Chapter 11
Historical Enlargements and Image Reversal

This chapter also discusses tinting and age deterioration.


It is unfortunate that many people, including some writers, have the misconception that photographic enlarging is an advanced technology that appeared late on the scene. Several writers were under the impression that in the early days of photography enlargements weren't possible, so if you wanted an 8x10 print you needed a negative of the same size.

Not so. Enlargers have existed from the beginnings of photography. Sir John Herschel described his in 1839; it even had a lens corrected for spherical aberration. That same year Talbot patented an enlarger for his calotypes. Draper enlarged Daguerreotypes with a copy camera in Massachusetts during the winter of 1839‑1840. By 1857 full‑figure portraits six feet tall were being made and Woodward's solar enlarger was in widespread use.

It is true that most early photographers preferred large plate cameras. William Henry Jackson is famous for hauling a 20x24 inch glass‑plate camera across the western mountains of the United States on muleback in 1875 and making superb contact prints. Possibly a lighter and smaller camera would have enabled Jackson to take even more breathtaking pictures. On the occasion of his ninetieth birthday in the middle 1930's Jackson was presented with a Leica 35mm camera. He remarked (National Geographic, Vol 175m No. 2, February 1989, p230.) "If I'd had one of these on the Hayden Survey, I'd have made many more pictures and lived longer." Yet Ansel Adams often used the 8 x 10 inch format for many of his classic pictures. Adams had a choice that Jackson did not. The transition from Jackson's 90 pound camera to the one pound miniature in less than a lifetime gives talent a wider scope but does not substitute for it.

Enlarger Light Sources

Enlargers cost money and not all photographers felt they were a business necessity. Exposures were lengthy before the days of fast bromide paper, and light sources were a problem. Inventors tried every kind of artificial light: candles, lamps burning kerosene, whale oil, coal gas, and acetylene; battery powered carbon arc lights; hydrogen‑oxygen limelight. The latter consisted of a cylinder of lime (calcium carbonate), heated in a gas or hydrogen‑oxygen flame. It produced a brilliant white light much superior to the yellow light of kerosene. It was first used for general illumination in 1826, and in 1841 to illuminate subjects for calotypes. Some photographers used acetylene thirty years after Edison invented the electric lamp in 1879, either because their places of business were not electrified, or simply because they thought the results were better. Also, early incandescent light bulb filaments were too large to be placed at the focus of a parabolic reflector to produce a parallel beam.

The sun was the cheapest light when it was shining. New York was much better than Boston for solar work; England was terrible, and much of the European continent was not much better. Exposures of forty five minutes for albumen paper were common, and because of the changing direction of the sun, enlargers or mirrors had to be adjusted every five minutes for uniform exposure. Fires were common, too, since the enlarger lens could act as a burning glass if it was not carefully focused and aimed. Sometimes clockwork was used to keep the enlarger pointed correctly, like an astronomical telescope, but apprentices were cheaper.

The quality of early enlargements was generally inferior to most modern results because of grain and lens aberrations, but sometimes these qualities were an aid in impressionistic work. Enlarged portraits, however, were best viewed at a distance. The history of enlarging is well documented. Ostroff's paper [108] is quite comprehensive; good descriptions can also be found in Eder [48], Gernsheim [61], Gilbert [65], Newhall [105], and Taft [140].

Image Reversal
The property of camera lenses that produces a reversed image is basically simple but often misunderstood. Many writers assume that what they term 'left‑to‑right reversal' is self evident to readers. But why 'left‑to‑right': why not 'top‑to‑bottom?' Users of 35mm reflex cameras see a normal non‑reversed image and their final prints come out the same way. Why? The users of view cameras and studio cameras are constantly aware that the images on their focusing screens are upside down; are their lenses somehow inferior to 35mm camera lenses? These questions are relevant to collectors because nineteenth century photographs may be negatives, negative/positives, direct positives, transfers, copies, or reversed by mirrors or prisms.

Camera lenses translate each picture element in a scene from its original position with reference to the center axis of the lens to a corresponding position on the focal plane on the opposite side of the axis. The lens acts as a crossover point for light rays from the scene.

Consider the focal plane to be occupied by transparent film (or the ground glass of a view camera), and view it from the position of the person taking the picture. The image is reversed both left‑to‑right and top‑to‑bottom. All this observer has to do is to stand on his or her head and everything looks normal (except possibly the photographer). Users of view cameras seldom do this in public, but there is an occasional temptation to do so. Lenses are symmetrical about their optical axes, so turning the camera upside down is no help.

