METHOD OF TAKING PICTURES FOR THE STEREOSCOPE.
However perfect be the stereoscope which we employ, the effect which it
produces depends upon the accuracy with which the binocular pictures
are prepared. The pictures required for the stereoscope may be arranged
in four classes:—
1. The representations of geometrical solids as seen with two eyes.
2. Portraits, or groups of portraits, taken from living persons or
animals.
3. Landscapes, buildings, and machines or instruments.
4. Solids of all kinds, the productions of nature or of art. */
_Geometrical Solids._
Representations of geometrical solids, were, as we have already seen,
the only objects which for many years were employed in the reflecting
stereoscope. The figures thus used are so well known that it is
unnecessary to devote much space to their consideration. For ordinary
purposes they may be drawn by the hand, and composed of squares,
rectangles, and circles, representing quadrangular pyramids, truncated,
or terminating in a point, cones, pyramids with polygonal bases, or
more complex forms in which raised pyramids or cones rise out of
quadrangular or conical hollows. All these figures may be drawn by the
hand, and will produce solid forms sufficiently striking to illustrate
the properties of the stereoscope, though not accurate representations
of any actual solid seen by binocular vision.
If one of the binocular pictures is not equal to the other in its
base or summit, and if the lines of the one are made crooked, it is
curious to observe how the appearance of the resulting solid is still
maintained and varied.
The following method of drawing upon a plane the dissimilar
representations of solids, will give results in the stereoscope that
are perfectly correct:—
[Illustration: FIG. 43.]
Let L, R, Fig. 43, be the left and right eye, and A the middle point
between them. Let MN be the plane on which an object or solid whose
height is CB is to be drawn. Through B draw LB, meeting MN in _c_; then
if the object is a solid, with its apex at B, C_c_ will be the distance
of its apex from the centre C of its base, as seen by the left eye.
When seen by the right eye R, C_c′_ will be its distance, _c′_ lying
on the left side of C. Hence if the figure is a cone, the dissimilar
pictures of it will be two circles, in one of which its apex is placed
at the distance C_c_ from its centre, and in the other at the distance
C_c′_ on the other side of the centre. When these two plane figures are
placed in the stereoscope, they will, when combined, represent a raised
cone when the points _c_, _c′_ are nearer one another than the centres
of the circles representing the cone’s base, and a _hollow_ cone when
the figures are interchanged.
If we call E the distance between the two eyes, and _h_ the height of
the solid, we shall have
E
AB:_h_ = ——— : C_c_,
2
_h_E 5_h_
and C_c_ = ————, or ————,
2AB 4AB
which will give us the results in the following table, E being 2½, and
AC 8 inches:—
Height of
object.
BC = _h_ AB = AC - _h_ C_c_
Inches.
1 7 0.179
2 6 0.4166
3 5 0.75
4 4 1.25
5 3 2.083
6 2 3.75
7 1 8.75
8 0 Infinite.
If we now converge the optic axes to a point _b_, and wish to ascertain
the value of C_c_, which will give different _depths_, _d_, of the
_hollow_ solids corresponding to different values of C_b_, we shall have
E
A_b_:—— = _d_:C_c′_,
2
_d_E
and C_c′_ = ————— AB,
2
which, making AC = 8 inches, as before, will give the following
results:—
Depth.
C_b_ = _d_ A_b_ = AC + _d_ C_c′_
Inches.
1 9 0.139
2 10 0.25
3 11 0.34
4 12 0.4166
5 13 0.48
6 14 0.535
7 15 0.58
8 16 0.625
9 17 0.663
10 18 0.696
11 19 0.723
12 20 0.75
The values of _h_ and _d_ when C_c_, C_c′_ are known, will be found
from the formulæ
2AB · C_c_
_h_ = ————————————,
E
2AB · C_c′_
and _d_ = ————————————.
E.
As C_c_ is always equal to C_c′_ in each pair of figures or dissimilar
pictures, the depth of the _hollow_ cone will always appear much
greater than the height of the raised one. When C_c_ = C_c′_ = 0.75,
_h_:_d_ = 3:12. When C_c_ = C_c′_ = 0.4166, _h_:_d_ = 2:4, and when
C_c_ = C_c′_ = 0.139, _h_:_d_ = 0.8:1.0.
When the solids of which we wish to have binocular pictures are
symmetrical, the one picture is the reflected image of the other, or
its reverse, so that when we have drawn the solid as seen by one eye,
we may obtain the other by copying its reflected image, or by simply
taking a copy of it as seen through the paper.
When the geometrical solids are not symmetrical, their dissimilar
pictures must be taken photographically from models, in the same manner
as the dissimilar pictures of other solids.
