In setting up the sunshine recorder, the axis should be placed in such a position that the angle which it makes with the base is the same as the altitude of the place where the observations are made. This is readily done by loosening the bolts at either side. The drum is removed, the celluloid sheet unwound by means of the key which holds it in place, the sensitive strip put in position, and the sheet again wound up. Strips of a special sensitive paper upon which the hours are indicated are furnished by the makers of the instrument, but it has been found preferable to use solio strips in order to facilitate comparison with the standards. The drum is placed on the axis, and is screwed up until it just escapes the collar at the top of the spiral. The clock is wound and started, and the outer cylinder put on so that the proper hour mark coincides with the index on the front of the base.
As a sunshine recorder, the instrument gives a perfect record, in which the varying intensities are readily recognizable. Since the cylinder moves one-half inch in an hour, and the slit is .01 of an inch, the time of each exposure is 72 seconds. This gives a very deep color on the solio paper, which results in a serious error in making comparisons with the standard. On account of the hood, diffuse light is not recorded when it is too weak to cast a distinct shadow. It seems probable that this difficulty will be overcome by the use of a flat disk containing the proper slit, and in this event the instrument will become of especial value for measuring the diffuse light of layered formations. The celluloid sheet constitutes a source of error in sunlight on account of the reflection which it causes. This can be prevented by using the instrument only on sunny days, when the protection of the sheet can be dispensed with.
81. The selagraph. This instrument is at present under construction, and can only be described in a general way. In principle it is a simple photometer operating automatically. It consists of a light-tight box preferably of metal, which contains an eight-day lever clock. Attached to the arbor of the latter is a disk 7 inches in diameter bearing on its circumference a solio strip 1 cm. wide and 59 cm. long. The opening in the box for exposure is 6 mm. square and is controlled by a photographic shutter. The latter is constructed so that it may be set for 5, 10, or 20 seconds, since a single period of exposure can not serve for both sun and shade. The shutter is tripped once every two hours, by means of a special wheel revolving once a day. Each exposure is 6 mm. square, and is separated by a small space from the next one. Twelve exposures are made every 24 hours, and 84 during the week, though, naturally, the daytime exposures alone are recorded. Comparisons with the multiple standard are made exactly as in the case of the simple photometer. The selagraph is made by the C. H. Stoelting Co., Chicago, Illinois.
Standards
82. Use. The light value of each exposure is determined by reference to a standard. When the photometer carries a permanent standard, each exposure is brought to the tint of the latter, and its value is indicated by the time ratio between them. Thus, if the standard is the result of a 5–second exposure to full sunlight at meridian, and a reading which corresponds in color requires 100 seconds in the habitat concerned, the light of the latter is twenty times weaker or more diffuse. Usually, the standard is regarded as unity, and light values figured with reference to it, as .05. With the selagraph such a use of the standard is impossible, and often, also, with the photometer it is unnecessary or not desirable. The value of each exposure in such case is obtained by matching it with a multiple standard, after the entire strip has been exposed. The further steps are those already indicated. After the exact tint in the standard has been found, the length of the reading in seconds is divided by the time of the proper standard, and the result expressed as above.
83. Making a standard. Standards are obtained by exposing the photometer at meridian on a typically clear day, and in the field where there is the least dust and smoke. Exception to the latter may be made, of course, in obtaining standards for plant houses located in cities, though it is far better to have the same one for both field and control experiment. Usable standards can be obtained on any bright day at the base station. Indeed, valuable results are often secured by immediate successive sun and shade readings in adjacent habitats, where the sun reading series is the sole standard. Preferably, standards should be made at the solstices or equinoxes, and at a representative station. The June solstice is much to be preferred, as it represents the maximum light values of the year. Lincoln has been taken as the base station for the plains and mountains. It is desirable, however, that a national or international station be ultimately selected for this purpose, in order that light values taken in different parts of the world may be readily compared.
