFFERT [J. S. M. P. E.

(1) Complete exposure in recording and printing corresponding to
the upper extreme of the straight-line portion of the H&D curve;

(2) the transmission of the dense part of the record is negligible;
and (3) the over-all gamma is equal to unity.

Maurer 4 has investigated the problem from the point of view of
determining the conditions of exposure and development that
give the least film loss over the required frequency range. From
his experimental results he concluded that the process is relatively
independent of the negative gamma and that for motion picture
positive, a density of 1.3 in both the negative and the print gave
the best compromise between volume and frequency characteristic.

More recently, Dimmick 5 has investigated the problem in some-
what the same manner except that he was interested in determining
the conditions which would give a maximum of signal; that is, a
minimum of film loss at 6000 cycles per second. He found that
when using Eastman motion picture positive film and D16 developer,
the greatest volume resulted from a print having a density of 1.6 and
a gamma of 2.00 printed from a negative whose density and gamma
were 1.8 and 2.18, respectively.

Neither of the two latter workers investigated the non-linear
distortion present. The writers are not aware of any experimental
work having been published on this phase of the problem, and as
the employing of a wider range of frequencies has created a some-
what greater interest in this question, it appeared desirable to sub-
ject the problem to a systematic study. The results below were
obtained on regular motion picture positive film used both as the sound
recording negative and the printing stock. The frequency range
investigated ranged from 100 to 4000 cycles per second for a series of
negatives ranging in density from 0.6 to 2.1. Several negatives were
made, so that in each case there were groups of negatives all of the
same density but ranging in gamma from 1.0 to 2.0. Here again
for each density and gamma there were several negatives in each
group ranging in modulation from 1 to 10 db. below 100 per cent.
From each one of these negatives was made a set of prints of densities
0.6, 0.9, 1.2, 1.5, and 2.0, respectively, at a fixed gamma of 2.0.
The developer used in all cases was D16.

The image of the recording aperture was rectangular, its height
on the film being 0.0005 inch. Great care was exercised in obtaining
uniform illumination along the length of the image, and also in ob-
taining a very sharp edge and a minimum of illumination on the
film beyond the geometrical edge of the image. The non-linear

Oct., 1933]

WAVE ANALYSIS OF SOUND RECORDS

325

distortion introduced by the electrical system and the recording
galvanometer was considered sufficiently small to be neglected.

The method of analyzing the records was similar to that previously

TABLE I

Second and Third Harmonic under Various Negative and Print Conditions

Print Density

V V

S S a 1
s fa !

Q O n,

0.6

0.9

1.2

1.5

2.0

o
e -o
o j:
% 3
w H

n -o
o >:

l e

Harmonic

a T3

a '

vi H

o
a -a

O i-

8 3
w h

1 -H

tn H

0.6 1.0 100

4.3 3.0

1.5 3.4

2.5 3.5

2.8 4.4

10.2 3.7

1.0 1.0

6.2 1.5

5.3 1.6

2.9 2.2

2.9 3.3

9.7 3.1

0.63 1.5

3.3 0.7

1.0 3.1

1.1 3.9

5.5 5.0

11.5 4.3

1.0 1.5

6.3 0.7

4.2 2.1

2.8 2.6

4.2 4.0

8.3 4.5

1.43 1.5

7.1 1.2

6.6. 1.0

7.3 1.5

3.4 2.4

0.9 3.6

1 . 58 1.5

8.5 1.3

7.5 1.1

6.5 0.7

2.5 1.6

3.3 2.1

0.85 2.0

2.4 2.4

0.8 3.0

2.5 4.2

8.4 4.4

13.8 4.0

1.0 2.0

5.1 2.7

2.4 2.3

0.8 4.7

4.7 4.5

9.4 5.5

1.3 2.0

7.1 0.9

4.5 2.3

3.1 0.4

5.0

5.6 5.2

1.5 2.0

6.8 0.5

6.1 0.8

4.5 1.3

4.2 2.0

2.0 3.7

2.1 2.0

8.6 1.1

7.5 1.1

6.9 0.7

6.1 0.3

5.0 1.6

0.8 1.0 2500

7.3 1.4

0.4 1.4

4.9 2.0

10.4 1.9

1.0 1.0

9.0 0.7

3.6 0.6

1.6 1.3

3.3 2.2

13.0 0.6

0.64 1.5

1.5 1.5

3.8 1.1

10.7 2.5

14.0 1.6

1.0 1.5

7.2 0.5

2.5 0.6

2.5 1.4

9.1 2.8

15.5 2.3

1.2 1.5

12.0 1.3

9.8 1.2

3.6 0.7

3.6 1.4

8.9 0.6

1 . 55 1.5

12.8 0.5

11.1 2.8

10.0 2.1

7.4 2.5

0.8 1.5

0.8 2.0

2.2 1.2

5.9 2.1

11.8 1.0

17.0 2.1

1.0 2.0

5.7 1.8

3.6 2.1

8.2 0.9

13.2 2.6

22.6 0.7

1.2 2.0

6.0 0.7

4.3 0.7

4.0 2.5

8.2 1.5

17.2 2.7

1.54 2.0

10.9 1.1

8.3 0.7

5.6 2.2

3.1 1.5

9.2 2.8

2.15 2.0

11.1 1.0

14.5 1.6

9.5 0.3

9.4 1.1

13.1 0.4

0.75 1.0 4000

1.6 1.6

1.3 0.9

5.9 1.3

14.1 0.7

1.13 1.0

9.3 1.4

4.3 1.7

4.1 1.2

6.9 1.1

18.2 1.2

0.70 1.5

1.4 0.5

7.3 1.1

13.7 1.1

20.5 0.8

1.05 1.5

6.7 0.9

3.9 0.8

3.0 1.7

10.1 0.2

16.3 0.7

1.62 1.5

10.7 0.5

7.5 0.8

3.5 1.2

4.4 0.9

8.2 1.9

1.34 1.5

12.4 0.5

7.3 0.4

5.1 0.7

2.1 1.6

7.7 1.2

1.66 1.5

12.7 1.2

11.2 0.7

8.5 1.7

5.0 0.4

2.1 0.5

0.8 2.0

1.5 1.8

6.3 1.5

8.8 1.1

15.8 1.2

1.0 2.0

2.6 0.8

3.8 1.3

7.7 1.7

13.6 0.8

21.4 1.4

1.25 2.0

6.7 0.6

4.5 0.6

2.6 1.1

9.4 1.4

19.0 1.1

1.46 2.0

10.3 3.7

8.3 0.8

4.0 0.5

1.3 1.3

8.5

2.1 2.0

14.7 3.4

11.3 0.0

7.1 1.0

4.9 0.4

4.7 0.4

326 SANDVIK, HALL, AND STREIFFERT [J. S. M. P. E.

used by two of the authors in analyzing variable density records. 6
The results are given in Tables I and II.

Table I shows the values of the second and third harmonic found
for a wide variety of negative and print conditions at frequencies of
100, 2500, and 4000 cycles per second. In compiling this table,
values of higher harmonics were omitted, as under conditions which
give results approaching satisfactory quality their values are less
than one per cent of the fundamental amplitude. A study of the
table shows that for a given print density and negative gamma the
second harmonic goes through a minimum, while in general the third
harmonic remains sensibly constant or decreases progressively with
higher negative densities, particularly at the higher print densities.

A consideration of the appearance of the records made under
various conditions indicates the reason for these two effects. For a
given printing light, such as to produce a density of 1.5, say, in the
positive image printed through the relatively clear negative areas, a
noticeable density will be produced by printing through the blackened
part of the negative, provided the negative density is low. This, of
course, leads to a lowering of modulation of the print transmission.
The nature of the image at low negative densities is such as to have
little or no filling in of the valleys of the waves. When sufficient
printing exposure to produce a density of the order of 1.5 at a gamma
of 2.0 is given in this region, the developed image shows considerable
spreading instead of coming to a narrow peak as required for perfect
reproduction. This effect flattens the peaks of the positive wave
while leading to even harmonic distortion, the second being the
only one of practical importance, since values of the fourth and
higher harmonics are less than one per cent except under negative
and print conditions causing sufficient loss of modulation of print
transmission as to be useless in practice. As the negative density
increases, the amount of printing through on the positive peaks
decreases, leading to a corresponding decrease in the second harmonic,
the value going to zero for some particular negative and print
condition. As the negative density is further increased, the valley
of the print begins to fill in, causing a flattening of this side of the
wave and re-introducing even harmonic distortion in a phase 180
degrees from that occurring on prints made from lower negative
densities. If the filling in became appreciable while there was still
some printing through, then the wave -form would be flattened at
both ends. Since a distorted wave-form symmetrical with respect
to the axis contains only odd harmonics, the existence of this phe-

Oct., 1933]

WAVE ANALYSIS OF SOUND RECORDS

327

nomenon should lead to a peak in the third and fifth harmonics at the
point of minimum even harmonics. The absence of any such ten-
dency shows that under the conditions of minimum even harmonic
distortion the reproduced wave-form is a faithful copy of the original
signal, since the amount of third harmonic which is present due to
lack of perfection of the photographic image in both peaks and valleys

TABLE II

Modulation of Print Transmission in Decibels below 100 Per Cent

Negative
Density

Negative Fre-
Gamtna quency

0.6

0.9

Print Density
1.2 1.5

2.0

0.6

1.0 100

5.2

5.7

9.4

12.4

18.4

1.0

4.0

4.7

5.8

11.3

0.63

1.5

5.5

6.9

9.1

14.1

21.8

1.0

1.5

4.7

3.9

4.4

4.6

9.9

1.43

1.5

4.X)

3.2

3.4

4.9

1.58

1.5

4.0

3.2

3.1

3.7

0.85

2.0

4.4

4.7

6.6

8.4

15.9

1.0

2.0

3.9

4.0

4.3

6.0

10.2

1.3

2.0

4.1

3.7

3.5

4.2

6.1

1.5

2.0

3.9

3.1

3.2

3.4

3.7

2.1

2.0

3.8

3.1

3.2

0.8

1.0 2500

6.8

7.0

9.1

12.5

1.0

6.5

5.7

6.8

8.2

15.3

0.64

1.5

7.3

8.4

11.1

15.4

1.0

1.5

6.1

5.7

6.2

7.9

13.8

1.2

1.5

6.5

3.9

5.0

5.2

7.6

1.55

1.5

8.3

5.8

4.9

6.0

5.2

0.8

2.0

6.2

6.6

8.2

11.4

1.0

2.0

5.7

6.6

8.3

13.4

1.2

2.0

5.7

5.4

5.2

5.5

12.0

1.54

2.0

6.2

4.4

5.2

6.6

2.15

2.0

8.2

6.4

5.2

4.7

5.2

0.75

1.0 4000

8.6

8.4

9.9

14.4

1.13

1.0

8.6

7.8

7.6

9.5

13.7

0.70

1.5

9.1

9.6

12.8

16.5

1.05

1.5

8.1

6.8

7.3

8.6

14.0

1.62

1.5

8.9

7.3

5.3

6.9

8.7

1.34

1.5

9.4

7.1

6.9

8.1

10.5

1.66

1.5

10.8

9.4

7.8

7.6

0.8

2.0

8.2

8.9

10.4

14.4

1.0

2.0

8.2

8.0

8.9

11.7

16.7

1.25

2.0

8.0

7.8

7.4

8.3

11.1

1.46

2.0

8.9

7.6

7.3

7.0

8.2

2.1

2.0

11.4

9.9

9.1

8.4

7.7

328 SANDVIK, HALL, AND STREIFFERT [J. S. M. P. E.

is negligible under the conditions existing for minimum even har-
monic distortion.

Table II shows the amount of peak modulation of print trans-
mission obtained for the negative and print conditions tabulated.
Since no correction was made for wave-form, the values may be
slightly in error in cases where there is considerable harmonic content ;
but since these cases are also far enough from maximum signal condi-
tions to make them valueless in practice, no corrections were applied.

As in Table I, for a given negative gamma, there is a value of
negative density giving maximum modulation for each print density,
although the maximum region is very broad at a frequency of 100.
The film losses found under the best conditions are 3.1, 4.4, and 7.0
db. at 100, 2500, and 4000 cycles per second, respectively. Since
the modulation of input signal was 1 db. below 100 per cent, and
since the film base itself causes approximately 1 db. loss, the actual
losses are about 1, 2.4, and 5 db., for the three frequencies 100,
2500, and 4000, respectively.

Table III shows the results obtained from the analysis of the
negative records made in the same way as the print analyses. The
harmonic values here are the combined second and third harmonic
amplitudes, being the square root of the sum of the squares of the
two quantities. The last column shows the modulation of negative
transmission in decibels below 100 per cent, and in general it can be
noted that the harmonic content is directly related to the loss of
output signal. This is especially true at the higher frequencies,
where the filling in of the valleys of the record cuts down the ampli-
tude of the signal, by flattening the bottom of the wave. This
introduces considerable second harmonic distortion into the values
given. It is noticeable that it is possible to obtain a better print
from the standpoint of both modulation and harmonics than can be
obtained as a negative, particularly at the higher frequencies, where
the even harmonics can be balanced out in printing as noted in con-
sideration of Table I.

The general properties of these data can now be more readily dis-
cussed by the aid of a few curves. Thus, Fig. 1 shows the relation
between the total harmonic content and the print density for prints
made from a group of negatives, A, B, C, D, and E, whose densities
were 0.85, 1.0, 1.3, 1.5, and 2.1, respectively, at a frequency of 100.
These curves show that the harmonic content is a function of the
print density, and wherever a sufficient range of print densities was

Oct., 1933] WAVE ANALYSIS OF SOUND RECORDS 329

TABLE III

Characteristics of Negative Records. Total Harmonic in Percentage of Funda-
mental. Output Modulation in Decibels below 100 Per Cent Used as Print

Density

Gamma

Frequency Harmonic

Modulation

0.6

1.0

100 5.5

4.3

1.0

4.9

3.6

0.6

1.5

4.3

5.4

1.0

1.5

4.7

3.5

1.4

1.5

7.1

3.6

1.6

1.5

11.1

3.0

0.85

2.0

4.1

5.0

1.0

2.0

5.8

3.7

1.3

2.0

5.9

3.6

1.5

2.0

6.0

3.6

2.1

2.0

7.9

3.9

0.5

1.0

1000 3.2

4.9

0.72

1.0

4.1

4.2

0.9

1.5

3.5

4.4

0.65

1.5

4.6

4.2

0.80

1.5

3.2

3.5

1.25

1.5

5.1

4.3

0.8

1.0

2500 6.9

4.6

1.0

11.3

5.2

0.65

1.5

5.8

4.4

1.0

1.5

10.0

4.2

1.2

1.5

14.0

5.2

1.55

1.5

13.6

7.3

0.8

2.0

4.6

4.0

1.0

2.0

7.7

4.3

1.2

2.0

12.4

3.9

1.55

2.0

13.9

4.6

2.1

2.0

14.4

8.0

0.75

1.0

4000 13.5

6.8

1.13

1.0

13.0

10.1

0.70

1.5

7.5

6.0

1.05

1.5

13.3

6.5

1.35

1.5

14.0

7.3

1.65

1.5

17.8

9.9

0.8

2.0

5.8

5.7

1.0

2.0

10.7

5.8

1.25

2.0

13.9

5.8

1.5

2.0

15.2

7.0

2.1

2.0

17.6

10.4

used the amount of harmonic content passes through a well-defined
minimum. The position of the minimum with respect to print

330

SANDVIK, HALL, AND STREIFFERT [J. S. M. P. E.

density depends on the negative density. The curves indicate that
for the lowest harmonic content the print density should be equal to
or slightly less than the negative density. Fig. 2 shows a similar
set of curves from records made at a frequency of 4000 cycles per
second. Again, it is seen that the position of this minimum point
depends on the density of the negative from which the prints were
made, and that its position again falls at a point where the densities
of the negative and the print are about equal. It should be observed
that, in general, the harmonic content increases at a somewhat
greater rate above than it does below the optimal value.

A to tl I.* t.ft t.ft Z.O
PRINT Def*\TV

FIG. 1 . Harmonic distortion at 100 cycles per second as a function
of print density for various negative densities. Negative densities
0.85, 1.0, 1.3, 1.5, and 2.1 for curves A, B, C, D, and E, respectively.
Negative gamma 2.0.

Fig. 3 shows the relation between print density and modulation
in print transmission or volume of prints made from negatives
A, B, C, D, and E, as above, that is, densities of 0.85, 1.0, 1.3, 1.5,
and 2.1, respectively, at a frequency of 100 cycles per second. These
curves show a definite relation between the print density and the
modulation. It is seen that modulation, like harmonic content,
is a function of the print density, and that its optimal value falls at a
point where the print density is a little less than that of its negative.
The conditions which produce the smallest amount of non-linear
distortion are fortunately in close agreement with the conditions

Oct., 1933]

WAVE ANALYSIS OF SOUND RECORDS

331

which produce the greatest modulation in the print transmission,
that is, output volume.

That this should be true becomes more or less evident when one
considers the structure of the developed photographic image. That
is, the loss in modulation is due primarily to light scattered beyond
the geometric edge of the image of the recording aperture, thereby
causing a "filling-in" in the valleys of the waves. Now any process
of printing and developing which cancels out this "filling-in" will
also affect a partial or complete restoration of the original wave-form.
The remarkable feature is the degree of cancellation which is ob-

D
\

O .Z A A A i-O .Z 1.4 1.6 1.8 ZO
PRINT DENSITY

FIG. 2. Harmonic distortion at 4000 cycles per second as a func-
tion of print density for various negative densities. Negative den-
sities 0.8, 1.0, 1.25, 1.5, and 2.1 for curves A, B, C, D, and E, re-
spectively. Negative gamma 2.0.

tained by a proper choice of print density. Thus, from a dense
negative which is filled in sufficiently so that the modulation in the
transmission is very low, a print can be made by proper choice of
print density, whose modulation of transmission approaches nearly
the theoretical maximum value.

Fig. 4 shows a set of curves similar to those of Fig. 3, but for a
frequency of 4000. At this frequency the print density for maximum
modulation is better defined than at a frequency of 100. In all cases
studied, the conditions for maximum modulation and minimum

332

SANDVIK, HALL, AND STRELFFERT [J. S. M. P. E.

harmonic content are met by making the print density equal to or
slightly less than the negative density.

The actual records made included four modulations for each
recording condition, namely: 1, 3, 6, and 10 db. below full modula-
tion. These records were all analyzed, and a study of the results
indicates that under satisfactory recording conditions there is little
change in the harmonic distortion for different recording modulations.
The curves of Fig. 5 show some of the effects obtained. Curve E,
the highest, was taken at a frequency of 4000, negative density 1.2,
print density 1.5, and shows a decreasing distortion as does the second

.0 8
^

02 4 .6 .8 I.O \.t. 1.4 I <b 1.8 Z.O

PRINT DE.NS\TV

FIG. 3. Modulation as a function of print density for various
negative densities; 100 cycles per second. Negative densities 0.85,
1.0, 1.3, 1.5, and 2.1 for curves A, B, C, D, and E, respectively.
Negative gamma 2.0.

curve A for a frequency of 100, negative density of 1.5, and print
density of 1.5. The other curves, B, C, and D, show approximately
constant values of about 2 to 4 per cent total harmonic, the
curves being for 1000 cycles, negative density 1.3, print density
1.5; 4000 cycles, negative density 1.5, print density 1.5; and
4000 cycles, negative density 2.0, and print density 2.0. For
abnormal negative or print conditions, the harmonic values may
decrease somewhat with decreasing modulation, particularly at
lower frequencies. In view of the fact, however, that these abnormal
conditions also introduce greater losses in modulation of print trans-

Oct., 1933]

WAVE ANALYSIS OF SOUND RECORDS

333

mission, they were not investigated further. It might be noted here
that even under rather bad conditions of harmonic distortion, the
volume distortion at all frequencies remains negligible, the output
levels keeping exactly the same ratio as the input levels.

From the standpoint of surface noise alone, the density in the
exposed area of the sound track should be infinite, or at least suffi-
ciently high so that the amount of light transmitted by it is negligible.
For practical purposes, this condition is sufficiently well fulfilled in
the neighborhood of a density of 1.5. At this value of negative and
print density the percentage of total harmonic content is nearly at

.Z A .G> 8 t.O 1.1. 14. I.G> 1.6 ZO

PRINT OE.NSITV

FIG. 4. Modulation as a function of print density for various
negative densities; 4000 cycles per second. Negative densities 0.8,
1.0, 1.25, 1.5, and 2.1 for curves A, B, C, D, and E, respectively.
Negative gamma 2.0.

its minimum value, as will be seen by again referring to Figs. 1 and 2.
Now referring to Figs. 3 and 4, one sees that a density of 1.5 is also
approximating the most favorable over-all value from the point of
view of film loss. It may be found necessary to alter this condition
slightly in order to get the best possible response at the frequencies
8000 and 10,000. Such a modification, if any, will very probably
be toward lower negative and print densities. However, any such
modification would not materially affect either the modulation or
harmonic content, since the above data show that by changing the
print density to conform to the density of the negative, one can obtain

334

SANDVIK, HALL, AND STREIFFERT [J. s. M. P. E.

equally satisfactory results over a wide range of densities. It would,
however, affect the relative level of the surface noise somewhat.
If, however, the conditions which lead to the least film loss at the
highest frequencies to be reproduced are not in agreement with the
conditions which give the smallest amount of distortion in the fre-
quency range where the harmonics generated by non-linear distor-
tions would be transmitted by the reproducing systems, then a
further compromise must be made. It is generally agreed that a
certain amount of harmonic distortion may be present before it

- 10

I B
5

I 6
?4

Q \ t. % 4 * 1 &9tO

INPUT IN OE.C.IBE.U.S BE.I-OW IOO ?. MODLH-KTION

FIG. 5. Harmonic distortion as a function of recording amplitude :
A, 100 c.p.s., negative density 1.5, print density 1.5; B, 1000 c.p.s.,
negative density 1.3, print density 1.5; C, 4000 c.p.s., negative den-
sity 1.5, print density 1.5; D, 4000 c.p.s., negative density 2.1, print
density 2.0; E, 4000 c.p.s., negative density 1.2, print density 1.5.
All negatives at gamma 2.0 except B, which is at gamma 1.5.

becomes apparent to an observer. The actual amount which can be
present before becoming perceptible is a subject which has led to a
certain amount of controversy among investigators, the amounts
being variously stated from 2 per cent to 10 per cent. If one assumes
the latter value to be the superior limit, considerable latitude is
allowed in negative and print conditions.

The results given here should serve only as a general basis on which
to operate, making the necessary modifications to suit any particular
case. Thus, for example, the quality of the image of the recording

Oct., 1933] WAVE ANALYSIS OF SOUND RECORDS 335

aperture, the ratio of light on the film within the geometric boundary
of the image to the light in the sound track outside that boundary
as well as the distribution of illumination within the boundary
and outside the boundary, and the type of film and developer used,
would modify the above results somewhat.

In general, it might be stated that the process is not critical to
over-all gamma, in fact that the term over-all gamma has very little
significance. The gammas of the negative and the print within wide
limits are of importance primarily because they serve as a measure
of the density that will be obtained for a given predetermined ex-
posure. For any given negative density developed to any gamma
lying within relatively wide limits, there is a corresponding print
density which gives the smallest amount of non-linear distortion.
There is also a print density which produces the greatest modulation
in the print transmission or output volume. The print conditions
required to satisfy these two conditions are in general mutually
agreeable.

REFERENCES

1 HARDY, A. C.: "The Rendering of Tone Values in the Photographic Re-
cording of Sound," Trans. Soc. Mot. Pict. Eng., XI (1927), No. 31, p. 475.

2 COOK, E. D.: "The Aperture Effect," J. Soc. Mot. Pict. Eng., XIV (June,
1930), No. 6, p. 650.

3 FOSTER, D.: "The Effect of Exposure and Development on the Quality of
Variable Width Sound Recording," J. Soc. Mot. Pict. Eng., XVII (November,
1931), No. 6, p. 749.

4 MAURER, J. A.: "The Photographic Treatment of Variable Area Sound
Films," J. Soc. Mot. Pict. Eng., XIV (June, 1930), No. 6, p. 636.

6 DIMMICK, G. L. : "High Frequency Response from Variable Width Records
as Affected by Exposure and Development," J. Soc. Mot. Pict. Eng., XVII (No-
vember, 1931), No. 6, p. 766.

6 SANDVIK, O., AND HALL, V. C.: "Wave- Form Analysis of Variable Density
Sound Recording," J. Soc. Mot. Pict. Eng., XIX (October, 1932), No. 4, p. 346.

DISCUSSION

MR. COOK: Mr. Sandvik has referred to some work which I did at one time
on a problem related to the subject which he has investigated. Because I have
been asked many times why experimental results differ from the values sup-
posedly predicated in my paper, it would seem desirable to call attention to the
assumptions underlying that work. Infinite sensitivity of the photographic
material was assumed; in variable amplitude film, all exposed parts were to be
uniformly opaque, all clear parts uniformly transparent. The reason for that
assumption lay in the fact that if the edges of the exposed parts were assumed
to be of a variable density, as exists in practice, some laborious numerical inte-

336 SANDVIK, HALL, AND STREIFFERT

grations would be encountered; and at the time the work was done, some time
in 1927, it was more important commercially to know an upper limit for the
loss of fundamental output than to be able to calculate exactly the harmonic
content. Since the above idealizing assumptions made the integrations simple,
the upper limits for these latter effects were also given, but the more general prob-
lem was given up.

Since that time Donald Foster has found occasion to extend the analysis
further. His investigation yields much more exact results in practical problems.
I have had occasion to check Mr. Foster's work approximately and find good
agreement between test results and his predictions. Particularly good agreement
was found between test and calculation for the variation of average transmission.
Now, I should like to ask Mr. Sandvik, first, how were the prints made from the
original negatives? I would assume that they were contact prints made in
a printing frame, since that would seem to be the obvious way to make them.
Second, was the distribution of light along the length of the aperture, as well as
transversely, examined; and if so, what was found? Third, on what type of
recorder were the original records made? And last, what was known about the
wave-shape of the oscillator used? The amount of distortion in some cases was
small enough to make one wonder about that element or even the amplifiers
used.

MR. SANDVIK: Prints were made both in printing frames, and on a con-
tinuous contact printer in order to see whether the results obtained by the two
methods differed. We found that when care was taken to insure good contact
and eliminate slippage on the continuous contact printer, the results in the two
cases were essentially the same.

As to the uniformity of illumination across the image of the recording aperture,
we measured the illumination along the longer dimension, and made it as uni-
form as possible along the entire length of the image. The maximum deviation
at any small section of the image from the average of all other similar sections
was of the order of one to two per cent. As to the other dimension, I can't give
you any definite information. Due to diffraction at the two edges the distri-
bution of illumination across the slit image is, of course, not uniform.

MR. PALMER: Are these harmonics produced by a spreading of the image,
due to a longer time of development?

MR. SANDVIK: The chief cause of the distortions is due to the spreading of
the image. There are other secondary causes, such as aperture effect, which I
consider very small compared with the spreading within the emulsion.

Regarding the distortion due to spreading within the emulsion, it doesn't
make much difference how the density is obtained whether by exposure or by
development. Spreading would be of approximately the same order, although
since the sharpness of the edge or the density gradient is a direct function of the
gamma of development, it is better to get the density by development.

MR. BATSEL: What percentage of complete modulation was used in making
these measurements?

MR. SANDVIK: We found that for a "normal" sound track, throughout the
frequency range investigated, the distortion is practically independent of modu-
lations below 100 per cent.

MILITARY TRAINING AND HISTORICAL FILMS*
F. W. HOORN**

Summary. Some of the problems involved in developing a program of military
and civilian training films are briefly described, in addition to the manner of using
the films in peace-time and war-time training. The relation between the Signal
Corps and the storage of historical film in the New Archives Building at Washington,
D. C., is briefly referred to.

It is assumed that the interest of the S. M. P. E. in the subject
of military use of films is based 'upon two aspects of the activity.
If the subject has any significance, it is derived, first, from the relation
of the undertaking to national defense and, second, from its possible
value as a guide to the possibilities that may lie in the use of talking
films for educational purposes.

This assumption is made without any implication that the army
is now ready to point the way to the development of training films.
Such an implication would be far from the fact. Only a very small
portion of the army's resources have been or can be placed at the dis-
posal of this activity until such time as its development can begin to
show promise of dividends commensurate with the investment.
It is well known that the army's resources are so scattered over
a wide range of civil and military missions, that any further dilution
of its strength by embarking on a grand scale upon new activities
of unproved worth is not considered justifiable. Before any con-
siderable commitments are made in the field of training films, it
will probably be required that measurable results demonstrate their
value.

This preface is not offered in order to evade responsibility for the
many limitations of technic in the output available to date, but to
present a true picture of our little organization, as a basis for an
analysis of its possibilities.

In the development of any training film program there are two

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Captain, U. S. Signal Corps, Washington, D. C.

337

338 F. W. HOORN [J. S. M. P. E.

distinct phases : the building up of a library of films and the creation
of an outlet by providing projection facilities. This appears to have
resulted in a vicious circle in the civilian teaching field in that the lack
of each has retarded the growth of the other.

In the field of military training the scope of the subject matter is
not so broad, and consequently the vicious circle should be easier
to break down by the production of a reasonable number of films of
wide applicability. Specificially, when twenty or more sound film
subjects have been produced, it will probably be justifiable to begin
the program of supplying projectors.

At present the activity is in a development rather than a production
stage. The laboratory is small and the personnel limited. Whether
or not it will grow, and to what extent, depends upon future develop-
ments. Next year it is hoped that we shall be able to produce from
twelve to sixteen reels of training film negative, if unexpected per-
sonnel cuts are not imposed. While this seems a small amount, it
should be adequate as a basis for study and future planning.

Ever since the World War, the army has produced and utilized in a
desultory manner, training films on a fairly wide variety of subjects.
These films were, however, of the silent type, and were therefore
much more limited in scope than those at present possible. Only
subjects in which the picture was virtually self-explanatory could
advantageously be presented. Titles as a means of explanation in an
instruction film leave much to be desired. That one picture is worth
a thousand words is one of those sayings whose forceful brevity
often inclines one to exaggerate then- applicability. Instruction in
architecture might conceivably be given mainly by pictures, whereas
instruction in economics or philosophy could best be presented by
words. Most military subjects are such as to require both picture
and words. Keeping clearly in mind that motion is the essence of good
films and that verbalism is obnoxious, it is nevertheless somewhat
apparent that subjects involving personnel, materiel, and their
maneuver over terrain, can best be presented with the aid of verbal
explanation to support the picture.

For that reason the advent of sound films gave a slight impetus
to the training film idea and indicated the desirability of determining
their usefulness. A start was made in acquiring a minimum of essen-
tial equipment and in training personnel. This personnel training
was further handicapped by the military necessity for rotation of
officers in duty and station, thereby limiting specialization.

Oct., 1933] MILITARY TRAINING AND HISTORICAL FILMS 339

More publicity has been given to the undertaking than is warranted
by its magnitude. The idea of replacing the legendary hard-boiled
drill sergeant with a phantom screen instructor seemed a fairly wel-
come subject for speculation by the press. One commentator haz-
arded the thought that the screen would not go far toward creating
the "day-of- judgment atmosphere provided by these old-time
drillmasters."

Actually, the field of training films is much broader than close-
order drill. As far as present experience permits one to judge, it
appears that its greatest value will lie in the field of tactics, especially
the tactics of small units. For such a type of subject can more easily
be produced and, when finished, has a wider applicability than those
pertaining to the higher commands. More of the actual movements
of the unit can be shown and less reliance needs be placed on anima-
tion.

In the case of large units, maps and drawings or aerial views,
combined with animation devices for locating organizations and
activities on the ground, will be indispensable. Actual scenes will be
limited to those showing the activities of commanders and staffs,
and special activities which can be viewed at sufficiently close range
to make them distinct.

In all training subjects the technic of presentation must offset the
well-known limitations of camera and screen as to distance and narrow
angle of vision. In the drama the relative location of events is usually
not as important as are the emotions and characters of the par-
ticipants, while in the military films continuous orientation of the
observer as to the space relations of the things he sees is usually
essential.

At the beginning of our development program, it seems desirable
to produce as wide a variety of subjects as possible, in order to be
able to judge the susceptibility to treatment of each type. In the
second stage, effort will probably be concentrated on the production
of subjects for which there is the rnost widespread demand, inasmuch
as efficiency of the method of instruction depends in large degree on
the universality of the subject.

Experience is still somewhat limited as to the best method of
treatment of military subjects. At the outset, only the off-stage voice
type of film was contemplated. Since then, it has seemed desirable
to attempt field recording for the sake of heightening the interest by
the inclusion of dialog between the actors wherever it appears that

340 F. W. HOORN [J. S. M. P. E.

the instructional matter can be so presented. A natural sound film
on the subject of Communications in the Infantry Regiment has
just come through the shooting stage and is awaiting cutting and
editing.

This film will at least furnish information as to the feasibility of
this method with our present organization. If it is satisfactory, future
productions will probably be combinations of the descriptive talk
and the natural sound methods. Subjects involving technic and the
method of handling materiel can often be handled best by the de-
scriptive talk or off-stage voice, while subjects based on the decisions
and actions of personnel can be handled largely by natural sound
with occasional off-stage voice explanations. Considerable study
can be utilized devising ways of incorporating instructional matter
in the speech of the actors without undue sacrifice of naturalness.

The success or failure of instructional films, whether military or
civilian, will in the end depend on the instructional effectiveness
which can be developed and the relative efficiency on a cost basis
as compared with other methods. A few of the factors affecting
instructional value have been touched upon. Actually its value
can be determined only by trial and test with films produced under
existing limitations. Relative efficiency on a cost basis can be
estimated only by making certain broad assumptions as to the con-
ditions under which film will be projected. Such data as are avail-
able indicate that a film of wide applicability in the military service
may be shown throughout its life to a hundred thousand persons.
On this basis, the cost per individual per hour of instruction should
not exceed fifteen cents, and may be considerably less.

In order that the widest possible use may be made of the films in
time of peace, sound projection facilities should be made available
to the civilian components of the Army, namely, the Organized
Reserve, the R. O. T. C., and the National Guard. For this purpose,
the 16-mm. projector seems to be quite well suited. Any improve-
ments which can be made in the way of simplicity so that the pro-
jector can be operated by any one after looking over the instructions,
will be a step toward solution of the future projection problems.

Until such time as it seems advisable to supply portable projectors
in considerable number, the use of sound training films will be limited
to groups who can make use of projection facilities on nearby army
posts or to groups in the large cities near Corps Area Headquarters
where portable projectors can be made available on loan. Another

Oct., 1933] MILITARY TRAINING AND HISTORICAL FILMS 341

method of utilizing these films for the time being is to send a mimeo-
graphed copy of the descriptive talk with the film, to be read in approxi-
mate synchronism with the picture by someone conversant with
the subject. This method has been used in some instances with
success but its effectiveness under average conditions has not been
determined.

In the event of a major war, the training film program will probably
enter a third stage, namely, the hurried production of films on basic
subjects for mass instruction of men throughout the country accord-
ing to uniform standards, using the latest developments of the mo-
ment. An attempt to provide a complete library in advance of such
a contingency would not be warranted, for military methods change,
as does everything else, and many of the films would be obsolete
when needed. If and when such a need arises, the army will have to
lean heavily on the industry in organizing a producing unit, capable
of turning out, let us say, eight or ten reels a week.

Production of historical film lies in the past and in the future.
At present only a very limited amount of footage is being produced
that can in any sense be termed historical.

The termination of the World War left us with approximately
six hundred thousand feet of negative made in France. This film
includes action at the front and in the rest and training areas, ac-
tivities at supply depots, and debarkation centers.

The quality of the film both as to photography and subject matter
varies greatly. A great deal depended on the initiative and skill
of the individual cameramen. In many cases the scenes are longer
than necessary, and editing while shooting could have been accom-
plished to great advantage.

However, the A. E. F. file constitutes a valuable record that will be
of increasing interest as time goes on. That every means should be
taken to preserve this film, or the most valuable part of it, is self-
evident. A step in this direction was taken when the advice of a
technical committee was obtained in order to provide the best possible
conditions in a film storage vault of the Archives Building at Wash-
ington, D. C., where it is contemplated placing this film for indefinite
storage.

At the present time the World War film is used principally for loan
to the American Legion and other service organizations, patriotic
societies, and educational institutions. Our laboratory makes approxi-
mately a thousand loans a year to meet this demand.

342 F. W. HOORN

In the event of another war, the organization of the photographic
service should be such as to make possible a continuous record of
each tactical division, with animation to show their movements and
descriptive talk to supply further detail and color.

To accomplish this task would require a considerable personnel,
with prior training in the various specialties, administration, direc-
tion, laboratory supervision, and technical work. A central labo-
ratory well to the rear would do the film processing and probably
animation work. A photographic company would be included among
the Army troops of each Field Army. From that company would be
sent out nine divisional assignment units to cover the activities
of the combat divisions. If, say, four field armies were mobilized,
it is estimated that the photographic personnel required would be
not less than ninety officers and eight hundred men, almost all of
whom would have to come from the industry.

BOOK REVIEW

Light in Motion Picture Projection (La Lumiere dans la Projection Cinema-
tographique). JACQUES MARETTE. Gauthier-Villars Co., Paris, 1933.

The purpose of this book, as stated in the introduction, is to present a coherent
practical explanation of the factors in projection influencing the quality of the
projected motion picture. The treatment is rather elementary, but the book is
valuable because it fulfills the need for a well-rounded exposition of the subject
in language understandable to those engaged in practical projection.

Topics touched upon include the following : principles of vision, quality of the
film image, visual appreciation of the projected image, the effect upon picture
quality of screen brightness and stray light, optical properties of the screen, rear
projection, light sources, the large screen, etc.

C. E. IVES

SOCIETY ANNOUNCEMENTS

FALL MEETING OF THE SOCIETY, EDGEWATER BEACH
HOTEL, CHICAGO, ILL.

OCTOBER 16-18, INCLUSIVE

CONVENTION ARRANGEMENTS COMMITTEE
W. C. KUNZMANN, Chairman

J. H. KURLANDER M. W. PALMER

LOCAL ARRANGEMENTS COMMITTEE

R. F. MITCHELL, Chairman

J. E. JENKINS R. P. BURNS O. HOLMES

H. T. COWLING G. W. BAKER J. E. McAuLEY

H. H. DEVRY O. B. DEPUE J. H. GOLDBERG

A. WARMISHAM B. W. DEPUE O. F. SPAHR

PROJECTION COMMITTEE

H. GRIFFIN, Chairman

J. E. McAuLEY J. H. GOLDBERG

R. P. BURNS J. FRANK, JR.

Members of Local No. 110, I. A. T. S. E.