_Portraits of Living Persons or Animals._
Although it is possible for a clever artist to take two portraits, the
one as seen by his right, and the other as seen by his left eye, yet,
owing to the impossibility of fixing the sitter, it would be a very
difficult task. A bust or statue would be more easily taken by fixing
two apertures 2½ inches distant, as the two points of sight, but even
in this case the result would be imperfect. The photographic camera is
the only means by which living persons and statues can be represented
by means of two plane pictures to be combined by the stereoscope; and
but for the art of photography, this instrument would have had a very
limited application.
It is generally supposed that photographic pictures, whether in
Daguerreotype or Talbotype, are accurate representations of the human
face and form, when the sitter sits steadily, and the artist knows
the resources of his art. _Quis solem esse falsum dicere audeat?_
says the photographer, in rapture with his art. _Solem esse falsum
dicere audeo_, replies the man of science, in reference to the hideous
representations of humanity which proceed from the studio of the
photographer. The sun never errs in the part which he has to perform.
The sitter may sometimes contribute his share to the hideousness of his
portrait by involuntary nervous motion, but it is upon the artist or
his art that the blame must be laid.
If the single portrait of an individual is a misrepresentation of his
form and expression, the combination of two such pictures into a solid
must be more hideous still, not merely because the error in form and
expression is retained or doubled, but because the source of error
in the single portrait is incompatible with the application of the
stereoscopic principle in giving relief to the plane pictures. The art
of stereoscopic portraiture is in its infancy, and we shall therefore
devote some space to the development of its true principles and
practice.
In treating of the images of objects formed by lenses and mirrors
with spherical surfaces, optical writers have satisfied themselves by
shewing that the images of straight lines so formed are conic sections,
elliptical, parabolic, or hyperbolic. I am not aware that any writer
has treated of the images of solid bodies, and of their shape as
affected by the size of the lenses or mirrors by which they are formed,
or has even attempted to shew how a perfect image of any object can be
obtained. We shall endeavour to supply this defect.
In a previous chapter we have explained the manner in which images are
formed by a small aperture, H, in the side, MN, of a camera, or in the
window-shutter of a dark room. The rectangles _br_, _b′r′_, and _b″r″_,
are images of the object RB, according as they are received at the
same distance from the lens as the object, or at a less or a greater
distance, the size of the image being to that of the object as their
respective distances from the hole H. Pictures thus taken are accurate
representations of the object, whether it be lineal, superficial,
or solid, as seen from or through the hole H; and if we could throw
sufficient light upon the object, or make the material which receives
the image very sensitive, we should require no other camera for giving
us photographs of all sizes. The only source of error which we can
conceive, is that which may arise from the inflexion of light, but we
believe that it would exercise a small influence, if any, and it is
only by experiment that its effect can be ascertained.
[Illustration: FIG. 44.]
The Rev. Mr. Egerton and I have obtained photographs of a bust, in the
course of ten minutes, with a very faint sun, and through an aperture
less than the hundredth of an inch; and I have no doubt that when
chemistry has furnished us with a material more sensitive to light, a
camera without lenses, and with only a pin-hole, will be the favourite
instrument of the photographer. At present, no sitter could preserve
his composure and expression during the number of minutes which are
required to complete the picture.
But though we cannot use this theoretical camera, we may make some
approximation to it. If we make the hole H a quarter of an inch,
the pictures _br_, &c., will be faint and indistinct; but by
placing a thin lens a quarter of an inch in diameter in the hole
H, the distinctness of the picture will be restored, and, from the
introduction of so much light, the photograph may be completed in a
sufficiently short time. The lens should be made of rock crystal,
which has a small dispersive power, and the ratio of curvature of its
surfaces should be as six to one, the flattest side being turned to the
picture. In this way there will be very little colour and spherical
aberration, and no error produced by any striæ or want of homogeneity
in the glass.
As the hole H is nearly the same as the greatest opening of the pupil,
the picture which is formed by the enclosed lens will be almost
identical with the one we see in monocular vision, which is always the
most perfect representation of figures in relief.
[Illustration: FIG. 45.]
With this approximately perfect camera, let us now compare the
expensive and magnificent instruments with which the photographer
practises his art. We shall suppose his camera to have its lens or
lenses with an aperture of only _three_ inches, as shewn at LR in Fig.