If the transparent film is developed and fixed, we simply turn it right side up and call it a negative, as Fox Talbot did and thereby became immortalized. A 'negative' should really be called a 'negative transparency' to distinguish it from positive transparencies, or lantern slides. Printing a positive from a negative cancels lens reversal if it is done emulsion‑to‑emulsion. Light can be transmitted from either side: if it comes from the emulsion side, the projected image is reversed, as anyone knows who has given a lecture and found the lantern slide captions reversed on the screen. The emulsion side has to be away from the light source to avoid reversal of the projected image; that is, emulsion-to-screen.

Most photographic processors are careful to adhere to the printing rule (either for enlarging or contact printing): always print emulsion‑to‑emulsion. But who knows how many times the rule has been violated, either accidentally or intentionally for esthetic effect? All we can do is to be aware of the basic characteristics of the various nineteenth century processes and to be on the lookout for helpful clues.

Binoculars contain internal prisms, and single-lens reflex cameras contain both prisms and mirrors, to restore the viewfinder image to normal orientation. Reflex camera prisms turn out to require five sides, hence the name 'pentaprism.' The reason for five sides is not obvious: interested readers can find ray diagrams in books on geometrical optics, elementary physics, and even camera advertisements. View cameras could have pentaprisms, too, but they would be prohibitively large and heavy.

The human eye and television cameras also reverse the image. Television cameras contain electronic circuits that restore normal perspective; Mother Nature uses neural circuitry in the brain for inversion in lieu of pentaprisms or electronics.

Mirror reversal is a different phenomenon from lens reversal. It can be demonstrated without a darkroom. Just look at yourself in a mirror and put your right hand on your right cheek. The image of your hand is in the right side of the mirror as you face it, but on the left cheek of your image in the mirror. Standing on your head does not put your hand image back on the image of your right cheek. Flat mirrors do not form optical crossovers as lense do: mirrors work on the principle that the angle of incidence of light rays equals the angle of reflection. Of course it is possible to photograph an image in a mirror, and some very pleasing pictures have been published.

The reason that mirror reversal causes left-to-right but not top-to-bottom reversal has been the subject of a number of articles with varying degrees of clarity. Martin Gardner's book The New Ambidextrous Universe [58] has an excellent description, somewhat longer than Richard Feynman ("No Ordinary Genius, The Illustrated Richard Feynman", Edited by Christopher Sykes; W. W. Norton & Company, 500 5th Avenue, NY NY 10110, 1994, pages 36-38.) who explains it as essentially front-to-back reversal. There is even semantic confusion about the meaning of reversal. Interested readers who enjoy a good puzzle will find considerable food for thought in these two intriguing essays.

Lens reversal is more relevant to many kinds of pictures including the ones that are often described as reversed, such as Daguerreotypes, ambrotypes, and tintypes. These comments apply only to first generation pictures. Copies and enlargements of original Daguerreotypes and tintypes made by the respective original processes (not a transparency process) will be re-reversed, or right side round. The third generation will again be reversed, and so on. Presumably surviving specimens become increasingly rare at this point, but one never knows unless the provenance is certain. See Chapter 7 for a discussion of the 'non-reversed' tintypes bound into Estabrooke's 1872 book.

Copies and enlargements of ambrotypes that were made as ambrotypes may or may not be reversed because, being transparent, they could be flipped over during copying. Of course resolution and picture quality suffer with each succeeding generation. This should be apparent in those rare examples when specimens are available for side‑by‑side comparison.

Reversing prisms were sometimes used in front of the lenses of Daguerreotype and other studio cameras; the prisms usually had their hypotenuse sides silvered. There is no way to deduce from the picture whether this was done unless there is a reference object such as lettering or architecture. It is therefore incorrect to make the sweeping statement that all Daguerreotypes were reversed, even though most of them were.

The effect of reversal is obvious in the case of subjects containing lettering or well‑known landmarks and architecture. To collectors the presence or absence of reversal may be an important clue to the identification of a process, a date, or a photographer. But what about portraits: does it really matter which way the subject faced?

There is a famous and intriguing example of this question. It is the matter of the rather prominent wart on Abraham Lincoln's right cheek. If his portrait is printed from a reversed negative the wart will have changed sides; that is, it will be on the other cheek, not just on the other side of the picture. There are many pictures of Lincoln still preserved, and they were made by three processes: tintypes, Daguerreotypes, and collodion glass negatives. The first two produce reversed pictures (unless they were copied or taken through a prism or mirror) but collodion plates can be printed either way.

In Taft's book Photography and the American Scene there is a frontal portrait of Lincoln that shows the wart on his right cheek and no wart on the left cheek. The caption states that "the print was made from the original negative... by Alexander Gardner." On page 243 another portrait shows the same thing, also from a negative. But in Beaumont Newhall's The Daguerreotype in America there is a Daguerreotype portrait of Lincoln (Plate 104) that shows the wart on his left cheek ‑ the expected effect of Daguerrian reversal. This may seem trivial, but to serious students of history such minutia may be clues to important questions of subject and process identification.