LADIES COMMITTEE

MRS. O. B. DEPUE, Hostess

assisted by

MRS. R. F. MITCHELL MRS. B. W. DEPUE MRS. J. E. JENKINS

MRS. H. T. COWLING MRS. H. DEVRY MRS. A. WARMISHAM

OPENING OF CONVENTION

The Convention will convene at 12:30 P.M., Monday, October 16, in the North
Room of the Edgewater Beach Hotel, opening with an informal luncheon, during
which the members of the Society will be addressed by several prominent speak-
ers. Luncheon tickets may be procured at the registration headquarters;
charge, one dollar per plate. The morning preceding the luncheon will be de-
voted to registration, committee meetings, and various organization matters.
The convention registration fee will be three dollars.

The hotel may be reached from the La Salle St. station either by taxicab ($1.50),
by No. 51 bus going north on Michigan Avenue, or by elevated express train at
the Howard St. station.

343

344 ^ SOCIETY ANNOUNCEMENTS [J. S. M. p. E.

SESSIONS

All tephnical sessions and film programs will be held in the East Lounge of the
hotel, wh;ere also will be located the registration headquarters. The Berwyn
room win be provided for the Board of Governors and the technical committees.
Technical sessions will be held on Monday, Tuesday, and Wednesday afternoons,
and on Tuesday and Wednesday mornings. The film program, under the direction
of Mr. H. DeVry and Mr. S. A. Lukes, will be held on Monday evening, and will
include several recent outstanding productions, in addition to several reels of the
Century of Progress World's Fair, photographed by Mr. H. T. Cowling.

BANQUET AND DANCE

EXHIBIT OF NEW MOTION PICTURE APPARATUS

Arrangements are being made to hold an exhibit of newly developed motion
picture apparatus, in order to acquaint the members of the Society with the
newly devised tools of the industry. This exhibit will not be of the same nature
as the usual trade exhibit. There will be no booths, although each exhibit will
be allotted definite space, and all exhibits will be arranged in one large room.
Please direct requests for space to the General Office of the Society, 33 West
42nd St., New York, N. Y., stating the number and nature of the items to be

exhibited.

SPECIAL EXTENDED RATES

Excellent accommodations are assured by the management of the Edgewater
Beach Hotel, and minimum rates are guaranteed. Reservations, subject to
cancellation prior to the convention dates, must be made at once, on account of
the heavy advance registration incident to the large attendance at the Century
of Progress Fair. Special rates have been arranged by the hotel, whicft will be
effective during the stay of S. M. P. E. delegates and their guests in Chicago,
should they wish to visit the Fair before or after the Convention. Particular
attention is called to the special railroad rates applying to trains leaving New
York on Saturdays and Tuesdays; 10-day round trip charges: fare $33, lower
berth $13.50; in contrast with the regular rates on other days of the week of
$65.40 and $18, respectively.

CENTURY OF PROGRESS WORLD'S FAIR

LADIES' HEADQUARTERS

A reception suite will be provided for the use of the ladies attending the Con-
vention, and an attractive program for their entertainment is being arranged by

Oct., 1933]

SOCIETY ANNOUNCEMENTS

345

the Ladies' Committee. Upon arriving, the ladies should register immediately
at the reception room. The ladies are invited, weather permitting, *1t<j be the
guests of Mr. and Mrs. H. DeVry aboard their yacht, for a cruise on Lake Michi-
gan and a view of the beautiful waterfront of the World's Fair.

TENTATIVE PROGRAM

MONDAY, OCTOBER 16TH

The morning will be devoted to organization of the Convention, registration,
meetings of committees, etc.

12:30 a.m. Luncheon (for members and their families and friends). Addresses

by prominent speakers, to be announced later.
2:30 p.m. East Lounge: Business Session.

Opening of Convention; A. N. Goldsmith, President.

Report of the Secretary; J. H. Kurlander.

Report of the Treasurer; H. T. Cowling.

Report of the Convention Arrangements Committee; W. C Kunz-

mann, Chairman.
Society Business: election of Officers for 1933-34; proposals for

amendment of Constitution and By-Laws. (The importance of

the matters to be acted upon at this meeting makes a full voting

attendance very desirable.)
Report of the Membership and Subscription Committee; E. R.

Geib, Chairman.

Technical papers will be scheduled for this session, as time permits.
8:00 p.m. East Lounge: Exhibition of recent talking motion pictures, including

pictures of the Century of Progress Fair, taken by H. T. Cowling.

TUESDAY, OCTOBER 17TH

9 -.30 a.m. East Lounge: Technical Session.

"Wide Range Recording;" by F. L. Hopper, Hollywood, Calif.
"Acoustic Requirements for Wide Range Reproduction;" by S. K.

Wolf, Electrical Research Products, Inc., New York, N. Y.
"Wide Range Sound Reproduction;" by J. S. Ward and F. C.

Willis, Electrical Research Products, Inc., New York, N. Y.
Report of the Historical and Museum Committee; E. Theisen,

Chairman.
Report of the Committee on Laboratory and Exchange Practice;

R. F. Nicholson, Chairman.
"Sixteen-Millimeter Film and Film Recording Problems;" by J. O.

Baker, RCA Victor, Inc., Camden, N. J.
2:00 p.m. Group "A," East Lounge: Projection Session.

"New 35-Millimeter Portable Projector;" by H. Griffin, Interna-
tional Projector Corp., New York, N. Y.
"Continuous Projection in Motion Pictures;" by H. R. Menefee,

Pathe News, New York, N. Y.
"Automatic Change-Over Device;" by A. Pritchard, Coronado,

Calif.

346

SOCIETY ANNOUNCEMENTS

[J. S. M. P. E.

"The Control Frequency Principle;" by J. E. Jenkins, Jenkins &
Adair, Chicago, 111.

2:00 p.m. Group "B," General Session.

"Acoustic Materials;" by H. M. Baker, U. S. Gypsum Co., Chicago,

111.
"Some Suggestions on Motion Picture Laboratory Practice;" by

D. E. Hyndman and H. E. White, Eastman Kodak Co., New York,

N. Y.
"The Photographic Disk Reproducer;" by E. D. Cook, RCA Victor

Co., Camden, N. J.
"Recent Improvements in the Bell & Howell Automatic Printer;"

by R. F. Mitchell and A. S. Howell, Bell & Howell Co., Chicago, 111.
"The Rotambulator, a New Type of Camera Stand;" by J. A.

Dubray, Bell & Howell Co., Chicago, 111.

7:30 p.m. Ball Room: Convention Banquet.

Dancing, Motion Pictures, and Entertainment.

WEDNESDAY, OCTOBER 18TH

9:30 a.m. East Lounge: General Session.

"Film Noise in Sound-on-Film Reproduction;" by H. C. Silent,

Electrical Research Products, Inc., Hollywood, Calif.
Report of the Standards and Nomenclature Committee; M. C.

Batsel, Chariman.
"Educational Use of Sound Pictures" (with demonstration); by

H. B. Lemon, University of Chicago, Chicago, 111.
Report of the Projection Practice Committee; H. Rubin, Chairman.

2:00 p.m. Group "A," East Lounge: Illumination Session.

"A Non-Rotating D-C. High-Intensity Arc;" by A. C. Downes,

National Carbon Co., Cleveland, Ohio.
"Alternating-Current Arc Equipment;" by J. E. Robin, Palisades,

N.J.
"A New Type of Carbon Arc Broadside Lamp for Use in Motion

Picture Production;" by Peter Mole, Mole-Richardson, Inc.,

Hollywood, Calif.
"Economies of Projection Lamps;" by E. W. Beggs, Westinghouse

Lamp Co., Bloomfield, N. J.

2:00 p.m. Group "B," General Session.

"The New DeVry Camera;" by H. A. DeVry, Chicago, 111.
"Home Talkie Movies;" by C. F. Jenkins, Washington, D. C.
"Effect of Film Shrinkage on Sound Film Printing;" by J. Crabtree,

Bell Telephone Laboratories, New York, N. Y.
"Manufacturing Problems in Sound Picture Equipment;" by H. E.

Decamp, Western Electric Co., Chicago, 111.
"Some Applications of Sound Reproducing Equipment at the

Century of Progress Exposition;" by E. P. Kennedy, Electrical

Research Products, Inc., Chicago, 111.

Oct., 1933 J SOCIETY ANNOUNCEMENTS 347

ADJOURNMENT OF THE CONVENTION

NOTE: The Society of Motion Picture Engineers will not be responsible for
statements made by authors. It should be understood that this program is
tentative, and that the order of presentation of the papers is subject to change by
the Papers Committee.

Convention Committee, Papers Committee,

W. C. KUNZMANN, Chairman O. M. GLUNT, Chairman

STANDARDS COMMITTEE

An informal meeting of the group of the Standards Committee engaged in pre-
paring and collecting data on sprocket dimensions was held at the General Office
on September 12. Drawings and charts submitted by Mr. H. Griffin were
studied preparatory to submitting them to the Standards Committee as a whole
at its next meeting, which is to be held on September 27.

A complete revision of the S. M. P. E. Standards booklet has been prepared for
submission to the Society at the approaching convention in Chicago, October 16
to 18. The Committee is also communicating with the British Kinematograph
Society of London in regard to deciding upon the dimensions of a standard
film core.

PROJECTION PRACTICE COMMITTEE

At a meeting held at New York, N. Y., September 19, the preliminary arrange-
ment of the semi-annual report, to be presented at the Chicago Convention, was
decided upon, the tentative draft to be acted upon at the next meeting on October
4. The subjects of the report will include an exposition of the elementary
systems of sound projection, projection routine and maintenance, and precau-
tionary measures for avoidance of trouble in projection.

EXTREMELY

FNE GRAN

. . * I HAT, of course, is the first es-
sential in any film that is to serve
satisfactorily for projection back-
ground shots. Eastman Background
Negative has this prime requisite
. . . plus adequate speed . . . plus ex-
cellent processing characteristics.
In short, it is being demonstrated
every day that this new Eastman
film is ideally adapted to its im-
portant special purpose. Eastman
Kodak Company. (J. E. Brulatour,
Inc., Distributors, New York,
Chicago, Hollywood.)

EASTMAN

BACKGROUND NEGATIVE

JOURNAL

OF THE SOCIETY OF

MOTION PICTURE ENGINEERS

Volume XXI NOVEMBER, 1933 Number 5

CONTENTS

Page

Directional Effects in Sound Film Processing II

J. CRABTREE AND J. H. WADDELL 351

Analysis of Sound Quality with the Variable Density Recording

Method from Sensitometric Data

R. SCHMIDT AND- A. KUESTER 374

The Aperture Alignment Effect E. D. COOK 390

The Morgana Color Process J. A. DUBRAY 403

New Developments in Portable Gas-Electric Generators for

Motion Picture Lighting P. MOLE 413

A Silent Camera H. R. KOSSMAN 420

A Triplex Moviola for Editing Re-Recording. . J. O. AALBERG 426

Book Review , 428

Society Announcements . 429

JOURNAL

OF THE SOCIETY OF

MOTION PICTURE ENGINEERS

SYLVAN HARRIS. EDITOR

Board of Editors
J. I. CRABTREE, Chairman

O. M. GLUNT A. C. HARDY F. F. RENWICK

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers.

Publication Office, 20th & Northampton Sts., Easton, Pa.
General and Editorial Office, 33 West 42nd St., New York, N. Y.
Entered as second class matter January 15. 1930, at the Post Office at Easton,
Pa., under the Act of March 3, 1879. Copyrighted, 1933, by the Society of
Motion Picture Engineers, Inc.

Papers appearing in this Journal may be reprinted, abstracted, or abridged
provided credit is given to the Journal of the Society of Motion Picture Engineers
and to the author, or'authors, of the papers in question. The Society is not re-
sponsible for statements made by authors.

Officers of the Society

President: A. N. GOLDSMITH, 444 Madison Ave., New York, N. Y.
Past President: J. I. CRABTREE, Kodak Park. Rochester, N. Y.
Vice-President: W. C. KUNZMANN, Box 400, Cleveland, Ohio.
Vice-President: O. M. GLUNT, 463 West St., New York, N. Y.
Secretary: J. H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J.
Treasurer: T. E. SHEA, 463 West St., New York, N. Y.

Governors

E. COUR, 1029 S. Wabash Ave., Chicago, 111.
H. T. COWLING, 7510 N. Ashland Ave., Chicago, 111.
P. H. EVANS, 1277 East 14th St., Brooklyn, N. Y.
R. E. FARNHAM, Nela Park, Cleveland, Ohio.
H. GRIFFIN. 90 Gold St., New York. N. Y.
E. HUSE, 6706 Santa Monica Blvd., Hollywood, Calif.
W. B. RAYTON, Rochester, N. Y.

DIRECTIONAL EFFECTS IN SOUND FILM PROCESSING II*
J. CRABTREE AND J. H. WADDELL**

Summary. This paper represents a continuation of work previously reported
(J. Soc. Mot. Pict. Eng., XVIII, No. 2, February, 1932, p. 207) concerning
distortion of the photographic image resulting from directional currents of developer
generated by movement of the film in film processing machines. Simpler methods
of breaking up the directional currents than those previously described have been
examined. Characteristic curves from "directional pairs" of sensitometric strips,
and "densograms" of uniform exposures have been used for purposes of comparison.

Application of developer in the form of fine jets impinging upon the emulsion
surface at frequent intervals has been found largely to eliminate directional effects.
Violent agitation of the body of developer in the developing bath by injection of air
gave good results, but was inferior to the jet system. Propelling the film on edge in a
tray type machine offered no advantage over the horizontal film position. It is shown
that under conditions found to be good from the standpoint of directional effects,
distortions due to "Eberhardt effect" are present.

In an earlier paper by J. Crabtree, 1 of the same title, it was shown
that in continuous film processing machines the unidirectional
movement of the film generates a current of developer relative to
it and opposite in direction. This current causes the photographic
density at any point of the film to be affected by that preceding it,
gives rise to various distortions of sensitometric exposures, and
introduces distortion into the sound record. It was shown that this
current is very persistent but can be broken up by use of a multiple
squeegee device. It is obvious that any system of developer circula-
tion efficient enough to suppress the directional current will prevent
these effects.

Subsequent examination of sensitometric and flash exposures
developed in the machines used in industrial laboratories showed
that this directional effect is present in all commercial film processing
machines in various degrees, depending upon the type of machine
and the manner of developer circulation; being pronounced in nega-
tive development of positive film to low gammas and. less in positive

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Bell Telephone Laboratories, New York, N. Y.

351

352

J. CRABTREE AND J. H. WADDELL [J. S. M. P. E.

NOISSHN9NVal

NOISSIASNVdi

x z;

u oi

11s

NOlSSIMSNVdi

Nov., 1933] DIRECTIONAL EFFECTS IN PROCESSING 353

development, as would be expected from conclusions arrived at in the
paper referred to above.

Fig. 1 summarizes data obtained from such an examination.
It shows H&D characteristic curves derived from "direction pairs"*
of sensitometric exposures for both negative and positive develop-
ment in twelve different laboratories. The degree of separation be-
tween these curves gives an indication of the extent of the "direc-
tional effect."

In addition to the sensitometric exposures, lengths of film were
processed which had received exposures to light over a longitudinal
distance of twelve inches, across the entire width of the film. These
exposures were made in a printing frame under conditions known to
give uniformity of exposure. After processing, a "densogram" of each
strip was made on a recording densitometer, the scanning being done
at the normal location of the sound track. This results in a graph
in which values of light transmission appear as ordinates and longi-
tudinal distances along the film appear as abscissas. If development
is uniform, the light transmission will be constant and the graph is a
horizontal line. When directional currents exist, the leading end of
the exposure will attain a higher density than the following portions
and the graph of transmission will show a rising characteristic, the
degree of slope and the length of the curved portion being measures
of the degree of non-uniformity of the development.

In Fig. 1 the graphs of the flash exposures are shown above the
corresponding H&D characteristic curves. Those curves having the
same numeral are from the same laboratory. Eleven commercial
laboratories are represented, the twelfth (No. 2) being the Sound
Picture Laboratory of Bell Telephone Laboratories, Inc.

It is evident from the graphs that in all the laboratories for which
data are available directional effects are pronounced in negative
development.

The modifications of the sound record which result from these
effects are not easy to determine. An attempt was made, by har-
monic analyses of microphotometric traces of the developed record,
to measure the distortion produced by development of a 100-cycle
record under conditions known to produce directional effects. The
results were not entirely satisfactory and do not warrant conclusions,
and lack of time has since prevented their repetition.

* That is, two identical sensitometric exposures are processed, one with the
"toe"end, the other with "shoulder" end, leading through the developer.

354 J. CRABTREE AND J. H. WADDELL [J. S. M. P. E.

Another development defect arising from insufficient developer
circulation results from the surge of developer through the film per-
forations, causing local turbulence around such perforations and
occasioning an increase in the degree of development adjacent to
these points. This may give rise to a variation in density of the sound
track corresponding to the frequency of the perforations. A micro-
photometric trace of three frames of a 5000-cycle record having such
defect in a marked degree is shown in Fig. 2.

FIG. 2. Sprocket hole modulation due to development defects.

It is obvious that an entirely satisfactory development of a sound
film record can not be attained unless the circulation of developing
solution at the emulsion surface of the film is sufficient to overcome
the effects mentioned. An attempt was therefore made to determine
what might be an effective means of attaining the degree of circula-
tion essential to this requirement. This doubtless depends to a con-
siderable degree on the type of machine involved. While only one
type the Erbograph, a horizontal tray type was available at the
Sound Picture Laboratory for the experiments, it was hoped that
certain principles could be arrived at covering conditions in general.

Rapid interchange, of developer at the surface of the film may be
accomplished by:

(1) Continual wiping of the emulsion surface by squeegee, brush, or by jet
of developer;

(2) Sufficiently violent agitation of the mass of developer; by :

(a) rapid inflow and outflow,

(6) agitation by blades or paddles,

It was felt that squeegees and brushes should be avoided, if possible,
as representing potential causes of scratches, and additional parts
requiring servicing. It was also felt that avoidance of agitation
by moving mechanical parts should be sought.

Nov., 1933] DIRECTIONAL EFFECTS IN PROCESSING 355

Agitation by the bubbling of air might be considered undesirable
on account of the increased rate of oxidation of the developing rea-
gents induced thereby. Although this is undoubtedly the case with
highly alkaline positive developers, the aeration induced by the
bubbling of air would probably in some laboratories not greatly
exceed that normally suffered due to entrainment of air by the
developer pump circulating systems. In any event, the depletion of
developer ingredients from aeration is a fraction of that due to the
developing of the image.

In the metol-hydroquinone developers of low alkalinity (borax
developers), in general use for negative development, aeration ac-
tually increases the developing power of the bath for a time. This
has been shown to be due to the oxidation of the hydroquinone
sulfite to the sodium salt of hydroquinone monosulfonate, liberat-
ing sodium hydroxide, which increases the alkalinity of the bath
and therefore increases its developing power. The exhaustion of the
bath by development of the image is accordingly partially offset by
this increase of developing power, and it would seem possible that in a
bath intended for comparatively short life, as negative baths usually
are, the composition could be arranged so as to balance to some
degree the effect of aeration against that of exhaustion, and thus
attain a fairly constant developing power.

It is felt, therefore, that bubbling of ah- as a means of securing
thorough agitation of developer is a practical possibility and perhaps
worthy of further study.

The most promising means of circulation and agitation was
considered to be by jets. By allowing the developer to enter the
developing tank as a multiplicity of jets under a suitable head, a most
efficient agitation can be secured. By making the jets of developer
impinge on the emulsion surface of the film, which is submerged in
the body of the developer solution, they serve the additional purpose
of multiple squeegees. The developer tank may, however, be left
empty, the film being suspended in air and the developer applied to
the emulsion surface by the jet system, which results in a very rapid
change of developer at the film surface. Aeration is, of course, marked
in this latter case.

One Hollywood laboratory employs a machine of very ingenious
design in which the film travels in a horizontal direction but with the
emulsion plane vertical, i. e., film on edge. Although this arrange-
ment is made for purely mechanical reasons it was thought that

356 J. CRABTREE AND J. H. WADDELL [J. S. M. P. E.

gravity would assist in the removal of development reaction products
more readily than in the arrangement where the film surface lies in a
horizontal plane. It was thought worth while to study this plan.

The following ten conditions were compared in the Erbograph
machine at the Sound Picture Laboratory* from the standpoint of
perfection in results from development:

FIG. 3. Erbograph machine; developing trays, showing inlet pipe
and overflow weir.

Condition 1. No circulation of developer other than that generated by the
motion of the film : film lying in a horizontal plane.

Condition 2. Developer inlet by single pipe at one corner, outflow over wire
along one side (Fig. 3) : film lying in a horizontal plane.

FIG. 4. Developing tray, showing single system of sprays.

* Fig. 3 shows the normal arrangement of tray and film in this type of con-
tinuous processing machine.

Nov., 1933] DIRECTIONAL EFFECTS IN PROCESSING 357

Condition 3. As in (2), but film on edge.

Condition 4. Developer inlet as a series of jets issuing from holes VM inch in
diameter in a series of pipes secured to the bottom of tray (see Fig. 4 for arrange-
ment). Film and jets submerged; film horizontal ; jets in line. Rate of circula-
tion 9 gallons per minute.

Condition 5. Developer inlet as double set of jets; one set at bottom of tray
with jets directed upward as in (4) and a second similar set suspended above the
tray and directed downward on the top strands of the film (Fig. 5). Film sub-
merged. Total rate of circulation 16 gallons per minute.

Condition 6. As in (5), but film on edge.

Condition 7. As in (5), but film suspended in air, i. e., developer return by
pump arranged to keep the tray always empty.

Condition 8. As in (2), but air bubbled through the developer from the system
of pipes shown in Fig. 4: film horizontal. Air consumption not determined,
but in quantity sufficient to cause as violent ebullition as possible without spilling
the developer over the tray sides.

=-;^-r L^--l^ ^^~^^^- r ^^^= = ^^_J^ J.

FIG. 5. Developing tray elevation, showing double system of sprays.

Condition 9. As in (<?), but film on edge.

Condition 10. As in (5), but jets staggered so as to distribute the points of
contact of jet with film effectively over the whole width of the emulsion surface.

Test exposures for comparing development efficiency may be of
several types, but in these experiments consisted of:

(a) A series of "direction pairs" of sensitometric strips on E. K. positive film,
covering various widths of track from full film width to sound track width in
the region of the sound track. Of these, the results from the full and sound
track width exposures are plotted for both negative and positive development
in Figs. 6 to 15.

(b) A series of uniform exposures, twelve inches long and occupying the whole
width of the E. K. positive film. After development, these exposures were
scanned along the entire length of the image in the region of the sound track as
described earlier in connection with Fig. 1, and also scanned transversely from
one row of perforations to the other in several selected places.

The data for both negative and positive development in D76
and D16 developers, respectively are given in Figs. 6 to 15. Di-
rectional effects are shown in almost equal degree for both positive
and negative development, because the positive gammas were so
chosen as to lie on the steep portion of the time-gamma curve, thus

358

J. CRABTREE AND J. H. WADDELL [J. S. M. P. E.

t-
Z

\
1 s

1 "

ION OF

i 1

r -

3 1

s 1

r s

UNIFOR

v,\

NOISSIHSNVdl

bo ..
gco

I.S

'*j

Nov., 1933]

DIRECTIONAL EFFECTS IN PROCESSING

359

( s

/ "

1 x -

Q
^

-1

NOISSIMISNVai

NOISSINSNVU1

360 J. CRABTREE AND J. H. WADDELL [J. S. M. P. E.

facilitating discrimination between the respective performances of
the different conditions tested.

Perfect development will show no difference between direction
pairs of sensitometric strips, and the "densograms" of the uniform
exposures will be straight lines both for the longitudinal and trans-
verse scannings. The fact that in no case were these results fully
attained indicates how difficult it is to secure perfect development by
the use of the modern motion picture developing machine.

A careful analysis of the records reveals the fact that condition
10 (Fig. 15) gave definitely the best results from the standpoints
indicated above. In this case, jets of developer six inches apart
impinge on the surface of the submerged film during its entire travel
in the developer bath, and these jets are carefully staggered to insure
that the whole width of the film is impinged upon.

Condition 5 (Fig. 10) presents exactly the same set-up except that
in this case the jets were in line with the longitudinal axis of the film.
This caused a greater degree of development in the center than
toward the edges, as shown by the transverse scannings of both
negative and positive.

Reducing the number of jets to one bank (condition 4, Fig. 9)
gave an obviously inferior result, although this was a very decided
improvement on the system originally provided for in these machines
(condition 2, Fig. 7). Condition 4 (Fig. 9) also represents about
the average condition prevailing in the film processing industry.
Note particularly in condition 2 (Fig. 7) the extremely unsatisfactory
way in which the uniform exposures were developed. This effect
is even more pronounced in condition 1 (Fig. 6) where no circulation
was provided.

When the double-bank jet system was used with the film on edge
(condition 6, Fig. 11), the results were not nearly so satisfactory as
with the film horizontal. In this case the jets passed mostly between
the strands of film. Since the general developer circulation was
identical with that in condition 5 (Fig. 10), it is obvious that for the
best results the jets must impinge on the emulsion surface; that is,
they must act as a wiper or squeegee, breaking up the directional
current existing at the face of the film as did the squeegee device
described at the beginning of this paper.

No particular advantage over condition 5 (Fig. 10) and certainly
not over condition 10 (Fig. 15) is attained by removing the body of
liquid in the tray (condition 7, Fig. 12). This is probably the result

Nov., 1933] DIRECTIONAL EFFECTS IN PROCESSING 361

of a tendency of the developer to drain along the sagging loops of
film. With tight strands, preferably on edge, this method should
give good results, although the necessary mechanical arrangement
would be cumbersome. Since excellent results were ob tamed by con-
dition 10 (Fig. 15), method 7 was not developed further in view of the
added aeration produced by this system.

The violent agitation produced by bubbling air in conditions 8
(Fig. 13) and 9 (Fig. 14) gave fairly good results, better in the case
where the film is flat (condition 8} and where the bubbles, or the
current of liquid carried hi front of the bubble, strike the emulsion
surface, than where the film was on edge with the current of solution
and bubbles passing between. Although a very decided improvement
was secured by the use of air over the regular system, it was dis-
tinctly inferior to the double developer jet system (condition 10,
Fig. 15). Nevertheless, the use of ah- jets would be useful hi those
cases where for design reasons the developer jets are not convenient.

The running of the film on edge through the developer bath hi the
regular system (condition 5) produces no improvement over running
it hi a horizontal plane (condition 2}, indicating that under these
conditions gravity is of little importance in determining the course
of the reaction products of development.

From these results can be drawn the f ollowing conclusions :

(1) To break up the directional current in a motion picture developing
machine of the Erbograph type at linear film speeds of 20 to 100 feet per minute,
it is necessary that the surface of the film be "wiped" at frequent intervals by
some method either by squeegee, brush, or by a jet of developer. A jet system
is described herein which is simple to install, has no moving parts and is very
effective.

(2) Where such a system as the above is impossible of application, good
results can be obtained by violent agitation of the mass of developer by injected
air (or other gas, preferably inert).

(3) Arrangement of the film so as to enlist the aid of gravity in removal of
the reaction products from the developing image is of little or no benefit with film
moving at the speeds investigated.

It is necessary to point out that while the complete interchange of
developer at the surface of the emulsion will prevent the local effects
referred to, it does not follow that perfect development of the photo-
graphic image will thus be attained. There still remain certain
effects which result from the diffusion of the developer within the
emulsion layer itself. Most prominent among these effects is the
well-known "Eberhardt effect" which refers to the depression of

362

J. CRABTREE AND J. H. WADDELL [J. S. M. P. E.

K
Z

1
i

*
I

LONGITUDINAL DISTANCE

\ -

) 3

/ "

( *

TRANSVERSE DISTANCE

0123
LOG RELATIVE EXPOSURE

FIG. 9 (a). Condition 4, negative circulating system as
in Fig. 4 ; film flat and submerged.

NOISSINSNVtil

A1ISN30

rt S

8?a

G ..

co ro
c.Sf

O [id

3
oo

o- S

Nov., 1933]

DIRECTIONAL EFFECTS IN PROCESSING

363

\
\

I I

5 I

! 1

m ,
*
*
I

UJ \

1 \

1 I

TUDINAL DISTANCE

N ,

TRANSVERSE DISTANCE

O 1 2 ]
LOG RELATIVE EXPOSURE

FIG. 9 (6). Condition 4, positive circulating system as
in Fig. 4; film flat and submerged.

x \

"s\

I N
\ v -

)

NOJSSIHSNVbJ.

1 f\| c

A1MN3Q

a M

364

J. CRABTREE AND J. H WADDELL [J. S. M. P. E.

( "

( 1

( ^

NOISSIHSNVai

bC-O

B-B

r 3

tiv
e a

o o

UNIFORM EXPOSURES DIRECTION OF DEVELOPMENT

LONGITUDINAL DISTANCE

\ 3

V 2

) l

V *

}
) "

TRANSVERSE DISTANCE

0123
LOG RELATIVE EXPOSURE

FIG. 10 (c). Condition 5, negative circulating system a;
in Fig. 5; film flat and submerged.

NOICCinCNVMl

' -

A1ICN30

Nov., 1933] DIRECTIONAL EFFECTS IN PROCESSING

365

\ ^

5 i

TANCE

a
1

) g

) il

* I

(

Sfc

3 6
yj3
.fa 3

> rt
'2 a

'if

a w

p-iO

U.M

1
o

D
1

^ S

^ ?

TRANSVERSE DISTANCE

0123
LOG RELATIVE EXPOSURE

Fio. 10 (b). Condition 5, positive circulating system a
in Fig. 5; film flat and submerged.

s; N

o x

^^^

"**

s^X

' NOISS KHSNVU1

1 N C
A1I9N30

366

J. CRABTREE AND J. H. WADDELL

0123
LOG RELATIVE EXPOSURE

Fio. 13 (a). Condition 8, negative agitation by air, film C^
flat. ja

3
|

ki.

LONGITUDINAL DISTANCE

( !

f
1 5

(. '

/ ?

( *

NOISSIHSNVdl

i - o
A1ISN30

\ 2

\ ^

S /

u 1

ruDINAL

J ?

(

\ 2

) o

/ *

Fio. 12 (o

Nov., 1933]

367

NOISSIWSNVdl

Cond

bO
|

"3 .

3 >.

25
M ^

o >>

a g

^r.

c >o

5.tf

^ofe,

.S
u

368

J. CRABTREE AND J. H. WADDELL [J. S. M. P. E.

NOissinsNvai

a
'%*

M
I*
|i

u bo
C a

.*>
Ufa

..S
^

i i

\
<

rt bo
bo-O
V V

NOISSISNVdl

Nov., 1933]

DIRECTIONAL EFFECTS IN PROCESSING

369

NOISSIWSNVbl

s .

0<fl

+3 &

- )

O ..
5 "3

NOISSIWSNVbJL

+3 Si

-i

0) O

370

J. CRABTREE AND J. H. WADDELL [J. S. M. P. E.

density at the border between a light exposure and a heavier expo-
sure due to diffusion of reaction products outward from the latter
into the former when development is checked before the region of
gamma infinity is reached.

Consider, for example, a square-topped light wave exposure. In
Fig. 16, abed represents a latent image produced by a peak exposure

FIG. 16. Eberhardt effect.

of such a wave, and eadf and bghc those produced by the lesser
exposure of the troughs. When immersed in developer, the emulsion
absorbs an amount of liquid depending on its composition and
temperature, swelling by so doing (in D76 to about 400 per cent).
The emulsion layer is now a spongy network of cells retaining both the
light-sensitive material and the developer. As development pro-
ceeds within this cellular structure, the developer reagents are
depleted and reaction products are formed. A difference in compo-
sition, and therefore of specific gravity, osmotic pressure, etc., now
exists between the developer in abed, the developer hi the bath, and
also in the regions eadf and bghc, since less change has occurred here.
Interchange of new and exhausted developer therefore ensues by
diffusion at the surface plane ab and at the boundary planes ad and
be. This is not an instantaneous procedure, and it follows that
at any moment during active development, the developing power
of the developer in the emulsion layer will be less in the center of the
heavy exposure than at its edges. It follows again from this that the
resultant density of the uniform exposure will be greater at the edges
than at the center, assuming that development has not been allowed
to reach the region of gamma infinity where, of course, the center
portions will receive full development.
The graph of density for the developed image abed will appear as in

Nov., 1933]

DIRECTIONAL EFFECTS IN PROCESSING

371

kl at B (Fig. 16). In the area of lighter exposure, eadf, the developing
power will have been consistently higher hi the center than near the
boundary plane ad, since depleted developer emerging from abed
will reduce the developing power in this latter region. The resulting
density will appear as hi mn or m'n' at C (Fig. 16). The shape of the
developed density wave will then be as shown at D and the wave of
light transmission as shown at E. Moreover, since the average
density of the heavy image will be less than it should be, the ampli-
tude of the wave envelope will be less than complete development
would give, though it will be offset to some degree by the density
depression of the edges of the lighter exposure.

These phenomena actually occur in practice, as is seen from Fig.

:h M M h /!

nnn

B U

\ i U i U

FIG. 17. Eberhardt effect; (a) negative, gamma 0.6;
(6) print from (a), gamma 2.0; (c) negative, gamma 2.0.

17, which shows a microphotometric trace of a square-topped wave
of 1000 cycles made by exposure to positive film through a line
grating test object and machine developed to gamma 0.6 under
condition 10 hi the above tests. This is a graph of transmission
comparable to E in Fig. 16, and its resemblance thereto is readily
apparent. The variation in transmission of the dense portion of the
wave (bottom of figure) is not very pronounced in the graph; but
this is by reason of the geometry of the graph, the percentage trans-
mission being low in this region. A print from this negative on
positive film, developed to regular print gamma (Fig. 17b), brings
out the shape of this denser region of the negative, since in this case
the transmission peak of the print corresponds to the trough of the
negative wave shape.

The basic cause of such an effect in the negative as that discussed
is that the image is formed within high gamma material developed to a

372 J. CRABTREE AND J. H. WADDELL [J. S. M. P. E.

low gamma. It is the checking of development during this active
stage that is the cause of the local distortions. If the image is de-
veloped to an approximate gamma infinity, as in a sound print, the
Eberhardt effect does not occur. Fig. 17 'c shows a similar exposure
to that producing Fig. 17a but developed in D16 to gamma 2.0.
Here both trough and peak are flat; that is, the wave is without
distortion other than that produced by image spread due to emulsion
turbidity.

It is apparent that the over-all effect of the Eberhardt phenomenon
in a wave-shape such as that shown will depend to a considerable
degree on dimensional considerations. With an emulsion layer 0.001
inch thick swelling in the developer to 4 mils, a section of a 1000-cycle
print recorded at normal speed would show the dark and light regions
as rectangles 18 mils long and 4 mils thick. Developer diffusion from
the surface downward would dominate, and the general result of the
Eberhardt effect would be relatively small. At 4000 cycles, an
approximately square cross-section would be seen where lateral
diffusion in the emulsion would dominate and the Eberhardt effect
be more pronounced, and the frequency characteristic of the record
thus distorted. The Eberhardt effect, together with the directional
effects above described, will doubtless explain certain observed
peculiarities of frequency characteristics of sound records derived
from negatives processed in different laboratories.

The Eberhardt effect will probably be less severe in the record
from a sine wave of exposure than the flat-topped wave described,
since the changes in density are gradual. Such exposures are now
being considered from this standpoint.

Since the Eberhardt effect is directly related to emulsion thickness,
its degree would probably be reduced by decreasing the thickness
of the sensitive layer. It is known that reducing the emulsion thick-
ness improves the resolving power of the film; therefore a decrease
in this dimension of the emulsion layer of standard positive film would
lead to improvement in the fidelity of the sound record.

The Eberhardt effect has been discussed rather fully to forestall
an impression that the adoption of the particular methods of de-
velopment described in this memorandum will solve all problems
associated with the development of the sound negative and print.

REFERENCE

1 CRABTREE, J.: "Directional Effects in Continuous Film Processing," /. Soc.
Mot. Pict. Eng., XVm (Feb., 1932), No. 2, p. 207.

Nov., 1933] DIRECTIONAL EFFECTS IN PROCESSING 373

DISCUSSION

MR. R. C. HUBBARD : What effect has the rate at which the film passes through
the bath on the directional effect?

MR. J. CRABTREE: I neglected to mention that. That was the reason why
the series of time-gamma curves was given in each illustration. There are three
gammas, representing three different machine speeds. Judging from these curves
there seems to be no definite relation between the degree of directional distortion
and the speed of film propulsion over the range covered.

MR. J. I. CRABTREE: I should think if a flapping motion were imparted to the
film by an increase of speed, enough agitation would be produced to improve the
condition.

Most machines at the present time are of the vertical type and it is a question
whether these principles apply to that type of machine. In the case where the
film travels slowly, vertically, we have found that it is necessary to agitate the
developer. We have had the most success by blowing nitrogen into the developer.
Nitrogen is relatively expensive, but we have had some success with air, depending
upon the type of developer. We have made quite an extensive study of the effect
of aerating various developers on their photographic properties. One interesting
observation is that with the developers containing hydroquinone, an exhausted
developer is revived considerably by aeration; due to the fact that caustic soda
is liberated in the chemical reaction. That is, the reaction between oxidized
developer, sodium sulfite, and air, gives probably hydroquinone sulfonate
plus caustic soda. In the reaction, therefore, the aeration offsets to some extent
the diminution in developing activity caused by the loss of developing agent
through exhaustion. These remarks apply only to developers that contain
hydroquinone, and do not apply to developers containing only metol.

MR. J. CRABTREE: In the text I have suggested that it might be possible so
to balance the composition of the developer and the degree of aeration as to main-
tain an approximately constant developing power. As regards the flapping, I
don't know. There is very little such motion. I think it would be dangerous
from a breakage standpoint.

MR. COFFMAN: Did you find at any time that an occasional upturn of the
shoulder, which is characteristic of the directional effect of which you speak,
also occurs when the motion is in the opposite direction?

MR. J. CRABTREE: Yes.

MR. COFFMAN: Occasionally we find a sensitometric strip that shows that
characteristic upturn, for no apparent reason whatever.

MR. J. CRABTREE : I have no explanation of that so far.

MR. COFFMAN: On the lib sensitometer we occasionally find that one side of
a strip will show an upturn whereas the other sides turn down; the two sides of
the strip, obviously, have gone through the bath in the same direction.

MR. J. CRABTREE: I am sure it is something in the mechanics of the thing
that we have not run down yet.