45. If we cover the whole lens, or reduce its aperture to a quarter
of an inch, as shewn at _a_, we shall have a correct picture of the
sitter. Let us now take other _four_ pictures of the same person, by
removing the aperture successively to _b_, _c_, _d_, and _e_: It is
obvious that these pictures will all differ very perceptibly from each
other. In the picture obtained through _d_, we shall see parts on the
left side of the head which are not seen in the picture through _c_,
and in the one through _c_, parts on the right side of the head not
seen through _d_. In short, the pictures obtained through _c_ and _d_
are accurate dissimilar pictures, such as we have in binocular vision,
(the distance _cd_ being 2½ inches,) and fitted for the stereoscope. In
like manner, the pictures through _b_ and _e_ will be different from
the preceding, and different from one another. In the one through _b_,
we shall see parts _below_ the eyebrows, below the nose, below the
upper lip, and below the chin, which are not visible in the picture
through _e_, nor in those through _c_ and _d_; while in the picture
through _e_, we shall see parts above the brow, and above the upper
lip, &c., which are not seen in the pictures through _b_, _c_, and _d_.
In whatever part of the lens, LR, we place the aperture, we obtain
a picture different from that through any other part, and therefore
it follows, _that with a lens whose aperture is three inches, the
photographic picture is a combination of about one hundred and thirty
dissimilar pictures of the sitter, the similar parts of which are not
coincident_; or to express it in the language of perspective, _the
picture is a combination of about one hundred and thirty pictures of
the sitter, taken from one hundred and thirty different points of
sight_! If such is the picture formed by a _three_-inch lens, what must
be the amount of the _anamorphism_, or distortion of form, which is
produced by photographic lenses of diameters from _three_ to _twelve_
inches, actually used in photography?[46]
[46] See my _Treatise on Optics_, 2d edit., chap. vii. p. 65.
But it is not merely by the size of the lenses that hideous portraits
are produced. In cameras with two achromatic lenses, the rays which
form the picture pass through a large thickness of glass, which may
not be altogether homogeneous,—through _eight_ surfaces which may
not be truly spherical, and which certainly scatter light in all
directions,—and through an optical combination in which straight lines
in the object must be conic sections in the picture!
Photography, therefore, cannot even approximate to perfection till the
artist works with a camera furnished with a single quarter of an inch
lens of rock crystal, having its radii of curvature as six to one, or
what experience may find better, with an achromatic lens of the same
aperture. And we may state with equal confidence, that the photographer
who has the sagacity to perceive the defects of his instruments, the
honesty to avow it, and the skill to remedy them by the applications of
modern science, will take a place as high in photographic portraiture
as a Reynolds or a Lawrence in the sister art.
Such being the nature of single portraits, we may form some notion of
the effect produced by combining dissimilar ones in the stereoscope, so
as to represent the original in relief. The single pictures themselves,
including binocular and multocular representations of the individual,
must, when combined, exhibit a very imperfect portrait in relief,—so
imperfect, indeed, that the artist is obliged to take his two pictures
from points of sight different from the correct points, in order to
produce the least disagreeable result. This will appear after we have
explained the correct method of taking binocular portraits for the
stereoscope.
No person but a painter, or one who has the eye and the taste of
a painter, is qualified to be a photographer either in single or
binocular portraiture. The first step in taking a portrait or copying
a statue, is to ascertain in what aspect and at what distance from the
eye it ought to be taken.
In order to understand this subject, we shall first consider the
vision, with _one eye_, of objects of three dimensions, when of
different magnitudes and placed at different distances. When we thus
view a building, or a full-length or colossal statue, at a short
distance, a picture of all its visible parts is formed on the retina.
If we view it at a greater distance, certain parts cease to be seen,
and other parts come into view; and this change in the picture
will go on, but will become less and less perceptible as we retire
from the original. If we now look at the building or statue from a
distance through a telescope, so as to present it to us with the same
distinctness, and of the same apparent magnitude as we saw it at our
first position, the two pictures will be essentially different; all the
parts which ceased to be visible as we retired will still be invisible,
and all the parts which were not seen at our first position, but became
visible by retiring, will be seen in the telescopic picture. Hence
the parts seen by the near eye, and not by the distant telescope,
will be those towards the middle of the building or statue, whose
surfaces converge, as it were, towards the eye; while those seen by
the telescope, and not by the eye, will be the external parts of the
object, whose surfaces converge less, or approach to parallelism. It
will depend on the nature of the building or the statue which of these
pictures gives us the most favourable representation of it.
If we now suppose the building or statue to be reduced in the most
perfect manner,—to half its size, for example,—then it is obvious that
these two perfectly similar solids will afford a different picture,
whether viewed by the eye or by the telescope. In the reduced copy, the
inner surfaces visible in the original will disappear, and the outer
surfaces become visible; and, as formerly, it will depend on the nature
of the building or the statue whether the reduced or the original copy
gives the best picture.
