A filter gets its color because it passes light of its color more readily than other colors. Colors that are distant from it in the spectrum (red, orange, yellow, green, blue, indigo, violet) are attenuated by the filter. This allows black-and-white photographers to use a filter to render parts of a subject relatively lighter or darker, depending on whether their colors are spectrally near to or far from that of the filter.
To make up for the partial blockage of light by filters, manufacturers supply numbers called “filter factors,” by which the exposure time is to be multiplied. If these numbers are 2, 4, or 8, it’s easy to lengthen the shutter speed 1, 2, or 3 stops. However, it’s not useful to multiply a shutter speed by a number that isn’t a power of 2. Therefore, it’s best to remember how many stops or fractions of stops to open the lens (see figure 1).
After I focus my camera, I estimate the largest aperture that will provide enough depth of field. Making part of the picture fuzzy by opening the lens still farther (to accommodate a filter) is not a desirable option. Instead, I change the film-speed setting on the meter. (I also change the film speed to correct for close-ups or when minus development is planned. Making each of these corrections in the meter as the need arises is a convenient way to avoid forgetting to do one of them. Checking the film-speed setting becomes a part of every use of the meter, which isn’t a burden because a careful photographer would check it anyway.)
Metering and filters
Metering through a filter, rather than using a filter factor (expressed in stops), seems to me unwise. If the filter is a polarizer or one in the yellow-to- red range that darkens the sky, you may be led to expose more than necessary. Also, you have no assurance that the spectral sensitivity of the meter matches that of the film, although that may be too fine a point to worry about.
During the last couple of decades, many people have paid to have their Pentax meters modified so that their spectral sensitivity would match a popular film. On many occasions at field workshops, I metered nature subjects alongside students whose meters had been modified (mine hadn’t) and our meters always agreed. According to sales literature, the modified meters were supposed to render brightly colored cards differently, but we never encountered any such cards.
Yellow, deep yellow, orange, red, and dark-red filters block progressively more blue light and hence, darken blue skies. When I want to render a blue sky quite dark, I am often reluctant to use a red filter, because it also darkens green foliage, though not as much as the sky. Evergreen trees, which often tend toward blue-green and are common on mountains, may be made so dark by a red filter that shadow detail is lost. In such a case, a #15 deep-yellow filter is the strongest I use for darkening the sky. Yellow filters not only don’t darken green foliage, they may lighten it slightly, especially if it is a yellowish green. (I will refer to filters by the Wratten numbers given in figure 1.)
If the sun’s rays are perpendicular to the lens axis, the sky can be darkened as much as a #25 red filter would, with- out depressing values of green foliage, by combining a yellow filter with a polarizer. No doubt you have noticed that a blue sky looks increasingly white nearer the horizon. Apparently, much of this light was polarized when I made Dry Creek and Dunes, Great Sand Dunes, Colorado with a #8 yellow filter and a polarizer. Notice that the sky just above the dunes appears dark enough to separate nicely from the white clouds. Without the polarizer, the sky would no doubt have been rendered lighter there, even if the photo- graph had been made with a red filter.
If using two filters, add the number of stops of additional exposure each needs—in the above case, 2⁄3 + 11⁄3 = 2 for T-Max film.
When a polarizer is used, maximum darkening occurs in the part of a blue sky where a line from there to the lens makes a 90 ̊ angle with the sun’s rays. If the sun’s rays are perpendicular to the lens axis, this angle usually is close enough to 90 ̊ for all parts of the image to avoid substantial variations in tone in the sky. But if the lens focal length is less than normal, and especially if the composition is horizontal, this angle may be far enough from 90 ̊ in parts of the sky to make them noticeably lighter. In that case, a brief, tapered burn may be needed to make the sky print evenly. This can be done with two cards at right angles to one another. One card shelters the part of the image representing the ground while the other moves.