ANALYSIS OF SOUND QUALITY WITH THE VARIABLE

DENSITY RECORDING METHOD FROM

SENSITOMETRIC DATA*

R. SCHMIDT AND A. KUESTER**

Summary The manner in which quality controlling factors for reproduced sound
can be derived from the photographic characteristics of variable density sound records
is described. Density curves and transmission values serve to calculate and express
in quantitative terms: loudness, volume range, noise, and distortion, for a given
degree of electrical amplification. Normal and noiseless sound recording are dis-
cussed and described from the photographic point of view. The photographic process
is described in sensitometric terms, and it is shown how the tolerances within which
it has to be carried can be determined and controlled.

The quality of sound reproduction in motion picture theaters is
described only in more or less vague terms. It may be called faith-
ful, hollow, rough, harsh r etc., but the commonly used terms of criti-
cism do not refer to defects in proportional or quantitative figures.
It is desirable hi sound film technic to have means for giving numeri-
cal values which could characterize the quality or defects of repro-
duced sound.

Disregarding the electrical equipment which is of importance
when considering the quality of recording or reproducing, this paper
will try to describe the influence of the photographic process and to
analyze the elements which together control the quality of sound re-
production. The term quality can be resolved into the following
elements :

(1) Volume (maximum loudness).

(2) Range of volume.

(3) Frequency range.

(4) Non-linear distortions.

Important factors of the photographic process that control these
quality controlling elements are :

(1) Intensity and range of negative exposure.

(2) Sound negative film properties and development.

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** I. G. Farbenindustrie, Agfa Film Fabrik, Wolfen, Germany.
374

ANALYSIS OF SOUND QUALITY 375

(3) Positive exposure.

(4) Sound positive film properties and development.

Fig. 1 shows which of the four elements are influenced by these
photographic factors. It is the purpose of this paper:

(1) To determine quantitatively the influence of changes in the
photographic process on the four quality controlling elements.

(2) To establish the degree of tolerance hi the photographic proc-
ess hi order to direct and control it in a practical way.

Influence of Photographic Conditions on the Quality of Reproduction

Sound

Intensity
of negativ

Range
e exposure

Neg.
Film

Neg.
Dev.

Pos.

Exposure

Pos.

Film

Pos.
Dev.

Volume

Range of

Volume

Frequency

Range

Non-linear

Distortion

FIG. 1. Chart showing which of the four elements of sound quality are
influenced by the various photographic factors.

Although all the controlling elements refer to the ultimate acoustic
output, it will be shown that we can deduce the numerical values for
them from the sensitometric data of sound negative and positive
film, except for the frequency range. The latter can not be deter-
mined by merely sensitometric methods. It has been dealt with
previously 1 and will only be referred to in this paper.

The proper interpretation of the well-known photographic den-
sity and transmission curves will therefore serve to characterize the
quality of the reproduced sound.

I. CONTROLLING ELEMENTS WHICH CHARACTERIZE REPRODUCTION, AND
THEIR DEDUCTION FROM SENSITOMETRIC DATA

(1) Maximum tone volume (loudness). Loudness is a physiologi-
cal term and refers to an acoustic intensity. This intensity can be
measured; it is proportional to the square of the acoustic pressure.

376 R. SCHMIDT AND A. KUESTER [J. S. M. P. E.

A definition has been given in the European literature for a stand-
ard acoustic intensity as a physical unit. For loudness, however, it
has been found sufficient to characterize it by proportional figures:

/! />!* p,

L = 10 log - = 10 log = 20 log -
/o Pt* p t

The sensation of loudness on the human ear has been found to
vary with the frequency of the sound for the same physical intensity.
Fig. 2 shows curves of equal loudness plotted against frequency
and acoustic pressure. From the diagram it can be seen that:

1 ^ e at 1000 cycles has a loudness of 70 db.}

>Diff. 20 db.

0.1 r at 1000 cycles has a loudness of 50 db.j

1 -j^f at 125 cycles has a loudness of 60 db. 1

VDiff. 30 db.
0.1 ^r at 125 cycles has a loudness of 30 db. j

These figures show that the loudness at low frequencies decreases
more rapidly than the loudness at higher frequencies when the total
volume of reproduced sound is lowered.

The volume obtained with a definite amplification depends on the
transmission range of the sound record. It can be considered an
ideal record if the minimum density be 0, the maximum transmission
1, and the maximum density , minimum transmission 0. Such a
recording has a transmission range of 1, and would give the greatest
possible loudness.

If the volume control of the amplifier is set so that the sound
record with AF =1.0 gives a loudness of 70 decibels, it is possible to
calculate the loudness of each sound record of 1000 cycles for lower
transmissions. A transmission of 0.45 compared with a transmission
of 1.0 gives a loudness which is

20 X log = 7 db. lower
0.45

(2) Volume range and noise. It has been shown that the maxi-
mum loudness of the sound record with a definite amplification is
controlled by the transmission range A7\ The minimum loudness
obtainable with this same record without change of volume control

Nov., 1933]

ANALYSIS OF SOUND QUALITY

on the amplifier is limited by the ground noise. Recordings below
the noise level will be masked by the noise. The lowest audible
sound above the noise level depends on various factors, and also on
individuals. For practical purposes the volume range of reproduced
sound can be given as the difference between the maximum loudness
and the noise level.

-3
10

-4
10

FIG.

62 125 25O 500 WOO 2000 4OOO 9OOO

*" Cycles

2. Curves of equal loudness plotted against
frequency and acoustic pressure.

The desirable volume range for the reproduction is about 70 db.
In order to increase the range of a sound record, it is necessary to in-
crease the maximum loudness or to reduce the noise. The latter has
been done successfully, and will be explained from the photographic
point of view later in this paper.

Noise caused by the sound record must be attributed to fluctua-
tions in density or transmission, which may be caused by mechanical
or optical unevenness of the sound record, such as dust, scratches,
etc. To a certain extent the graininess and physical properties of the
emulsion will cause a similar effect. The variation in density which
causes the noise is about == 0.01 from the unmodulated.

The determination of the volume range from sensitometric data
can be done as follows :

Max. dens, of sound record, d m **.. = 1.25; Min. transm. /min. =0.056
Min. dens, of sound record, dadn. = 0.18; Max. transm. 2 ma x. =0.661

Transmission range AT =0.605

378 R. SCHMIDT AND A. KUESTER [J. S. M. P. E.

Unmodulated density of sound record du = 0.45

Max. dens, of noise level dtim**. = 0.46; Min. transm. /Wmin. =0.347
Min. dens, of noise level dnmin- =0.44; Max. transm. /nux. =0.361

Transmission range of noise record ATn =0.014

0.605

Volume range L = 20 log = 33 db.

0.014

(3) Frequency range. This is one of the controlling factors of the
quality of reproduction. As it can not be dealt with in sensitometric
data, it will not be described in this paper.

(4) Non-linear distortions. It is essential that a faithful sound
reproduction contain the same frequencies as the original sound. If
the reproduced sound contains additional frequencies, these must be
attributed to non-linear distortion, which is caused by non-linear re-
sponse of some part within the sound recording or reproduction proc-
ess. In this case, a sine wave on the recording end would be repro-
duced as a different and distorted wave on the reproducing end.
As a rule, it is possible to analyze such a distorted wave and determine
its components (Fig. 3). A wave of 250 cycles may thus be com-
posed of a group of sine waves of 250, 500, 750, 1000 cycles, etc.

FIG. 3. A 250-cycle wave ana-
lyzed into sine waves off requency
250 and 500.

It depends on the character of the response of the non-linear parts
of the system, as to which of these higher frequencies will be repro-
duced. The proportion of the amplitudes of the higher frequencies

Nov., 1933]

379

and the original frequencies is being used to give quantitative figures
for non-linear distortion.

Per cent distortion =

Non-linear distortion will cause harshness, and a variation of the
timbre. A large percentage of non-linear distortion of a single fre-
quency will create only harmonic overtones which do not influence
the quality. In practice, however, when various frequencies are
recorded and reproduced, it is possible that the overtones create an
inharmonic relation which may cause objectionable effects.

The variable factors of the photographic process are: exposure
and development of the sound negative, printing light, development
of the sound positive, control of the transmission of the sound track,
and the modulations of the light from the exciting lamp. From the
transmission curve non-linear distortion can be detected.

$0.6

0.4

I ' 2

u 2 4- 6 Q 1O

> Ne <j a tii/e Expo sure

FIG. 4. Light intensity of recording lamp vs. light energy
transmitted by sound track.

Fig. 4 shows the light intensity from the recording lamp which ex-
poses the film, plotted against the light energy transmitted by the
sound track, falling on the photoelectric cell. If this curve is a
straight line, non-linear distortion will be avoided; if it is not a
straight line, it is possible to draw the curve for the distorted fre-
quency, as shown hi Fig. 5.

380

R. SCHMIDT AND A. KUESTER

[J. S. M. P. E.

FIG. 5. The sine wave, below, projected to the non-
linear transmission curve, produces the distorted sine
wave at the left.

Fig. 5 shows a sine wave, a non-linear transmission curve for the
sound track, and the resulting distorted wave of the sound reproduc-
tion. This distorted curve can be analyzed and resolved into higher
frequencies by calculation or geometric construction (Fourier's
Analysis or various types of analyzers). Fig. 6 shows Mader's
analyzer.

FIG. 6. Mader's wave analyzer.

Nov., 1933]

ANALYSIS OF SOUND QUALITY

381

II. THE PHOTOGRAPHIC PROCESS

There are some factors affecting the quality of sound reproduction
which depend solely on the quality of the recording material; for in-
stance, resolving power or color sensitivity. In this paper we are
interested only in those factors which we can vary and which can be
adjusted during the photographic process. Those are:

(1) Density curve and unmodulated density of the sound negative.

(2) Density curve and unmodulated density of the sound positive.

(3) Density curve of the sound record. ) The over-all photographic charac-

(4) Transmission curve. ) teristics.

Fig. 7 shows the curve of the sound negative film, N; the curve of the
sound positive film, P; the curve of the sound record, R; and the
transmission curve of the sound record.

1.5

d A i.o-

0.5-

1.5

J><

T P

FIG. 7.

Curves of density of negative, N, positive, P, and over-all, R; and
transmission curve of sound record, T.

Only those parts of the curves which are actually used for the re-
cording are shown in the diagram. The transmission curve gives di-
rectly the data which are required to determine the various factors
which control the sound reproduction and which have been described
in the previous paragraph. The curves N, P, and R refer to the re-
cording, developing, and printing processes. It will be shown later
how the tolerances for these processes can be determined from those
curves.

The control of the photographic process in preparing photographic
sound records is done by sensitometric means. It is necessary that
development and exposure of the sensitometric strips correspond as
closely as possible to practical conditions. As the exposure for the
sound negative is about 1 /26,ooo second and that of the sound positive
about 1 /3o second, the corresponding sensitometric strips should be
given a similar time of exposure. The curve of the sound record
R can be obtained by direct printing of the sound negative sensito-

382

R. SCHMIDT AND A. KUESTER

[J. S. M. P. E.

metric strip or it can be derived from the positive sensitometric strip
P. It is necessary to analyze and examine only those parts of the
curves which are actually used in recording and reproduction. For
this purpose the exposure range of the recording light must be known
and the working conditions must be adjusted accordingly. If the
modulations are not linearly related to the fluctuations of the elec-
trical current from the microphone, the over-all transmission expo-
sure curve must be brought into a non-linear shape, so as to produce
an over-all linear result. This is shown in Fig. 8.

A B

FIG. 8. A: non-linear response; B: correction of
non-linear response by means of a non-linear transmission
curve.

In order to determine the tolerances within which the photographic
process should be carried out, the unmodulated density or the density
curve of the sound negative may be varied between limits which
are wider than would be met in the commercial laboratories. By
means of the transmission-exposure relationship, volume range and
distortion may be traced and compared. It can then be decided
whether or not they are tolerable for faithful reproduction. It can
also be easily understood that the straight-line portion of the sensito-
metric curve is the best suited and allows the widest tolerance.

In order to have the widest possible volume range, it has been
found necessary, and also possible, to lower the noise level of the
sound reproduction. Noise becomes more noticeable and disturbing
as the sound volume decreases. It is less objectionable at a high
volume. Therefore, it is not necessary to eliminate noise completely,
but only to reduce it in proportion to the volume. Photographic and

Nov., 1933]

ANALYSIS OF SOUND QUALITY

383

sensitometric data can serve to explain in what way and to what ex-
tent this may be achieved.

It has previously been shown that the ground noise is produced by
a variation in density of == 0.01 from the unmodulated. The varia-
tion in transmission for the noise record and the loudness of the noise,
therefore, decreases if the unmodulated density increases. Fig. 9
shows the loudness of the noise, plotted against unmodulated densi-
ties, with the amplifiers set to give a loudness of 70 db. for a trans-
mission of 100 per cent.

70'
60-
50-

40-
30
20\
10

Loudness in c/8

<J unmodul

1234 56789 10

FIG. 9. Decrease of noise with increasing unmodulated density;
T = 1, to give 70 db.

At an unmodulated density of 0.3 the noise will be recorded with
a density from 0.29 to 0.31. This corresponds to a transmission
range of 0.02, the variation between 0.51 and 0.49. If the unmodu-
lated density is increased to 1.3, the noise recording will vary between
the densities 1.29 and 1.31. The corresponding variation in trans-
mission is then 0.002, between 0.051 and 0.049. In this case an in-
crease of the unmodulated density from 0.3 to 1.3 will reduce the
variation in transmission for the noise record to 1 /w its original value,
which means that the noise level is reduced 20 log 10/1 = 20 db.

The loudness or amplitude of the sound record should not be varied
with noiseless recording. It is, therefore, necessary that the trans-
mission range for recorded sound be kept constant, while the un-
modulated density is varied.

384

R. SCHMIDT AND A. KUESTER

[J. S. M. P. E.

Relation between
Recording amplitude
Reproduced Volume
and Noise.

Loud ness in db

\ Jmi

I t

'nmod

Jma*

FIG. 10. Conditions for normal recording; the unmodulated exposure
and noise being constant.

Noiseless Recording.

Relation between
Recording amplitude
Reproduced Volume
and Noise.

o 10 to 30 40 so to 70
* Loud ness in db

10 II

FIG. 11. Conditions for noiseless recording; the unmodulated exposure
and noise decreasing with loudness. The effective volume of reproduction is
greater than that in Fig. 10.

Nov., 1933]

ANALYSIS OF SOUND QUALITY

385

0.8[-
T p 0.6

0.4

0.2

0.6 1-0 1.5 2.0

FIG. 12. Satisfactory transmis-
sion curves for normal recording : a
gives more distortion in noiseless re-
cording than c.

0.5 1.0

> Amplitude

FIG. 13. Curve showing increas-
ing distortion at higher amplitudes:
a, for curve a in Fig. 12, with nor-
mal recording; b, for curve a in Fig.
12 with noiseless recording.

Fig. 10 shows the conditions for normal and Fig. 11 the conditions
for noiseless recording. In Fig. -10, the unmodulated exposure is kept
constant; in Fig. 11, it decreases with decreasing sound amplitude,
in such a way that the minimum exposure remains constant when the

Exposure Range Jo:l
gamma ne( j = 0.60
gamma p OS = 2.00

fl)

d un

Pos.

(2)

0.8

0.6
0.4
0.2

075 TO TS 2.0

16 +'28=44

Noiseless Record

28 \ Normal Record

10 2O 3O 40 50 60 70

>. db

(3)

% Distortion

1 to

Norm
\ f

x v '

''Noiseless

(4)

0.5

FIG. 14. Chart for analyzing photographic and sensitometric effects on
character of reproduced sound.

386 R. SCHMIDT AND A. KUESTER [j. s. M. P. E.

amplitude or level varies. This is done by electrical means. It can
also be seen that the noise level in Fig. 10 is constant, whereas in Fig.
11 it decreases when the loudness decreases. The effective volume
range in reproduction is, therefore, greater in Fig. 1 1 than in Fig. 10.

The photographic process can be carried out in the same way for
noiseless recording as for normal recording, if the maximum loudness
is the same and the transmission curve is sufficiently linear. If, how-
ever, the transmission curve is not straight in its portion correspond-
ing to low transmission values, recordings with low amplitudes will
be greatly distorted with the noiseless recording method.

In Fig. 12 are shown two transmission curves which will give satis-
factory results with normal recording. The usual volume range can
be kept within the straight-line portion; distortion will therefore be
very little, and noticeable only at unusually high sound levels when
the curved part must be included for recording.

In Fig. 13, curve a shows the increase in distortion with higher
amplitudes for curve a in Fig. 12 when normal recording is being used.
If the same transmission curve is being used for noiseless recording,
the sound record of low and medium amplitudes uses the curved por-
tion of a in Fig. 12 ; only high amplitudes utilize the more linear por-
tion. In this case the distortion values are very high for low ampli-
tudes and decrease at higher amplitudes. . Noiseless recording, there-
fore, will in this case show more distortion than normal recording
(Curve b, Fig. 13).

If it is not possible to adjust the photographic process so as to
give a straight-line transmission curve, it is necessary for noiseless
recording to have it straight for low transmission values. It is less
objectionable when the portion having high transmission values is
not straight, as this will cause increased distortion only at the higher
levels.

HI. ADJUSTMENT AND CONTROL OF THE PHOTOGRAPHIC PROCESS

For an analysis of the quality of sound reproduction from sensito-
metric data it has been found necessary to determine :

(1) The relation of the photographic response to the recording
light: An average frequency must be recorded with low amplitude
and developed together with a sensitometric strip similarly exposed.
The maximum and minimum density of the wave record must be
read ; then the exposure corresponding to those densities can be found
from the sensitometric strip, the exposure of which is known. This

Nov., 1933]

ANALYSIS OF SOUND QUALITY

387

Expost.
gomn
gamn
d unn
d unn

1.5- 1.5-
Overall ,

k 1 ' "X ./% 1 '""

re Ranye 10:1
10. ney. = O.60
ta pos. = 2.O o
iod n =0.60
lodpos 0.50
(1)

as-
Pos. ,
/ P 06 ~

A t-Lii

14 + 28=42 ,

O 05 10 t's 3.0
*-I<jJ > J_

% Disforf/on
ord.

A 10-
. ^^. Norm

V^^, < Noiseless

! 28
1 ' Normal Recor

. . 1 1 , .

O 20 4O 60 30 /-,\

>- d b

, . 0.5 10 l.S
(4) ^.p

Overall

(a)

13+27 =

-I Noiseless Record
-\ Normal /?<

FIG. 15 (above). Chart for analyzing photographic and sensitometric
effects on character of reproduced sound; the negative exposure has been
reduced 40 per cent from the value given in Fig. 14.

FIG. 16 (below). Chart for analyzing photographic and sensitometric
effects on character of reproduced sound; the positive exposure has been
reduced 40 per cent from the value given in Fig. 15.

388 R. SCHMIDT AND A. KUESTER [J. S. M. P. E.

must be done repeatedly, increasing the amplitude to its limits. The
values obtained for exposure must be plotted against the electrical
current used for modulation. From the curves thus obtained the
straight-line portion only should be used.

(2) The maximum exposure to which the photographic process is
to be adjusted.

(3) The empirical adjustment of the photographic process:
From a straight-line transmission curve with a chosen range, the corre-
sponding density curve is drawn. From this curve, the sound positive
and sound negative curves can be determined. The degree of distor-
tion corresponding to various amplitudes of the signal and the trans-
mission range of the recording can be analyzed with the use of those
curves. If the constructed curves do not fall within the chosen trans-
mission range, this range should be decreased until the constructed
transmission curve matches closely enough the chosen curve.

If in recording, the exposure of the film is not in linear proportion
to the electrical current, it is necessary to determine a transmission
curve corresponding to this non-linear relation, as shown in Fig. 8.

(4) The tolerance within the photographic process : the condi-
tions are varied within certain limits and the quality controlling fac-
tors analyzed by means of the transmission curve.

(5) The standardization of sensitometric characteristics for prac-
tical use : if the most satisfactory conditions under which the photo-
graphic process must be carried out have been determined, they can
easily be controlled, as is common in laboratory practice, by (a) gamma
and unmodulated density of the sound negative; (&) gamma and
unmodulated density of the sound positive; (c) the development
characteristics of the positive film.

In practice, when applied to a definite recording process and a
standardized film quality, the determination of the necessary data is
far more simple than would be assumed from the above description.

As an example, it will be shown how definite information about
the photographic conditions and sensitometric data for characteriza-
tion of sound reproduction quality can be obtained from Figs. 14,
15, and 16, each of which contains:

(1) Exposure range.

Gamma of the sound negative and sound positive.

Unmodulated density of the sound negative and the sound positive.

(2) Density curves of sound negative film, positive film, of the printed sound
record and transmission curve of the sound record.

Nov., 1933] ANALYSIS OF SOUND QUALITY 389

(3) Maximum loudness and volume range for normal and noiseless recording.

(4) Proportional values for non-linear distortion at various modulations with
normal and noiseless recording.

In preparing charts of this kind for certain definite variations, it
can easily be decided whether or not they are tolerable for the process.
The conditions shown in Fig. 14 have been varied. In Fig. 15 the
negative exposure has been reduced 40 per cent, and in Fig. 16 the
positive exposure has been reduced 40 per cent. The charts show that
the reduction of the negative exposure has little influence under the
conditions characterized in Fig. 14. The reduction of the positive
exposure under the same conditions, however, causes considerably
more distortion, as shown in Fig. 16.

REFERENCE

1 SCHMIDT, R., AND KUESTER, A.: "Untersuchungen iiber die Aufzeichnungsg-
giite bei Tonaufnahmen (Study of High-Quality Recording of Photographic
Sound Records)," II, Agfa Veroeffentlichungen (Agfa Publications), p. 83.

THE APERTURE ALIGNMENT EFFECT*
ELLSWORTH D. COOK**

Summary. The use of wide frequency-range equipment has forced the designer
to consider many of the possible sources of distortion in order to determine limits
for their effects. The work described in this paper was performed to ascertain the
improvement provided by the new Bilateral sound track as well as to set allowable
tolerances for commercial equipments. It is shown that the amount of distortion
due to aperture alignment can be reduced readily to negligible amounts, particularly
when the Bilateral Sound Track is used, since here both sides of the sound wave vary,
thereby reducing the amplitude required for full modulation. The analysis has been
limited to variable amplitude film and is restricted to a sine wave.

The improvements introduced in recording and reproducing
processes in the past few years have made it a trifle difficult to
appreciate the number and importance of the individual steps by
which the progress has been made. Several rather recent public
demonstrations have shown that the advances have been material.
It is probable that very few additions to our fund of general knowledge
have been made which were not at least dimly suspected at an earlier
time; but the understanding of the problems in this field, together
with their importance and solutions, is on a much firmer foundation
today than was heretofore the case. Admittedly, many of the
improvements have been minor in themselves; but taken in the
aggregate, the contribution made by them has not been incon-
siderable. It has been necessary to investigate critically various
possible sources of distortion to determine their importance and to
ascertain a superior limit for their magnitudes.

When the Bilateral track 1 was commercialized, it was desired to
determine the amount of improvement obtained for those cases
where the recording and reproducing apertures did not fall at the
correct angle to the direction of motion of the film. The problem
of aperture alignment was recognized by several investigators rather
early in the history of film recording. Perhaps one of the first authors
to publish the results of his work was N. R. Stryker, 2 whose treat-

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** RCA Victor Co., Camden, N. J.

390

APERTURE ALIGNMENT EFFECT

391

ment was limited to variable density film. That part of the problem
will not be considered here. In variable amplitude film, the problem
becomes more difficult, as far as analytical treatment is concerned.
In fact, in this work, it has not been found possible to solve the
problem for a sine wave by a purely analytical treatment and the
general case of a complex wave has not been treated as yet.

Under the conditions of this analysis, it will be shown that mis-
alignment of the aperture needs cause little distortion in the steady-
state case except at the more extreme high frequencies; and, if
commercially realizable tolerances be met, this source of distortion
need not be serious in the case of a sound track of the Unilateral type,
whereas if a sound track of the Bilateral type be used, it may be
reduced to an altogether negligible quantity. In order to evaluate
the amount of distortion produced, when the aperture does not fall

FIG. 1. Unilateral sound track, completely modulated.

at the proper angle to the direction of film travel, certain idealizing
assumptions will be made.

The reproducer case will be considered under the assumptions
(1) that the film moves with a constant velocity V; (2) that a per-
fect sine wave had been recorded on the film in such a manner that
wherever light fell on the record it was uniformly and completely
exposed with perfect resolution ; (3) that the aperture was uniformly
illuminated and infinitely fine. The importance of this latter
consideration has been treated elsewhere, 3 ' 4 and hence this analysis

392 ELLSWORTH D. COOK [j. s. M. P. E.

may be simplified by omitting any consideration of it here. The
same analysis may be used when considering the recording problem.

The analysis was not carried beyond the point at which the aper-
ture became tangent to the recorded wave-front for two reasons:
first, that it is seldom necessary to consider frequencies higher than
is thus permitted, and second, that the labor involved in the
calculation is not an inconsiderable item.

If a sine wave of unit amplitude, and frequency w/2ir be recorded
as a Unilateral track modulated 100 per cent, the result is shown in
Fig. 1.

In that figure, linear distances along the abscissa are converted
to angles as follows :

(S)

The signal voltage developed by an ideal photoelectric cell receiving
light from the aperture, will be proportional to the length 5 uncovered
by the film record. It will be evident that the numerical work can
be confined to the width Y with advantage. Consideration of Fig. 2
will show that the numerical analysis can be confined to Fig. 1 if
several properly chosen values of the angle a be considered. The
results can be extended later to other percentages of modulation
for both the Unilateral and Bilateral sound tracks. Thus :

Y = A jl + S i n ??

in which

u = Y tan
Hence:

S = Y sec a = A sec a \ 1 + sin (* + Y tan a)

\ 1 + sin ^ (* + Y tan a) [ (1)

( ^ )

Attempts to segregate the variables analytically in equation (1)
have led to an expression in which the coefficients involved a series
of Bessel functions of the ordinate y and a trigonometric function
of the angle of aperture rotation a, but so far have failed to yield
a simple Fourier Series. Recourse was, therefore, had to the labori-
ous mechanical method of formulating the Fourier Series. This is
regrettable since the method of treating a complex wave is not revealed.

In the representative wave of Fig. 1, the width Y was calculated
at various points along the wave and resolved into a Fourier Series

Nov., 1933]

APERTURE ALIGNMENT EFFECT

393

for each of a sufficient number of values of the ratio r/\ which, in
itself, is determined by the angle a. The actual length of aperture
5 in any practical case is obviously a function of the actual width Y
and the secant of the actual angle a of aperture rotation. The
difference between S and Y in any practical case is quite negligible.
The final result for 100 per cent modulation is:

S = A sec

+ 0i sin ( 2?r - + 0"i ) + (h sin ( 4ir - + o-j I

o 3 sin ( 6*- - +
\ X

(2)

The extension of this result to any degree of modulation may be
accomplished in an obvious manner by considering the sound track
to be made up of three parts: an exposed region of constant width,
a 100 per cent modulated region, and a clear or unexposed region,

FIG. 2. Unilateral sound track.

each of which run the length of the record. In the consideration of
the Unilateral sound track, the total width will be assumed to be
2G. This will be made clear by reference to Fig. 2. It will be seen
that for the case of 100 per cent modulation, A = G.
By definition, let

where

- - 2 - (0

394

ELLSWORTH D. COOK

[J. S. M. P. E.

From equation 2, Y can be written as :

sin (u + <p + <n) + a* sin (2{ + <f>] +

[~Oo + 0i sin (w + >
L-|- dj sin (3 { co ~j~ <p\

Hence:

^m = [(G - A) + A {ao + ai sin (w + <f> + *i) + }]

If it be assumed that the percentage of modulation is m,
then (m) = A/G:

y - G[{(\ - w) + oowj + m \ai sin (w + <p + <n) + }] (4)

It will be found convenient to work on a percentage basis, in
which case S' m may be written :

S m sec a [{(1 m) + om} + w{ai sin (w + * + ffi) + ....}] (5)

FIG. 3. Effect of rotation of reproducer aperture on fundamental and
generated harmonics, for Unilateral sound track; a = coefficient of average
transmission; a\ = coefficient of fundamental; a^ = coefficient of second
harmonic; c = coefficient of third harmonic; r = Rm; m = percentage
of modulation.

Equation (5) can be looked upon as an expression for the length
of a stationary aperture, assumed set at an angle a to its normal
position, which is cut off by a wave of percentage modulation m
moving past it in a negative direction. This point of view, which
describes the way in which the sound record really operates, will be
found useful in work on the Bilateral sound track.

The diagrammatic representation shown in Fig. 4 is for a Bilateral
track with a modulation of m per cent. Sound tracks of that type
have several marked advantages over the Unilateral track shown
in Fig. 2. In addition to other improvements attained, it will be

Nov., 1933]

APERTURE ALIGNMENT EFFECT

395

found that many ill effects due to aperture rotation are materially
reduced by the use of such a sound track.

Consideration of Fig. 4 will show that S m is made up of the sum
of the separate intersections of two waves, each started with a
phase displacement of 180 degrees and traveling in opposite direc-
tions past a stationary aperture set at an angle a to its normal
position. Although it may not be so evident, that is also true of
the effective width ^ m of the aperture. Thus :

FIG. 4. Bilateral sound track.

It will be found desirable to express equation (4) in the following
form:

sin a + &2 sin 2co ) ~

... .

{(1 - m) +

+ m

_+ m{ci cos w + C2 cos 2w + ....}

Therefore

2{(1 -m)

(bi sin w + 61 sin (180 - co) )
m \ + ....+.... \

(6)

(7)

_+ m {ci cos w + c\ cos (180 w) + . . . }
Expanding the terms of equation (7) yields

( bi sin <o + Ci cos 2w
) + Cow} + w ( + b t sin 3w + c 4 cos 4co

(8)

396

ELLSWORTH D. COOK

[J. S. M. P. E.

Therefore, expressing S m on a percentage basis:

j 61 sin w + Ct cos 2w
sin 3w + +

S m

[{(1 - m)

om}

The numerical values of these coefficients for a sine wave record
under the assumptions prescribed above, are given in the curves of
Figs. 3 and 5.

At the risk of repetition, it is again pointed out that the means of
treating complex waves is not revealed from this work. The data
given by these curves do, however, allow the calculation of the fre-
quency characteristic for the fundamental, the generated second and

FIG. 5. Effect of rotation of reproducer aperture on fundamental and
generated harmonics, for Bilateral sound track; a = coefficient for average
transmission; bi = coefficient for fundamental; Ct = coefficient for second
harmonic; fe = coefficient for third harmonic; r = Rm; m = percentage
of modulation.

third harmonics, and the variation of the average transmission for
the reproducer for sound tracks of either type. In passing, it may
be remarked that in so far as this problem is concerned, it is fortunate
that the variation in average transmission does not become significant
within the commercially realizable frequency range with commer-
cially obtainable tolerances for an angle of rotation a. It might be
mentioned that variations of this factor due to other causes were
shown hi an earlier paper, 3 although only a superior limit to the
magnitude of the change was considered. Commercial results on
wide-range equipment have shown that as with many other by-
products of distortion, the effect can be controlled and made in-
conspicuous.

Nov., 1933 J

APERTURE ALIGNMENT EFFECT

Fundamental frequency

FIG. 6. Reproducer frequency characteristic, for Unilateral sound track:
infinitely fine reproducer aperture; film velocity, 90 ft. per min.; track
width, 0.070 in.; aperture rotation, 0.5 degree (solid curves), 0.25 degree
(broken curves) ; modulation 100%.

In Fig. 6 the frequency characteristic of a reproducer (or with
corresponding conditions, of a recorder) using a Unilateral track is
.shown together with the effect on the generated harmonics up to the
third order. In interpreting these curves, it should be noted that
the generated nth harmonic has a frequency n times that given by
the abscissa. The assumptions made, in addition to those already
mentioned, were a 0.070-inch sound track completely modulated,
an aperture rotation of 1 /2 degree, and a film speed of 90 feet per
minute. The dotted curves show the same conditions for an aper-
ture rotation of */4 degree.

Present-day commercial equipment, it should be noted, has a
tolerance of Vs degree for aperture rotation. Therefore, the rather

Fundamental frequency

FIG. 7. Reproducer frequency characteristic for Bilateral sound track:
same conditions as for Fig. 6.

398

ELLSWORTH D. COOK

[J. S. M. P. E.

FIG. 8. Reproducer wave-shape for Unilateral sound track: infinitely fine
reproducer aperture; modulation 100%; (a), r/\ = 0.049; (b), = 0.101.

* * I I /? 5 t * 8 *
FIG. 9. Same as Fig. 8: curve (a), r/\ = 0.160; curve (&), = 0.233.

* I I / I I

FIG. 10. Same as Fig. 8: curve (a), r/\ = 0.331; curve (b), = 0.481 (be-
yond tangency).

Nov., 1933] APERTURE ALIGNMENT EFFECT 399

great harmonic distortions shown in this figure under the assump-
tions for which it was calculated, do not exist in the commercial
equipment using the Unilateral sound track unless it be out of adjust-
ment. If such distortion were to exist, the reproduction would be
quite intolerable. Inspection of a protractor will show how simple
it would be to set an aperture to less than the angle for which the
curves were calculated. The figure, however, indicates what may
be expected if one be quite careless in aligning the aperture. By
comparison with the results for the new sound track, however, the
improvements made in this detail may be noted.

In Fig. 7, results for the Bilateral track under similar conditions
are shown. The considerable improvement is evident. The essen-
tial physical differences between the two tracks are that, whereas
for the results shown in Fig. 6 for the Unilateral sound track, the
aperture with l /% degree of rotation fell parallel to the wave-front
at approximately 10,000 cycles, such is not the case for the same
aperture rotation in Fig. 7 for the Bilateral sound track; since here
both edges of the effective sound track area vary, and consequently
their individual amplitudes are reduced, assuming the same total
track width and percentage of modulation.

Figs. 8 to 10, inclusive, show how, for the Unilateral track, the re-
producer wave-shape (i. e., variation in illumination passing to the
photoelectric cell through an infinitely fine reproducer aperture),
changes as the aperture rotation is increased. The curves are plotted
for various values of the factor r/\ under the assumptions mentioned
above. Fig. lOb is of special interest in that it is for the condition
of the resultant wave beyond the tangent condition. The tendency
toward a saw-toothed wave-shape for the Unilateral track is definitely
shown. The problems introduced by such wave-shapes falling on
the photoelectric cell are not essentially part of this investigation,
but their existence is noted.

Figs. 11 and 12 show somewhat similar conditions for the Bilateral
track, except that here the wave-shape for Fig. lib corresponds to
a much greater angle of aperture rotation than was the case for
Fig. 10a for the unilateral track, since both these figures are for the
condition of the aperture parallel to the recorded wave-front. It
will be noticed that the wave-shape for the bilateral track differs
radically from that of the Unilateral track. It is interesting to note
that it is difficult to judge relative amounts of distortion by inspec-
tion of wave-shapes.

400

ELLSWORTH D. COOK

[J. S. M. P. E.

From the details of the analytical treatment, one is struck at once
by the close apparent relation of this problem to that of reproducer
speed variation, which is so serious because of the introduction of
spurious new, and in general non-harmonic, frequencies not existent
in the original record. Careful distinctions must be made here since
in treating the problem of aperture alignment from this apparently

* * i *^ I * * $ 4

FIG. 11. Reproducer wave-shape for bilateral sound track: infinitely
fine reproducer aperture; modulation 100%; curve (a), r/X = 0.098;
curve (b), = 0.202; curve (c), = 0.320.

* 5 \ p \
FIG. 12. Same as Fig. 11: curve (a), r/X

11*1

0.466; curve (b), = 0.662.)

analogous point of view, the "speed" variation, being dependent on
the distance u or its equivalent angle 7 = 2ir(u/\) is related in
frequency and extent to the recorded frequency, with the result that
the "new" frequencies are now necessarily multiples of the recorded
frequencies, which is not necessarily true in the problem of true film
speed variation. For the Unilateral track, a fair first approximation

Nov., 1933]

APERTURE ALIGNMENT EFFECT

401

might be obtained by assuming that the ''speed" variation in the
record movement produces a displacement having a sine wave of
variation about an average value of phase displacement of
2ir(G tan a)/X, as denned in Fig. 2. The amplitude of the variation
would be dependent on the degree of modulation as well as on the
aperture rotation, while the "frequency" would depend upon the re-

t * a

ft I

FIG. 13. Required phase variation to produce a reproducer wave-shape
similar to that of Fig. 12 by variation of film velocity (y max . = 118.52 de-
grees).

FIG. 14. Wave-shape for recorder using Unilateral sound track with aperture
rotated a degrees.

corded frequency. Fig. 13 shows the equivalent phase variation y
for a 100 per cent modulated Unilateral track as a function of equiva-
lent time under the condition r/X = 0.331; that is, with the aperture
tangent to the recorded wave-front. In the case of the Bilateral
track, the discussion is more complicated.

402 ELLSWORTH D. COOK

If the recorder aperture has been rotated, the distortion may be
corrected by properly aligning the reproducer aperture. Of course,
no such procedure is recommended, but the fact remains that the
final distortion from the reproducer due to this effect is traceable to
the geometrical difference between the angles of rotation of the re-
corder and the reproducer apertures.

The preceding analysis, therefore, should be applicable to recording
as well. From Fig. 2 for the reproducer, where the recording was a
pure sine wave,

Y m = G ~1 + m sin (2 + y tan )

Assuming in the recording process, that the length of aperture 5
(or light beam) varies as a pure sine wave, the conditions are as shown
in Fig. 14.

In this case :

Y m = G 1 + m cos a sin (z y tan a)

Here it is supposed that m cos a is a new percentage of modulation,
and that z is the independent variable. In extending the Unilateral
recorder track analysis to include the case of the Bilateral recorder
track, the same procedure used in connection with the reproducer in
Fig. 4 is required. The unilateral wave is expressed as a sine and
cosine series in the variable co and the resultant wave is considered
as the sum of two traveling waves moving in opposite directions
past a stationary point where the variable of the first wave is w
and the variable (180 o>) is substituted for co to obtain the second
traveling wave.

In conclusion, the helpful criticisms of Mr. E. W. Kellogg are
gratefully acknowledged.

REFERENCES

1 DIMMICK, G. L., AND BELAR, H.: "Extension of the Frequency Range of
Film Recording and Reproduction," J. Soc. Mot. Pict. Eng., XIX (Nov., 1932),
No. 5, p. 401.

1 STRYKER, N. R.: "Scanning Losses in Reproduction," /. Soc. Mot. Pict. Eng.,
XV (Nov., 1930), No. 5, p. 610.

COOK, E. D.: "The Aperture Effect," J. Soc. Mot. Pict. Eng., XTV (June,
1930), No. 6, p. 650.