If we repeat the preceding experiments with _two eyes_ in place of
_one_, the building or statue will have a different appearance;
surfaces and parts, formerly invisible, will become visible, and the
body will be better seen because we see more of it; but then the
parts thus brought into view being seen, generally speaking, with
one eye, will have less brightness than the rest of the picture. But
though we see more of the body in binocular vision, it is only parts
of vertical surfaces perpendicular to the line joining the eyes that
are thus brought into view, the parts of similar horizontal surfaces
remaining invisible as with one eye. It would require a pair of eyes
placed vertically, that is, with the line joining them in a vertical
direction, to enable us to see the horizontal as well as the vertical
surfaces; and it would require a pair of eyes inclined at all possible
angles, that is, a ring of eyes 2½ inches in diameter, to enable us to
have a perfectly symmetrical view of the statue.
These observations will enable us to answer the question, whether or
not a reduced copy of a statue, of precisely the same form in all
its parts, will give us, either by monocular or binocular vision,
a better view of it as a work of art. As it is the outer parts or
surfaces of a large statue that are invisible, its great outline and
largest parts must be best seen in the reduced copy; and consequently
its relief, or third dimension in space, must be much greater in the
reduced copy. This will be better understood if we suppose a _sphere_
to be substituted for the statue. If the sphere exceeds in diameter the
distance between the pupils of the right and left eye, or 2½ inches, we
shall not see a complete hemisphere, unless from an infinite distance.
If the sphere is very much larger, we shall see only a segment, whose
relief, in place of being equal to the radius of the sphere, is equal
only to the versed sine of half the visible segment. Hence it is
obvious that a reduced copy of a statue is not only better seen from
more of its parts being visible, but is also seen in stronger relief.
_On the Proper Position of the Sitter._
With these observations we are now prepared to explain the proper
method of taking binocular portraits for the stereoscope.
The first and most important step is to fix upon the position of
the sitter,—to select the best aspect of the face, and, what is of
more importance than is generally supposed, to determine the best
distance from the camera at which he should be placed. At a short
distance certain parts of one face and figure which should be seen are
concealed, and certain parts of other faces are concealed which should
be seen. Prominent ears may be either hid or made less prominent by
diminishing the distance, and if the sight of both ears is desirable
the distance should be increased. Prominent features become less
prominent by distance, and their influence in the picture is also
diminished by the increased vision which distance gives of the round
of the head. The outline of the face and head varies essentially with
the distance, and hence it is of great importance to choose the best. A
long and narrow face requires to be viewed at a different distance from
one that is short and round. Articles of dress even may have a better
or a worse appearance according to the distance at which we see them.
Let us now suppose the proper distance to be _six feet_, and since it
is impossible to give any rules for taking binocular portraits with
large lenses we must assume a standard camera with a lens a quarter of
an inch in diameter, as the only one which can give a correct picture
as seen with one eye. If the portrait is wanted for a ring, a locket,
or a binocular slide, its size is determined by its purpose, and the
photographer must have a camera (which he has not) to produce these
different pictures. His own camera will, no doubt, take a picture for a
ring, a locket, or a binocular slide, but he does this by placing the
sitter at different distances,—at a very great distance for the ring
picture, at a considerable distance for the locket picture, and at a
shorter distance for the binocular one; but none of these distances are
the distance which has been selected as the proper one. With a single
lens camera, however, he requires only several quarter-inch lenses
of different focal lengths to obtain the portrait of the sitter when
placed at the proper distance from the camera.
In order to take binocular portraits for the stereoscope a binocular
camera is required, having its lenses of such a focal length as to
produce two equal pictures of the same object and of the proper size.
Those in general use for the lenticular stereoscope vary from 2.1
inches to 2.3 in breadth, and from 2.5 inches to 2.8 in height, the
distance between similar points in the two pictures varying from 2.30
inches to 2.57, according to the different distances of the foreground
and the remotest object in the picture.
Having fixed upon the proper distance of the sitter, which we shall
suppose to be _six_ feet,—a distance very suitable for examining a
bust or a picture, we have now to take two portraits of him, which,
when placed in the stereoscope, shall have the same relief and the
same appearance as the sitter when viewed from the distance of _six
feet_. This will be best done by a binocular camera, which we shall now
describe.