84. Kinds of standards. The base standard is the one taken at Lincoln (latitude 41° N.) at meridian June 20–22. This is properly the unit to which all exposures are referred, but it has been found convenient to employ the Minnehaha standard as the base for the Colorado mountains, in order to avoid reducing each time. Relative standards are frequently used for temporary purposes. Thus, in comparing the light intensities of a series of formations, one to five standards are exposed on the solio strip before beginning the series of readings. Proof standards are the exposed solio strips, which fade in the light, and can, in consequence, be kept only a few weeks without possibility of error. The fading can be prevented by “toning” the strip, but in this event the exposures must be fixed in like manner before they can be compared. This process is inconvenient and time-consuming. It is also open to considerable error, as the time of treatment, strength of solution, etc., must be exactly equivalent in all instances. Permanent standards are accurate water-color copies of the originals obtained by the photometer. These have the apparent disadvantage of requiring a double comparison or matching, but after a little practice it is possible to reproduce the solio tints so that the copy is practically indistinguishable from the original. The most satisfactory method is to make a long stroke of color on a pure white paper, since a broad wash is not quite homogeneous, and then to reject such parts of the stroke as do not match exactly. Permanent standards fade after a few month’s use, and must be replaced by parts of the original stroke. Single standards are made by one exposure, while multiple ones have a series of exposures filling a whole light strip. These are regularly obtained by making the exposures from 1–10 seconds respectively, and then increasing the length of each successive exposure by 2 seconds. Single exposures of 1–5 seconds as desired usually serve as the basis for permanent standards, but a multiple standard may also be copied in permanent form. Exposures for securing standards must be made only under the most favorable conditions, and the length in seconds must be exact. The use of the stop-watch is imperative, except where access may be had to an astronomical clock with a large second hand, which is even more satisfactory. The length of time necessary for the series desired is reckoned beforehand, and the exposures begun so that the meridian falls in the middle of the process.
Single standards are exceedingly convenient in photometer readings, but they are open to one objection. In the sunshine it is necessary to make instant decision upon the accuracy of the match, or the exposure becomes too deep. In the shade where the action is slower, this difficulty is not felt. For this reason it is usually desirable to check the results by a multiple standard, and in the case of selagraph records, where the various exposures show a wide range of tint, light values are obtainable only by direct comparison with the multiple standard. The exact matching of exposure and standard requires great accuracy, but with a little practice this may be done with slight chance of error by merely moving the exposure along the various tints of the standard until the proper shade is found. The requisite skill is soon acquired by running over a strip of exposures several times until the comparisons always yield the same results for each. The margin of error is practically negligible when the same person makes all the comparisons, and in the case of two or three working on the same reading the results diverge little or not at all. The efficient difference for light is much more of a variable than is the case with water-content. It has been determined so far only for a few species, all of which seem to indicate that appreciable modification in the form or structure of a leaf does not occur until the reduction in intensity reaches .1 of the meridian sunlight at the June solstice. The error of comparison is far less than this, and consequently may be ignored, even in the most painstaking inquiry.
Readings
85. Time. The intensity of the light incident upon a habitat varies periodically with the hour and the day, and changes in accord with the changing conditions of the sky. The light variations on cloudy days can only be determined by the photometer. While these can not be ignored, proper comparisons can be instituted only between the readings taken on normal days of sunshine. The sunlight varies with the altitude of the sun, i. e., the angle which its rays make with the surface at a given latitude. This angle reaches a daily maximum at meridian. The yearly maximum falls on June 22, and the angle decreases in both directions through the year to a minimum on December 22. At equal distances from either solstice, the angle is the same, e. g., on March 21 and September 23. At Lincoln (41° N. latitude) the extremes at meridian are 73° and 26°; at Minnehaha (39°) they are 75° and 28°. The extremes for any latitude may be found by subtracting its distance in degrees north of the two tropics from 90. Thus, the 50th parallel is 26.5° north of the tropic of Cancer, and the maximum altitude of the sun at a place upon it is 63.5°. It is 73.5° north of the tropic of Capricorn, and the minimum meridional altitude is 16.5°.