4 FOSTER, D.: "The Effect of Exposure and Development on the Quality of
Variable Width Photographic Sound Recording," /. Soc. Mot. Pict. Eng., XVII
( Nov., 1931), No. 5, p. 749.

THE MORGANA COLOR PROCESS*
J. A. DUBRAY**

Summary.- The B & H "Morgana" process is an additive color process. Each
successive picture frame is analytically photographed through a red and a blue-green
filter alternately. The conventional color filter wheel has been replaced by an oscillat-
ing element that brings the proper filter into position between the lens and the film at
each exposure. Regular panchromatic reversal film is used. The normal photo-
graphic speed is 24 picture frames per second though other speeds can be used.

During projection two successive frames move forward and one backward, or in
reverse, in the following order: 12; 1-2-3; 2-3-4; 3-4-5; etc. The result is that,
although the film is running at a linear speed of 24 frames (! 2 / 3 feet) per second, 72
frames are alternating at the aperture during the same length of time, each picture
frame being projected three times on -the screen. This accrued projection speed
eliminates color flicker and greatly reduces color fringing. A conventional filter
wheel rotating before the projection lens at a speed of 2160 rpm. synthetically pro-
duces the impression of color during projection. Photographic filters are now avail-
able for panchromatic reversal 16-mm. film for daylight or tungsten filament in-
candescent bulbs, selected for proper analysis of color during the photographic process
in accordance with the light radiation characteristics of the source of light used.

"Morgana" is the trade-name of an additive 16-mm. color process
recently announced by the Bell & Howell Company. Its basic prin-
ciples are well known, 1 and have been applied more or less success-
fully to the cinematic reproduction of colors practically since the
beginning of motion pictures. The process is fundamentally based
on analyzing the object photographically into two records of wide and
complementary spectral bands. One record (picture frame) selects
the so-called "warm" colors, the yellows and the reds; and the other,
the "cold" colors, the violet and green. This selection is obviously
attained by using colored screens or filters, which absorb the un-
wanted colors for each picture frame, and transmit to the film those
that are selected. The film must necessarily be sensitive to all the
light radiations of the visible spectrum.

The result in the picture print (or film finished by the reversal
process) is that the multitude of densities that concur to form the

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Bell & Howell Co., Chicago, 111.

403

404 J. A. DUBRAY [J. S. M. P. E.

image and that correspond, for each successive picture frame, to the
particular radiations transmitted by the selective filters, are in-
versely proportional to the brilliance of the hues that concur to make
the object a colored one. It is quite obvious that the filters will be
orange-red for transmitting the "warm" colors, and the complemen-
tary blue-green for the "cold" colors.

The fidelity of color reproduction depends mainly upon three fac-
tors:

(1) The color-sensitivity characteristic of the photographic
emulsion.

FIG. 1. Exterior view of Filmo Morgana
camera.

(2) The selective characteristic of the photographic and projec-
tion filters relative to the character of the source of light.

(5) The instrument (camera and projector) must be so con-
structed as to fulfill the requisites for analyzing, and subsequently
synthesizing, the color radiations of the object.

It is not within the scope of this paper to discuss the first two
factors, since many data are already available on these subjects.
Suffice it to say, that panchromatic emulsions must, of necessity, be
used ; and that the filters used for photographing must be so selected

Nov., 1933 J THE MORGANA COLOR PROCESS 405

as to coordinate the color-sensitivity characteristic of the film with
the character of the light used for photographing. Thus, different
pairs of filters must be used when photographing with a tungsten
filament lamp, and when photographing in daylight. For projection,
the color transmission characteristic of the filter is chosen with re-
gard to the light radiated by tungsten filament lamps, which are
used exclusively for 16-mm. projection.

It may not be amiss to mention here that since the faithfulness
with which color is rendered depends upon the above-mentioned
physical characteristics, and since these characteristics vary some-

FIG. 2. Front view; lens removed show-
ing oscillating filter.

what and are difficult, if not impossible, to control with regard to the
visual characteristics of the human eye, the results obtained can only
approximate the true colors of the object. The approximation at-
tained, however, is quite close, and, at any rate, very pleasing to the
eye.

The designing of the camera for the Morgana process did not pre-
sent any great difficulties or involve radically distinctive features
except as to its portability and compactness.

Fig. 1 shows the external appearance of the Filmo Morgana camera.

406 J. A. DUBRAY [J. S. M. P. E.

It will be noticed that the cumbersome filter-carrying wheel usually
attached to cameras to be used to take alternate frames through
different filters, has been replaced by an oscillating filter carrier
(Fig. 2). The red and the blue filters are alternately brought into
position between the lens and the film by a to-and-fro motion of the
carrier which takes place during the periods of shutter occultation.

The advantages of this method of alternating the filters are : com-
pactness and correspondingly improved portability of the camera;
immobility of the filter during the periods of exposure; protection
of the filter against possible damage due to external causes; and the

FIG. 3. Camera head showing slot through which
color filters are inserted.

facility with which the filters may be removed, permitting the camera
to be used for ordinary black-and-white cinematography. Black-
and-white pictures may be made with the same roll of film, since
panchromatic film is used and the films are processed in the cus-
tomary way.

The ease with which filters can be withdrawn or inserted in their
carrier is illustrated in Fig. 3. By rotating the knurled rim of the
camera head, the slot A is brought to the position shown in the illus-
tration and the filter holder is easily withdrawn or inserted, as the

Nov., 1933]

THE MORGANA COLOR PROCESS

407

case may be. A twist of the knob closes the slot and prevents the
filter from moving out of place or stray light from entering the camera
when the filter is not in position and the camera is being used for
black-and-white work.

Fig. 4 illustrates the inside mechanism of the camera head, which
does not differ from that of the similar model of the Filmo camera
except for the cam shown at A which controls the oscillating motion
of the filter carrier.

Since the sensation of color is attained during projection by the
subjective composition of successive complementary frames, it is

FIG. 4.

Camera head mechanism; A, the cam that actu-
ates the filter carrier.

obvious that each frame should be identical to its complement.
This is not the case, however, and the change of position of a moving
object, due to the lapse of time between the taking of two successive
(complementary) frames, produces the well-known effect called
"fringing." This effect is, fortunately, objectionable only when the
object that is photographed is moving rather rapidly across the field
of view of the camera and quite close to it.

It is evident that the shorter the time between any two comple-
mentary frames the less noticeable will the fringing be. This would

408 J. A. DUBRAY [J. S. M. P. E.

naturally suggest that the taking speed be increased. On the other
hand, such increase of speed involves an objectionable consumption
of film and limitations with regard to the exposure time. It is recom-
mended for satisfactory results that the Morgana camera be operated
at a normal speed of 24 frames per second. However, it is not com-
pulsory to maintain this rated speed, but higher or lower speeds can
be used to produce slow-motion effects or for photographing inani-
mate objects. The Morgana camera is therefore equipped with a
control by means of which its speed can be varied from 8 to 32 pic-
ture frames per second.

FIG. 5. Morgana projector.

No limits are imposed as regards focal length of lenses and dia-
phragm opening, except for the necessary increase of exposure as
compared with black-and-white cinematography, due to the absorp-
tion of the filters and the increase of the taking speed. This in-
crease of exposure is, however, relatively small, as it corresponds to a
factor of 4x for daylight and of 3x for Mazda illumination.

In the projector (Fig. 5) the conventional color-filter wheel has
been retained for convenience of construction as well as for con-
venience in synchronizing the proper sector of filter with the picture
frames corresponding to it.

Nov., 1933] THE MORGANA COLOR PROCESS 409

Fig. 6 is a view of the filter wheel, showing the two color sectors.
Each sector is composed of smaller sectors (two for the red and three
for the blue) and transparent spacings, the color transmission char-
acteristic and area of each sector being chosen so that the aggregate
effect upon rotation will be correct for the color transmission of the
taking filter and the color radiation characteristics of the projection
lamp.

The most striking feature of the Morgana projector is to be found
in the manner in which the motion of the film is controlled at the
projector gate.

FIG. 6. Color filter wheel showing the red and
blue-green sectors.

Since the pictures are taken at a normal speed of 24 frames per
second, it would be necessary only to project them at that speed,
care being taken that the proper filter is placed before its correspond-
ing frame. However, taking into consideration that the final effect
depends upon the ability of the eye to superimpose each pair of com-
plementary pictures, and that the colors of the object are analyzed
in terms of only two primary components, it is found that a projection
speed of 24 frames per second involves physiological difficulties
known as "color flicker" or "color bombardment," which are extremely

410

J. A. DUBRAY

(J. S. M. P. E.

disagreeable and fatiguing. It is obvious that the greater the
rapidity of alternation of the two complementary colors the less
noticeable the flicker will be.

In order to achieve greater projection speed the movement of the
Morgana projector is so designed that each forward movement of
two successive frames at the projector aperture is followed by a
backward or reverse movement of one frame, so that each picture
frame is projected three times instead of once. The rate of projec-
tion at the aperture is therefore 72 frames per second, although the
linear motion of the film at the feed and take-up sprockets corre-

COLOR

ALTERNATION

TIME

EXPOSURE

FIG. 7. Chart showing the alternation of color filters and the triple
projection of each film.

spends to a running speed of only 24 picture frames per second (7.2
inches).

Fig. 7 shows graphically the relation between film movement, time
of exposure, and color. The letters R and B indicate the color of the
projection filters, red and blue-green, and the arrows indicate the di-
rection of the movement. The graph illustrates the alternation of
the red and blue-green filters, and the triple projection of each frame.

The intermittent movement of the Bell & Howell regular Filmo
projector is so designed that for every frame "pulled down," the
single-bladed shutter revolves three times. In the Morgana projec-
tor, motion is imparted to the film for every revolution of the shutter,
the cycle of motion being as stated above, causing a movement of

Nov., 1933] THE MORGANA COLOR PROCESS 411

two frames in the forward direction and one in the reverse direction.

Fig. 8 shows details of the Morgana projector gate with the lens
and lens holder removed. At A is the pull-down feeding finger. At
B is the reverse feeding finger, and at C is a pilot pin that engages the
film perforation for each one of its motions, in order to register each
frame at the projector aperture.

Fig. 9 illustrates the cams and double shuttle that control the
motion of the feeding fingers and pilot pin, again designated by the
letters A, B, and C.

FIG. 8. Detail of projector gate and registering mechanism.

SUMMARY

The obvious objections to the Morgana process are that it is a two-
color process, and that color fringing is experienced in photograph-
ing close-ups in fast motion. The Morgana process is practicable
despite these objections, because of its manifold advantages.

First, and most important, it allows any lens, from a wide-angle to a
telephoto, to be used on the camera; and, perhaps even more im-
portant, it allows any number of duplicates to be made, a vital ne-
cessity for industrial and educational applications.

Even though it is a two-color process, the only colors that are
really lost are the deep purples, the magentas, and the rich yellows.

412 J. A. DUBRAY

Flesh tones are exceptionally good, much better than heretofore ob-
tained, to our knowledge, with any two-color process.

The backing-up phase of the projection is radically new in projec-
tion practice, and is responsible for the ability to show ordinary move-
ment without any apparent trace of flicker or objectionable color fring-
ing, so that for practical purposes, bearing in mind the slight limitations
of the process, it is quite satisfactory for industrial and educational
applications.

FIG. 9. Details of cam and double shutter.

An advantage of the process is the fact that considerably less light
is needed for color photography, and that large color pictures can
be projected quite readily. With the 400-watt lamp now widely used,
an 8 X 10-ft. picture of adequate brilliance for an audience of one to
three or four hundred is quite satisfactory. Some interesting work is
being done with this new process, including time-lapse work with
growing flowers, medical cinematography, and the like.

REFERENCE

1 MATTHEWS, G. E.: "Processes of Photography in Natural Colors," /. Soc. Mot.
Pict. Eng., XVI (Feb., 1931), No. 2, p. 188. "Photography of Colored Objects,"
Eastman Kodak Co., Rochester, N. Y.

NEW DEVELOPMENTS IN PORTABLE GAS-ELECTRIC
GENERATORS FOR MOTION PICTURE LIGHTING*

PETER MOLE**

Summary. A newly developed gas-electric generator, recently made available
to the industry, is described. There are two types, the small units having a ca-
pacity of 40 kw. and the larger ones a capacity up to 160 kw.

The gas-electric generators described have unique features of voltage control, which is
an essential requirement when generators are used in the field with incandescent
lamps as a load. They have been designed to have minimum weights for their ca-
pacities and are more silent in operation than any units heretofore built.

It is the purpose of this paper to describe some developments
that have been made in the past year toward producing suitable
gas-electric generators for supplying current used in motion picture
production on location. During the earlier years of sound motion
pictures, the limitations imposed by the recording operations re-
stricted the taking of pictures almost exclusively to sound-proofed
stages. In the past two years technologic advance in recording
equipment and operation has made it entirely practicable to carry
on location work.

In order that the shots taken of scenes away from the studios
may attain to the same high quality as those taken in the studios,
it is necessary for both day and night work on location to utilize
the advantages of lighting by artificial means. The supplementing
of daylight with artificial illumination is designated in studio ver-
nacular as "booster" lighting.

In most cases the current for booster lights may be supplied by
generators having capacities up to 60 kw. The lighting of night
scenes to a large extent parallels that of lighting stages, large values of
current in this case being required. Current for location lighting
may be obtained from either of two sources. Where power com-
panies have transmission lines reasonably close to the location,
those lines may be tapped and portable motor-generator sets con-
nected to them to produce the 11 5- volt direct current, which is

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Mole-Richardson, Inc., Hollywood, Calif.

413

414

PETER MOLE

[J. S. M. P. E.

standard in the industry. Many desirable locations isolated from a
commercial supply require the use of electric generators driven
by prime movers.

Gas engine driven generators are not new to the industry. In
the days of silent pictures many satisfactory plants were built, using
aeronautical engines connected to direct-current generators and
mounted on trucks.

FIG. 1. 40-kw, gas-electric generator set.

The advent of sound pictures made obsolete most of the gas-electric
generators of the old type. The engines heretofore used were princi-
pally war-time Liberty, Curtis, OX 5, and Hispano engines. The
valve actions were always noisy, and the engines themselves in most
cases had fallen into disrepair and were unreliable.

As location work was revived, the small booster lighting plants
were developed, and the better types of automobile motors were
found suitable. Very satisfactory plants have been built utilizing
Chevrolet, Cadillac, Lincoln, and other engines that could develop

Nov., 1933] NEW DEVELOPMENTS IN GENERATORS 415

sufficient power to suit the various requirements. The plant illus-
trated in Fig. 1 is capable of delivering 350 amperes at 125 volts as
operated under motion picture conditions, and under continuous
operation will deliver 250 amperes at that voltage. The essential
characteristics of the booster plant are light weight, minimum size, re-
liability, and quietness of operation. The plant, the interior of which
is shown hi Fig. 2, is equipped with a Ford V-8 engine, capable of
producing 70 hp. at 3250 rpm. This motor was selected because of

FIG. 2. Interior view, 40-kw, gas-electric generator set.

its compactness and the reliability that it has demonstrated for motor
car use. It is connected to an electric generator which is specially
designed to match the power output of the motor. This generator
attains its rated voltage at 1200 rpm. Its armature shaft is flexibly
connected to the engine crank -shaft. The frame of the generator is
rigidly attached to the end bell of the generator so that the engine-
generator assembly may be mounted on a three-point support.
The supporting points rest on suitable rubber pads to reduce the
transmission of sound. The radiator and fan are mounted on the

416 PETER MOLE tf. S. M. P. E.

generator end of the frame to avoid having to conduct the heat
radiated by the engine and its exhaust pipes over the generator.
The generator is self-ventilated, and in many hours of service has
proved well adapted for the purposes for which it had been designed.

This little power plant is six feet long, two feet wide, and three
feet, ten inches high, weighing only 2100 pounds, so that it may be
readily transported from place to place on a truck carrying additional
equipment. It is covered completely by a sheet metal housing
insulated with sound-proofing material; the housing is made in
sections to facilitate repairing.

The unique feature of the plant is the system of voltage control
and engine governing. The use of filament lamps has made close

FIG. 3. 160-kw., gas-electric generator set.

control of the voltage very necessary, since a sudden rise of voltage,
due to a reduction of the load, tends to burn out the lamps. During
the nominal shooting period it is essential that the voltage be main-
tained constant, in order that the illumination of the units be uni-
formly maintained. Hand control is inadequate, because the least
inattention on the part of the operator might allow wide voltage
fluctuations and consequent detrimental effects.

In taking sound pictures it is essential that the generating equip-
ment operate as silently as possible, as often the locations are in
canyons and valleys, the reverberation characteristics of which make
noisy equipment impracticable. Adequate muffling of the exhaust
and the use of suitable sound-insulated housing in these modem

Nov., 1933]

NEW DEVELOPMENTS IN GENERATORS

417

plants have made it possible to operate the sets reasonably close to the
sound recording equipment.

The development of high-power equipment for use on large night
locations presented a more difficult problem than the designing of
suitable booster equipment, because the greater power involved
necessitated heavy equipment and engines of large displacement;
yet, for convenience in such operations, portability must be maintained.

Fig. 3 illustrates the latest development in the larger portable
gasoline generators. The plant is capable of delivering a current of

FIG. 4. Interior view, 160-kw., gas-electric generator set.

1400 amperes at 125 volts intermittently, and is capable of de-
livering 1000 amperes continuously. To deliver this power a gasoline
motor of 270 hp. is used.

The unit shown is equipped with a Hall-Scott model 168 Invader
Engine having six cylinders of 5V2-inch bore by 7-inch stroke, operat-
ing with a compression of 100 pounds. Operating on standard
gasoline it delivers 270 hp. at 2100 rpm., and 248 hp. at 1800 rpm.
Although probably other motors might have been used, this one was

418 PETER MOLB [J. S. M. P. E.

selected because a good opportunity had been afforded to observe
its operation for over a year, a large number of motors of this type
being used in water-taxis in Los Angeles harbor. It was also of
advantage that the manufacturer's plant was located at Oakland,
California, which made close contact possible when making certain
changes required for the service. The engine was primarily designed
for marine use, but the Hall-Scott engineers re-designed the crank
case and removed the reverse gear that had been a part of the stand-
ard unit, making the engine adaptable to our requirements. /

The generator in this plant was especially designed by the Generffl
Electric Company to match the power curve of the gas engine.
It is very compact, weighing only 2100 pounds. The generator
attains its rated voltage at 1200 rpm., and delivers 1000 amperes
continuously at 1600 rpm. No difficulty has been encountered
in generating 1450 amperes under the usually intermittent demands
of picture production. The generator is practically flat-compounded
at 1800 rpm., and performs well within the speed range noted above.
The large commutator provides ample surface for brush contact.
The generator is self -forced-ventilated.

The motor is mounted on a sub-frame supported on 3 points, and is
connected by a flexible coupling to the generators. The job has been
carefully engineered to provide accurate alignment under all operating
conditions. The radiator is of the sectional type, and consists
of two separate cores of six sections each, connected to headers at the
top and bottom, the cores being separated by a two-inch air space ;
the air is circulated through the radiator by a fan of the aeronautical
type, driven by a variable-speed motor supplied with current by
the main generator. By varying the speed of the cooling fan, the
temperature of the water may be maintained at 180F. at the water-
jacket outlet, this temperature being correct for most efficient
operation.

Since the motor was designed for marine use, it was necessary to
add a centrifugal pump, so as to provide the volume of water re-
quired for radiator cooling. The engine is supplied with full-force
feed lubrication with oil that is filtered and cooled to 150 degrees by a
water-cooled heat interchanger.

The speed of the engine is controlled by a centrifugal governor
adjustable to various speeds at the controlling panel. In addition
to providing normal regulation of voltage by compounding the
generator, a voltage regulator has been installed. All controls are

Nov., 1933] NEW DEVELOPMENTS IN GENERATORS 419

concentrated in a panel on which are mounted the ignition and aux-
iliary switches, a tachometer, water and oil temperature indicators,
charging ammeter, hand throttle, oil-pressure gauge, generator
ammeter and voltmeter, shunt field rheostat control, voltage-
regulator relay, and circuit breaker operating lever.

The engine is mounted in a closed compartment, which is venti-
lated through louvers by the carburetor intake air. The walls of
the compartment are sound-proofed with suitable insulating material ;
and since the engine has been carefully designed to operate quietly,
sounds that would be detrimental have been reduced to a minimum,
permitting operation under heavy load within 200 feet of the micro-
phone on open locations.

Mufflers are mounted in a compartment above the engine, the
exhaust manifold connecting to three mufflers of large capacity.
Exhaust noises have been reduced to a point where they are not
objectionable. The generator has been placed between the engine
and the radiator in order that the temperature of the generator may
not rise above the optimum value.

This 175-kw. set is mounted on a tandem wheel truck, which
carries ample gasolene in its tanks for ten hours of operation under
maximum load. With the exception of the gasoline tank the entire
plant is self-contained, and may be lifted from the truck. This is an
essential feature because these plants are often used on ships, in
baggage cars, in scenes taken from trains, and other unusual places.

A SILENT CAMERA*
H. R. KOSSMAN**

Summary. After reviewing briefly some of the difficulties of excluding noise
from sound cameras, a silent camera, recently developed, is described in detail. This
camera makes unnecessary the use of blimps and possesses all the features necessary
and desirable in modern camera construction.

Before sound became wedded to the motion picture, the important
requirements of a camera were steadiness, lightness in weight, sim-
plicity of operation, and compactness. But with the advent of sound,
camera manufacturers, as well as cameramen and sound recordists,
realized that the cameras were far too noisy. The question of
minimizing or possibly of eliminating the camera noise then became a
major problem. In trying to eliminate the noise of the movement,
some manufacturers changed the design of the intermittents. Some
omitted the pressure plates or the pilot pins, or both. But it soon
became apparent that those changes would not accomplish their aim.
In many cases, the camera became better from the point of view of
sound recording but at the expense of the quality of the picture.

Early in 1928 enterprising cameramen began to place their cameras
in sound-proofed booths, which had hulls like battleships, the main
difference between these cages and the battleship being that it would
have been easier to manoeuvre the battleship into firing position than
to place the booth in line for every new shot that had to be taken.
To make matters worse, sometimes two or three cameras were placed
in one booth.

The sound-proofed booths caused so much discomfort that even-
tually some cameramen, in desperation, removed their cameras from
the booths and covered them mountain high with blankets. That
was supposed to represent an advance over the sound-proofed booth.
One of the chief difficulties with the system was that the cameraman
did not have any access to the camera. The next step was the
"blimp."

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Andre Debrie, Inc., New York, N. Y.
420

A SILENT CAMERA

Although the blimp represented a vast improvement, many of the
cameras had outside magazines which necessarily made the blimps
tremendously large. The monster blimps built by the various
studios were aptly named: they looked like blimps; although they
could neither fly nor float. Quite the contrary, they were tre-
mendously heavy and cumbersome. Their instability and tendency
to tip over, generally at the most inopportune moment, caused the

FIG. 1. Exterior view of camera; (.4) knob for shifting
ground glass, (B) focusing dial, (C) focusing tube (Sx), (D)
speed indicator dial, () footage and turn counter, (F)
locket and switch with electromagnetic cut-out, (G) hand
crank and opening for removing motor, (H) automatic fade
and lap dissolve, ( J) hand fade, (K) knob for locking case.

cameramen and the assistants no end of trouble. As quickly as a
new style of blimp was built and tried, so quickly was it discarded for
some reason or other. They generally violated the four cardinal
requirements of a successful camera: steadiness, lightness in weight,
simplicity, and compactness. It was more difficult to follow focus
and to make other vital adjustments of the camera while in operation,
and they were so unwieldly that it was impossible to use them for

422

H. R. KOSSMAN

[J. S. M. P. E.

certain kinds of shots where close-ups in restricted areas are often
advantageous.

The first Debrie sound camera, Parvo T, having 1000-foot maga-
zines was furnished wjjth a blimp having outside controls, which
represented quite an advance on account of its unusually small
weight and the ease of handling it. But a camera was visualized
which would be compact, comparatively light in weight, and in which
such vital parts as pilot pins and pressure plates would not have to be
sacrificed in order to produce a quietly running camera.

FIG. 2. Interior view of camera, front; (L)
ground glass, (Af) intermittent pressure plate, (N)
registering pins, (0) film punch, (P) pivot for gate
and ground glass, (Q) pan adjuster, (R) tilt adjuster.

To solve the problem an entirely different approach was made.
From the start it was decided not to silence the camera, but rather
to construct a silent camera. That meant that different raw ma-
terials would have to be used, such as lightweight metals for elimi-
nating vibration, specially tested for their sound-carrying qualities;
different gear materials; different methods of lubricating; the
mounting of various important parts in rubber ; gears would have to

Nov., 1933]

A SILENT CAMERA

423

run entirely in oil; the casing would have to be independent of the
camera; and the walls of the casing made of a special bakelite and
rubber product.

In the camera so developed, the movement has a pull-down of 18
degrees; the pilot pins are direct acting; and the intermittent pres-
sure plates are made of rubber composition. This last feature is of
utmost importance, because of the increasing use of panchromatic and

FIG. 3. Interior view of camera, left side;
(Q) pan adjuster, (R) tilt adjuster, (T) knob for
closing shutter, ( U) oil level, ( V) take-ups.

superpanchromatic film, the emulsion coating of which is very sensi-
tive. The intermittent pressure plates are designed to avoid scratch-
ing the film in the aperture, by relieving the pressure on the film when
it is in motion.

The shutter opening is 150 degrees with 7 openings ranging from 20
to 150 degrees. The camera is equipped with an automatic dissolve
mechanism, which operates over 72 frames, the camera being stopped

424

H. A. KOSSMAN

LT. S. M. p. E.

automatically by the operation of an electrical contact at the end of
the fade. A hand fade is also provided, so that the length of the fade
can be made to suit any condition. The focusing scale is at the back
of the camera, and is made to suit lenses of various focal lengths.
This is accomplished by changing the angle of the focusing helix of the
various lens sleeves. Focusing is done directly on the film, or on
ground glass, with the focusing tube at the back of the camera.

FIG. 4. Interior view of camera, right side; (V)
take-ups, (W) switch for automatic fade, (X) auto-
matic switch for anti-buckling device, ( Y) gear shift
for motors of 1500 or 2400 rpm., (Z) knobs for set-
ting footage and turn counters to zero.

Besides the focusing scale and the focusing tube, the footage and
image counter as well as the speedometer and the main switch are
located at the back of the camera. The latter is of the electromag-
netic type, and is integral with the plug for connection to the power
supply. In order to reverse the motor the female plug is turned
around. The plug is plainly marked so that no mistakes are possible.
The switch is connected to an anti-buckling contact, located on the

Nov., 1933J A SILENT CAMERA 425

lower sprocket, and the electrical cut-out contact of the automatic
fade. The anti-buckling contact causes the switch to cut out when
the film is subjected to the slightest amount of tension, or when the
film has run out.

The motor is mounted in the interior of the camera. It is a 3-
phase, a-c., 220-volt motor, 1 / z & hp.; motors of other types can be
used. Special attention has been given to the casing and mounting
of the Super Parvo. The casing is made of layers of bakelite, rubber,
and felt, so as to be sound-proof. The frame of the camera
is mounted on rubber blocks so as to prevent vibrations from reaching
the casing. The gears, as well as the intermittent movement, are
entirely enclosed and run in oil. Three oil-cups take care of all the
lubrication of the camera.

The front of the casing is closed with a sheet of optical glass, the
frame of which carries the rods for an outside extension to accommo-
date the sun bellows, iris, and other front attachments.

The camera measures 19 by 14 by 10 inches, which is about one-
third the size of a representative blimp. The weight is about 80
pounds, including the motor and magazines. A 1000-foot magazine
is located on each side of the camera.

A special friction type tripod is used, which weighs 25 pounds.
The tripod head has a F-shaped slide, in which the camera slides and
locks automatically. Perfect balance is assured by a new type of
hydraulic balancer.

A TRIPLEX MOVIOLA FOR EDITING RE-RECORDING*
J. O. AALBERG**

Summary. A multiple sound head moviola, designed for lessening the time
required to synchronize and check effects tracks, is described. It consists of one
standard viewing machine and three sound reproducers coupled together by flexible
shafts and clutches.

The illusion of realism in sound motion pictures is continually
being enhanced by the addition of suitable sound effects. Also
many producers are adding musical accompaniment to portions of
their productions in the attempt to add to the picture's entertain-
ment qualities. These modifications are accomplished by re-re-
cording. Next in importance to the selection of realistic effects and
appropriate music is the synchronizing of those effects with the
corresponding action. When from three to ten tracks must be
synchronized correctly to a frame, the time and expense involved
become appreciable.

To quicken the task of synchronizing and checking effects tracks,
a multiple sound head moviola, shown in Fig. 1, was assembled.
It consists of one standard viewing machine and three sound repro-
ducers coupled to each other through flexible shafts and clutches.
Additional sound heads might be added to meet special requirements.
A quarter horse-power induction motor drives the sound head next to
the viewing head, allowing use of the sound head alone or with the
viewing head when only one track is being worked. As more tracks
are to be worked, the additional sound heads are coupled together.
Provision is made for running split film, as shown in Fig. 2. The sound
head may be adjusted for 35-mm. film by sliding one flange of each
of the rollers toward the right and removing the detachable rail from
the center of the film slide. A pocket for storing the rail is provided
on the supporting frame of each sound head. The amplifier is of
conventional design, employing two 247 pentodes in its output
stage to feed a dynamic speaker which, for convenience, is mounted

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** RKO Studios, Hollywood, Calif.
426

A TRIPLEX MOVIOLA

427

(a) (6)

FIG. 1. Triplex moviola: (a) front view; (6) rear view.

FIG. 2. Close-up of sound head.

428 J. O. AALBERG

on the machine. Complete alternating-current operation of the
amplifier is provided, including polarizing potentials for the photo-
electric cells and energy for the exciter lamps. The housing that
forms the base of the sound heads contains the amplifier and trans-
formers. The outputs of the sound heads are directly connected to
the amplifier, and the individual levels are adjusted by varying the
brightness of the exciting lamps. These controls are shown at
the left hi Fig. la. A volume control is provided for adjusting the
combined levels. The assembly runs either forward or backward,
enabling one to go back a short distance to re-check a sequence
without re-winding and re-threading. Lacking all the complications
and the mass of selsyn or interlocking motors commonly used to
synchronize a number of sound heads, this device stops and starts
quickly, a very desirable feature when tracks must be moved a frame
at a time and checked until found correctly synchronized. The entire
machine occupies a floor space of only two feet by four feet, and is
mounted on casters to facilitate moving it from room to room.

The need for such a device became apparent during the assembling
of effects tracks for the Radio picture, King Kong. After some ex-
perience, it was possible for the film editors to mix their sound tracks
on this machine and make all necessary checks without running in a
projection room, as had formerly been the procedure. The triplex
moviola was built for the RKO Studios by the Moviola Company of
Hollywood.

BOOK REVIEW

A New System of Cinematography in Relief (Due Nuovi Systemi di Cinema tog-
rafia in Rilievo). DOTT. ING. GUIDO JELLINEK. Liberia Editrice Politecnica,
Milan, 1932. In this monograph, which is hailed on the cover as "a revolution
in the technic of the cinema," two proposed systems of projection in relief are
discussed, and some crude experiments in support of the theories developed are
described. The two "new" systems are, first, photographing the object by films
in a series of different planes and, second, the use of a lenticular screen of the
Lippmann type. The discussion and the experimental apparatus pictured serve
to emphasize the fact that solutions of the problem along those lines run to great
complexity of machinery and extreme refinement of apparatus adjustment. It
is not clear that the author has carried his studies or his experiments far enough
actually to establish the validity of his methods, even if they could be made
practicable. Thus it is not clear how a series of images in different planes are
to be prevented from blanketing each other; and in his proposed utilization of
the Lippmann principle the author does not appear to realize that the images
as obtained would be pseudoscopic instead of stereoscopic.

H. E. IVBS

Society Announcements

CHICAGO CONVENTION, OCTOBER 16-18, 1933

The semi-annual Convention of the Society was held at Chicago, 111., October
16-18, with headquarters at the Edgewater Beach Hotel. The success of the
meeting was due in large measure to the efforts of Mr. W. C. Kunzmann, chair-
man of the Convention Arrangements Committee; Mr. O. M. Glunt, chairman of
the Papers Committee, and his associates; and to the following individuals and
firms:

Mr. H. Griffin, of the International Projector Corp., for providing and installing
the projection equipment; Electrical Research Products, Inc., for supplying the
sound reproducing equipment; the Bausch & Lomb Optical Co., for supplying
the projection lenses and the balopticon; the Strong Electric Co., for the recti-
fiers; Mr. J. E. MacAuley, for the Peerless projection lamp; National Theater
Supply Co., for the projection booth, reels, rewinds, etc.; and the Day Light Screen
Co., for the projection screen. Thanks are also due to the Chicago Film Board
of Trade; to Messrs. S. A. Lukes arid T. Maloy, of Local No. 110, of Chicago;
to Mr. J. H. Goldberg; to Messrs. W. Immerman and L. Lipstone, of Balaban &
Katz, for providing the banquet entertainment and engaging the orchestra;
and to Mr. B. Pearlman, of the National Theater Supply Co. Mr. and Mrs.
H. DeVry very kindly entertained the ladies visiting the Convention on a cruise
on their yacht, viewing the Chicago beach and the World's Fair from the lake.
Films for projection on Monday and Tuesday evenings were provided through
the courtesy of Metro-Goldwyn-Mayer Corp., Paramount Publix Corp., RKO
Pictures, Inc., United Artists, Fox Film Co., and Mr. H. T. Cowling.

At the luncheon that opened the Convention on Monday, October 16, addresses
were delivered by Mr. C. F. Strodel, executive officer of Balaban & Katz; Mr.
Donald P. Bean, of the University of Chicago; Dr. Allen D. Albert, of the Cen-
tury of Progress Exposition; and Mr. W. H. Strafford, of Chicago.

PROGRAM
MONDAY, OCTOBER 16

The morning was devoted to the organization of the Convention, registration,
meetings of Committees, etc.

12:30 p.m. Luncheon (for members, their families, and friends.)
Speakers: Mr. C. F. Strodel, Executive Officer, Balaban & Katz; Mr. Donald
P. Bean, University of Chicago; Dr. Allen D. Albert, Century of Progress
Exposition; Mr. W. H. Strafford, of Chicago.

2:30 p.m. East Lounge. Business Session.

Opening of Convention. A. N. Goldsmith, President.
Report of the Secretary, J. H. Kurlander.
Report of the Treasurer, H. T. Cowling.

429

430 SOCIETY ANNOUNCEMENTS [J. S. M. P. E.

Report of the Convention Arrangements Committee, W. C. Kunzmann,
Chairman.

Society Business: election of officers for 1933-34; proposals for amendment of
the Constitution and By-Laws.

Report of the Membership and Subscription Committee, E. R. Geib, Chairman.

"A Brief History of the Kinetograph, the Kinetoscope, and the Kineto-phono-
graph," by W. K. L. Dickson, La Haule, Jersey, Channel Islands, England.

"The Control Frequency Principle," by J. E. Jenkins, Jenkins & Adair, Chi-
cago, 111.

8:00 p.m. East Lounge. Exhibition of recent motion pictures, including pictures
of the Century of Progress Fair, taken by Mr. H. T. Cowling.

TUESDAY, OCTOBER 17

9:30 a.m. East Lounge. Wide Range Session, Dr. A. N. Goldsmith presiding.

"Wide Range Recording," by F. L. Hopper, Electrical Research Products, Inc.,
Hollywood, Calif.

"Acoustic Requirements for Wide Range Reproduction," by S. K. Wolf, Elec-
trical Research Products, Inc., New York, N. Y.

"Wide Range Sound Reproduction," by F. C. Willis, Electrical Research
Products, Inc., New York, N. Y.

Report of the Historical and Museum Committee, E. Theisen, Chairman.

Report of the Committee on Laboratory and Exchange Practice, R. F. Nichol-
son, Chairman.

"Some Practical Applications of Acoustics in Theaters," by G. W. Baker and
M. A. Smith, U. S. Gypsum Co., Chicago, 111.

2:00 p.m. East Lounge. Photographic Session, Mr. J. I. Crabtree presiding.

"Recent Improvements in the Bell & Howell Automatic Printer," by R. F.
Mitchell and A. S. Howell, Bell & Howell Co., Chicago, 111.

"Effect of Film Shrinkage on Sound Film Printing," by J. Crabtree, Bell Tele-
phone Laboratories, Inc., New York, N. Y.

"Film Noise in Sound-on-Film Reproduction," by H. C. Silent, Electrical Re-
search Products, Inc., Hollywood, Calif.

"Further Investigation of Ground Noise in Photographic Sound Records," by
O. Sandvik, Eastman Kodak Co., Rochester, N. Y.

"Color Photography for Industrial and Business Films," by R. H. Ray and
H. W. Cress, Ray-Bell Films, Inc., St. Paul, Minn.

"Automatic Change-Over Device," by A. Pritchard, Coronado, Calif.

7:30 p.m. Ball Room. Convention Banquet
Dancing, motion pictures, and entertainment.

WEDNESDAY, OCTOBER 18

9:30 a.m. East Lounge. General Session, Mr. R. F. Mitchell presiding.

"The Use of Talking Pictures as an Additional Tool at the University of
Chicago, by H. B. Lemon, University of Chicago.

Nov., 1933] SOCIETY ANNOUNCEMENTS 431

"Sixteen-Millimeter Film and Film Recording Problems," by J. O. Baker,

RCA Victor Co., Camden, N. J.
"Sprocket Dimensions for Visual and Sound Projection Equipment," by H.

Griffin, International Projector Corp., New York, N. Y.
Report of the Committee on Standards and Nomenclature, M. C. Batsel.

Chairman.

Report of the Projection Practice Committee, H. Rubin, Chairman.
"A New 35-Mm. Portable Projector," by H. Griffin, International Projector

Corp., New York, N. Y.
"The New DeVry Sound Recording Camera," by H. DeVry, Chicago, 111.

2:00 p.m. East Lounge. General Session, Mr. R. E. Farnham presiding.

"Manufacturing Problems Involved in the Production of Sound Picture Equip-
ment," by H. E. DeCamp, Western Electric Co., Chicago, 111.

"The Rotambulator a New Type of Camera Stand," by J. A. Dubray, Bell &
Howell Co., Los Angeles, Calif.

"Economics of Advertising Projector Lamps," by E. W. Beggs, Westinghouse
Lamp Co., Bloomfield, N. J.

"A Non-Rotating D-C. High-Intensity Arc," by D. B. Joy and A. C. Downes,
National Carbon Co., Cleveland, Ohio.

"A New Type of Carbon Arc Broadside Lamp for Use in Motion Picture Pro-
duction," by P. Mole, Mole-Richardson, Inc., Hollywood, Calif.