_The Binocular Camera._
This instrument differs from the common camera in having two lenses
with the same aperture and focal length, for taking at the same instant
the picture of the sitter as seen at the distance of _six_ feet, or any
other distance. As it is impossible to grind and polish two lenses,
whether single or achromatic, of exactly the same focal length, even
when we have the same glass for both, we must bisect a good lens,
and use the two semi-lenses, ground into a circular form, in order
to obtain pictures of exactly the same size and definition. These
lenses should be placed with their diameters of bisection parallel to
one another, and perpendicular to the horizon, at the distance of 2½
inches, as shewn in Fig. 45, where MN is the camera, L, L′ the two
lenses, placed in two short tubes, so that by the usual mechanical
means they can be directed to the sitter, or have their axes converged
upon him, as shewn in the Figure, where AB is the sitter, _ab_ his
image as given by the lens L, and _a′b′_ as given by the lens L′. These
pictures are obviously the very same that would be seen by the artist
with his two eyes at L and L′, and as
ALB = _a_L_b_ = _a′_L′_b′_,
the pictures will have the same apparent magnitude as the original, and
will in no respect differ from it as seen by each eye from E, E′, E_a_
being equal to _a_L, and E′_a′_ to _a_L.
[Illustration: FIG. 46.]
Since the publication in 1849 of my description of the _binocular
camera_, a similar instrument was proposed in Paris by a photographer,
M. Quinet, who gave it the name of _Quinetoscope_, which, as the Abbé
Moigno observes, means an instrument for seeing M. Quinet! I have not
seen this camera, but, from the following notice of it by the Abbé
Moigno, it does not appear to be different from mine:—“Nous avons été à
la fois surpris et très-satisfait de retrouver dans le _Quinetoscope_
la chambre binoculaire de notre ami Sir David Brewster, telle que nous
l’avons décrite après lui il y a dix-huit mois dans notre brochure
intitulée _Stéréoscope_.” Continuing to speak of M. Quinet’s camera,
the Abbé is led to criticise unjustly what he calls the limitation
of the instrument:—“En un mot, ce charmant appareil est aussi bien
construit qu’il peut être, et nous désirons ardemment qu’il se répand
assez pour récompenser M. Quinet de son habileté et de ses peines.
Employé dans les limites fixées à l’avance par son véritable inventeur,
Sir David Brewster; c’est-à-dire, employé _à reproduire des objets
de petite et moyenne grandeur, il donnera assez beaux résultats_.
_Il ne pourra pas servir, evidemment, il ne donnera pas bien l’effet
stéréoscopique voulu, quand on voudra l’appliquer à de très-grands
objets, on a des vues ou pay sages pris d’une très-grande distance;
mais il est de la nature des œuvres humaines d’être essentiellement
bornées._”[47] This criticism on the limitation of the camera is
wholly incorrect; and it will be made apparent, in a future part of
the Chapter, that for objects of all sizes and at all distances the
binocular camera gives the very representations which we see, and that
other methods, referred to as superior, give unreal and untruthful
pictures, for the purpose of producing a startling relief.
In stating, as he subsequently does, that the angles at which the
pictures should be taken “are too vaguely indicated by theory,”[48] the
Abbé cannot have appealed to his own optical knowledge, but must have
trusted to the practice of Mr. Claudet, who asserts “that there cannot
be any rule for fixing the binocular angle of camera obscuras. _It is
a matter of taste and artistic illusion._”[49] No question of science
can be a matter of taste, and no illusion can be artistic which is a
misrepresentation of nature.
[47] See _Cosmos_, vol. ii. pp. 622, 624.
[48] _Id._ vol. vii. p. 494.
[49] _Id._ vol. iii. p. 658.
When the artist has not a binocular camera he must place his single
camera successively in such positions that the axis of his lens may
have the directions EL, EL′ making an angle equal to LCL′, the angle
which the distance between the eyes subtends at the distance of the
sitter from the lenses. This angle is found by the following formula:—
½_d_ 1.25
Tang. ½A = ———— = ————
D D
_d_ being the distance between the eyes, D the distance of the sitter,
and A the angle which the distance between the eyes, = 2.5, subtends at
the distance of the sitter. These angles for different distances are
given in the following table:—
D = Distance A = Angle formed
of Camera by the two
from the directions
Sitter. of the Camera.