The first time I combined a polarizer with another filter was with a #25 red when photographing a white- washed Greek church against a blue sky. I wanted the sky to be rendered black, since the print might look weak if its darkest value were gray. The sky was a rather pale blue because of the presence of a lot of white light, presumably polarized. The polarizer got rid of that, leaving a strong blue that responded well to the red filter. Even so, the sky would only have been dark gray if I had placed the brightest whitewash on zone VIII and given normal development. Instead, I placed it on Zone VI, which made the blue sky fall low enough to insure that it would print black. N+2 development brought the highlights back up to Zone VIII.
Lightening tones with filters
For a filter to substantially lighten the rendition of a certain color, the filter must match that color rather closely. Then, when the extra exposure required by the filter is given, light of that color readily passes through the filter, giving more exposure to the negative than is given by parts of the subject that have colors at the other end of the spectrum. It’s common for efforts to lighten something by using a filter to be disappointing because of lack of saturation of the color in the subject or because the colors didn’t match well enough.
The strong B+W 091 deep-red filter used for Red Dahlia (figure 2B) had a great effect because the dahlia was a very similar deep red. Much of the red light from the flower passed through the filter, so the three stops extra expo- sure the filter required served to “overexpose” the flower petals. Green light from the leaves was partly blocked, rendering them somewhat darker. If the background had been a saturated blue, it would have been rendered practically black. During the 21-second exposure with the filter, the hot lights caused one of the leaves to move, a common problem when using these lights for long exposures of plants.
Calculations for Red Dahlia began at ISO 400, with the expectation of N+1 development. The close-up correction of 21⁄3 stops brought the film speed setting down to 80. The meter suggested 2 seconds without a filter (Figure 2A), which was increased to 3 seconds for reciprocity correction. The film speed was cut in half three times, from 80 to10, for the exposure with the deep red filter and the indicated exposure of 15 seconds was increased to 21 seconds to correct for reciprocity departure (see B&W Reci- procity Departure Revisited, PT July/August 2003). The resulting negatives have similar density ranges, but the filter lowered the density of an area of the leaf from .61 to .48, and raised the density of a petal from .94 to 1.17.
In figure 1, I have treated the Wratten #29 and the B+W 091 filters the same way, although I have an old published chart that calls for a correction of 41⁄3 stops for the #29. I suspect that 41⁄3 is a typographical error and that 31⁄3 was intended. B+W has specified a 3-stop correction for their 091, and experience has supported this. If you have a #29, I suggest trying a 3 or 31⁄3 stop correction for a subject that includes a dark gray, textured tree trunk on an overcast day, to see if this correction adjustment is suitable for the #29. Make one negative without a filter, placing the tree trunk on Zone III, and another negative with the filter and 3 stops more exposure. The exposure increase should be done with the aperture, to avoid possible shutter inconsistency.
Filters in practice
Figure 3A shows a scene photographed on T-Max 400 with no filter. For the other three versions, I gave 2 stops more exposure with a #11 yellow- green filter, 1 stop more with a #15 deep-yellow filter, and 3 stops more with a B+W 091 deep-red filter. To show the effects of the filters, I tried to print so that the neutral card was rendered as the same shade of gray in each print. There is some inaccuracy in the magnitudes of the changes because the matching of the card images was done by eye, rather than with a reflection densitometer, but all of the changes are in the expected direction. The same variable-contrast printing filter was used for all four prints, although the density range of negative A was about .10 greater than the others.
With the #11 yellow-green filter (B), the sky is slightly darker than with no filter (A). Lightening of green vegetation is most easily seen in the grass, the peony leaves next above, and in the shrubs just above the card. The #15 deep-yellow filter (C) darkened the sky a little more, and made the vegetation a little lighter than with no filter (A), but not quite as light as with the #11 yellow-green filter (B).
The B+W 091 deep-red filter used for version D produced the darkest sky. If the portion of sky that was included had been farther above the horizon and hence, more blue, all three filters would have had more effect on it and the B+W 091 would have made the sky very dark. It darkened the green leaves and grass somewhat, an effect we saw in figure 2.
Normally, I use a filter at the yellow end of the spectrum to reduce haze if parts of a scene are far away. I don’t carry a blue filter, which could be used to emphasize haze. Simply using no filter has made haze sufficiently visible on the rare occasions when I wanted it to be more prominent.