"A New Carbon for Use in Photography," by A. C. Downes, National Carbon
Co., Cleveland, Ohio.

ADJOURNMENT OF THE CONVENTION

Note : The Society of Motion Picture Engineers will not be responsible for state-
ments made by authors.

Convention Committee, Papers Committee,

W. C. KUNZMANN, Chairman. O. M. GLUNT, Chairman

BOARD OF GOVERNORS

At a meeting held at the Edgewater Beach Hotel at Chicago, October 15,
amendments were formulated for presentation to the Society on the following
day, for effecting a general revision of the administrative policies of the Society.
These included changes in dues, admission fees, subscription fees, and the appoint-
ment of five vice-presidents whose work it would be to maintain a close super-
vision over the various phases of the Society's activities, to coordinate the work
of the various committees, and thus, in the interests of economy, to relieve the
General Office of a considerable part of its labors. The proposals include also
the plan to establish three grades of membership, to take the place of the two
now in force. The three grades would be known as Fellow, Active, and Associate.

The amendments were voted upon by the general Society on Monday, October
16; the proposed amendments of the By-Laws being approved contingent
upon the subsequent approval, by letter ballot, of the proposed amendments of

432 SOCIETY ANNOUNCEMENTS

the Constitution. Complete drafts of the proposals and the discussion held on
them at the meeting, will be mailed to all Active members in the near future,
together with a letter ballot form for voting upon them.

ELECTION OF NEW OFFICERS

The returns of the letter ballots cast recently by the Active membership,
counted at the Chicago Convention, indicated the following results:

A. N. GOLDSMITH, President.

O. M. GLUNT, Junior Vice-President.

J. H. KURLANDER, Secretary.

T. E. SHEA, Treasurer.

H. GRIFFIN, Governor.

W. B. RAYTON, Governor.

The remainder of the officers and Board of Governors are as listed on the
reverse of the Contents page of this issue of the JOURNAL. In addition, the
Board of Governors elected Mr. H. T. Cowling to serve as governor in the place
of Mr. W. C. Hubbard, recently deceased, for the remainder of his term.

JOURNAL

OF THE SOCIETY OF

MOTION PICTURE ENGINEERS

Volume XXI DECEMBER, 1933 Number 6

CONTENTS

Page

A Brief History of the Kinetograph, the Kinetoscope, and the

Kineto-Phonograph W. K. L. DICKSON 435

A Sixteen-Millimeter Portable Sound-on-Film Projection Equip-
ment C. R. HANNA, P. L. IRWIN, AND E. W. REYNOLDS 456

A Light- Weight Single-Film Recording System for Newsreels

and Travelogues C. R. SAWYER 466

A Non-Intermittent High-Speed 16-Mm. Camera

F. E. TUTTLE 474

The Cine-Kodak Special

J. STOIBER, O. WITTEL, AND F. E. TUTTLE 478

Portable Recording Equipment D. CANADY 483

Unoccupied Motion Picture Fields W. H. SHORT 494

New Motion Picture Apparatus 510

Society Announcements 517

Report of Treasurer 519

Index June to December, 1933 520

JOURNAL

OF THE SOCIETY OF

MOTION PICTURE ENGINEERS

SYLVAN HARRIS, EDITOR

Board of Editors
J. I. CRABTREE, Chairman

O. M. GLUNT A. C. HARDY F. F. RENWICK

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers.

Publication Office, 20th & Northampton Sts., Easton, Pa.
General and Editorial Office, 33 West 42nd St., New York. N. Y.
Entered as second class matter January 15. 1930. at the Post Office at Easton,
Pa., under the Act of March 3. 1879. Copyrighted. 1933. by the Society of
Motion Picture Engineers, Inc.

Officers of the Society

President: A. N. GOLDSMITH, 444 Madison Ave., New York, N. Y.
Past President: J. I. CRABTREE, Kodak Park, Rochester, N. Y.
Vice-President: W. C. KUNZMANN, Box 400, Cleveland, Ohio.
Vice-President : O. M. GLUNT, 463 West St., New York, N. Y.
Secretary: J. H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J.
Treasurer: T. E. SHEA, 463 West St., New York, N. Y.

Governors

E. COUR, 1029 S. Wabash Ave., Chicago, 111.
H. T. COWLING, 7510 N. Ashland Ave., Chicago, 111.
R. E. FARNHAM, Nela Park, Cleveland, Ohio.
H. GRIFFIN, 90 Gold St., New York, N. Y.
E. HUSE, 6706 Santa Monica Blvd., Hollywood, Calif.
W. B. RAYTON, Rochester, N. Y.
H. G. TASKBR, 41-39 38th St., Long Island City, N. Y.

A BRIEF HISTORY OF THE KINETOGRAPH, THE KINETO-
SCOPE AND THE KINETO-PHONO GRAPH*

W. K. LAURIE DICKSON**

In the year 1879 at the age of 19 I had read much of a Mr. Edison
in America and his scientific experiments, and so wrote to him to
inquire whether he would take me on his staff of experimenters
(Fig. 1). His reply was not encouraging. It read as follows:

Menlo Park, N. J
March 4, 1879.

William Kennedy Laurie Dickson,
Care of Mrs. Aubin,

2 Tregunter Road. London. W.
Dear Sir,

Your favor of the 17th ult. has just been rec'd.

I cannot increase my list of employees as I have concluded to close my works
for at least 2 years, as soon as I have finished experiments with the electric light.

Very truly,

T. A. Edison.

However, in spite of this, I persuaded my mother and sisters to
pull up stakes, and after a stormy crossing we landed in New York
and continued down to Richmond, Virginia, by the Old Dominion
S. S. Line. After residing there for two years, we youngsters made
for New York City early in 1881. I took my book of credentials,
etc., to show to Mr. Edison at his office at 65 Fifth Avenue, in case I
should be lucky enough to gain an interview.

My reception was unique. "But I told you not to come, didn't
I?" said Mr. Edison. I agreed, but told him I couldn't have done
otherwise after reading about the work in which he was engaged. He
watched my face while turning my testimonials over, until I had to
remind him please to read them. He only replied, "I reckon they

* Requested and recommended for publication by the Historical Committee.
Presented at the Fall, 1933, Meeting at Chicago, 111., at which meeting Mr.
Dickson was elected an Honorary Member of the Society.
** Montpelier House, Twickenham, Middlesex, England.
EDITOR'S NOTE: Mr. Dickson was born in France of Scotch parentage at
Chateau St. Buc, Minihic-sur-Ranse, in 1860.

435

436

W. K. LAURIE DICKSON

[J. S. M. P. E.

are all right; you had better take your coat off and get to work."
I had won.

He then gave me a note to Mr. Charles Clark, chief mathematician,
and another to Mr. W. S. Andrews, superintendent of the Goerk St.
testing and experimental department of the Edison Electric Works,
under whose able and kindly tutelage I secured a good knowledge of
what was wanted. The following year, with Mr. Edison's approval,
Mr. W. S. Andrews gave me his place while he traveled through the
United States planning and erecting electric light and power stations.

FIG. 1. William Kennedy Laurie Dickson.

My tests and experiments under Mr. Edison's direct instructions
were indescribably interesting. We attempted to arrive at a fixed
standardization of all electrical apparatus for home and power
stations, such as type of dynamo, motors, lamps, meters, etc. One
test or series of experiments stands out very clearly in my mind.
I had the good fortune to help Mr. Edison to determine the meaning
of the "Edison effect," or first concept of the famous "valve" used now
in radio apparatus.

Dec., 1933] BRIEF HISTORY OF THE KINETOGRAPH 437

In 1885 Edison took me away from the Electric Works at Goerk
St. to assist him in his private laboratory at Newark, N. J., where I
was given research problems to work on. In 1887 Mr. Edison,
who knew that I was keen on photography, disclosed his favorite
scheme of joining his phonograph to pictures taken photographically
with a device like the Zoetrope.

FIG. 2. Reproduction of original letter by
T. A. Edison (Courtesy Century Magazine).

He was then erecting his large laboratory at Orange, N. J., in which,
as soon as completed, I was allowed to select two large rooms : namely,
No. 5 on the 1st floor for the kinetophonograph experiments, and
No. 14 above for magnetic ore separation work, analysis, etc.

As to animated photography, Fig. 2 is a reproduction of a letter
by Mr. Edison, in his own handwriting, regarding his conception of

438 W. K. LAURIE DICKSON [j. s. M. P. E.

the work he wished me to cany out for him.* This proved entirely
successful in the end by giving our kinetograph the double duty of
taking the films and reproducing them on a screen in the simplest
possible way. If the kinetograph could take good, steady pictures
it followed that the same pictures could be projected as they eventu-
ally were by using a smaller sprocket wheel to allow for the slight
shrinkage of the film after developing and fixing.

By his order, however, projection was put aside, and our experi-
ments were concentrated on a "money earner," the kinetoscope, which
as it proved left the field open for all.

EARLY EXPERIMENTS WITH CYLINDER RECORDS

Edison's idea, as disclosed to me in 1887 at the Newark Labora-
tory, was to combine the phonograph cylinder or record with a similar
or larger drum on the same shaft, which drum was to be covered
with pin-point microphotographs which of course must synchronize
with the phonograph record (Fig. 3). I pointed out to him that in
the first place I knew of no medium that was sensitive enough to
take microphotographs at so rapid a rate while running continuously
on the same shaft.

"Well, try it; it will lead to other things," was Edison's reply.
I did so soon as I got to his new laboratory, then being erected at
Orange, N. J.

Before making the drum, which was to fit over the phonograph
shaft, I made a small micro camera, using various objectives or lenses
taken from one of my microscopes to produce the pin-head photos.
In this micro camera I tried Daguerre's process on highly polished bits
of silver and developed in the usual way. The subject I used was a
lantern slide of Landseer's stag for all these comparative single still
pictures.

The time of exposure was about three-quarters of a minute. Of
course, this method was soon abandoned. Next I tried silver nitrate
on wet collodion, using an exposure of 10 seconds, which was finally
shortened to 5 seconds. Then I had a light drum made and pro-
duced a few spirals of pictures on a dead slow shaft. These, even
with ammonia acceleration, proved a failure. So I increased the

* This letter was reproduced as a foreword to an article by the author and his
sister, Antonia Dickson, which appeared in Century Magazine, 48 (1894), p. 206.
Also published in "History of the Kinetograph, Kinetoscope, and Kinetophono-
graph" by the same authors; 1895. Albert Bunn, New York, N. Y.

Dec., 1933] BRIEF HISTORY OF THE KINETOGRAPH 439

size of the aluminum drum and of the pictures, and coated the
drum with a bromide of silver gelatin emulsion ; and would have ob-
tained a fairly good result but for some chemical action which took
place between the aluminum and the emulsion. That made me try
a glass drum and a one-opening rapid shutter.

My second batch of emulsion was light struck, owing to the night-
watchman's bursting in at 2 A.M., which so disgusted me that I just
slotted the aluminum drum and wrapped a sheet of Carbutt's stiff
sensitized celluloid over it. This proved quite satisfactory and
did away with my home-made emulsion coatings. The pictures were
sharp and good, and to save time in making prints or positives I
turned the negative into a positive effect with bichloride of mercury.
A reproduction of one of these sheets of 1 / 4 -inch pictures may be
seen in the History of the Kinetograph, etc. (See footnote, p. 438, also
Fig. 4.)

FIG. 3. Sketch of Edison's first method of synchron-
izing phonograph with picture records.*

To view these small pictures was another matter. If run con-
tinuously the result, of course, would be a black streak, unless seen
through a slotted disk. To obtain full illumination it occurred to
me to rig up a small Geissler spiral tube without a shutter.

These pictures, when viewed through a low-power microscope, were
fairly good in spite of the curvature of the drum. A disk was tried
next, to avoid the difficulty arising from the curved drum surface.
However, for the sake of simplicity, we returned to the celluloid
sheet and drum. The drum at one end had pins projecting exactly
opposite each picture (Figs. 4 and 5). The pins came in contact
with a stiff primary wire from an induction coil, whereas the secondary

* EDITOR'S NOTE: This sketch and several others illustrating the article were
prepared by Mr. Dickson at the request of the Historical Committee. Originals
are on file at the Los Angeles Museum.

440

W. K. LAURIE DICKSON

[J. S. M. P. E.

FIG. 4.

Courtesy T. Ramsaye

Drum for inspecting microscopic pictures on film strip shown on right.

fed the Geissler tube placed directly over or a little to one side of
these pictures. Owing to the extreme rapidity of the flash of light
from the Geissler tube, when compared with the movement of the
drum of pictures, the images appeared to stand still and were sharply
defined.

I have not mentioned some of my earlier failures, such as the use
of a vertical disk which, however, being flat, got rid of the distortion
of the drum. Also, I need not go into a detailed description of my
glass drum illuminated intermittently from the inside by a back-
and-forth straight shutter.

I was glad to get away from drums, disks, etc., and a hopelessly

FIG. 5. Sketch of images on Carbutt film wound around drum and viewed
with the aid of light from a Geissler tube.

Dec., 1933] BRIEF HISTORY OF THE KINETOGRAPH 441

limited number of pictures, looking forward some day to getting
decent lengths or strips of film from Blair or Carbutt.

EXPERIMENTS WITH SHORT FILM STRIPS

My next attempt, after abandoning drums and the like (early 1888),
was to proceed with narrow strips of Carbutt celluloid, 18 inches
long, notched on the top, and impelled intermittently by a clock
escapement movement (Fig. 6). A rotating shutter and a P/Vhich
focus lens were used. The pictures were l /\ inch square. This tenta-
tive test or experiment seemed to be leading me in the right direc-
tion, as will be shown. On trying to join these strips, the usual
trouble was that the joints stuck in the frame or open guide, which,
however, we made as springy as possible.

While wandering through our museum of showcases containing
many hundreds of models of Edison's inventions, I caught sight of

FIG. 6. Sketch of Carbutt film with notched edge.

his perforated paper automatic telegraph. By the end of a week
my capable mechanic, William Heisse, had made me a perforator
which made two round holes to a picture. The sprocket drive had
a single row of sprockets to fit the newly punched Carbutt strips.

The escapement movement then in use was soon found to be
much too slow to satisfy persistency of vision in the stopping and
starting which we found imperatively necessary in the endeavor to
get a quick change and a long rest. So we adopted the Maltese cross,
and after some modifications found it to answer our purpose very
well (Fig. 7). In less than a month we had a good working camera.
This occurred in the autumn of 1888.

The pictures were taken horizontally, but were only l / 2 inch in
size on Carbutt celluloid strips. Though the strips shot through

442 W. K. LAURIE DICKSON [J. S. M. P. E.

the gate at a good rate, we didn't mind how often we threaded up
to enjoy our success. The longest we could make was about 40
inches, or 3 joints of 14 inches each. This apparatus was finished,
but neither Carbutt nor Blah- could supply us with thinner or longer
strips.

EASTMAN'S FILM AND ITS APPLICATION

Toward the close of the year 1888 it was rumored that the Eastman
Company was experimenting on a new product for their cameras, and
that it would be shown at the New York Camera Club by Mr. Geo.
Eastman's representative. At the end of the meeting, which I
attended, I approached the demonstrator, explained what we wanted
and asked for the 2 by 4 inch sample to show to Mr. Edison. The

Cjt&MMu*

MMMflj

FIG. 7. Types of intermittent movements used 1887-1890.

representative quickly grasped the situation and its great possibilities,
and invited me to come out to Rochester to see Mr. Eastman, which
I did the next day. I knew then that we should reach our goal if
the Eastman Company could supply this new product in good lengths.
When I showed Mr. Edison my new find his smile was seraphic;
"Good," he said, "we can now do the trick just work like hell."

On reaching the Rochester Works, Mr. Geo. Eastman received
me most courteously; and after a long talk he took my arm, guiding
me through his long darkrooms while he touched on some of his
ambitions. He showed considerable enthusiasm as to the new possi-
bilities of the kinetograph requirements.

Dec., 1933]

BRIEF HISTORY OF THE KINETOGRAPH

443

The central table was covered with long sheets of plate glass,
carefully fitted together in lengths. The atmosphere was pregnant
with the fumes of amyl acetate, and I was glad to get out of it. He
then took me into another room and gave me many experimental
samples wrapped in red and black paper for my tests; and invited me
to come whenever I liked. That was the beginning of many years of
friendship, strengthened by our mutual interest. Mr. Eastman was
at all times ready to concede to my wishes, supporting me in my
efforts to attain my great purpose for Mr. Edison.

On my return to Orange, my assistant and I were soon able to test
some of these short samples of which we had a
good supply, thanks to Mr. Eastman's gener-
osity. However, we found that we should re-
quire greater sensitiveness and less coarseness.
The silver grains were too apparent when mag-
nified.

A few weeks elapsed before I saw Mr. Eastman
again and explained our difficulties, which were
remedied principally by reducing the coarse-
ness of the silver bromide. This change proved
most satisfactory. When perforating the film,
however, the sprocket wheel often broke through
and tore the film. I had to ask Mr. Eastman
whether he could make his base tougher and
less brittle, which he did.

Meanwhile we had to use this rapid nega-
tive for our positive prints ; and although they
lacked pluckiness, we partly overcame it by us-
ing potassium bromide in our developing bath
to reduce its sensitiveness. This caused me to
apply again to Mr. Eastman for a less sensitive product or emulsion
similar to that used in lantern slide work, which we ultimately re-
ceived.

We were, however, much troubled with "frilling," and often a
gelatinous mass of pictures was left at the bottom of our developing
or fixing troughs, while the base remained on the drum, a situation
which we found very trying and necessitated further conferences on
this matter. Mr. Eastman, however, managed to overcome this

* EDITOR'S NOTE : This scene was made apparently on film samples supplied
Mr. Edison by Mr. Eastman before the Kodak product was announced publicly.

FIG. 8. Print
from one of first
successful Edison
negatives on East-
man film (Approxi-
mate date. May,
1889).*

444 W. K. LAURIE DICKSON [j. s. M. P. E.

difficulty in part. An early film of a horse shoeing scene (May, 1889)
shows partial frilling (Fig. 8).

About this time we received six rolls of improved negative film
50 feet long, and later some slow positive film. All these samples
and experiments were made exclusively for us by Mr. Eastman, who
took an ever-increasing interest in what we were doing.

To return to the work in hand, we had to devise certain essentials,
such as a circular film cutter or trimmer, a perforator, a clamp with
steady pins to fit the punch holes, to use in joining the films with a
thin paste of the base dissolved in amyl acetate, which, I suppose, is
still commonly used. Room 5 at the Edison Laboratory was used
principally for our photographic experiments and mechanical work.
The precision workshop was located near our room.

To take these photographs or strips, our camera or kinetograph
had to be carried down to a small improvised platform placed against
our ore-milling outhouse. A bright sunny -natured Greek, Sacco
Albanese by name, was one of my very earliest victims, figuring
mostly in the l / t -mch, and later in the * /2-inch, pictures. Draped in
white, he was made to go through some weird antics. (See Fig. 4.)
Some time later, early in 1890, I persuaded Mr. Fred. Ott,
Edison's chief mechanic, to give me an exhibition of sneezing.

The famous firms of Bausch & Lomb and Messrs. Gundlach, of
Rochester, N. Y., produced some fine lenses for the work, and I have
always felt that they should be brought into this early history of the
kinetograph for the perfection of their work and their patience with
me and my demands. The lenses were at last standardized as
2Vs-inch mean focus. With such lenses we made our second kineto-
graph.

Our first outside subject was the dancing and wrestling bears,
which we took early in 1889 in the open yard of the Edison Labora-
tory, long before any studio was erected. I found these performing
bears on the main street of Orange, N. J., and persuaded their Italian
owners to follow me to the laboratory to have their photos taken.
After a few rehearsals my assistant was ready with a pieced sample of
film and the picture was soon made.

My drawer full of samples and tests was destroyed after I left
America, and there remain only a few scraps of originals and repro-
ductions and my booklet The History of the Kinetograph, etc. I may
add also that the studio and the "Black Maria" were razed to the
ground as of no historical interest, as it was deemed by my successor.

Dec., 1933] BRIEF HISTORY OF THE KlNETOGRAPH 445

The "Horse Shoeing Scene" of May, 1889, was taken outside with
an improvised background, as were other scenes taken prior to that
date. (See Fig. 8.)

SYNCHRONIZATION OF THE EDISON PHONOGRAPH AND KlNETOGRAPH

In the midst of this work Mr. Edison went abroad, leaving me his
instructions. I accompanied him to the steamer. As the boat
glided out I saw Mr. Edison leaning over the railing, his fists to
his eyes to imitate the viewing of the pictures in our experimental
kinetoscope. I understood the pantomine to mean that I was to
have the kinetoscope completed before his return. A rough model of
the instrument was constructed.

The gearing up of the kinetograph with the phonograph kept us
busy day and night, experimenting until both were done in time. A
smaller sprocket wheel had to be made for the row of perforations
to fit the teeth better, owing to the shrinkage resulting from develop-
ing and fixing the negative film, which we had to use for all purposes.

On my return from the boat I persuaded my friend Mr. Charles
Batchelor, who was in charge of the laboratory, to build an outside
studio to my specifications, combining a sliding glass roof to let in
the sunlight unobstructed, for photographing kinetograph subjects
before a black or other suitable background.

Attached to this room, which was about 18 by 20 ft. in size, were
two darkrooms one for punching, trimming, joining the films,
and printing the positives; the other for developing, fixing, washing,
and "glycerining" * the films. These operations were done by using
large, black, enameled drums adjustably suspended at each end when
immersed in long, shallow troughs.

The films were spirally wound around these drums and the ends
clamped to hold the film in place. When deemed to be thoroughly
developed, the drum was carried to a similar trough to revolve in
water coming from a spray over the length of the film or drum. The
used water was carried away by an overflow from the trough (Fig. 9).
The film was then carried to the fixing trough and back to the washing
arrangement, thence to the glycerine trough, and dried before a fan
while revolving on the motor-driven drum.

Method of Making Prints. As to our method of printing negatives,
I had a large 8- or 10-inch sprocketed drum made, geared to run

* A bath of glycerine and water, used to render the film more flexible.

446

W. K. LAURIE DICKSON

[J. S. M. P. E.

slowly, over which the films came in contact, the unexposed film being
under the negative and the pins engaging both films. A small pea-
lamp and reflector were placed above the negative. A square of
ground glass was interposed between the light and the film, and the
light was regulated by a small slide resistance to give the right ex-

FIG. 9.

Printing and developing equipment
used at Edison laboratory.

posure. Two spools on each side were used, geared to pick up the
negative and positive films (Fig. 9).

Above these darkrooms I had another room for projecting tests
with the kinetograph and phonograph (Fig. 10). Having succeeded
in devising a very simple method of doing this, I hastened to get my
assistant to take a short film, using me as the subject, hi combina-
tion with the phonograph, to prove that Mr. Edison was right in
supposing such a thing could be done as predicted in his preamble in
the History of the Kinetophone, etc.

Dec., 1933]

BRIEF HISTORY OF THE KINETOGRAPH

447

Mr. Edison's return to his laboratory took place Oct. 6, 1889.*
Within the hour I had him by the arm and led him to the new
studio and kinetophone exhibit, on which we had been working day
and night. On seeing the studio, Mr. Edison asked, "What's that
building?" I explained its necessity. "Well, you've got cheek;
let's see what you've got." We went in. Gradually his face lit up,
the clouds of disapproval were dissipated and finally dispelled. I had

FIG. 10. Projection room (about 1894), showing phonograph
attached to kinetoscope for synchronization of sound and picture.
(Century Magazine, 18, 207, 1894).

placed him in a chair in the upper projecting room to witness his
first "talkie," or exhibit, of the kinetophone. For a wonder, the
exhibition was good. No breakdown of the film occurred nor did the
Zeiss arc lamp sputter. There was much rejoicing. Edison sat with

* It has been said that Mr. Edison sent me long, weekly cables and letters dur-
ing his absence, to instruct me further as to what information he had gathered
when in Paris from others working on these lines ; which, of course, was absolutely
incorrect. I wish to emphasize the fact that Mr. Edison never during his ab-
sence communicated with me either by cable or letter.

448 W. K. LAURIE DICKSON [j. s. M. P. E.

the eartubes to the phonograph. My assistant started the arc lamp
and removed the metal sheet interposed between the arc and the film.
The phonograph motor controlled the projecting kinetograph.

I was seen to advance and address Mr. Edison from the small 4-
foot screen; small, because of the restricted size of the room. I
raised my hat, smiled, and said, "Good morning, Mr. Edison, glad
to see you back. Hope you like the kinetophone. To show the
synchronization I will lift my hand and count up to ten." I then
raised and lowered my hands as I counted to ten. There was no
hitch, and a pretty steady picture. If the pictures were steady
in the taking, why not in the reproduction on the screen ?

A rough description of the method adopted to synchronize the two
instruments may be useful. The only modification made to the

jfTfi^ 77TT, -A

L. . ,,/ -

FIG. 11. Sketch of arrangement used to synchronize phonograph record and
motion picture film.

kinetograph, was to place a ratchet wheel at one end of the driving
shaft. Thus one end of the shaft held the sprocket wheel which en-
gaged the perforated film, and the other end held a magnetic escape-
ment device, which was controlled and timed through a relay and
battery from an extra commutator collar on the phonograph motor
shaft. The impulses electrically received through the ratchet wheel
were spaced at l /z-inch intervals for each phase or picture (Fig. 11).
This arrangement gave very good synchronization and was extremely
simple though a little slower than when using the Geneva movement
which, of course, was put temporarily out of gear for these kineto-
phone demonstrations.

- A little later, about a year, I fancy, Mr. Edison devised an ingenious
stopping and starting device, which took the place of the Maltese
cross or Geneva movement for a time. A horizontal iVj-inch disk

Dec., 1933]

BRIEF HISTORY OF THE KINETOGRAPH

449

FIG. 12. Interior of first studio (1889), showing models 1 and 2 of the
kinetoscope. (From History of the Kinetoscope, etc., by W. K. L. Dickson
and Antonia Dickson. A. Bunn, N. Y., 1895.)

with one slot at its edge was centered on a shaft to run continuously.
A disk of similar size with a nose or pin projecting and running verti-
cally was driven forward on a slip collar. The nose rested and slid

FIG. 13. Interior view of the kinetoscope (Talbot, F. A. : "Moving
Pictures," Lippincott's, Philadelphia, Pa., 1923).

450

W. K. LAURIE DICKSON

[J. S. M. P. E.

along the slotted horizontal disk; and as the vertical disk revolved,
the nose or pin in the horizontal disk would fall rapidly through the
slanting slot (Fig. 7). The film was therefore moved along, since it
was engaged at the other end of the shaft by means of a sprocket
wheel.

I waited for Mr. Edison's further approval of my model of the
kinetoscope that day on his return, before going ahead with a standard
model. He was quite enthusiastic, and the next day, Oct. 7, 1889,

A >> ft ft

FIG. 14.

Evolution of standard film width and
picture size.

he brought in several visitors among them Samuel Insull, to see the
"Wonderbox." A photograph of the studio interior taken a few
weeks later shows both the No. 1 and 2 kinetoscope models (Fig. 12).
AlthougJi this apparatus has often been described, I will venture to
repeat a description.

THE FINISHED KINETOSCOPE

In the first place, the film or positive which was 47 feet long, ran
continuously and not intermittently as in the kinetograph.

Dec., 1933]

BRIEF HISTORY OF THE KINETOGRAPH

451

The kinetoscope mechanism was built up on a small platform, and
consisted of an open gate through which the film ran, drawn by the
sprocket wheel (Fig. 13). A 10-inch shutter revolved at high speed.
The shutter was slotted in one place giving a 3 /ie-inch opening to
admit the light from a small electric lamp after passing through the
film. The light rays from the 8-volt lamp were converged by means
of a reflector, causing the rays to cross through
the Yie-inch shutter opening to obtain a maxi-
mum of light.

The film, after leaving the gate and sprocket
wheel, passed down and up over several velvet
covered rollers or spools in an endless band. The
last spool was weighted and controlled by a
spring to take up any slack. The film used in
the first model had one row of perforations. The
pictures were x / 2 inch in width (Figs. 14 and 15),
and the shutter opening was so set as to meet a pic-
ture centrally. . To increase the magnification of
the image, I further devised a two-eye plano-
convex lens to increase the size of the image. As
this whole device was approved by Edison, I
forged ahead, and soon was using a larger film
perforated along both edges. The mechanism
was driven by a small 8-volt motor, using
storage batteries. The manufacture of this
device was started in 1893 and was soon carried
out in quantities for exhibitors.* Meanwhile,
the projecting kinetograph remained in status quo,
to my great regret.

At the end of the year 1889, I increased the
width of the picture from Va inch to 3 / 4 inch,
then, to 1 inch by 3 / 4 inch high (Fig. 15). The
actual width of the film was ! 3 / 8 inches to allow for
the perforations now punched on both edges, 4 holes to the phase or
picture, which perforations were a shade smaller than those now in
use. This standardized film size of 1889 has remained, with only
minor variations, unaltered to date (Fig. 16).

* The first commercial showing of this device was at the Holland Brothers
Peep Show Parlour, 1155 Broadway, New York, N. Y., April 14, 1894 ("History of
the Motion Picture" by T. Ramsaye, Simon & Schuster, N. Y., 1925).

FIG. 15. Print
from length of early
narrow width film
made with kineto-
graph (perforations
cut off when first
reproduced in book
form).

452

W. K. LAURIE DICKSON

[J. S. M. p. E.

The demand for films for the kinetoscopes was somewhat of a prob-
lem as the machines were being manufactured. Just at this time I
was switched off to carry out some further experiments on magnetic
ore-milling machinery. However, I never failed to remind Mr.
Edison of his risk in not patenting all these "movie" devices. I
think it was Mr. Frank L. Dyer who at last persuaded him to move in
the matter, after I had talked it over with him. This, however, was
not done until a year or so later, while others anticipated us by a

-U.vtvbrorw.xy
dfv;v\i> a.S<c0* -

FIG. 16. Types of mechanisms and film used in the
kinetograph.

patent date though we could prove priority of invention and actual
working models, etc.

THE MAKING OF COMMERCIAL KINETOSCOPE PRINTS

On my resuming movie work (which I did off and on), I had to do
my best to secure attractive subjects for the kinetoscopes. It was

Dec., 1933] BRIEF HISTORY OF THE KINETOGRAPH 453

only at certain times of the day that I could get such pictures in the
studio with its glass roof slid back, so I decided in 1891 to build
another studio revolving on wheels to meet the sun at all hours, which
some wag later dubbed the "Black Maria."* This had a stage at one
end, backed by a tunnel to give velvety black effects, and at the other
end was a darkroom fitted to change the films.

The kinetograph camera was fixed on an adjustable table which
we could wheel out on rails for focusing close-ups or whole stage
effects. The roof was made to fold back to admit the sun after the
building had been moved around (Fig. 17). Here we took most of the
pictures, although some were made at the old studio and a few in the
open.

The outside pictures were taken against a long or very wide gray
painted wooden wall so that when the subject moved out of the pic-
ture we would still follow it by keeping it centrally framed against an

FIG. 17. "The Black Maria" motion picture studio; designed by Dickson
in 1891, completed Feb., 1892.

even, gray background, as was necessary when photographing Duncan
C. Ross' Horsemanship or Mounted Sword Fencing.

The camera end of the "Black Maria" used to be swung around and
such pictures taken through a small window inside the darkroom.
The lighting and stage were so great an improvement, that I repeated
several of the subjects I had taken earlier in the 1889 studio, such
as the organ grinder and monkey, Sandow, and some others. I can
give only an inadequate list of subjects, most of which were taken in

* Completed on Feb., 1892, at a cost of $637.67.

454 W. K. LAURIE DICKSON [j. S. M. P. E.

the "Black Maria," built in 1891-92. A partial list of subjects taken
at Edison's laboratory between 1889 and 1895 follows:

Trick Dog Teddy and other Dog and trick Cats.

Madame Bertoldi, contortionist.

The Gaiety Girls.

Colonel Cody's (Buffalo Bill) Shooting Skill.

Colonel Cody and his Sioux Indians.

Sioux Ghost Dance.

Sandow in Feats of Strength.

Texan Cowboy Throwing Lassos (in the open).

Alcide Capitaine.

Mexican Knife Thrower.

Madame Armand Ary.

Fencing Bout Experts.

John Wilson, the Tramp.

Boxing Cats.

Sheik Hadji Tahar Summersaults, etc.

Walten and Slavin (long and short) comedians.

Japanese Dancers.

Chinese Opium Den Police Raid a comic.

Milk White Flag (a play).

These subjects were taken on our full width film, having double
perforations, four to a phase as used to this day, and standardized by
me for Edison at the end of 1889.

In conclusion, when I left Mr. Edison's laboratory in 1895, having
accomplished the problems assigned to me, I joined my friends
Messrs. H. N. Marvin, E. B. Koopman, and Herman Casler* to
carry out a new method of reproducing motion by means of a pack of
cards, which I had devised shortly after I had left Mr. Edison. This
device we called the "Mutoscope." The cards were viewed both as a
peep show and as booklets given out by manufacturers for publicity
purposes. A syndicate was immediately formed, and our master
mechanician, Mr. Herman Casler, worked out his famous "punch as
you go" taking camera.** When it was ready, I managed to secure
some very stirring pictures.

* The Biograph Co., sometimes known as K.M.C.D. (after the initial letter
of the founder's names) was formed in the summer of 1896. Their first commercial
showing was on Oct. 13, 1896, at Hammerstein's Olympic Music Hall, in New
York.

** The Biographic camera had an intermittent movement consisting of a double
mutilated, rubber-covered roller. When portions of the roller having the
thicker radii came together on revolving, the film was moved forward.

Dec., 1933] BRIEF HISTORY OF THE KINETOGRAPH 455

Later, I secured pictures with this equipment in London, of
Queen Victoria's Jubilee; in Rome, of Pope Leo XIII; and in Africa,
of the Boer War. These films were exhibited in London and the
United States. The pictures of the Boer War were shipped back
periodically and exhibited first at the Palace Theater in London.

In conclusion, it is with considerable pleasure that I look back on
these days and night sat Edison's laboratories; of strenuous work,
defeat, and triumph. My friendship and close association with
Edison will always remain as a happy memory shadowed by his loss.
I still watch with keen interest the development of this great industry
Edison's dream materialized.

BIBLIOGRAPHY

"The Life and Inventions of Edison," a series of 124 articles by A. and W
K. L. Dickson. gassier' s Magazine, 3 (Nov., 1892), p. 3; et seq.

"Edison's Invention of the Kineto-Phonograph," by A. and W. K. L. Dickson.
Century Magazine, 48 (1894), p. 207. Also other articles for the Century and
Harper's Magazines.

"History of the Kinetograph, Kinetoscope, and Kineto-Phonograph," by A.
and W. K. L. Dickson. Albert Bunn, New York, N. Y., 1895.

"The Biograph in Battle," by W. K. L. Dickson. Unwin, Paris and London,
1900.

A SIXTEEN-MILLIMETER PORTABLE SOUND-ON-FILM
PROJECTION EQUIPMENT*

C. R. HANNA,** P. L. IRWlN.f AND E. W. REYNOLDS f

Summary. A portable sound-on-film projection equipment using 16-mm. film is
described. The film is standard with the exception that one row of sprocket holes is
omitted to provide space for the sound track. The projector is only slightly larger than
the average silent picture projector. A detailed description of its mechanical, electri-
cal, and optical features is given.

The complete equipment is mounted in three carrying cases, one for the projector,
one for the amplifier, and one for the loud speaker and screen. The projector case
serves also as a sound-proof housing when the equipment is in operation. The re-
wind, splicer, cables, spare tubes and lamps, and the films are located in the case for
the loud speaker and screen. Each of the carrying cases weighs approximately 40
pounds, making the total weight of the equipment 120 pounds.

For some time there has been a need for a small projection equip-
ment suitable for home or portable use. Until recently, efforts to
obtain a relatively light equipment have been directed, for the most
part, toward the disk type of machine ; but because of the necessarily
large size of the wax record none of these equipments has been very
small. This paper describes a complete sound-on-film projection
equipment, using 16-mm. film, the total weight of the equipment
being 120 pounds. The apparatus to be described was developed in
1930 to meet the needs of the Westinghouse Company for educa-
tional purposes within its organization and for advertising.

The following requirements were sought in developing this appara-
tus: first, it should be simple, rugged, and easily operated by the
average person; second, the quality of sound reproduction should
be equal to that of the modern radio or electric talking machine.

THE FILM

The film is the standard 16-mm. film, with the exception that only
one row of sprocket holes is employed so as to leave room for the

* Received July 1, 1933.

** Research Laboratories, Westinghouse Electric & Mfg. Co., Pittsburgh, Pa.
t Formerly with the Westinghouse Electric & Mfg. Co.
456

16-MM. PROJECTION EQUIPMENT

457

sound track on the opposite edg^,of the film. Fig. 1 shows an enlarged
section of sound film of this type. The speed of the film is 24 frames
or 7.2 inches per second, corresponding to the same number of frames
or 18 inches per second in the 35-mm. film. At that reduced speed
it is, of course, a great deal more difficult to obtain response at the
higher frequencies, because of the reduced length of wave along the
film. For example, a frequency of 5000 cycles corresponds to a
wavelength of 0.0014 inch with the small film as compared to 0.0035
inch with the 35-mm. film. It has been found possible, however,
with care in design and focus of the optical systems, and by providing
adequate mechanical filtering for ob-
taining uniform film velocity, to attain
very good quality of reproduction at
the reduced speed.

THE PROJECTOR

Fig. 2 is a photograph of the front
side of the machine. The two reels
are mounted coaxially at the top, and
the three feeding sprockets are all
mounted on one shaft below. The path
of the film is as follows: after coming
off the back reel, it passes around the
sprocket (a, Fig. 2) nearest the center
plate, and thence to the picture gate
with a loop, which is skewed one posi-
tion toward the observer so as to be in
line with the middle feed sprocket. An
intermittent claw at the bottom of the

picture gate moves the film at that point, after which the film takes
the form of a horizontal loop at the bottom of the machine before
passing vertically upward through the sound gate. From the
sound gate it passes around two rollers, one of which is connected
to a free flywheel on the opposite side of the center plate. The fric-
tion between the film and the roller drives the flywheel with-
out slippage, and thus the effective mass of the film is increased.
The middle sprocket, which pulls the film at this point, is flexibly
mounted on its shaft by means of a helical spring inside the sprocket.
After passing it, the film takes the form of a skewed loop, shifting one
position toward the observer and under the front sprocket, which is a

FIG. 1. 16-mm. sound film.

458 HANNA, IRWTN, AND REYNOLDS [J. S. M. p. E.

hold-back sprocket (b, Fig. 2). This places the film in line with the
front or take-up reel.