5 inches, 28° 6′
6, 23 32
7, 20 14
8, 17 46
9, 15 48
10, 14 15
11, 13 0
12, 1 foot, 11 54
13, 11 0
14, 10 17
15, 9 32
16, 8 56
17, 8 24
18, 7 56
19, 7 31
20, 7 10
24, 2 feet, 5 58
30, 4 46
36 inches, 3 feet, 3 59
42, 3 25
48, 4 feet, 2 59
54, 2 39
60, 5 feet, 2 23
72, 6 feet, 1 59
84, 7 feet, 1 42
96, 8 feet, 1 30
108, 9 feet, 1 20
120, 10 feet, 1 12
The numbers given in the greater part of the preceding table can be
of use only when we wish to take binocular pictures of small objects
placed at short distances from cameras of a diminutive size. In
photographic portraiture they are of no use. The correct angle for a
distance of _six_ feet must not exceed _two_ degrees,—for a distance
of _eight_ feet, _one and a half_ degrees, and for a distance of
_ten_ feet, _one and a fifth_ degree. Mr. Wheatstone has given quite
a different rule. He makes the angle to depend, not on the distance
of the sitter from the camera, but _on the distance of the binocular
picture in the stereoscope from the eyes of the observer_! According
to the rule which I have demonstrated, the angle of convergency for
a distance of _six feet_ must be 1° 59′, whereas in a stereoscope of
any kind, with the pictures _six_ inches from the eyes, Mr. Wheatstone
makes it 23° 32′! As such a difference is a scandal to science, we
must endeavour to place the subject in its true light, and it will
be interesting to observe how the problem has been dealt with by
the professional photographer. The following is Mr. Wheatstone’s
explanation of his own rule, or rather his mode of stating it:—
“With respect,” says he, “to the means of preparing the binocular
photographs, (and in this term I include both Talbotypes and
Daguerreotypes,) little requires to be said beyond a few directions as
to the proper positions in which it _is necessary_ to place the camera
in order to obtain the two required projections.
“We will suppose that the binocular pictures are required to be seen
in the stereoscope at a distance of eight inches before the eyes, in
which case the convergence of the optic axes is about 18°. To obtain
the proper projections for this distance, the camera must be placed
with its lens accurately directed towards the object successively in
two points of the circumference of a circle, of which the object is the
centre, and the points at which the camera is so placed must have the
angular distance of 18° from each other, exactly that of the optic axes
in the stereoscope. The distance of the camera from the object may be
taken arbitrarily, for so long as the same angle is employed, whatever
that distance may be, _the picture will exhibit in the stereoscope
the same relief_, and be seen at the same distance of eight inches,
_only_ the magnitude of the picture will appear different. Miniature
stereoscopic representations of buildings and full-sized statues are,
therefore, obtained merely by taking the two projections of the object
from a considerable distance, _but at the same time as if the object
were only eight inches distant_, that is, at an angle of 18°.”[50]
[50] _Phil. Trans._, 1852, p. 7.
Such is Mr. Wheatstone’s rule, for which he has assigned no reason
whatever. In describing the binocular camera, in which the lenses
must be only 2½ inches distant for portraits, I have shewn that the
pictures which it gives are perfect representations of the original,
and therefore pictures taken with _lenses_ or cameras at any other
distance, must be different from those which are seen by the artist
looking at the sitter from his camera. They are, doubtless, both
pictures of the sitter, but the picture taken by Mr. Wheatstone’s rule
is one which no man ever saw or can see, until he can place his eyes
at the distance of _twenty inches_! It is, in short, the picture of a
living doll, in which parts are seen which are never seen in society,
and parts hid which are always seen.
In order to throw some light upon his views, Mr. Wheatstone got “a
number of Daguerreotypes of the same bust taken at a variety of
different angles, so that he was enabled to place in the stereoscope
two pictures taken at any angular distance from 2° to 18°, the former
corresponding to a distance of about six feet, and the latter to a
distance of about eight inches.” In those taken at 2°, (the proper
angle,) there is “an undue elongation of lines joining two unequally
distant points, so that all the features of a bust appear to be
exaggerated in depth;” while in those taken at 18°, “there is an undue
shortening of the same lines, so that the appearance of a bas-relief is
obtained from the two projections of the bust, the apparent dimensions
in breadth and height remaining in both cases the same.”
Although Mr. Wheatstone speaks thus decidedly of the relative effect
produced by combining pictures taken at 2½° and 18°, yet in the very
next paragraph he makes statements entirely incompatible with his
previous observations. “When the optic axes,” he says, “are parallel,
_in strictness_ there should be no difference between the pictures
presented to each eye, and in this case _there would be no binocular
relief_, but I find _that an excellent effect_ is produced when the
axes are nearly parallel, by pictures taken at an inclination of 7° or
8°, and even a difference of 16° or 17° _has no decidedly bad effect_!”
That Mr. Wheatstone observed all these contradictory facts we do
not doubt, but why he observed them, and what was their cause, is
a question of scientific as well as of practical importance. Mr.
Wheatstone was not aware[51] that the Daguerreotype pictures which
he was combining, taken with _large lenses_, were not pictures as
seen with two human eyes, but were actually binocular and multocular
monstrosities, entirely unfit for the experiments he was carrying on,
and therefore incapable of testing the only true method of taking
binocular pictures which we have already explained.
[51] Mr. Wheatstone’s paper was published before I had pointed out the
deformities produced by large lenses. See p. 130.