Such an arrangement of film feeding mechanism is very compact
without, in any way, making the threading operation more difficult
than it would be with sprockets and reels on separate shafts. Other
advantages of the arrangement will be described later.

Fig. 3 shows the picture gate and intermittent movement. The
gate has flexible steel shoes under the tension of a cantilever spring
at the bottom for holding the film against the slightly curved back-

FIG. 2. View of front of projector.

plate. Such a construction causes the pressure to be uniform
throughout the arc of contact between the shoes and the film and
serves, also, to allow the film to be pushed forward by the inter-
mittent claw in case the perforations do not register at the start.

The intermittent movement consists of a heart-shaped cam of
novel design surrounded by a square follower. Two adjacent sides
of the follower are in the form of cantilever springs, which press
against the cam so as not to allow any lost motion. These cantilevers
are deflected slightly outward by the cam so as to provide a spring
force which is in excess of the maximum force of inertia and friction

Dec., 1933]

16-MM. PROJECTION EQUIPMENT

459

of the reciprocating parts and the film. So constructed, the inter-
mittent operates very quietly. It is also cheap to manufacture be-
cause the tolerances of the cam may be greater than if the usual
type of follower were used. The cam is designed to move the claw
down in a 75-degree movement of the cam with respect to the fol-
lower. This is reduced to an actual 66-degree movement of the
cam, however, because of a 9-degree clockwise movement of the
rocker and follower opposite in direction to the counter-clockwise

FIG. 3. Picture gate and intermittent movement.

cam movement. The placing of the rocker between the cam and
the claw is responsible for the opposed rotation. It results also in
another advantage: namely, that the radius of curvature of the orbit
of the claw may be made any desired value. For example, if the
rocker is midway between the center of the cam and the claw, the
orbit of the claw will be a straight line for a considerable distance.
In the design of this machine, the rocker is placed one-third the
distance between the center of the cam and the claw with the result
that the curvature of the claw orbit approximates the curvature of
the picture gate.

A two-blade shutter is mounted behind the center plate on the same
shaft as the cam. The angle of the blades is 75 degrees, so as to

460

HANNA, IRWIN, AND REYNOLDS

[J. S. M. P. E.

occult the light completely during movement of the film. Fig. 4 is
a schematic diagram of the picture optical system. Condenser C\
focuses an enlarged image of the filament of lamp L at the lens Ct,
which in turn focuses the evenly illuminated surface of C\ on the film
F after reflection from the mirror M . The condenser C 3 re-images the
enlarged filament image at a point midway between the two lenses
of the objective 0. The shutter 5 is placed close to C\, which is a
conjugate focus of the film in approximately a 1 to 1 ratio. This
allows complete shutting-off of the useful illumination with a shutter

FIG. 4.

C 3 F O

Picture optical system.

movement no greater than the height of one frame, namely, 0.3 inch.
This type of system was necessary because of the fact that there is a
fairly great distance between the lamp and the film. Aside from the
loss of light at the mirror and at the surfaces of the two extra lenses
in the condenser system, this arrangement has as great an optical
efficiency as any short optical system.

Fig. 5 is a close-up photograph of the sound optical system and
sound gate. A heavy filament 10-volt, TVs-amp, lamp is used for
the source so as to allow it to be operated on a 60-cycle line without
objectionable hum. A small condenser lens immediately in front
of the lamp illuminates a slit, which is reduced in a 4 to 1 ratio by the
objective at the left end of the optical system barrel. The image
at the film is 0.0005 by 0.050 inch. Convenient arrangements for
orienting the slit and for adjusting the image focus are provided.

Dec., 1933] 16-MM. PROJECTION EQUIPMENT 461

Fig. 5 also shows in detail the mechanical filtering system for main-
taining a uniform velocity of the film past the light image. The
sound gate is similar to the picture gate, but has a smaller radius of
curvature so as to hold in perfect focus the unsupported sound track
at the edge of the film. The use of flexible shoes under tension for
pressing the film against the curved back-plate is of greater impor-
tance here than in the picture gate, because such construction pro-
vides very smooth friction, with the result that the filtering system
may be of smaller dimensions than would otherwise be the case.

FIG. 5. Sound gate and optical system.

This friction provides tension on the film of sufficient magnitude to
cause adequate driving friction between the film and the flywheel
roller without the use of pressure rollers for holding the film against
the flywheel roller. The avoidance of pressure rollers is very impor-
tant, because they would produce torque pulsations which only a
much larger flywheel could resist. As was stated at the beginning,
the center sprocket which pulls the film at this point of the film path
is resiliently connected to its shaft. The purpose of the flexible
mounting is twofold: first, if there are any irregularities in the speed
of the shaft, they will be readily absorbed by the spring; second, if

462

HANNA, IRWIN, AND REYNOLDS

[J. S. M. P. E.

the sprocket teeth do not exactly fit the perforations of the film be-
cause of shrinkage, the sprocket, because of its flexibility and its
small inertia compared to that of the flywheel, may readily adjust
its motion to that of the uniformly moving film. Thus the sprocket
moves with a steady motion, plus a pulsating motion of exactly the
right character to transmit substantially uniform motion to the film ;
in other words, the flexible mounting of the sprocket prevents shaft
and sprocket tooth pulsations from reaching the film.

As in all other filtering systems made of springs and inertia mem-
bers, it is necessary that the system be damped to prevent oscilla-

FIG. 6. Rear view of projector.

tion. For that purpose, a lubricated leather brake is placed inside
the sprocket so as to provide viscous damping for relative movements
between the sprocket and its shaft. The purpose of the front or
hold-back sprocket in the system is to form a film loop for isolating
the spring sprocket from any pulsations which might be transmitted
through the film by the take-up reel.

With this combination of refinements, it is possible to maintain
extremely constant film speed with the use of a flywheel only 5 inches
in diameter and 1 inch in thickness. The uniformity of speed is
such that with good piano records, for example, there is no noticeable
"twang" or "fuzzing" of high notes during reproduction.

Dec., 1933]

16-MM. PROJECTION EQUIPMENT

463

Fig. 6 shows the rear of the machine. The motor is a single-phase.
110-volt, 60-cycle induction type with an external resistance for
obtaining split-phase in the stator. Its pinion drives the large
double idler gear, whose bearing is the outside of the bronze bushing
through which the flywheel shaft passes. The gear on the sprocket
shaft meshes with the smaller of the two gears of this double idler,
and rotates at 180 rpm. The intermittent gear meshes with the
larger gear of the idler and rotates at 1440
rpm. A spring take-up belt passes around
a small pulley which is integral with the
idler gear. Although the flywheel appears to
be gear driven, it will be remembered that it
is free-floating in its bearing, and driven only
by the friction of the film as the latter passes
around the roller on the opposite side of the
assembly.

In this construction it will be seen that there
are but five gears in all, and that the motor
is mounted upon the base, where its vibration
will cause the least harm. Lubrication is
effected through oiler wicks on the sprocket
shaft, flywheel shaft, and the intermittent
shaft; and through oiler holes on the motor,
rocker arm shaft, and the rocker arm slide.
The intermittent cam and the spring sprocket
damper are packed with vaseline and require
only infrequent attention.

Although no provisions are made for film
guides and loop formers, an arrangement for
automatically threading the machine is pro-
vided for those who do not care to thread by

hand. Fig. 7 shows a form of film splice, which is made by a special
punching on the trailing end of one film and the leader of the next,
so that a film already in the machine may be used to lead through
a subsequent film. To make convenient use of this splicing ar-
rangement an automatic stop is provided, so as to halt the machine
at the instant the end of the film leaves the loaded reel.

Fig. 8 illustrates the electric circuit for starting and stopping the
machine manually, and also the arrangement for automatically stop-
ping the machine in the manner described. A friction brake mounted

FIG. 7. Film splice
for self-threading.

464

HANNA, IRWIN, AND REYNOLDS

[J. S. M. P. E.

on the back reel provides the proper amount of torque for holding
together a pair of relay contacts so long as the reel is rotating. In the
diagram this pair of contacts is marked Holding Switch on Reel.
The instant the reel stops, the contacts open and the machine shuts
down. A new reel may then be inserted on the shaft and its leader
attached to the trailing end of the preceding film. After being led
through the machine, the splice is unhooked and the film is at-
tached by hand to an empty reel placed at the front. At the end of a
performance, a short strip of film is led into the machine and left
there for threading the first film of a subsequent showing.

110 V AC

110 V A.C

PROTECTOR LAMP

MOTOR

HOLDING SWITCH
ON REEL

FIG. 8. Control circuit.

It will be seen from the circuit diagram that both motor and pic-
ture lamps are turned on at the same time, this arrangement obviat-
ing the necessity of a fire shutter.

THE COMPLETE EQUIPMENT

Fig. 9 is a photograph of the projector mounted in a sound-proof
carrying case. A chimney at the top of the lamp house and ventilat-
ing holes in the side of the case provide adequate cooling for continu-
ous operation without a fan. The chimney is lined with sound ab-
sorbent material to reduce noise transmission. When the complete
equipment is set up for operation, the amplifier case is rigidly fastened

Dec., 1933]

16-MM. PROJECTION EQUIPMENT

465

to the back of the projector case so that the photo-cell will be in line
with the sound optical system for any angle of tilt of the equipment.
The light passes through holes in the two carrying cases, the photo-
cell being mounted on the amplifier. The latter is operated by any
110- volt, 60-cycle supply, no batteries being used at all.

The screen is 15 by 20 inches, but for large audiences a separate
roll screen 3 by 4 feet is used. The case in which the loud speaker
and the small screen are mounted has a removable back with rewind
and splicer mounted upon it. In the same case are provisions for

FIG. 9. The projector in sound-proof carrying case.

carrying eight 400-foot reels of film, spare tubes and lamps, and the
cables.

Acknowledgment is made to K. A. Oplinger and C. N. Batsel for
their work in developing recording apparatus and for making the
films, to S. Sentipal for the design of many features of the projector,
to E. H. Greibach who worked with one of the authors on the design
of the mechanical filtering system, and to K. A. Oplinger and W. O.
Osbon for working out the convenient portable arrangement of the
several components of the equipment.

A LIGHT-WEIGHT SINGLE-FILM RECORDING SYSTEM
FOR NEWSREELS AND TRAVELOGUES*

C. R. SAWYER**

Summary. A portable sound recording system is described, which weighs less
than 100 pounds, is easily operated, and produces high-quality results when used in
combination with a single-film recording camera.

Since the beginning of sound recording for motion pictures, it has
been recognized that the equipment for newsreel work should differ
in many respects from that which would be the most satisfactory for
studio recording. However, the urgency of producing such equip-
ment made it advisable to adopt the available apparatus, with as
little modification as practicable. So in most cases lightness was
sacrificed to reliability of operation.

While at that time it was quite possible to picture the conditions
which would be encountered in the studio, those in the newsreel field
were more difficult to anticipate. Therefore, it was felt advisable to
allow reasonable factors of safety in including apparatus which
might be needed. This was done largely because the newsreel man is
frequently much more distant from his base than is the sound man
in a studio or on location. With the knowledge gained from actual
field experience and with the development of such new apparatus as
the permanent magnet light valve 1 and the moving coil microphone, 2
it was now feasible to improve the Western Electric Newsreel equip-
ments radically from the standpoints of weight, size of equipment,
ease of operation, and quality of recording. In describing this new
equipment in some detail, the improvements made and the reasons
for them will be pointed out.

Power Supply and Motor Drive. The first point to be considered
is that of the power supply, since heretofore it has been the control-
ling factor in the weight of equipments. To permit greater flexibilitv

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Electrical Research Products, Inc., New York, N. Y.
466

NEWSREELS AND TRAVELOGUES 467

in operation, it is almost compulsory that batteries be used for the
power whether it be for the amplifier supply or for the motor drive
of the camera.

The use of a single film for both sound and picture negatives is a
logical practice to follow. Only the one motor for the camera is
required, instead of the two required by the double-film system:
namely, one for the recording machine and one for the camera.
In addition, the motors for an interlocking system are larger and less
efficient. The additional battery required makes the total motor
battery weight of the double-film system probably three times as
great as that for single-film. It must be recognized that the speed
control is not as constant as with regulating apparatus, but experience
has shown that for speech and for most newsreel pick-ups of music the
regulation that can be obtained with a single motor of proper design is
satisfactory.

While for the same power output 1 10-volt motors may be slightly
lighter than 12-volt motors, a 110- volt battery of the size required for
this power drive would weigh very much more than the corresponding
12-volt battery. No particular gain in battery weight, however,
would result in reducing the voltage to the ordinary 6-volt automobile
battery; therefore a 12-volt airplane type storage battery was se-
lected for the motor and sound lamp supply.

Microphone. The introduction of the moving coil microphone has
made possible the elimination of the transmitter amplifier, with a
resultant reduction in weight. This gain would be partially thrown
away were it necessary to mix the outputs of two microphones.
However, it has been found that in newsreel work it is practically
never necessary to "mix," since the background sounds picked up in
the single microphone are usually too loud rather than too weak.
The conditions encountered in sound newsreel work are usually such
as to make the mixing of sound from two or more microphones ex-
tremely inconvenient. Several newsreel sound men have stated that
they have all they can do to operate one microphone well, without
attempting to "mix" two microphones.

In travelogue work where sounds, particularly of descriptive lec-
tures, may have to be introduced, it is usually advisable to "dub"
these at the studio by re-recording, rather than to record them in the
field. This does not mean, however, that cases do not occur where an
additional microphone, located at a different point from the principal
one, would be desirable. The equipment is therefore planned for the

468

C. R. SAWYER

[J. S. M. P. E.

use of either of two microphones which may be selected by a switching
key located at the amplifier.

Amplifier Assembly. We may now consider the amplifier assembly
to which the input and output cords are connected. The earlier
shortcomings of this type of apparatus have been overcome by using
new light-weight transformers developed at the Bell Telephone
Laboratories, by the use of tubes having relatively low power de-
mands, and by carefully designing the circuit from the battery supply
standpoint. As a result, it is possible to include the amplifier and all
its batteries in one case weighing less than 35 pounds, and still leave
room for the storage and transportation of connecting cords, monitor-
ing headphones, and a microphone. This enables the sound man

FIG. 1. Amplifier assembly (rear view).

to carry this one light case to a remote location and have all the neces-
sary apparatus readily available for his pick-up work. The taking
of a talking newsreel picture can be started almost immediately upon
arrival at the scene of an important news event, which might be lost
if there were delays in setting up for operation.

The amplifier assembly consists essentially of four parts, not
counting the cover. The first is the outside case, which has a sepa-
rate compartment for the A -battery opening at the back of the case
(Fig. 1). When the cover of the case is raised (Fig. 2) there are seen,
from left to right, the 5-battery unit which contains six 22V2-volt
batteries, the amplifier unit, and the control unit. This assembly is
15 inches long, 12 inches high, and 6 8 /s inches wide.

Dec., 1933] NEWSREELS AND TRAVELOGUES 469

Amplifier Unit. For convenience, the amplifier unit, weighing
10 pounds, may be removed from the main case and placed upon any
convenient support or carried on the arm. For this purpose, the
amplifier is connected to the batteries in the case by a 10-foot plug-
ended cord, one of the plugs connecting to a jack in the back of the
amplifier and the other to a jack in the control unit. The micro-
phones are connected to two special clamping jacks mounted on the
face of the amplifier unit. Between these two jacks is a small switch

FIG. 2. Amplifier assembly showing cover raised.

which determines and indicates which microphone is connected to the
input of the amplifier.

On the face of the amplifier are also mounted the volume-indicator
meter, a volume-indicator dial switch, a volume-control dial, and a
jack into which may be inserted the plug of the headset used for
monitoring. Access to the five vacuum tubes used in this amplifier is
obtained by opening half of the hinged face-panel; access to the
remaining equipment in the amplifier and to the C-batteries required
by it, may be obtained by unscrewing the four screws holding the
outside case in place.

470 C. R. SAWYER (J. S. M. P. E

Filament current for the five 264-A vacuum tubes used in the
amplifier is derived from a single 2- volt storage cell or two dry cells.
The current to these filaments, which are connected in parallel, is
controlled by a single rheostat in the control unit. The amplifier has
a frequency-gain characteristic which is flat within 2 db. from 40 to
10,000 cycles, the gam is approximately 95 db., and the maximum
single-frequency output approximately 10 db. with respect to six
milliwatts. This output is sufficient for the operation of the perma-
nent-magnet light valve used in this system, and the gain is ample for
practical newsreel work. For newsreel operation, it is usually desir-
able to droop the response below 200 cycles per second by means of a
simple network mounted inside the amplifier.

Control Unit. In the control unit is a single meter with a switch
and resistances so arranged that the sound lamp current, the amplifier
filament current, and the amplifier plate voltage may be read sepa-
rately. The battery supply to the amplifier passes through this
control unit, and rheostats are provided for regulating the filament
and lamp currents. There is also placed in this unit a low-pass filter,
which reduces the high-frequency currents fed to the light valve,
thereby reducing the probability of overloading the valve.

Camera Battery Case. The current for the sound lamp in the modu-
lator unit is provided by the 12- volt battery used for the camera
motor drive. However, for convenience of adjustment, the regulat-
ing rheostat for the lamp current is placed in the control unit, the
necessary leads being brought back to this unit from the motor bat-
tery case. Because of this connection, it is also convenient to carry
the audio-frequency leads through this battery case. The complete
weight of this battery case is 48 pounds and its dimensions are 13 Vs
by 6 8 / 8 by 12 inches. One special 10-ft. cord transmits speech,
lamp, and motor currents from the battery case to the camera
assembly.

Modulator Unit. A proportionate reduction in weight has been
made at the modulator unit by use of the permanent magnet light
valve and by the substitution of a new small sound lamp. Based on
these two improvements, a complete redesign of the modulator unit
(Fig. 3) was made, with the result that the modulator unit and sound
gate, with their mounting plate, now weigh less than 4 pounds. The
mounting plate slides into the back of the camera and can be readily
removed when not in use. Electrical connections to the modulator
unit are made by means of four pin plugs which slip into correspond-

Dec., 1933] NEWSREELS AND TRAVELOGUES 471

ing jacks on the camera. It is assumed that the sound camera will
be provided with a small terminal plate which will also mount the
drive-motor rheostat and possibly a lamp switch, although the lamp
is also controlled by a switch at the amplifier.

The cover of the modulator unit is held in place by a single thumb-
nut, which is permanently attached to it so that it will not be mis-
placed when the cover is removed. This thumb-nut functions with a
screw fastened to a bracket that also helps to protect the sound lamp.
The essential parts of the unit itself are condenser and objective
lenses, light valve, and the lamp, which is mounted upon a separate

FIG. 3. Modulator unit with cover removed.

removable bracket. Two motions suffice to remove this lamp and
bracket, and two more will put in an extra bracket with a prefocused
lamp. This makes it possible to change the sound lamp within a half-
minute in case of an accidental burnout during a take. The light
valve is of the permanent magnet type, with the ribbons perma-
nently clamped once they are tuned. This light valve can be re-
moved and replaced readily, being held in place by a spring latch with
a locking cam to prevent accidental opening. It will be noted that in
this modulator unit there are no screw connections. This, it is
expected, will practically eliminate loose connections in this important
part of the circuit.

472

C. R. SAWYER

[J. S. M. P. E.

The operating period of a set of dry batteries or a charge of the
storage batteries may be of interest. The 2-volt storage battery and
the 135-volt .6 -batteries which are included in the amplifier assembly
weigh five pounds each. These power sources total ten pounds in
weight, and will operate the amplifier for 13 hours. This will permit
the recording of about 35,000 feet of film which, for a newsreel man, is
approximately a month's work. The 12-volt storage battery, which
furnishes power for the sound lamp and the camera driving motor, has

FIG. 4.

Light-weight single-film recording system,
complete with sound camera.

sufficient capacity to operate approximately 2 1 /z hours, which will
permit the recording of about 12,000 feet of film. For normal news-
reel operation it needs to be recharged only two or three times a
month.

The normal sound recording part of the equipment will then consist
of a microphone and its tripod, the complete amplifier assembly, the
motor battery case with camera cord, the modulator unit, the moni-
toring headset, and two connecting cords, one from the microphone to
the amplifier and the other from the amplifier to the battery case.

Dec , 1933] NEWSREELS AND TRAVELOGUES 473

The total weight of this equipment is less than one hundred pounds.
Even when the camera and its tripod are added (Fig. 4), there should
be no difficulty in transporting the system by hand or as personal
baggage by train, automobile, or even by mule-back.

REFERENCES

1 PERREAULT, G. E.: "A Permanent Magnet Light Valve," Bell Laboratories
Record, 10 (Aug., 1932), No. 12, p. 412.

* JONES, W. C., AND GILES, L. W. : "A Moving Coil Microphone for High
Quality Sound Reproduction," J. Soc. Mot. Pict. Eng. XVH (Dec.. 1931), No. 6,
p. 977.

A NON-INTERMITTENT HIGH-SPEED 16-MM. CAMERA*

F. E. TUTTLE**

Summary. A rather simple form of non-intermittent 16-mm. camera that seems
to be especially well suited for high-speed motion-picture photography has been de-
veloped. Film under tension is pulled continuously by a sprocket across a gate that
has an aperture enlarged in the vertical dimension. The image is displaced opti-
cally to follow the film by means of a uniformly rotating plane parallel plate of glass
located between the lens and the film. Framing is accomplished by small uniformly
rotating blades, the edges of which follow the frame lines of the picture down as the
film moves.

The camera can be run at speeds as high as 2500 pictures a second. Two forms
of the camera have been developed for use with the precision liming clock, one designed
for race timing by the Kirby system, having an operating speed of 120 frames per
second, the other using a camera with a maximum speed of 2500 frames per second.

For some time technicians have realized that there are fruitful
advantages to be gained in the study of motion, if only some practical
method were available for revealing to the eye motions that occur too
rapidly to be seen under normal conditions. Probably the first
instrument to be used for the purpose was the stroboscope which,
however, was limited mainly to the study of recurring phenomena.
With the development of sources of intermittent light of great
intensity for illuminating the subject in a manner similar to that of
the stroboscope, high-speed photography has become possible.
However, equipment for such photography has not been made
available to many people interested. Moreover, a mechanism for
driving the film past an aperture, similar to that used in the camera
under discussion, is required.

A number of high-speed cameras of the continuous motion type
have been constructed, a few of which operated successfully; the
remainder either enjoyed short-lived success or failed utterly. Un-
fortunately, they were all too costly or too bulky for general lab-
oratory use. The fundamental principles of some of them were dis-
cussed in a previous paper. 1

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Development Department, Eastman Kodak Company.
474

16-MM. CAMERA

475

Of all the methods of optically displacing an image, the one that
apparently lends itself to the design of a continuous camera is the
method using the rotating plane parallel glass plate. This plate,
rotating uniformly, shifts the image to follow the film for short
intervals remarkably well. This method is used in the high-speed
camera described here.

Fig. 1 shows the essential parts of the camera. The plane parallel
glass plate A rotates about a center perpendicular to the plane of the
paper. Light from the lens G, passing through this plate, is displaced

FIG. 1. Schematic arrangement of the high-speed, non-intermittent camera.

almost linearly for a part of a revolution of the plate. The disk
shutter E then cuts off the light until the opposite face of the plate
is in position to give the necessary approximation of linear dis-
placement of the image for satisfactory definition. The film in the
gate C moves continuously past an aperture, at a rate equal to that
of the image displacement during the open period of the shutter.
The height of the aperture is slightly greater than that of a film
frame. The film is pulled over the gate by the lower sprocket D,
and is held taut by the spring-mounted sprocket D'. B and B' are

476

F. E. TUTTLE

[J. S. M. P. E.

special framing devices which cause the frame line to follow the film
during the exposure period. All moving parts in the camera rotate
at a constant speed. The parts operate at speeds up to approxi-
mately 250 to 300 pictures per second. For higher speeds, as great

FIG. 2. Showing the manner of using the camera with the Western Electric
clock system of splitting seconds.

as 2500 pictures per second, the rotating framing blades B and B'
are replaced by stationary framers with only a slight loss in framing,
and the disk shutter by a barrel shutter F (Fig. 2) of low inertia
and with little tendency toward distortion. The rotating parts are

Dec., 1933) 16-MM. CAMERA 477

limited, then, to the barrel and a glass plate, the two sprockets, and
gears necessary for driving those parts. It has been found that the
majority of high-speed camera subjects are self -framing or may be
made so by controlling the illumination; or, may be framed at the
subject itself.

The field of usefulness of the high-speed camera is increased
greatly when used in conjection with a very accurate clock developed
by the Western Electric Company. The Kirby "two-eye" method 2
employed records simultaneously on each picture the image of the
clock dials and the exposure of the subject. Fig. 2 shows how the
image of the clock dials H, directly below the camera, passes through
the clock lens tube /, is reflected at right angles by the prism J, and
passes through a small section of the glass plate A where it is dis-
placed to follow the film K. The smallest division on the clock dial
is two one-thousandths of a second, making it possible to read it to
one one-thousandth part of a second.

Our own use of the two speed-types of high-speed camera has
shown them to be of great service in studying mechanical problems
and has convinced us that they are powerful tools in making visible
any form of motion. Many subjects of popular trend have also
been photographed, with very interesting and instructive results.

REFERENCES

1 TUTTLE, F., AND REID, C. D. : "The Problem of Motion Picture Projection
from Continuously Moving Film," /. Soc. Mot. Eng., XX (Jan., 1933), No. 1,
p. 3 ; reprinted from /. Opt. Soc. of America, 22 (Feb., 1932), No. 2, p. 39.

2 FETTER, C. H. : "A New Way of Splitting Seconds," /. Soc. Mot. Pict. Eng.,
XX (April, 1933), No. 4, p. 332; reprinted from Bell Telephone Quarterly, XI
(Oct., 1932), No. 4, p. 293.

THE CINE-KODAK SPECIAL*
J. STOEBER, O. WITTEL, AND F. E. TUTTLE**

Summary. The new 16-mm. Cine-Kodak Special and a new tripod made by
the Eastman Kodak Co. are described. Features of the camera include a dissolve
shutter, a turret which carries any two of a series of interchangeable lenses, a reflex
focusing finder, a spring motor with long uniform run characteristics, controlled speed
from 8 to 64 frames per second, interchangeable film chambers having capacities of
100 and 200 feet, single-exposure trip, single-frame-per-turn and eight-frame-per-
turn hand crank or motor drive shafts, and masks for double exposure and effect pho-
tography. The tripod has twist lock legs, and a smoothly operating panoraming and
tilting head.

Ten years ago the Eastman Kodak Company placed on the market
the first 16-mm. motion picture camera. At that time the problem
was to make the taking of home motion pictures certain and simple.
Several sub-standard film cameras had been introduced before, but
none had proved very satisfactory because of the user's difficulty in
obtaining consistently good results. The controlled film reversal
process made relatively certain the achievement of uniformly good
photographic quality, and, when coupled with elementary taking
equipment, promoted in the public the feeling that the taking of
home movies was easily within its ability.

Continued improvement in emulsions and in the reversal process,
together with the advent of Kodacolor, has attracted increasing
numbers of workers in various fields. Engineers, research laboratory
technicians, surgeons, biologists, sales organizations, amateur cinema
clubs, and the like, have all found application for 16-mm. pictures.

Contact with these workers directly and through observations of
processed film has shown that the serious worker is not only desirous
of, but is ready for, more elaborate picture- taking equipment. Com-
pilation of these observations has resulted in the design of the Cin-
Kodak Special, a 16-mm. motion picture camera which is capable of
producing any of the results desired.

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Development Department, Eastman Kodak Co., Rochester, N. Y.
478

THE CINE-KODAK SPECIAL

479

The Cine-Kodak Special is built in two units. One unit con-
tains the spring motor operating mechanism and controls; the other
consists of either the 100-foot or the 200-foot capacity film chamber.
Removing the film chamber from the motor unit and replacing it with
another film chamber containing a different type of film or a different
length of roll is a very simple task which requires but a few seconds.
It is unnecessary to remove the camera from the tripod to accomplish
this change-over. During the operation, not a single frame of film is
fogged because of a shield which must cover the gate aperture before
the chamber can be removed. In order to prevent the loss of film,
the camera drive mechanism is locked until the aperture shield on the

FIG. 1. Cine-Kodak Special with 200-ft. film chamber.

new chamber has been opened. Each film chamber is equipped with a
film footage indicator which shows the amount of unexposed film.

Fig. 1 shows the camera with the 200-foot film chamber and the
operating control panel at the side of the spring motor unit. By
means of the frame speed control A , the camera speed can be varied
from eight frames to sixty-four frames per second, when driven by the
spring motor. B is the spring motor crank. The camera runs forty
feet of film at one spring winding. A warning bell signals when the
spring is nearly tight and again when it is nearly run down. C is an
eight-frame hand crank or electric motor drive shaft. If it is desired
to run more than forty feet of film without interruption, an over-

480

STOIBER, WITTEL, AND TUTTLE

[J. S. M. P. E.

running clutch releases the spring drive and permits the operator to
use this small hand crank after the spring motor has run down. With
the speed control set, the governor furnishes a very nice brake for
hand cranking. With this hand crank the film can be wound back-
wards in the camera for making double exposure shots, overlap dis-
solves^/c. D is a single-frame trip. With the spring motor wound up,
the operator can make continuously as many as fifteen hundred single-
frame exposures for animation work. One frame is exposed with each
revolution of the single-frame hand crank shaft E, which also may be

FIG. 2. Cine-Kodak Special, showing reflex
finder.

used as a motor drive shaft. This feature makes the Cin-Kodak
Special very well suited for the making of growth study pictures and
the like. F is the variable shutter lever. The camera can be oper-
ated with the shutter one-quarter open, one-half, or wide open. Each
change of shutter position, therefore, gives an exposure change equal
to one full lens stop. The shutter on the camera can also be closed or
opened gradually while the camera is running, permitting fades and
overlap dissolves. G is the operating release for the camera.

Dec., 1933]

THE CINE-KODAK SPECIAL

481

The Cine- Kodak Special is equipped with a turret which accommo-
dates two lenses of the interchangeable type. Available for the
Special are the one-inch //1. 9, 15-mm.//2.7 wide-angle lens; two-inch
//3.5 lens; three-inch, 4y 2 -inch; and six-inch //4.5 telephoto lenses.
Each lens carries its own straight-through finder sight.

The top of the Cine-Kodak Special is shown in Fig. 2. H is a reflex
focusing finder which uses as an objective the lens that is in position
for taking pictures. When using this finder, a mirror is brought into

FIG. 3. Cine-Kodak Special with 100-ft.
film chamber mounted on special tripod.

position between the back of the lens and the film, which reflects the
image to a ground glass in the finder. Since the ground glass oc-
cupies the same position in the reflected beam as the film does in the
straight-through beam, it is accurate for focusing and for showing the
field covered by any lens. Starting the camera moves the mirror out
of the path of the light from the lens to the film. / is a mask slot
in the film chamber. A set of six masks is furnished with the camera:
a pair of vertical and a pair of horizontal half masks, a circular mask,

482 STOIBER, WITTEL, AND TUTTLE

and an oval mask. / is a geared footage counter which is essential
for double-exposure work.

In Fig. 3 the camera is mounted on a Cine*-Kodak Special tripod.
Although the tripod has been designed particularly for the Cine-
Kodal Special, it can be used with all types of amateur cameras. It
is extremely rigid, easily adjustable, and very light in weight. The
tripod has vertical and horizontal panoraming heads. The camera
may be pointed straight up or straight down, and locked firmly in any
position. Tension on the panning and tilting heads is adjustable.
The twist-lock legs can be locked at any desired height.

The problem of making a camera with the features of the Special
was complicated by the necessity for keeping the total bulk and
weight to a minimum. The camera is 2 3 / 4 by 5 by 8 inches. When
loaded with one hundred feet of film, it weighs only 9 l /2 pounds.

DISCUSSION

MR. FARNHAM : Does the dissolve occur in a definite number of frames or does
it depend on how fast the dissolve lever is pushed down?

MR. TUTTLE: The dissolve is manually operated. The time that the dis-
solve takes depends on the operator's use of his finger. The shutter is a variable
shutter, adjustable, and the camera can be run with the shutter wide open, half-
open, or quarter-open, allowing the exposure to be changed without changing the
lens stop. It is useful in Kodacolor, where the lens stop can not be changed, and
it saves putting in an adjustable diaphragm or using neutral density filters.

MR. KALLMAN: When you put the mask in, is it visible behind the ground
glass?

MR. TUTTLE: No. It is visible through the straight-through finder in which
each lens carries its own finder. These masks have red tops, which can be seen
through the finder, giving warning as to whether you are doing it right or not
at the time.

MR. MITCHELL: Is the footage dial the type that has a lever resting on the roll
itself and measures the footage of the roll?

MR. TUTTLE: Yes.

MR. MITCHELL: Does it back up when you back up the camera too?

MR. TUTTLE: Yes.

PORTABLE RECORDING EQUIPMENT*
DON CANADY**

Summary. After reviewing briefly the requirements of portable and semi-
portable recorders in general, a semi-portable and a portable recorder employing
the Zetka-Canady glow lamp are described. The features involved in the design
of the types A and Q glow lamps are discussed, with particular reference to the
electrical and optical characteristics of the systems.

Increasing competition among independent producers of motion
pictures has resulted in a demand for two types of sound recording
equipment. For several reasons, not the least of which is a financial
one, the independent producer does not feel warranted in purchasing
two complete recording channels, one a permanent rack type in-
stallation, and the other of the so-called "suitcase," or trunk type,
for location work. This condition has resulted in various combina-
tions, of which the most satisfactory are described below.

(1) Equipment Installed Permanently in Sound Truck. Where
garage space is available immediately adjacent to the studio, this
method practically amounts to employing a mobile recording room,
the equipment being constantly available for location work. A
source of power for operating the recorder and the camera may be
provided for in the truck, or the power may be obtained from com-
mercial lighting mains. The main disadvantage is that the recordist
is usually compelled to work "blind," and in certain types of re-
cording, this may be a serious handicap. The use of an extension
mixer surmounts this difficulty; but since it involves not only
additional equipment, but also additional personnel, it usually
devolves upon the recordist to "do the best he can."

(2) Semi-Portable Equipment. Installations of this type require
that both the amplifier and the recorder be transferred readily from
the studio to the sound truck, and vice versa. The amplifier, which
is built into a carrying case, is installed on a special shock-proof

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Canady Sound Appliance Co., Cleveland, Ohio.

483

484 DON CANADY [j. s. M. P. E.

mounting in the truck. As a general rule, 6- volt storage batteries
supply the power, duplicate batteries and controls being provided
for in the studio and the truck. Experience has shown that this
type of installation is adaptable to probably 90 per cent of the work
encountered.

(3) Semi-Portable, without Truck. With this arrangement, the
power supply and controls are contained in cases, so that the field
of action is extended to cover practically all kinds of work except
those in which weight becomes a deciding factor.

(4) Portable Equipment. In this case, a much lighter and more
compact amplifier is required, although no essentials, either in
performance or operation can be sacrificed. The recorder must
also be of light weight, and the portability of the equipment must
be such that it can be moved for use practically anywhere that a
professional motion picture camera may be taken.

SEMI-PORTABLE RECORDER

The new semi-portable recorder described here, together with the
other standard units of equipment, is adaptable to any of the fore-
going combinations, as well as to others which may be required.
The recorder without the magazine weighs only 75 pounds.

The motor and silent coupling are completely enclosed, although
adequate provision has been made for ventilation. The glow lamp
holder assembly is likewise enclosed by a substantial cover which
serves the dual purpose of protecting the glow lamp from breakage,
and the recordist from accidental contact with the leads.

Experience has shown that, as far as recording is concerned, there
is a direct relation between the number of sprockets and the fre-
quency of "jams" or "buckles." Simplicity of the film path has
therefore been sought in this equipment. Entering the recorder
from the loaded side of the magazine, the film passes over one side of
the main driving feed sprocket, and is engaged by several teeth to
assure positive feed. A loop to guard against sprocket pull is next
provided, after which the film passes over the specially ground roller
which carries it past the recording beam. A mechanical filter on the
drive for the next sprocket, together with a heavy, well-balanced
flywheel on the roller shaft, assure a uniform speed of the film past
the recording beam.

After passing over the filter sprocket, which engages several
sprocket holes in the film, another loop is provided before the film

Dec., 1933 J PORTABLE RECORDING EQUIPMENT 485

passes over the right side of the main drive sprocket and thence into
the take-up side of the magazine. Contact with the film is reduced
to a minimum, and scratches and "static" marks are thus avoided.
Either Bell & Howell, or Canady film magazines may be used.

Idler Control. With the exception of the idler on the filter sprocket,
all idlers are opened and closed by the turn of a single knob. In
either position, they are locked.

Glow Lamp Holder Assembly. Since this unit is involved in the
operation of threading and unthreading, it has received special
attention, to the end that these operations might be performed more
quickly and easily. The glow lamp is held firmly within a slotted
cylinder provided with a knurled ring, which, when tightened, locks
the tube securely in the assembly. Also on this cylinder are a mi-
crometer adjusting ring and a lock ring. The inner face of the
adjusting ring rests squarely against a machined surface on the case
when the assembly is in the operating position, thus fixing the clear-
ance between the glow lamp and the film.

By simply turning the knob the entire assembly is moved, with all
adjustments unchanged, into either the operating position or the
threading or inspection position. The preparation of the recorder
for threading is thus reduced to turning three knobs. The hori-
zontal adjustment of the light beam is made by means of two counter
locking screws.

Synchronizing Light. Adjoining the glow lamp holder assembly
is the synchronizing or marking light. This lamp is totally different
from all other "sync" lights in that the flash is produced by passing
a spark between two electrodes. The spark is imaged on the film in
line with, but on the opposite side from, the recording beam. The
synchronizing mark made with the sparking device is a single sharp
striation as compared with the blurred mark produced by an incan-
descent lamp which requires a finite time both to come up to full
brilliance and to cool after the current is cut off.

A succession of marks can be made along the film at various inter-
vals by making connections as desired to a built-in commutator. This
light has proved to be dependable and trouble free; and since there
is no filament, it does not need periodic replacement.

Automatic Stop. An automatic stop is provided which shuts off
the motor in case of a buckle in time to prevent an appreciable
amount of film from piling up. It is believed that this equipment
is the first to incorporate such a feature. Contacts are enclosed

486 DON CANADY [j. s. M. P. E.

on the outside of the recorder. The arm which actuates the con-
tacts does not bear on the film except in case of a buckle, and does
not interfere with threading or unthreading.

AMPLIFIERS

Semi-Portable Amplifier. The amplifier, which is contained within
a case 26 in. long, 15 in. high, and 5 in. deep, weighs 25 pounds,
complete; and in addition to the amplifier circuit, contains a three-
position mixer of an improved type, a monitor circuit, a volume
indicator, and a control circuit for the glow lamp. Metering facilities
are also provided for all plate and filament circuits.

The various units, as well as the wiring, have been specially

FIG. 1. Portable recorder and amplifier.

treated to protect them against extreme climatic conditions that
portable equipment must often endure.

Performance curves show a substantially flat characteristic from
25 to 10,000 cycles, and an output in excess of -(-15 db. without dis-
tortion. Due to the full-range characteristics of the Canady-Zetka
glow lamp, no compensation is used or needed. Amplifiers which
compensate for deficient response of the glow tube inevitably amplify
the harmonics of the low or medium tones out of proportion to those
of the high tones; and thus the true characteristics of the tones are
lost even before they reach the glow lamp.

Each stage of the amplifier is adequately filtered against inter-
action, and also as a precaution against "motorboating," when used
with run-down batteries,

Dec., 1933] PORTABLE RECORDING EQUIPMENT 487

An adjustable low-pass filter is provided for use under certain
conditions when it is necessary to reduce or eliminate wind noise,
background, or other undesirable sounds.

Portable Amplifier. Portable amplifiers have been variously
described as being of the trunk type, or suitcase type, but this one is
more nearly the size of a handbag. The case measures 18 by 12 by 5
inches, outside dimensions, and the complete amplifier weighs 20
pounds. Since a single microphone input is provided, the gain con-
trol of the amplifier is the only volume control needed (Fig. 1).

FIG. 2. Interior of portable recorder.

By means of a plug and jacks, filament and plate currents may be
read, and also the glow lamp current. A separate meter is provided
for the volume indicator. The current to the glow lamp is controlled
from the panel of the amplifier, and the cables from the microphone,
batteries, and glow lamp are brought in through non-interchangeable
connectors. Headphone monitoring is provided.

While nothing essential has been omitted, size and weight have
been reduced in this model to the lowest point consistent with re-
liable operation.

488

DON CANADY
PORTABLE OR MODEL A RECORDER

[J. S. M. P. E.

Designed to fulfill the demands of portable recording, where
weight and reliability are primary considerations, the model A re-
corder produces excellent recordings (Fig. 2).

The film path through the recorder is similar to that in the larger
model (Fig. 3). Idlers, however, are individually controlled. The
glow lamp holder is provided with micrometer adjustments and a
damping ring to hold it in the operating position. Light from
the glow lamp reaches the film through an optical slit of quartz, with

FIG. 3. Interior of studio recorder.

an aperture of 0.0008 by 0.250 inch. The glow lamp holder will
accommodate any standard recording lamp. As in the larger model,
either Canady or Bell & Howell magazines may be used.

Microphones and Cables. Microphones of the condenser type
with built-in, two-stage amplifiers are standard equipment, and have
proved satisfactory and reliable under varying and adverse conditions.
Small current, non-microphonic tubes are used in the microphone
amplifier, thus reducing background noise, and permitting the use of
a longer microphone cable without increasing the supply voltage.

Dec., 1933] PORTABLE RECORDING EQUIPMENT 489

Battery Case. The new developments in vacuum tubes have made
possible the use of smaller size B batteries than formerly, and the
complete high-voltage supply for the plates of the main and micro-
phone amplifiers and also of the glow lamp, is contained in a case
measuring 22 by 10V2 by IP/2 inches, which can be easily carried
by one man.

Life figures, based on recorded footage, or hours of operation are
not available, but where the Canady-Zetka glow lamps have been
used, the batteries have lasted six weeks to three months, depending,
of course, upon the extent to which they have been used.

Where the equipment is to be used in a studio, either permanently
or part of the time, a special rectifier unit is available, permitting
operation from the lighting mains.

Storage Battery. One 6-volt, 80-ampere-hour storage battery
supplies the filament currents of both the microphone and main
amplifiers, and under average conditions, will do so for a week to ten
days before recharging is necessary. Where a 6-volt source is em-
ployed for the recorder, a 150-ampere-hour battery is required, so
as to permit the equipment to be used away from its base for three
or four days without danger of the voltage's falling below a use-
ful value.

Motors. Standard motors for driving the recorder and camera
are of the 110- volt, 60-cycle, synchronous type. Where 220- volt,
3-phase alternating current is available, interlocking motors can be
supplied.

To fulfill the need for a reliable drive that would not require a
commercial source of power, a special unit has been developed.
On the same shaft with the 6-volt motor, and in a common housing,
is mounted a 60-cycle alternator. The camera motor is of the syn-
chronous type, its speed being controlled by means of a rheostat in the
field circuit of the 6-volt motor. A direct-reading frequency meter
is so mounted as to be plainly visible at all times; and, due to the
extended scale of the instrument, variations in speed may be readily
noticed and corrected. Due to the absence of moving parts, this
type of speed indicator is reliable and is unaffected by changes of
temperature.

GLOW LAMPS

Type A Glow Lamp. The two outstanding systems for producing
a variable density sound track on film are the mechanical vibrating

490

DON CANADY

[J. S. M. P. E.

system and the glow lamp recording system. Up to the present time
the difficulty of producing glow lamps which are sufficiently uniform
and which fulfill the other requirements imposed have resulted
in the adoption of the mechanical vibrating system by some of the
leading producers. In this system it has been possible mechanically
to peak the response of the device in order to compensate for the loss
at high frequences caused by the finite width of the slit used.

In the glow lamp recording system there is one outstanding ad-
vantage over the mechanical vibrating system, in that there is ab-
solutely no inertia to be overcome. On the other
hand, losses at high frequencies due to slit width
must be compensated for in the amplifier.

Up to the present time the advantages of the glow
lamp recording system have been veiled in practical
usage by a number of shortcomings in the glow lamps
themselves, which can be listed as follows :

(1) Low light-intensity ; requiring that the lamp be used with
the mechanical slit, or its equivalent, as close to the film as possi-
ble, with the attendant troubles of slit-clogging, etc.

(2) Too large wattage consumption in the lamp ; causing ex-
cessive heating of the bulb and element, and consequent blacken-
ing and obstruction of light.

(3) Very short and indefinite life ; due to blackening of the
bulb or change in the gas pressure or surfaces of the elec-
trodes.

(4) Lack of uniformity; requiring individual calibrations and
adjustments for each lamp.

(5) Erratic electrical characteristics during life of the lamp;
requiring constant watching and readjustment.

(6) Low sensitivity; requiring a comparatively large output
recording from the recordin g amplifier.

lamp.

The design and development of a glow lamp which

would obviate these objections has proved a difficult and tedious
problem (Fig. 4).

As the primary object of the glow lamp is to radiate actinic light
through an exceedingly small aperture, the lamp must be so de-
signed that the glowing spot within the bulb is greatly concentrated
and located as closely as possible to the end of the bulb through
which the light is to pass. In the Zetka lamp a special glass is used,
which is particularly transparent to violet and ultra-violet rays.
At the point where the light is required to pass, the bulb is blown

FIG. 4. Zetka

Dec., 1933]

PORTABLE RECORDING EQUIPMENT

491

very thin, thus further assisting in the transmission of that highly
important spectral range.

To avoid premature blackening of the tube it was found vital to
select electrodes having the minimum tendency to vaporize under
operating conditions. It was also found important to select a mix-
ture of gases and to determine a gas pressure, the combination of
which would reduce the vaporization to the lowest possible point.
Furthermore, the choice of the gases and their pressure are deter-
mined by the desired light intensity. Naturally, the life of a lamp
will depend upon the conditions under which it is operated; but
the new Zetka glow lamp is capable, under ordinary conditions, of

14
\Z
10
8
6
4
Z

FIG.

hi
K

I
<

VOIJTS

K> 3*0 40O A SO 5OO 560

5. Static impedance characteristic of Zetka recording lamp.

recording upward of 25,000 feet of sound track. Some recordings
have been made as long as 60,000 feet.

Every effort has been exerted to make the lamp as rugged as
possible in order to withstand safely the shocks and abuse of shipping
as well as those of recordings under "field" conditions. The latest
design of lamp has a maximum outside bulb diameter of 1 inch, and
the over-all length from the tip of the bulb to the end of the contact
prongs is approximately 5 3 /s inches. It is mounted in a standard
UX-199 base, only two of the prongs of which are used for electrical
connection.

The voltage-current curve of the lamp is shown in Fig. 5, the
linearity of which is very unusual with gas-filled devices. This
feature is very important from an electrical standpoint because it
indicates a constant impedance (5000 ohms) of the lamp during the

492 DON CANADY [j. s. M. P. E.

signal "swings." Duplication of these lamps has been found per-
fectly feasible when proper care is taken at every stage of their
manufacture. The frequency characteristic is shown in Fig. 6.

It is evident that if the lamp is operated at 550 volts, 8 milli-
amperes, the heat dissipated within the bulb is only 4.4 watts, which
is not sufficient to cause undue concern. When operated under such
conditions the maximum permissible voltage swing is about 100
volts, thus placing a power demand upon the amplifier of only
three-tenths of a watt.

Under those conditions it is easily possible to expose positive
film when the lamp is placed close to the film with the usual 0.001-
inch, or even smaller, mechanical or quartz slit. When used with an
optical system, the light intensity is quite adequate for VD film;

__ rW rVW f \J \J W

FIG. 6. Frequency characteristics of recording lamps. 1. Old style
"Glow lamp." 2. "Q" Recording lamp.

with a calculated slit, at the film, of about 0.0008 inch. The lamp
can be used with any standard amplifier having suitable means for
controlling modulation.

Glow Lamp Circuit. In accord with the usual practice, it is es-
sential that a series resistance of at least 10,000 ohms be connected
in the circuit of the lamp at all times. A variable resistance ranging
from 10,000 to 50,000 ohms will be found satisfactory.

Type Q Glow Lamp. Light from glow lamps of the type A may be
transmitted to the film by a mechanical slit, a quartz slit, or an
optical train. The mechanical slit tends to collect emulsion and
block the light. The quartz slit is more efficient, but due to the
extreme thinness of the quartz, it must be handled carefully. The
optical system, of course, avoids the mechanical difficulties, but at
the expense of light transmitting efficiency. The comparative

Dec., 1933] PORTABLE RECORDING EQUIPMENT 493

ability of various materials to transmit the actinic light of the glow
lamp is shown in the following table.

Relative Transmission of Various Materials to Glow Lamp Radiation

Transmission
Material (per cent)

Lead glass 20

Lime glass 26

Nonex 38

"E.J." Pyrex 40-46

Corex 68

Quartz 88

Since the losses are cumulative, and increase rapidly according to
the number of elements, it is apparent that even very high-grade
optical systems introduce serious losses. The problems of spherical
and chromatic aberration, and coma must be solved by introducing
additional lenses and thus increasing the transmission loss still
further. To compensate for the additional loss, the glow lamp
must be operated at a considerably greater brilliancy. This means
a large current, a higher operating temperature, a consequent re-
duction in the life of the lamp, and shorter battery life when batteries
are used.

A recent development by J. B. Zetka surmounts these difficulties,
while retaining the advantages of the lamp. Known as the type Q
glow lamp, it incorporates both the lens and the slit in the structure
of the tube itself. Corex glass is used for this section of the tube.
The light aperture or slit measures 0.0008 by 0.1 25 inch. Improve-
ments in the internal construction of the tube have resulted in ample
film exposure with a lamp current of six milliamperes. Life records
have repeatedly shown from 75,000 to 100,000 feet of exposed film.

The single system of recording directly on the picture negative
at the base of the camera has been purposely neglected in this paper.
The shortcomings of the single system have been so serious as to lead
to its abandonment by the majority of producers. The larger grain
of negative film causes undesirable background noise, and raises the
lower limit of sound which may be recorded, while the necessity for a
common laboratory treatment of sound track and picture inevitably
results in a compromise which fails to realize the maximum possi-
bilities of either. It is also impracticable to cut the common negative,
where "flashbacks," inserts, etc., are desired.

UNOCCUPIED MOTION PICTURE FIELDS'
WILLIAM H. SHORT**

Summary. A description of the work and purposes of the Motion Picture Re-
search Council is presented, followed by an interpretation of some of the results of
this work as regards (l) child receptivity, (2) memory, (3) emotional stress, (4) atti-
tude, and (5) behavior patterns induced by viewing motion pictures. The unexploited
motion picture fields are said to be those of (/) teaching pictures, (2) juvenile entertain-
ment (3) entertainment for a large class of adults who do not apparently attend the
motion picture theaters for various reasons, and (4) the creation of good-will for the
future in respect of adult education, documentary or historical films, and scientific
films.

The Motion Picture Research Council is a body of more than three
hundred men and women, widely distributed geographically, and
somewhat equally divided between the three allied groups of social
workers, social scientists, and socially minded lay men and women.
Miss Jane Addams of Hull House (Chicago, 111.) is dean of our social
workers group. Perhaps our social scientist group would agree that
Dr. Wesley C. Mitchell is first among equals hi that company. From
the group of socially minded citizens I should not know whom to men-
tion first but included hi the list are the folio whig: John H. Finley,
Henry Sloane Coffin, Henry Fairfield Osborn, E. R. A. Seligman,
Stephen S. Wise, Francis J. McConnell, Alanson B. Houghton,
Thomas J. Watson, Gerard Swope, Frederick Peterson, Clinton
Rogers Woodruff, William Danforth, J. Lionberger Davis, Charles M.
Moderwell, Charles W. Gilkey, Henry J. Fisher, Herbert S. Houston.
The Chairman of the Council is Dr. John Grier Hibben, President
Emeritus of Princeton University.!

The purpose of the Council, organized five and a half years ago,
has from the first been definitely and positively constructive, and
friendly to the motion picture art. All were convinced at the begui-

* Presented at the Spring, 1933, Meeting at New York, N. Y.
** Director, Motion Picture Research Council, New York, N. Y.

t Since deceased, and succeeded by Pres. A. Lawrence Lowell of Harvard
University.
494

UNOCCUPIED MOTION PICTURE FIELDS 495

ning both that the motion picture is here to stay and that it is po-
tentially a tremendous boon to the human race. For these reasons,
the social workers felt the need, as an equipment for dealing with
motion picture problems, for a broad fund of scientifically obtained
information regarding the potentialities and actualities of motion
pictures, especially in relation to childhood and youth. Miss Ad-
dams said that she had been serving on motion picture committees
for a generation, and that little or no progress had been made just
because such knowledge was not available.

The social science group was impressed with the fact that in the
motion picture art we have a new and major tool of civilization for the
social use and guidance of which their fraternity has an especial
responsibility. Dr. Mitchell early gave expression to this convic-
tion in the following words:

Motion Pictures are one of the most powerful influences in the "making of
mind" at the present time. They affect great masses of people, and they affect
those masses during the impressionable years of childhood and youth. The in-
dustry has developed on a strictly commercial basis into one of our most con-
spicuous branches of business, demonstrating that moving pictures meet an in-
tensely felt interest. So far, the only social guidance which has been exercised
has been of the negative sort a censorship designed to remove objectionable
features. Constructive guidance has been notably slight.

Obviously the moving pictures are doing things to our thoughts and feelings.
It is high time we discovered what those things really are. There is no problem
that could be of greater concern to those who are interested in the quality of our
future citizens.

The lay group of the Council shares both the scientific and the
citizenship interests of the other two groups. Together they present
a united front of interest and desire to serve in the development of
the new motion picture art in the broadest and most socially useful
ways.

The Motion Picture Research Council does not oppose censorship,
but does believe that the time has come for something other and better
than that. It was in this faith that it had its origin and is doing its
work. The ultimate and constructive aim of the Council, has, from
the beginning, been to develop a plan and recommendations for the
motion picture art, as broad and scientifically helpful in every respect
as, through four centuries of use, the art of printing has become.
And then to devise and stimulate leadership that will enable the mo-
tion picture art to achieve this high and honorable destiny.

496 WILLIAM H. SHORT [J. s. M. P. E.

THE PAYNE FUND RESEARCH

But the Council thought that it ought first, by careful and patient
research, to find out whether it was correct in the belief that in the motion
picture art civilization has a new tool of high intellectual and social
potency. It, therefore, invited and received the assistance of the
Payne Fund, an agency "interested in enterprises connected with the
radio, motion pictures, and reading in their relations to the education
of children and youth." The Payne Fund, in turn, organized an
Educational Research Committee of qualified psychologists and
sociologists from the social science faculties of the Universities of Chi-
cago, Iowa State, Ohio State, New York, Columbia, Yale, and Penn-
sylvania State College. A program of research was developed by
prolonged conference of social workers and the members of this
Educational Research Committee ; budgets were drawn ; the required
funds were deposited with university treasurers; and the several re-
search men went to work, under a commission without fear or favor
to get the facts in their respective sectors of this broad field, and to
pursue then" research until amply abundant data had been accumu-
lated to substantiate the conclusions at which they might arrive.

For four years, without interference by publicity or otherwise yet
cooperatively the research went on. Under the guidance of Dr.
W. W. Charters the research guns were kept on the targets. Findings
have been obtained that, with unimportant exceptions, fit together
like the parts of a well-planned structure to make a body of scientifi-
cally obtained data which, taken together, has still greater validity
than does any one of its workmanlike parts. The Macmillan Com-
pany is now printing the findings in nine scientific volumes and one
summary volume. Within a few days these will begin to be made
available. The summary volume, in popular language, is by Henry
James Forman. Let me summarize in the briefest possible way, just
a few of these findings; and state some important conclusions to
which, in my own judgment, they point. In due time, as I have
already said, the Council itself will make and publish its well-con-
sidered conclusions and recommendations.

(1) Child Receptivity to Movie Ideas. Even very young children
take in, to a remarkable degree, what they see on the screen. Second-
and third-grade children (7 to 9 years of age) take in 60 per cent of
what a group of young college professors, graduate students, and
their wives take in ; and the percentage of retention increases rapidly
as the children become older. This goes with every class of ideas;

Dec., 1933] UNOCCUPIED MOTION PICTURE FIELDS 497

and in the case of both general information only incidentally connected
with the film story, and the specific content of the story there isn't
much difference between the two.

It has been generally supposed that many classes of ideas in the
films go quite over the heads of children. Well, they don't; they
all register to a significant though not identical degree, even with
the second-graders. They remember best such things as sports,
crimes, acts of violence, general action, humor anything that ap-
peals to the emotions. The child who is quite unable to give you a
connected story about the film can and does, when appropriate ways
of testing are used, show that his intellectual takings-in have been of
such inclusive sort and high percentages as I have just stated. Just
as the very young child who could not give you a dozen connected
sentences, either verbally or on paper, of what he has seen in his own
home, has nevertheless acquainted himself there with a thousand
folk ways and learned the rudiments of speech; so does he take in
and store away the promiscuous contents of the films.

(2) Remembering. Do children remember what they see in the
movies, or is it in mind today and gone tomorrow? Here the inves-
tigators were slated for much the biggest surprise of the research.
The classic investigation previously made to find out how long chil-
dren remember what they learn from books had shown only about
thirty per cent retention after the lapse of a month.

These experiments with the films show that at the end of a month
the child recalls nearly, or quite, one hundred per cent of what
he knew on the morning after seeing the picture, and not in-
frequently more than one hundred per cent sometimes as much as
one hundred ten per cent. After six weeks, the average retention was
still ninety per cent, and after three months practically the same
sometimes even more. The second-grade children beat the professors
at remembering.

The only explanation for this lasting impression seems to be that
the impact made by the films is of a different sort from that made by
books a dramatic, emotional impression, akin to a vivid experience.
You and I, at this moment even though in middle life or beyond
can recall with vividness and complete detail, certain dramatic oc-
currences of childhood days; we couldn't forget them if we lived a
thousand years. Well, the movie recollections of children are of the
same order visual, dramatic, sometimes poignant experiences. As
about three thousand persons of varying ages were studied in this

498

WILLIAM H. SHORT

[J. S. M. P. E.

memory investigation, by means of twenty thousand testings and
eight hundred thirteen thousand items attempted, the findings re-
ported may safely be accepted as established.

(3) Emotional Stresses. In addition to intellects which take in
and remember, we also have an equipment of emotions. Any one
who has seen mature men and women laughing and weeping over an
imaginary screen story does not need other proof that the motion
picture is a peculiarly effective instrument of emotional appeal. If,
besides, he has been on a Saturday or Sunday afternoon within
earshot of a theater that was showing a Western serial to a house

FIG. 1 . Schematic arrangement of the psychogalvanometer.

full of children, he will have reason to suspect that children are at least
not less affected emotionally than adults.

Well, he will be quite right about it. Exhaustive experiments
were made by recognized experts. These were carried on both in
the laboratory and under regular theatrical conditions. Children
and adults were used, boys and men, girls and women. Some six
thousand four hundred readings were made with a psychogalvanom-
eter, two thousand one hundred fifty of them in the movie theater.
Heart-beat was also counted by appropriate mechanical devices.

The psychogalvanometer consists of a Wheatstone bridge, W, (Fig. 1), one arm
of which is connected to the electrodes E, into which the child under observation
dips the index and third fingers of his left hand. As the balance of the bridge is
disturbed, the deflection of the galvanometer G causes a beam of light to oscillate
across a roll of photographic film. The adjustable-speed motor M drives the film

Dec., 1933] UNOCCUPIED MOT-ION PICTURE FIELDS 499

past the aperture, deriving its power from the service lines 5. The transformer,
also connected to S, has its secondary connected to the vibrating bar T t ,
which interrupts the current through the lamp L, producing a time line on the
film in half -seconds. A second galvanometer, G, is connected to another bridge
Wk, across which are connected a dry-cell and a pneumocardiograph, B. The
latter is a carbon button fastened to a leather strap to be attached to the arm of
the child, a coiled spring maintaining the pressure on the skin uniform. The
light reflected by the galvanometer connected to WH impinges upon the same film
as that to which the oscillating beam is reflected by the galvanometer connected
to W t . The bridges are used to balance the galvanometers and set their deflec-
tions to zero at the beginning of each test.

What was learned? Briefly, that in the case of slapstick stuff,
adolescents are roughly under twice as much emotional stress as
adults; and young children, between six and eleven years, three
times as much. No horror or mystery pictures were used in the ex-
periment, yet, in certain scenes, young children registered from eight
to ten times the emotional intensity of the adults. The adult has
had experience that enables him to say to himself, "That isn't
real; that isn't true." To the child, what he sees is both real and
true.

In cases of love scenes and those of suggestive sex, children nine
years of age appeared to react with about the same intensiveness as
adults. From twelve to fifteen, all children reacted to a still greater
degree. During the years following puberty sixteen to eighteen
years of age the emotional stresses caused by sex pictures are at
their maximum twice as great as is experienced by adults. And
that is considerable.

The rate of heart beat was found to be profoundly affected. The
experimenters did not venture to use pictures of horror or mystery in
any of their experiments with children only those of quite common
variety. Yet, in watching these, the heart beat ran up repeatedly
from a normal of seventy-five or eighty to one hundred twenty-five
and one hundred forty. Using The Mysterious Dr. Fu Manchu, the
heart beat of a mature woman, selected for her normal and equable
temperament, jumped up frequently to one hundred fifty and one
hundred sixty-eight, and on single occasions to one hundred eighty
and one hundred ninety-two.

Study of sleep disturbance was also made throughout a two and
one-half year period, during which the sleep motility of one hundred
seventy boys and girls from six to eighteen years of age was observed.
Twenty experimental beds were electrically equipped so that every
movement was recorded in a distant room on revolving drums of

500 WILLIAM H. SHORT [J. S. M. P E.

paper. These beds were occupied on three hundred forty-seven
nights a total of six thousand six hundred fifty child-nights of
sleep that is, for fifty-nine thousand five hundred eighty hours.

The records of movements during every minute of that time were
carefully studied. It was found that after attending the "mine run"
of movies at a neighboring theater, so situated that the children
could be home and in bed by nine o'clock, the average restlessness of
girls was increased fourteen per cent and that of boys twenty-six per
cent. This restlessness continued for several nights before normal
sleep was restored. No mystery or horror pictures were seen during
the sleep study. But numerous instances are recorded by other ob-
servers in which such films resulted in broken sleep and frightful
nightmares for weeks and even months.

There were, as you see, three separate and distinct tests of emo-
tional stresses experienced by children. There were direct objective
measurement by means of the psychogalvanometer; indirect objec-
tive measurement by observation of the heart-beat; objective mea-
surement of resulting sleep disturbance by the electric hypnograph.
These, both separately and collectively, indicate the presence in
children of states of high emotional excitement as the result of movie
attendance.

The arrangement of the hypnograph is illustrated in Fig. 2. It consists of a
bar, C, arranged to move up and down, according to the motion of the bed-springs,
past brushes / and 2, across which is connected a resistance. Inserted into the bar
are inlays, so that the circuit is broken as brush 2 passes from one segment to the
next. The operation is as follows: when the child asleep in bed moves, the bar
C moves up or down ; when brush 2 rests on insulation, the current must pass
through the shunt resistance, and, as a result, becomes too weak to actuate the
electromagnet that produces the record on the tape. When brush 2 rests on
conducting material, the resistance is short-circuited, the current is increased, and
the electromagnet operates, producing a dip in the otherwise linear record. The
tape moves at a constant speed, so that the duration of bodily movements can
be determined and studied. The system is so adjusted as to record such move-
ments as those of the forearm when the elbow is bent, or of turning the head, with-
out being so sensitive as to record heart -beat and respiratory movements. The
signal light indicates when the bed is unoccupied, or when a fault occurs in the
circuit; if the circuit is open, or if the bed is unoccupied and the brushes rest on
the bakelite section of bar C, the relay operates and the lamp is lighted.

(4) Attitude Studies. These had to do with the effectiveness of
motion pictures as an instrument for changing the social attitudes of
children in either socially approved or socially disapproved direc-

Dec., 1933]

UNOCCUPIED MOTION PICTURE FIELDS

501

tions. An attitude is a habitual, or relatively fixed, emotional re-
sponse to a given idea or situation. It was found that with a single
film the attitude could be changed much as the line of mercury is
run up or down in a thermometer by the application of heat or cold;
and that two or three films of a similar theme were much more ef-
fective than one.

Sons of the Gods, a film that shows the Chinese in a favorable light,
made a whole townfull of children much more favorable to the Chi-
nese than they were before, while Harold Lloyd's Welcome Danger

FIG. 2. Schematic arrangement
of the hypnograph.

carried Chinese stock, already low, to a still lower level. Birth of a
Nation resulted disastrously to the standing of the negro among the
children of the Illinois town where it was shown. The pro-negro
film Hallelujah, if it had been used, might have moved popular senti-
ment toward the negro equally far in the other direction. Extensive
tests made with films on different subjects thoroughly established
the effectiveness of the film as an instrument of propaganda, even
though it was not so intended by the producer.

A.n effort to measure the total influence on children of the hundreds

502 WILLIAM H. SHORT [J. S. M. P. E.

or thousands of films they had seen was not very successful. In
view of the conglomerate and conflicting nature of film content it was
a heroic undertaking at best. But quite aside from that, the reasons
for its comparative failure are easy to see as one looks back upon it.
The proposal was to discover and measure the differences between
a group of non-movie-going children and an equated group of movie
goers. No group of children unaffected by movie attendance and
environment could be found. So, a group who went infrequently had
to be compared with one that went more often; that was not con-
vincing. In the second place, the technic used was lacking in sensi-
tiveness when compared with the technic used in determining the
influence of a single film. Finally, it could not be determined by
the technic whether the effects discovered were the result of selection
or of movie-going.

The failure of the study does not, of course, cast any doubts on
the validity of the results of the other experiments that did success-
fully measure the changes in attitude caused by single films, or by
groups of two or three films of like theme. It has been completely
established that the film is a highly effective instrument for changing
the social attitudes in great masses of people, and in either socially
approved or disapproved directions.

(5) Behavior. The study of behavior patterns and conduct, as
molded by the films, was so extensive that I can not attempt here
to give even the slightest summary. It shows that the movie theater
is an educational system, a school of conduct it may be desirable
conduct, in socially helpful directions; it may be objectionable con-
duct, leading to vulgarity, delinquency, and crime. What the young
child sees on the screen he accepts with much the same trustfulness
that the fledgling in the nest displays in accepting food from the
mother bird. A South American gentleman of education and cul-
ture, who had been several times in the eastern United States, said
to a North American friend that in the visit he was then about to make
he would land at a California port and see "wildwest." What this
gentleman had seen on the screen had had validity for him. How
much more does it have validity for the child!

This attitude of trustfulness toward what is seen on the screen is
not much shaded until well into high-school years, if even then.
Occasionally an entire and startling change in personality takes place
in children and youth under the influence of a screen star who is ad-
mired. Such influences on conduct reach their maximum effective-

Dec., 1933] UNOCCUPIED MOTION PICTURE FIELDS 503

ness in the disorganized areas of great cities peopled by the foreign-
born and their children. Among such un-Americanized populations
the movie is shown to be one of the major resources from which to
become acquainted with what they conceive often mistakenly;
sometimes tragically to be American ideals and ways. What the
child sees on the screen he goes out and day-dreams about, imitates in
his play, follows in his love-making, becomes a devoted worshipper of
the screen stars who guide him into life.

ASSUMPTION OF MOVIE VALUES SUSTAINED BY RESEARCH

The Motion Picture Research Council, as I have already said, set
out five years ago in the belief that the new art of the motion picture
is, in fact, a highly important tool of civilization. I think we can
now agree, brief and inadequate as the foregoing summary has neces-
sarily been, that that assumption has been sustained by the research.
More than sustained, says the Research Director of the Payne Fund
studies in his introduction to the forthcoming summary volume by
Forman. "The Motion Picture," he says, in reviewing the evidence,
"is powerful to an unexpected degree, in affecting the information,
attitudes, emotional experience, and conduct patterns of children."

UNOCCUPIED FIELDS FOR MOVIE EXPLOITATION

I assure you that I have introduced only the smallest fraction of
the evidence which the research will soon submit for public scrutiny
and evaluation. May I not assume, on the basis or evidence already
submitted, and on the deliberate judgment of the Research Director
just quoted, that we are convinced of this fact, namely, that in the
motion picture we are dealing with a sharp-edged tool, which, in the
public interest should be used with exceedingly great care and good
judgment? If so, it is enough for my present purpose.

These data, scientifically ascertained as they have been, lead me
and I believe that they will lead the public to the conclusion that
there are unoccupied fields for the exploitation of the motion picture
of even greater importance than any that has yet been cultivated;
that those fields demand immediate attention; and that the scope,
the usefulness, the reputation, and the business opportunities of the
motion picture will be vastly enlarged thereby.

Where do these unoccupied fields lie? Like the Acres of Diamonds
in Russel Conway's famous lecture, they are under our very feet.

504 WILLIAM H. SHORT [J. S. M. P. E.

We have been looking at them and talking about them without really
believing that they were there.

(1) Teaching Pictures. It is hardly necessary to say that I do not
have in mind mere dry-as-dust facts recorded on films instead of on
printed pages the film has no advantage in a competition of that
kind. What I do have in mind is almost a new kind of teaching
the dramatic presentation of truth. There is a growing realization
of the fact that in the education of children the school has been ad-
dressing itself too exclusively to the intellect and not enough to imagi-
nation and emotions. The film is a foreordained instrument with
which to re-dress this balance. Can any one see a convincing reason
why the dramatic art should not be drawn upon to supplement present
methods of instruction ?

Again, we all understand that so far the school is creating a bad
mess in its attempt to make good citizens. In the light of the moral
debacle of the present day, few will be disposed to question that.
Has some one said that morality is ethics touched with emotion?
Could any teaching medium be more ideally adapted than is the
motion picture art for adding emotion to ethical teaching, and thus
to emphasize the qualities of citizenship rather than those of grafters,
gangsters, and racketeers? I have already pointed out the fact that
an effectively made film necessarily becomes an instrument of propa-
ganda, whether or not the producer so intends it. Let us have
films for use in the schoolroom intelligently aimed at the production
of character. Whether they are made by or for the State I care not
which we need them, and need them now.

Yet, however important the foregoing considerations may be, the
main point of the argument for the film hi education is this. From a
single viewing, second- and third-grade children gained an increase in
knowledge in the field covered by the film ranging from fifteen to
twenty-seven per cent; fifth- and sixth-grade children, an increase
ranging from three to fifty-six per cent; children in the first two high-
school years, an increase ranging from thirteen to sixty-seven per
cent. These percentages of knowledge-increase do not appear to be
exceptional, but typical and usual. The retention for all those grades
over periods of weeks, and even months, is not very much under one
hundred per cent; sometimes even more than one hundred per cent.
Is it not clear that a teaching medium with which such results are
attained can not longer be ignored in the classroom? A glance at

Dec., 1933] UNOCCUPIED MOTION PICTURE FIELDS 505

the vast organization and equipment used in the text-book trade will
show what that means for the film industry.

(2) Juvenile Entertainment. Some attempt which I refrain
from enlarging on here has been made to have the public believe
that child attendance at the movie is a minor and infrequent mat-
ter. That is a serious error. The Payne Fund study of child at-
tendance is much the most extensive and careful that has ever been
made. It shows that the habit of going to the movies is almost uni-
versal among children. Two or three years ago children five to eight
years of age were attending, on the average, twenty- two shows a year,
seeing therefore about sixty-six films besides trailers. Children from
eight to nineteen years of age were attending, on the average, one
show a week, seeing 150 films a year. The weekly movie audience of
thirteen years and under, in continental United States calculated on
a total weekly attendance, by ages, of 77,000,000 was eleven and a
quarter millions. Children and adolescents constitute thirty-one and
one-half per cent of our total population, but thirty-seven per cent of
the movie audience. That is in normal neighborhoods. In congested
parts of large cities the proportion is very much greater. Child at-
tendance no doubt rises and falls proportionally with the rise and fall
of the total motion picture audience. But it is universal, constant,
enormous, nation-wide. Children and adolescents provide approxi-
mately 37 per cent of the total movie audience, however the size of
that audience may vary.

While children are entertained by the movies, they do not go to
the movies for entertainment in the same sense as do tired and jaded
adults. While children pay some attention to the film story, it is not
in that that their primary interest lies they may miss the story, or
misunderstand it, but at the same time be intensely interested in the
film. While children sometimes get a moral from the film, it is often
overlooked, and may be quite a different moral from that which the
adult gets. It is a wholly mistaken idea that a moral at the end of
a film will for the child redeem unwholesome subject matter.

The child's primary interest is in the incidents of the film. He goes
to the movies to enlarge his world. When he first opened his eyes in
infancy it was on a world of which he knew nothing at all. His first
interest was in the folk-ways or mores of his own home. He wanted
to learn what things are, what they are for, how older people do
things. A little later, he extended his interest to the neighborhood,
but for a good many years he will be neither a moralist nor a philo-

506 WILLIAM H. SHORT [j. s. M. P. E.

sopher. He is, rather, an observer. He is accumulating the inci-
dents, ideas, and ways of doing things, out of which he will later build
a moral (or immoral) system for himself, and a way of life.

In the movies, he finds a new, a larger, an entrancingly interesting
world, which he can observe and from which he can learn. "Where
should I learn to love, if not from the movies?" asked a young man
of college age. He revealed the chief interest that takes children and
youth to the movies to learn a thousand interesting things that
neither home, neighborhood, nor school has given them a chance to
learn.

In view of this omnivorous interest in the contents of the films,
and in view of the findings already reported that of every category
of movie ideas even second-grade children take in on an average 60
per cent of what superior adults take in, and then remember it almost
100 per cent over very long periods the matter of what they see in
the movies becomes one of the very greatest import.

As everybody knows, there are a number of film previewing agen-
cies that pass on the audience-fitness of films for adults, family,
adolescents, and young children. Mr. Will Hays' annual report,
submitted the 27th of March, 1933, states that of 476 feature pic-
tures previewed during 1932 by seven previewing organizations, 413
(86.7 per cent) were "variously endorsed for family, adult, or child
entertainment" by "one or more of these committees."

But manifestly it is more important for the parents of children to
know from how many and what percentages of these films approval
for child consumption was withheld by those previewing committees.
Being "variously endorsed for family, adult, or child," to quote Mr.
Hays, is not enough. And being approved for one classification
perhaps for adults does not offset a possibly unanimous disapproval
for children.

One of the Committees represents the Women's University Club
of Los Angeles, Calif., a constituent member of the Association of
University Women. What does that previewing group of university
women have to say about the fitness for children of the film output of
1932? They endorse only 19.8 per cent, and to get that high a per-
centage one has to include films which they pronounce only as "fair,"
"passable," "probably good," or "probably harmless." The number
that receive their enthusiastic endorsement is small indeed.

What, now, do children see when they go to the movies?

(a) Love, sex, and crime are the subjects of three-fourths of the

Dec., 1933] UNOCCUPIED MOTION PICTURE FIELDS 507

feature films. In recent films, romantic love plays a smaller part,
and sex and crime a larger part than ten years ago the world of the
movie is largely one of sex and crime.

(6) The films appear to be excessively concerned with crime.
Crime and violence have lately constituted 40 per cent of film content.
Of 115 feature films selected at random "only 26 were free from some
sort of crime." In the remaining eighty-nine a total of four hundred
forty-nine crimes were depicted, four hundred six being actually
committed, and forty-three others attempted. Of these four hun-
dred forty-nine crimes, eighty-three were murders and homicides,
fifty-four committed and twenty-nine others attempted; fifty-nine
were assaults; thirty-six gambling; twenty-five threatening with
weapons; twenty-one kidnappings; twenty carrying concealed weap-
ons; eighteen hold-ups, seventeen successful and one thwarted; twelve
verbal threats to kill; eleven fighting with weapons; eleven suicide
attempts, nine of them fatal.

(c) Life-goals shown in the movies are, on the whole, far from
inspiring. In these same one hundred fifteen films, five hundred
seventy-four life-goals or ambitions, are set forth. Only nine per
cent of them seem to be socially useful. The success striven for is
usually personal and selfish. It is very seldom that life-ambitions
in the movies are such as we have been accustomed to place before
our youth for admiration.

The contents of the films are, you see, a long way from being a cross-
section of life. The world of the movies, instead of being better than
what we see around us, is worse very much worse. It is decidedly
a cock-eyed world that one sees in the movies. Knowing, as the
country will now know, how much of this sort of thing the children
take in from the films and how tenaciously they remember it; how
uncritical they are toward it ; how powerfully it stirs their emotions ;
how it shapes their attitudes and controls their conduct; I can not
believe that they will continue to tolerate child attendance at movies
of the present kind. Why should they? Why should the motion
picture industry want them to?

What the industry wants, I assume, is business. The business of
supplying juvenile books for children is a distinct, a large, and a reason-
ably remunerative branch of the publishing trade. It can be made
an equally large and remunerative branch of the film trade. Juvenile
film classics can be produced at small expense children like simple
rather than elaborate things. Juvenile film classics can be shown

508 WILLIAM H. SHORT [j. s. M. P. E.

from year to year, for there is a large and steady crop of young chil-
dren. They will earn a profit for as long a period as do the plates of a
child's book. The children need and must have films, for, in the ex-
pressive words of Dr. Mitchell already quoted, they "meet an intensely
felt interest."