Had Mr. Wheatstone combined pictures, each of which was a correct
monocular picture, as seen with each eye, and as taken with a small
aperture or a small lens, he would have found no discrepancy between
the results of observation and of science. From the same cause, we
presume, namely, the use of multocular pictures, Mr. Alfred Smee[52]
has been led to a singular method of taking binocular ones. In one
place he implicitly adopts Mr. Wheatstone’s erroneous rule. “The
pictures for the stereoscope,” he says, “are taken at two stations, at
a greater or less distance apart, according to the distance at which
they are to be viewed. For a distance of 8 inches the two pictures are
taken at angles of 18°, for 13 inches 10°, for 18 inches 8°, and for 4
feet 4°.” But when he comes to describe his own method he seems to know
and to follow the true method, if we rightly understand his meaning.
“To obtain a binocular picture of anybody,” he says, “the camera must
be employed to take half the impression, and then it must be moved
in the arc of a circle of which the distance from the camera to the
_point of sight_[53] is the radius for about 2½ inches when a second
picture is taken, and the two impressions conjointly form one binocular
picture. There are many ways by which this result may be obtained. A
spot may be placed on the ground-glass on which the point of sight
should be made exactly to fall. The camera may then be moved 2½ inches,
and adjusted till the point of sight falls again upon the same spot
on the ground-glass, when, if the camera has been moved in a true
horizontal plane _the effect of the double picture will be perfect_.”
This is precisely the true method of taking binocular pictures which we
had given long before, but it is true only when small lenses are used.
In order to obtain this motion in the true arc of a circle the camera
was moved on two cones which converged to the point of sight, and Mr.
Smee thus obtained pictures of the usual character. But in making
these experiments he was led to take pictures _when the camera was in
continual motion backwards and forwards for 2½ inches_, and he remarks
that “_in this case the picture was even more beautiful than when the
two images were superimposed_!” “This experiment,” he adds, “is very
remarkable, for who would have thought formerly that a picture could
possibly have been made with a camera in continual motion? Nevertheless
we accomplish it every day with ease, and the character of the likeness
is wonderfully improved by it.” We have now left the regions of
science, and have to adjudicate on a matter of opinion and taste. Mr.
Smee has been so kind as to send me a picture thus taken. It is a good
photograph with features enlarged in all azimuths, but it has no other
relief than that which we have described as monocular.
[52] _The Eye in Health and Disease_, by Alfred Smee, 2d edit. 1854,
pp. 85-95.
[53] This expression has a different meaning in perspective. We
understand it to mean here the point of the sitter or object, which is
to be the centre of the picture.
A singular effect of combining pictures taken at extreme angles has
been noticed by Admiral Lageol. Having taken the portrait of one of
his friends when his eyes were directed to the object-glass of the
camera, the Admiral made him look at an object 45°! to the right,
and took a second picture. When these pictures were placed in the
stereoscope, and viewed “without ceasing, turning first to the right
and then to the left, the eyes of the portrait follow this motion as
if they were animated.”[54] This fact must have been noticed in common
stereoscopic portraits by every person who has viewed them alternately
with each eye, but it is not merely the eyes which move. The nose, and
indeed every feature, changes its place, or, to speak more correctly,
the whole figure leaps from the one binocular position into the
other. As it is unpleasant to open and shut the eyes alternately, the
same effect may be more agreeably produced in ordinary portraits by
merely intercepting the light which falls upon each picture, or by
making an opaque screen pass quickly between the eyes and the lens,
or immediately below the lens, so as to give successive vision of
the pictures with each eye, and with both. The motion of the light
reflected from the round eyeball has often a striking effect.
[54] _Cosmos_, Feb. 29, 1856, vol. viii. p. 202.
From these discussions, our readers will observe that the science, as
well as the art of binocular portraiture for the stereoscope, is in a
transition state in which it cannot long remain. The photographer who
works with a very large lens chooses an angle which gives the least
unfavourable results; his rival, with a lens of less size, chooses,
on the same principle, a different angle; and the public, who are no
judges of the result, are delighted with their pictures in relief, and
when their noses are either pulled from their face, or flattened upon
their cheek, or when an arm or a limb threatens to escape from their
articulation, they are assured that nature and not art is to blame.
We come now to consider under what circumstances the photographer may
place the lenses of his binocular camera at a greater distance than 2½
inches, or his two cameras at a greater angle than that which we have
fixed.
1. In taking family portraits for the stereoscope, the cameras must be
placed at an angle of 2° for 6 feet, when the binocular camera is not
used.
2. In taking binocular pictures of any object whatever, when we wish
to see them exactly as we do with our two eyes, we must adopt the same
method.