(3) Entertainment for Adults. Mr. Terry Ramsaye, a few years
ago, estimated the movie-going part of the people of the United States
to be 35,000,000 there are, of course "repeaters" that swell the
weekly attendance much beyond that figure. Ramsaye may have
underestimated the number who had the movie habit in 1925 or
1926 let us assume that it is larger than that now. Let us add 15,-
000,000 to Ramsaye's estimate of 35,000,000, and call it 50,000,000.*
Let us assume also that there are 20,000,000 others too young, too
old, or otherwise unable to go. There are still fifty-five millions left
who could, but do not go a pretty sizeable movie audience, larger
by five millions than we assume to be going now, even after adding
fifteen millions to Ramsaye's estimate.

It is replied that they do not attend when a good picture is offered
to them. The adequate reply to that is that a generation has been
spent in convincing that intelligent audience that there is little or
nothing for them in the movies. Also that a careful study of motion
picture advertising made by the Payne Fund shows that the industry
is not geared up to appeal to that audience. I can not deal further
with this matter, but I am ready to defend the thesis that that intelli-
gent group which has heretofore been ignored or flouted is the largest
and most profitable audience in the United States for motion picture
entertainment; that it has been alienated; and that it can be won
back by making and exhibiting adult drama.

What should be the nature of the adult drama, it has not been a
province of the Payne Fund research to determine. Personally, I
believe that the industry will have to learn how to deal effectively with
other dramatic material than the movie triology of love, sex, and
crime. Perhaps it will need to study more assiduously than it has
yet done, the nature of the motion picture as an art. I am quite
sure that it will have to discard the formula of producing chiefly for
the moron and the sophisticate. Whatever may be true of present
movie-goers, I am certain that among the fifty-five million non-goers is

* Dr. Howard T. Lewis, Prof, of Marketing in the Graduate School of Business
Administration at Harvard University, says in The Motion Picture Industry
(1933) that the movie-going part of the population in the U. S. is 25 per cent.

Dec., 1933] UNOCCUPIED MOTION PICTURE FIELDS 509

a vast potential audience of intelligent and clean-minded persons.
Nor do I despair of the present movie audience. Is the motion pic-
ture industry interested in this unreached audience of fifty-five mil-
lions, largely adults? I think that it should be interested.

(4) Good-Will. There are other unoccupied fields for the films
but less likely to return immediate profits. They are, however,
of a sort that far-sighted publishing houses are accustomed to culti-
vate both for their good-will value and for the prospect of ultimate
profit as well. The research just completed shows that the film is as
effective in imparting information to adults as to children; and that
the field of adult education in this country appears to be destined to
have a rapid development. Documentary films such as Cavalcade
have an important function. In addition, there is the whole field
of the scientific film.

Surely the movie art has yet many and rich realms to conquer at
which as yet it has scarcely more than glanced. It is high time
manifestly for clear-sighted and high-minded citizens to take coun-
cil concerning the extension of the usefulness of the motion picture art
into these vast unoccupied fields. Scientific research has fully estab-
lished its claim to equal place at the side of its sister art of printing.
This new art belongs to the people not to any group of men just
as printing, the electric light, the telephone, the radio, the internal
combustion engine, the aeroplane, the automobile products of
American genius belong to the people. No one of them is more
basic, and none so replete with social values as is the motion picture
art.

I would be the last to say a single word against the showmen whose
quick understanding and enterprise gave the engineers their oppor-
tunity to perfect the mechanical and technical equipment of the
motion picture art, and to present it to their generation complete and
adequate for the high services of which we have been speaking.
Rather, let my last word be one of regret that others besides show-
men have been so slow to see the social values of motion pictures;
and my hope and belief that this slowness of vision will soon be
amply atoned for. The larger understanding and appreciation of
this new and marvelous art that will come as a result of the Payne
Fund Research, must surely help powerfully to that end. To that
end, also, the Motion Picture Research Council hopes to do its bit
in alliance with the many other like-minded groups throughout the
country.

NEW MOTION PICTURE APPARATUS

MIRROR TELEPHOTO SYSTEM*

The problem of gaining "distance" in motion picture photography, in con-
tradistinction to that of achieving close-ups, has led to the development of tele-
photo objectives. When it is required that the focus be increased to a large ex-
tent, and if the aperture be kept within the usual range, the length of the required
attachment and the diameter of the front lens become rather impracticable.

The Askania Werke, Berlin, has recently developed a mirror telephoto system
similar in principle to the mirror system used in astronomical instruments for
lengthening the focus. It consists of a large ring mirror perforated at the center,
and a small round mirror mounted opposite the larger one. Incoming rays of
light from the object being photographed are reflected from the main mirror to
the smaller mirror, which, in turn, reflects them to the film in the camera through
the hole in the ring mirror. The system is arranged in a tube to be mounted in
front otf the camera, a small lever extending to the outside of the tube for making
fine focal adjustments of the small mirror, which moves axially (See Figs. 1 and 2.)

The distance between the two mirrors is about one-fifth the equivalent focus ;
the rays are converged after being reflected by both mirrors in the same way as
though they had passed through an objective having a focus five times as great
as the distance between the two mirrors. The system can be used with motion
picture cameras as well as with still cameras.

SPECIAL SPEED CAMERAS*

A high-speed camera has been developed by the Askania Werke, Berlin, using
continuously running film, which permits taking photographs on 16-mm. film at
the rate of 1000 pictures per second; or 2000 pictures per second if half-sized
frames are used. A special disk with 32 lenses is mounted behind the standard
objective lens, and is rotated by means of a simple belt drive at a speed of 3750
rpm . The film speed is about 65 feet per second.

For higher speeds a drum camera has been designed, in which a number of
pictures are taken on a strip of film 85 or 150 cms. long. The film strip is laid
against the inside rim of the drum and pressed firmly against the rim by the
centrifugal force developed when the drum is rotated at adjustable speeds up to
6000 rpm.; with the larger drum and half or quarter-size pictures, the speed
can be increased to 28,000 pictures per second. Only one objective is used.

* Described at a meeting of the Chicago section of the S. M. P. E. by H. P.
Xiemann, American Askania Corp.
510

NEW MOTION PICTURE APPARATUS

511

A multiple camera, following the original plan of Muybridge in 1877, consists
of 12 individual plate cameras mounted on a large disk. Between the objective
and the photographic plates a disk having 13 slits rotates at a high speed and acts
as a shutter for all 12 cameras. At a shutter speed of 6000 rpm. 14,000 pictures
per second are taken. With twice the number of objectives and plates, and the
same speed of rotation, the number can be increased to 58,000 pictures per
second.

FIG. 1. Exterior view of mirror telephoto system.

FIG. 2. Schematic arrangement of mirror telephoto system.

AUTOMATIC CHANGE-OVER DEVICE*

The Apasco automatic change-over device represents an effort to substitute a
mechanical-electrical device for visual means of depending upon the continuity of
motion picture projection from reel to reel. It consists of two magazines 20

* Apasco System, Hollywood, Calif.

512

NEW MOTION PICTURE APPARATUS [J. S. M. P. E.

inches square and 5 inches deep one for each projector (Fig. 3). A series of
twelve rollers has been arranged to accommodate the 18 feet of film required for
the change-over. The film is cut at a point 6 ft. 4 in. from the last scene of the
picture ; but in the event that the cast of characters occurs at the end of the reel
then the film must be cut at that distance from the 19th frame beyond the end of
the sound record.

The projectors are threaded in the usual manner; then the slack from the reel
is threaded around rollers 1,2,3, and 4; between rollers 5 and 6, turning roller 5
a complete turn to the left; and thence around rollers 7, 8, 9, 10, 11, and 12, to
the reel. Various modifications of the threading are provided for equipment with
fast or slow speed pick-up.

As the end of the film leaves the reel, roller 12 drops, starting the oncoming

FIG. 3. Automatic change-over device.

machine. Sufficient film has been wound around the rollers to allow the speed of
the oncoming machine to attain the required value. As the end of the film
leaves roller 6, roller 5 unwinds to its normal position, making a momentary con-
tact that operates the dousers and changes over the sound by means of a specially
constructed double-throw magnetic switch.

THE LUMEN ARC*

The Lumenarc is a high-intensity, fairly large-current arc discharge occurring
between hot cathodes at relatively low tube voltage; in contrast with the conven-
tional low-intensity, low-current neon tube discharges occurring between cold

* Described at a meeting of the West Coast Section of the S. M. P. E. by E. O.
Erickson, Electrical Products Corp.

Dec., 1933]

NEW MOTION PICTURE APPARATUS

513

electrodes at high tube voltages. The tubing is larger than neon sign tubing,
and the luminous intensity of the positive column of the discharge is much
greater.

Lumenarc is available in three colors; red, blue, and green. The red, or orange-
red, color is obtained by using neon gas, the blue, by the excitation of mercury
vapor in the presence of several of the rare gase.s, both colors being employed
with clear glass tubing. The green color is attained in the same way as the blue,
except that an amber or noviol (no-violet transmitting) glass is used. Pyrex is
used for tubes of all colors, in diameters of 18 to 25 mm. and lengths up to 12 feet.

The current ranges from 1 to 2 amperes, depending upon the nature of the
gaseous path of the discharge, at 600 volts for a 6-ft. length. For indoor use,
a gas mixture is used that permits the operation of 6-ft. lengths on commercial
220- volt circuits, or 3-ft. lengths on 110- volt circuits. The tubes are of the

FIG. 4. Semi-professional Filmo camera.

double-cathode type, and high reactance supply circuits are required to stabilize
the arc on account of its negative-impedance characteristic.

SEMI-PROFESSIONAL FILMO CAMERA*

The design of this camera (Fig. 4) has been based on that of the Filmo turret-
head camera, so that its new features can be applied to the previous models
Filmo 70D and TOD A. The camera is designed for 16-mm. film, its capacity
having been increased by the addition of an external 200-ft. magazine, which
does not interfere with alternative use of the camera by the internal loading of

* Bell & Howell Co., Chicago, 111.

514 NEW MOTION PICTURE APPARATUS [J. S. M. P. E.

100-ft. spools. It is driven either by a 12-volt or 110- volt synchronous motor,
as desired.

An 8 to 1 crank, usable when the motor is removed, permits continuing a scene
after the spring motor has run down. The camera governor controls the speed
accurately at any desired value between 8 and 64 frames per second, regardless
of whether electric motor, hand crank, or spring motor is used.

For lap dissolves and double exposures, the film may be moved backwards by
means of the hand crank. A newly designed range-finder is built into the camera
door, and subject distances may be quickly and accurately determined. The
range-finder is accurate to within one-fourth of an inch, for example, at a distance
of three feet. Single-frame exposures may be made by locking the starting
button gravity catch out of the operating position.

MICROSCOPIC MOTION PICTURES*

An attachment has been developed for making microscopic motion pictures with
the Bell & Howell 16-mm. motion picture camera and any ordinary microscope.

FIG. 5. Equipment for making microscopic motion pictures.

The device consists of a horizontal tube mounting a split-beam prism, which de-
flects about 90 per cent of the available photographic light in a parallel ray into
the standard 1-inch //3.5 camera lens, which remains set at infinity. The re-
mainder of the light passes up the microscope tube, set at 160 mm., over which
fits a finder-sleeve fitted with a mask that shows the limits of the field being
photographed by the camera. (See Fig. 5.)

The reduced amount of light reaching the eye makes it easy to observe the
object that is being photographed and to keep it in sharp focus by means of the
fine adjustment of the microscope itself. The third part of the accessory set-up
is an adjustable camera stand that raises and lowers the camera to the exact
height made necessary by the particular object under the microscope objective.
It is heavy enough to hold the camera rigid and parallel to the microscope tube.

* Bell & Howell Co., Chicago, 111.

Dec., 1933] NEW MOTION PICTURE APPARATUS

THE EYE-EASY EDITOR*

515

A new type of editing device for 16-mm. film, known as the Eye-Easy Editor,
projects a large "still" of the frame being inspected, and thereby relieves eye-
strain and facilitates editing.

FIG. 6. The "eye-easy" editor.

FIG. 7. S-O-F Animatophone.

* Victor Animatograph Corp., Davenport, Iowa.

516 NEW MOTION PICTURE APPARATUS

It is equipped with a special prism and projecting head which permit the pic-
ture to be projected any convenient distance and viewed right side up. It is also
equipped with a rewind which may be used with or independently of the projector
head and the built-in splicer which is also a part of the standard equipment
(Fig. 6). An interesting feature of the splicer is that it is heated to a given tem-
perature to insure quick and positive bonding of film.

Available as an extra item is a small film-pack camera which attaches to the
editor in place of the prism. The camera is complete with film- pack adapter,
ground glass, and camera lens. It is of the fixed exposure type and easy to
operate.

ANIMATOPHONE*

Threading and operating the Victor S-O-F projector (Fig. 7) are no more
complicated than with a silent projector. The sound head, comprised of exciter
lamp, lens, sound gate, photoelectric cell, and threading rolls, is side-mounted on
the support base of the projector and occupies a space of only 2'/ by 4'/ by 6
inches. The amplifier (5 tube) is mounted at the rear of the projector and oc-
cupies a space of only 6 by 7 by 8 inches. Auditorium speaker and 50-foot cord
are housed in a removable side of the projector carrying case. The entire equip-
ment in carrying case weighs only fifty pounds. Sound volume and picture
illumination are sufficient for comparatively large school and church auditoriums.

* Victor Animatograph Corp., Davenport, Iowa.

SOCIETY ANNOUNCEMENTS
WILLIAM K. L. DICKSON

At the last meeting of the Board of Governors, held at Chicago on Oct. 15,
1933, Mr. William K. L. Dickson, pioneer motion picture engineer who collabo-
rated with Thomas A. Edison in the early work on motion picture projectors,
was nominated for the grade of Honorary Membership in the Society of Motion
Picture Engineers. On the following day, Oct. 16, 1933, at the opening of the
Chicago Convention, the general membership of the Society voted unanimously
to approve the nomination made by the Board. The Society felicitates Mr.
Dickson, and will send to him shortly a formal certificate of Honorary Member-
ship endorsed by the officers of the Society and the Board of Governors. (A
history of Mr. Dickson's work is presented in this issue of the JOURNAL, on
page 435.)

NEW YORK SECTION

The results of the recent balloting for officers of the Section for the year Oct. 1,
1933, to Oct. 1, 1934, were as follows:

Chairman, H. G. TASKER

Sec.-Treas., D. E. HYNDMAN (reflected)

Manager, M. C. BATSEL (reelected)

The second Manager of the Section is J. L. Spence, whose term expires Oct.
1, 1934.

PACIFIC COAST SECTION

The results of the recent balloting for officers of the Section for the year Oct. 1,

1933, to Oct. 1, 1934, were as follows:

Chairman, E. HUSE (reelected)
Sec.-Treas., G. F. RACKETT (reelected)
Manager, W. C. HARCUS

The second Manager of the Section is J. A. Dubray, whose term expires Oct. 1,

1934. The Annual Business Meeting and Dinner of the Section were held on
Nov. 7, 1933, at which time the new officers were inducted and a paper was
presented on the technical aspects of the Hollywood Planetarium, now being
constructed, by Mr. W. Hartman.

517

518 SOCIETY ANNOUNCEMENTS

CHICAGO SECTION

The results of the recent balloting for officers of the Section for the year Oct. 1.

1933, to Oct. 1, 1934, were as follows:

Chairman, E. COUR
Sec.-Treas., C. H. STONE
Manager, B. E. STECHBART

The second Manager of the Section is O. B. Depue, whose term expires Oct. 1,

1934. A meeting of the Section was held on Nov. 16, 1933, at which time a
paper entitled "The Use of the Weston Photographic Exposure Meter in Cinema-
tography" was presented by Mr. P. A. Westburg.

CONSTITUTIONAL AMENDMENTS

There have recently been mailed to the Active members of the Society the
following items: (1) a pamphlet describing the proposed amendments of the
Constitution and By-Laws as presented at the Chicago Convention, Oct. 16-18,
1933; (2) a transcript of the discussion that followed the presentation of the
proposals at the Convention; (3) a copy of the present Constitution and By^
Laws; and (4) an official ballot for voting on the Constitutional amendment.
The proposed amendments of the By-Laws had already been approved by the
Society at the Chicago Convention subject to the subsequent approval of the
proposed amendments of the Constitution; this balloting, therefore, pertains
only to the proposed Constitutional amendments. The Active members are
urged to return their ballots to the General Office of the Society as quickly as
possible: they must be returned not later than Jan. 14, 1934, in order to be
counted.

SOCIETY OF MOTION PICTURE ENGINEERS

REPORT OF THE TREASURER
FOR THE PERIOD OCT. 1, 1932, TO SEPT. 30, 1933

Balance, September 30, 1932
Receipts during Period

Dues and Fees

Dues of active members
Dues of associate members
Dues of sustaining members
Admission fees

Publication Income
Journal sales
Reprints
Advertising

Other Income

Interest on bank balances
Certificates, badges and binders
Convention 'receipts
Miscellaneous receipts

Disbursements during Period

General Expenses

Convention expenses
Office rent and expenses
Salaries

Officers expenses
Committees and Sections
Contingency
Reporting discussions
General Society expenses

Publication Expenses
Journal
Reprints

Balance, September 30, 1933

Manufacturers' Trust Co.
Genesee Valley Trust Co.
Rochester Savings Bank
Bowery Savings Bank

$18,200.86

$4,712.87
2,980.39
2,200.00
225.00 $10,118.26

2,478 . 60
1,045.79
1,666.81

589 . 54
1 1 . 40

961 . 00
25 . 43

918.08

3,052 . 73

.7,100.16

283 . 21

1,061 .55

87.35

161.40

101 .40

5,469 . 18
789 . 14

5,191.20

1,587.37 16,896.83

$35,097.69

12,765.88

6,258.32 19,024.20

$16,073.49

1,275.64
5,320.26
7,565.02
1,912.57 $16,073.49

519

AUTHOR INDEX, VOLUME XXI

Author
AALBERG, J. O.

AGNEW, P. G.
BAKER, J. O.

(and BATSEL, C. N.)
BATSEL, C. N.

(and BAKER, J. O.)
BEGGS, E. W.

(and PALMER, M. W.)

BEST, G. M.
BRADLEY, I. L.

CANADY, D.
CHAMBERS, G. A.

(and WRATTEN, I. D.)

COOK, E. D.
CRABTREE, J.
CRABTREE, J.

(and WADDELL, J. H.)
CRABTREE, J. I.

(and RUSSELL, H. D.)
CRABTREE, J. I.

(and PARKER, H., JR.,

and RUSSELL, H. D.)
DICKSON, W. K. L.

DOWNES, A. C.

(and JOY, D. B.)
DUBRAY, J. A.
EVANS, R.

FARNHAM, R. E.

HALL, V. C.

(and SANDVIK, O.,
and STREIFFERT, J. G.)

520

JULY TO DECEMBER, 1933

Issue Page

A Triplex Moviola for Editing Re-
Recording Nov. 420
National Standardization in America Oct. 261
Sound Recording and Reproducing Using

16-Mm. Film Aug. 161

Sound Recording and Reproducing Using

16-Mm. Film Aug. 161

Professional Motion Picture Photography
with High-Intensity Short-Life In-
candescent Lamps Aug. 126
Economies in Sound Film Processing Sept. 236
Voice and Personality in the Motion

Pictures Sept. 209

Portable Recording Equipment Dec. 483

The Eastman Type 116 Sensitometer as a
Control Instrument in the Processing
of Motion Picture Film Sept. 218

The Aperture Alignment Effect Nov. 390

Sound Film Printing Oct. 294

Directional Effects in Sound Film Proc-
essing II Nov. 351
An Improved Potassium Alum Fixing

Bath Containing Boric Acid Aug. 137

Some Properties of Two-Bath Developers
for Motion Picture Film July 21

A Brief History of the Kinetograph, the
Kinetoscope, and the Kineto-Phono-
graph Dec. 435

A New Alternating-Current Projection

Arc Aug. 116

The Morgana Color Process Nov. 403

The Sound Film Program of the United

States Department of Agriculture Sept. 224

The Use of Mazda Lamps for Color

Photography Aug. 166

Wave-Form Analysis of Variable Width

Sound Records Oct. 323

INDEX

521

HANNA, C. R.

(and IRWIN, P. L.,
and REYNOLDS, E. W.)
HOORN, F. W.
HUSE, E.

IRWIN, P. L.

(and HANNA, C. R.,
and REYNOLDS, E. W.)
IVES, H. E.

JONES, L. A.

JOY, D. B.

(and DOWNES, A. C.)
KOSSMAN, H. R.
KREUZER, B.
KUESTER, A.

(and SCHMIDT, R.)

LASKY, M.

(and RUBIN, B.)
LEWIN, W.

MOLE, P.

PALMER, M. W.
(and BEGGS, E. W.)

PARKER, H., JR.

(and CRABTREE, J. I.,
and RUSSELL, H. D.)

REYNOLDS, E. W.
(and HANNA, C. R.,
and IRWIN, P. L.)

RIDGWAY, D. W.

RUBIN, B.

(and LASKY, M.)

RUSSELL, H. D.

(and CRABTREE, J. I.)

RUSSELL, H. D.

(and CRABTREE, J. I.,
and PARKER, H., JR.)

SANDVIK, O.

(and HALL, V. C.,
and STREIFFERT, J. G.)

A Sixteen-Millimeter Portable Sound-on-

Film Projection Equipment Dec. 456

Military Training and Historical Films Oct. 337
Sensitometric Control in the Processing of

Motion Picture Film in Hollywood July 54

A Sixteen-Millimeter Portable Sound-on-

Film Projection Equipment Dec. 456

An Experimental Apparatus for the Pro-
jection of Motion Pictures in Relief Aug. 106

A Historical Summary of Standardization
in the Society of Motion Picture
Engineers Oct. 280

A New Alternating- Current Projection

Arc Aug. 116

A Silent Camera Nov. 420

Radio City Sound Equipment Sept. 181

Analysis of Sound Quality with the
Variable Density Recording Method
from Sensitometric Data Nov. 374

A Practical Method and Photometer for

Controlling Exposures in Photography Aug. 154

Photoplay Appreciation in the Nation's

Schools July 9

New Developments in Portable Gas-
Electric Generators for Motion Picture
Lighting Nov. 413

Professional Motion Picture Photography
with High-Intensity Short-Life In-
candescent Lamps Aug. 126

Some Properties of Two-Bath Developers

for Motion Picture Film July 21

A Sixteen-Millimeter Portable Sound-on-

Film Projection Equipment Dec. 456

The Preselection of Takes for Processing

from Exposed Undeveloped Negative Sept. 230
A Practical Method and Photometer for

Controlling Exposures in Photography Aug. 154
An Improved Potassium Alum Fixing

Bath Containing Boric Acid Aug. 137

Some Properties of Two-Bath Developers

for Motion Picture Film July 21

Wave-Form Analysis of Variable Width

Sound Records Oct. 323

522

INDEX

SAWYER, C. R.

SCHMIDT, R.

(and KUESTER, A.)

SCHWARTZ, R. P.

(and TUTTLE, H. B.)
SHORT, W. H.
STOIBER, J.

(and WITTEL, O.,
and TUTTLE, F. E.)
STREIFFERT, J. G.,

(and SANDVIK, O.,
and HALL, V. C.)
THEISEN, E.
TUTTLE, C.

TUTTLE, F. E.
(and WITTEL, O.,
and STOIBER, J.)

TUTTLE, F. E.

TUTTLE, H. B.

(and SCHWARTZ, R. P.)
WADDELL, J. H.

(and CRABTREE, J.)
WITTEL, O.

(and STOIBER, O.,

and TUTTLE, F. E.)
WRATTEN, I. D.

(and CHAMBERS, G. A.)

A Light-Weight Single-Film Recording

System for Newsreels and Travelogues Dec. 466

Analysis of Sound Quality with the
Variable Density Recording Method
from Sensitometric Data Nov. 374

The New Cine-Kodak Special in Medi-
cine July 3

Unoccupied Motion Picture Fields Dec. 494

The Cine-Kodak Special Dec. 478

Wave-Form Analysis of Variable Width

Sound Records Oct. 323

The History of the Animated Cartoon Sept. 239
Distortion in the Projection and Viewing

of Motion Pictures Sept. 198

The Cine-Kodak Special Dec. 478

A Non-Intermittent High-Speed 16-Mm.

Camera Dec. 474

The New Cine-Kodak Special in Medi-
cine July 3

Directional Effects in Sound Film Proc-
essing II Nov. 351

The Cine-Kodak Special Dec. 478

The Eastman Type 116 Sensitometer
as a Control Instrument in the Proc-
essing of Motion Picture Film Sept. 218

CLASSIFIED INDEX, VOLUME XXI

JULY TO DECEMBER, 1933

Acting

Voice and Personality in the Motion Pictures, I. L. BRADLEY, No. 3 (Sept.).

p. 209.
Animation

The History of the Animated Cartoon, E. THEISEN, No. 3 (Sept.), p. 239.
Aperture

The Aperture Alignment Effect, E. D. COOK, No. 5 (Nov.), p. 390.
Apparatus

New Motion Picture Apparatus, No. 6 (Dec.), p. 510.
Arcs, Projection

A New Alternating- Current Projection Arc, D. B. JOY and A. C. DOWNES,

No. 2 (Aug.), p. 116.
Artistic Considerations

Voice and Personality in the Motion Pictures, I. L. BRADLEY, No. 3 (Sept.),
p. 209.

Cameras

A Silent Camera, H. R. KOSSMAN, No. 5 (Nov.), p. 420.
The Cine-Kodak Special, A. WITTEL, J. STOIBER, and F. E. TUTTLE, No. 6

(Dec.), p. 478
A Non-Intermittent High-Speed 16-Mm. Camera, F. E. TUTTLE, No. 6 (Dec.),

p. 474.

New Motion Picture Apparatus, No. 6 (Dec.), p. 510.
A Brief History of the Kinetograph, the Kinetoscope, and the Kineto-Phono-

graph, W. K. L. DICKSON, No. 6 (Dec.), p. 435.
Cartoons

The History of the Animated Cartoon, E. THEISEN, No. 3 (Sept.), p. 239.
Change-Overs

New Motion Picture Apparatus, No. 6 (Dec.), p. 510.
Color Cinematography

The Morgana Color Process, J. A. DUBRAY, No. 5 (Nov.), p. 403.
The Use of Mazda Lamps for Color Photography, R. E. FARNHAM, No. 2

(Aug.), p. 166.
Committee Reports

Projection Practice Committee, No. 1 (July), p. 84; No. 2 (Aug.), p. 89;

No. 4 (Oct.), p. 347.

Standards Committee, No. 1 (July), p. 84; No. 4 (Oct.), p. 347.
Report of the Non-Theatrical Equipment Committee, No. 1 (July), p. 16.

Densitometry (See Sensitometry)

523

524 IOT>EX [j. s. M. P. E.

Development of Motion Picture Film

Some Properties of Two-Bath Developers for Motion Picture Film, J. I.

CRABTREE, H. PARKER, JR., and H. D. RUSSELL, No. 1 (July), p. 21.
Directional Effects

Directional Effects in Sound Film Processing II, J. CRABTREE and J. H.

WADDELL, No. 5 (Nov.), P- 351.
Distortion

Distortion in the Projection and Viewing of Motion Pictures, C. TUTTLE, No.
3 (Sept.). p. 198.

Economies

Economies in Sound Film Processing, G. M. BEST, No. 3 (Sept.), p. 236.
The Preselection of Takes for Processing from Exposed Undeveloped Negative,

D. W. RIDGWAY, No. 3 (Sept.), p. 230.
Editing

A Triplex Moviola for Editing Re-Recording, J. O. AALBERG, No. 5 (Nov.),

p. 426.

New Motion Picture Apparatus, No. 6 (Dec.), p. 510.
Electrical Machinery and Equipment

New Developments in Portable Gas-Electric Generators for Motion Picture

Lighting, P. MOLE, No. 5 (Nov.), p. 413.
Exposure

A Practical Method and Photometer for Controlling Exposures in Photography,
M. LASKY and B. RUBIN, No. 2 (Aug.), p. 154.

Fixing of Motion Picture Film

An Improved Potassium Alum Fixing Bath Containing Boric Acid, H. D.
RUSSELL and J. I. CRABTREE, No. 2 (Aug.), p. 137.

General

National Standardization in America, P. G. AGNEW, No. 4 (Oct.), p. 261.
Voice and Personality in the Motion Pictures, I. L. BRADLEY, No. 3 (Sept.),

p. 209.
The Sound Film Program of the United States Department of Agriculture,

R. EVANS, No. 3 (Sept.), p. 224.

Military Training and Historical Films, F. W. HOORN, No. 4 (Oct.), p. 337.
Photoplay Appreciation in the Nation's Schools, W. LEWIN, No. 1 (July), p. 9.
Unoccupied Motion Picture Fields, W. H. SHORT, No. 6 (Dec.), p. 494.
Generators
New Developments in Portable Gas-Electric Generators for Motion Picture

Lighting, P. MOLE, No. 5 (Nov.), p. 413.
Glow Lamps
Portable Recording Equipment, D. CANADY, No. 6 (Dec.), p. 483.

Historical

Military Training and Historical Films, F. W. HOORN, No. 4 (Oct.), p. 337.
A Historical Summary of Standardization in the Society of Motion Picture

Engineers, L. A. JONES, No. 4 (Oct.), p. 280.
The History of the Animated Cartoon, E. THEISBN. No. 3 (Sept.;, p. 239.

Dec., 1933] INDEX 525

A Brief History of the Kinetograph, the Kinetoscope, and the Kineto-Phono-
graph, W. K. L. DICKSON, No. 6 (Dec.), p. 435.

Illumination in Photography

Professional Motion Picture Photography with High-Intensity Short-Life

Incandescent Lamps, M. W. PALMER and E. W. BEGGS, No. 2 (Aug.), p. 126.
The Use of Mazda Lamps for Color Photography, R. E. FARNHAM, No. 2

(Aug.), p. 166.
New Developments in Portable Gas-Electric Generators for Motion Picture

Lighting, P. MOLE, No. 5 (Nov.), p. 413.

Lamps for Photography

Professional Motion Picture Photography with High-Intensity Short-Life

Incandescent Lamps, M. W. PALMER and E. W. BEGGS, No. 2 (Aug.), p. 126.
The Use of Mazda Lamps for Color Photography, R. E. FARNHAM, No. 2

(Aug.), p. 166.

Lighting

Professional Motion Picture Photography with High-Intensity Short-Life

Incandescent Lamps, M. W. PALMER and E. W. BEGGS, No. 2 (Aug.), p. 126.
New Developments in Portable Gas-Electric Generators for Motion Picture

Lighting, P.' MOLE, No. 5 (Nov.), p. 413.

Medical Photography

The New Cine-Kodak Special in Medicine, H. B. TUTTLE and R. P. SCHWARTZ,
No. 1 (July), p. 3.

Meters

A Practical Method and Photometer for Controlling Exposures in Photography,
M. LASKY and B. RUBIN, No. 2 (Aug.), p. 154.

Microscopy

New Motion Picture Apparatus, No. 6 (Dec.), p. 510.

Military

Military Training and Historical Films, F. W. HOORN, No. 4 (Oct.), p. 337.

Newsreels

A Light-Weight Single-Film Recording System for Newsreels and Travelogues,
C. R. SAWYER, No. 6 (Dec.), p. 466.

Non-Intermittent Equipment

A Non-Intermittent High-Speed 16-Mm. Camera, F. TUTTLE, No. 6 (Dec.),
p. 510.

Non-Theatrical Equipment

Report of the Non-Theatrical Equipment Committee, No. 1 (July), p. 16.
New Motion Picture Apparatus, No. 6 (Dec.), p. 510.

Obituary

Walter Akemann, No. 2 (Aug.), p. 176.

William C. Hubbard, No. 2 (Aug.), p. 176; No. 3 (Sept.). p. 254.

526 INDEX [j. s. M. P. E.

Photometers

A Practical Method and Photometer for Controlling Exposures in Photography,
M. LASKY and B. RUBIN, No. 2 (Aug.), p. 154.

Portable Equipment

A Sixteen-Millimeter Portable Sound-on-Film Projection Equipment, C. R.

HANNA, P. L. IRWIN, and E. W. REYNOLDS, No. 6 (Dec.), p. 456.
New Developments in Portable Gas-Electric Generators for Motion Picture

Lighting, P. MOLE, No. 5 (Nov.), p. 413.
A Light- Weight Single-Film Recording System for Newsreels and Travelogues,

C. R. SAWYER, No. 6 (Dec.), p. 466.

Portable Recording Equipment, D. CANADY, No. 6 (Dec.), p. 483.

Preselection of Takes

Economies in Sound Film Processing, G. M. BEST, No. 3 (Sept.), p. 236.
The Preselection of Takes for Processing from Exposed Undeveloped Negative,

D. W. RJDGWAY, No. 3 (Sept.), p. 230.

Printing

Sound Film Printing, J. CRABTREE, No. 4 (Oct.), p. 294.

Processing

Economies in Sound Film Processing, G. M. BEST, No. 3 (Sept.), p. 236.
Directional Effects in Sound Film Processing II, J. CRABTREE and J. H.

WADDELL, No. 5 (Nov.), p. 351.
The Preselection of Takes for Processing from Exposed Undeveloped Negative,

D. W. RIDGWAY, No. 3 (Sept.), p. 230.

Processing, Control of

The Eastman Type 116 Sensitometer as a Control Instrument in the Processing

of Motion Picture Film, G. A. CHAMBERS and I. D. WRATTEN, No. 3 (Sept.),

p. 218.
Sensitometric Control in the Processing of Motion Picture Film in Hollywood,

E. HUSE, No. 1 (July), p. 54.

Projection, General Information

A New Alternating-Current Projection Arc, D. B. JOY and A. C. DOWNES,

No. 2 (Aug.), p. 116.
Distortion in the Projection and Viewing of Motion Pictures, C. TUTTLE, No. 3

(Sept.), p. 198.
A Sixteen-Millimeter Portable Sound-on-Film Projection Equipment, C. R.

HANNA, P. L. IRWIN, and E. W. REYNOLDS, No. 6 (Dec.), p. 456.
An Experimental Apparatus for the Projection of Motion Pictures in Relief,

H. E. IVES, No. 2 (Aug.), p. 106.
A Brief History of the Kinetograph, the Kinetoscope, and the Kineto-Phono-

graph, W. K. L. DICKSON, No. 6 (Dec.), p. 435.

Projection Practice

Report of Projection Practice Committee, No. 1 (July), p. 84; No. 2 (Aug.),
p. 89; No. 4 (Oct.), p. 347.

Dec., 1933] INDEX 527

Sensitometry
The Eastman Type 116 Sensitometer as a Control Instrument in the Processing

of Motion Picture Film, G. A. CHAMBERS and I. D. WRATTEN, No. 3 (Sept.),

p. 218.
Sensitometric Control in the Processing of Motion Picture Film in Hollywood,

E. HUSE, No. 1 (July), p. 54.
Analysis of Sound Quality with the Variable Density Recording Method from

Sensitometric Data, R. SCHMIDT and A. KUESTER No. 5 (Nov.), p. 374.

Sixteen-Millimeter Equipment
Sound Recording and Reproducing Using 16-Mm. Film, C. N. BATSEL and

J. O. BAKER, No. 2 (Aug.), p. 161.
The Cine-Kodak Special, O. WITTEL, J. STOIBER, and F. E. TUTTLE, No. 6

(Dec.), p. 478.
A Non-Intermittent High-Speed 16-Mm. Camera, F. E. Tuttle, No. 6 (Dec.),

p. 474.
A Sixteen-Millimeter Portable Sound-on-Film Projection Equipment, C. R.

HANNA, P. L. IRWIN, and E. W. REYNOLDS, No. 6 (Dec.), p. 456.

Sound Recording, Variable Density Method

Sound Recording and Reproducing Using 16-Mm. Film, C. N. BATSEL and J. O.

BAKER, No. 2 (Aug.), p. 161.
Analysis of Sound Quality with the Variable Density Recording Method from

Sensitometric Data, R. SCHMIDT and A. KUESTER, No. 5 (Nov.), p. 374.

Sound Recording, Variable Width Method

Sound Recording and Reproducing Using 16-Mm. Film, C. N. BATSEL and J. O.

BAKER, No. 2 (Aug.), p. 161.
Wave-Form Analysis of Variable Width Sound Records, O. SANDVTK,

V. C. HALL, and J. G. STREIFFERT, No. 4 (Oct.), p. 323.

Sound Reproduction, General Information
Sound Recording and Reproducing Using 16-Mm. Film, C. N. BATSEL and

J. O. BAKER, No. 2 (Aug.), p. 161.

Radio City Sound Equipment, B. KREUZER, No. 3 (Sept.), p. 181.
Light- Weight Single- Film Recording System for Newsreels and Travelogues,

C. R. SAWYER, No. 6 (Dec.), p. 466.

Portable Recording Equipment, D. CANADY, No. 6 (Dec.), p. 483.
A Brief History of the Kinetograph, the Kinetoscope, and the Kineto-Phono-

graph, W. K. L. DICKSON, No. 6 (Dec.), p. 435.

Standardization

National Standardization in America, P. G. AGNEW, No. 4 (Oct.), p. 261.
A Historical Summary of Standardization in the Society of Motion Picture
Engineers, L. A. JONES, No. 4 (Oct.), p. 280.

Standards Committee
No. 1 (July), p. 84; No. 4 (Oct.), p. 347.

Stereoscopy

An Experimental Apparatus for the Projection of Motion Pictures in Relief.
H. E. IVES, No. 2 (Aug.), p. 106.

528 INDEX

Studio Equipment

A Triplex Moviola for Editing Re-Recording, J. O. AALBERG, No. 5 (Nov.).

p. 426.
Studio Lighting

Professional Motion Picture Photography with High-Intensity Short-Life
Incandescent Lamps, M. W. PALMER and E. W. BEGGS. No. 2 (Aug.), p. 126.

Technical Cinematography

The New Cine-Kodak Special in Medicine, H. B. TUTTLE and R. P. SCHWARTZ,

No. 1 (July), p. 3.
Telephotography

New Motion Picture Apparatus, No. 6 (Dec.), p. 510.
Theater Equipment

Radio City Sound Equipment, B. KRBUZER, No. 3 (Sept.), p. 181.
Travelogues

A Light-Weight Single-Film Recording System for Newsreels and Travelogues,
C. R. SAWYER, No. 6 (Dec.), p. 466.

Viewing Devices

A Triplex Moviola for Editing Re-Recording, J. O. AALBERG, No. 5 (Nov ),
p. 426.