3. If a portrait is wanted to assist a sculptor in modelling a statue,
a great angle might be adopted, in order to shew more of the head. But
in this case the best way would be to take the correct social likeness,
and then take photographs of the head in different azimuths.
If we wish to have a greater degree of relief than we have with our
two eyes, either in viewing colossal statues, or buildings, or
landscapes, where the deviation from nature does not, as in the human
face, affect the expression, or injure the effect, we must increase
the distance of the lenses in the binocular camera, or the angle of
direction of the common camera. Let us take the case of a colossal
statue 10 feet wide, and suppose that dissimilar drawings of it about
_three_ inches wide are required for the stereoscope. These drawings
are _forty_ times narrower than the statue, and must be taken at such
a distance, that with the binocular camera the relief would be almost
evanescent. We must therefore suppose the statue to be reduced _n_
times, and place the semi-lenses at the distance _n_ × 2½ inches. If
_n_ = 10, the statue 10 feet wide will be reduced to ¹⁰/₁₀, or to 1
foot, and _n_ × 2½, or the distance of the semi-lenses will be 25
inches. With the lenses at this distance, the dissimilar pictures of
the statue will reproduce, when combined, a statue one foot wide, which
will have exactly the same appearance and relief as if we had viewed
the colossal statue with eyes 25 inches distant. But the reproduced
statue will have also the same appearance and relief as a statue a foot
wide reduced from the colossal one with mathematical precision, and it
will therefore be a better or more relieved representation of the work
of art than if we had viewed the colossal original with our own eyes,
either under a greater, an equal, or a less angle of apparent magnitude.
We have supposed that a statue a foot broad will be seen in proper
relief by binocular vision; but it remains to be decided whether or
not it would be more advantageously seen if reduced with mathematical
precision to a breadth of 2½ inches, the width of the eyes, which
gives the vision of a hemisphere 2½ inches in diameter with the most
perfect relief.[55] If we adopt this principle, and call B the breadth
of the statue of which we require dissimilar pictures, we must make
_n_ = B/2½, and _n_ × 2½ = B, that is, the distance of the semi-lenses
in the binocular camera, or of the lenses in two cameras, must be made
equal to the breadth of the statue.
In concluding this chapter, it may be proper to remark, that unless
we require an increased relief for some special purpose, landscapes
and buildings should be taken with the normal binocular camera, that
is, with its lenses 2½ inches distant. Scenery of every kind, whether
of the picturesque, or of the sublime, cannot be made more beautiful
or grand than it is when seen by the traveller himself. To add an
artificial relief is but a trick which may startle the vulgar, but
cannot gratify the lover of what is true in nature and in art.
_The Single Lens Binocular Camera._
As every photographer possesses a camera with a lens between 2½ and 3
inches in diameter, it may be useful to him to know how he may convert
it into a binocular instrument.
In a cover for the lens take two points equidistant from each other,
and make two apertures, _c_, _d_, Fig. 43, ²/₁₀ths of an inch in
diameter, or of any larger size that may be thought proper, though
²/₁₀ is the proper size. Place the cover on the end of the tube, and
bring the line joining the apertures into a horizontal position.
Closing one aperture, take the picture of the sitter, or of the
statue, through the other, and when the picture is shifted aside by
the usual contrivances for this purpose, take the picture through the
other aperture. These will be good binocular portraits, fitted for
any stereoscope, but particularly for the Achromatic Reading Glass
Stereoscope. If greater relief is wanted, it may be obtained in larger
lenses by placing the two apertures at the greatest distance which the
diameter of the lens will permit.
[55] It is only in a horizontal direction that we can see 180° of the
hemisphere. We would require a circle of eyes 2½ inches distant to see
a complete hemisphere.
_The Binocular Camera made the Stereoscope._
If the lenses of the binocular camera, when they are whole lenses, be
made to separate a little, so that the distance between the centres of
their inner halves may be equal to 2½ inches, they become a lenticular
stereoscope, in which we may view the pictures which they themselves
create. The binocular pictures are placed in the camera in the very
place where their negatives were formed, and the observer, looking
through the halves of his camera lenses, will see the pictures united
and in relief. If the binocular camera is made of semi-lenses, we
have only to place them with their thin edges facing each other to
obtain the same result. It will appear, from the discussions in the
following chapter, that such a stereoscope, independently of its being
achromatic, if the camera is achromatic, will be the most perfect of
stereoscopic instruments.
The preceding methods are equally applicable to landscapes, machines,
and instruments, and to solid constructions of every kind, whether they
be the production of nature or of art.[56]
[56] See Chapters X. and XI.