The English standard candle is a spermaceti candle made. to burn 120 grains per minute with a flame height of 45 milli

meters.

The Hefner-Alteneck amyl acetate lamp is a simple openflame lamp burning amyl acetate. The metal tube surrounding the wick requires to be made exactly according to standard, and the height of the flame is adjusted to be exactly 40 millimeters. The light is somewhat red in color, which is a disadvantage, but on account of the exactness with which it can be reproduced it has been adopted as standard in Germany.

Other standards which are in use will be found described in books on photometry.

VELOCITY OF LIGHT.

801. Early Experiment.-It has been seen that light seems to pass from the source to the eye in straight lines. This at once suggests inquiry whether or not the eye instantly experiences the sensation of light when a candle is uncovered.

The earliest attempt to solve this question was made by the Florentine Academy after a method proposed by Galileo. A light on an eminence was uncovered and flashed to a station on a distant hill where a second observer also having a covered light was watching. As soon as the flash was seen by the second observer he uncovered his light, sending an answering flash back to the first station. The first observer was to note the exact time between the uncovering of his light and the sight of the return flash. The experiment showed that if any time at all was required for light to travel from one station to the other it was too short to be detected by that method.

802. Roemer's Discovery. The first evidence that light required an appreciable time to pass from one point to another was obtained by the Danish astronomer Roemer, in 1676, by the following method:

The first satellite of Jupiter passes into the planet's shadow and disappears or is eclipsed every time it revolves around the planet. Some years before Roemer's discovery Cassini had carefully determined the periodic time of the satellite and had prepared tables showing when the eclipses might be expected

to take place for several years ahead. On comparing these tables with the recorded times of observed eclipses Roemer found that they were observed sooner than predicted when the earth was on the side of its orbit nearest to Jupiter, and later than predicted when it was on the opposite side. He concluded that the discrepancy was due to the velocity of light; for evidently if it takes ten minutes for light to cross the earth's orbit from B to C, then an eclipse would be seen 10 minutes later if the earth

FIG. 476.

Sun

were at C than if it were at B. The observations indicated that light requires 16 minutes to cross the whole of the earth's orbit, or approximately 8 minutes to go from the sun to the earth or, more exactly, 498 seconds to traverse the 92,900,000 miles between sun and earth, making the velocity of light in interplanetary space 186,600 miles or 300,200 kilometers per second.

803. Bradley's Discovery.-No further evidence of the velocity of light was obtained until 1727 when the English astronomer Bradley discovered that the stars in any given part of the heavens were apparently displaced from their mean positions by an exceedingly small amount which depended on the position of the earth in its orbit. The explanation of this phenomenon, which is known as aberration, was finally suggested to him by the observation that the position of a flag on a small boat depended on the velocity and direction of motion of the boat as well as on the wind. He said to himself that the apparent direction in which light comes to the earth from a star must be affected by the velocity of the earth, just as the apparent direction of a breeze to a man in a boat depends on the motion of the boat.

For suppose V (Fig. 477) is the velocity of a breeze at right angles to the motion of a boat which is moving forward with velocity v, then the motion alone would cause a breeze equal and opposite to v, and the resultant breeze as felt in the boat appears to come in the direction R. So if V is the velocity of light from a star and v is the velocity of the earth, a telescope must be pointed forward in the direction R in order that the

V

R

U

FIG. 477.

light may come down the axis of the instrument. The apparent direction of the star in that case differs from its true direction by the angle x, such that

Or, looking at the matter in another way, suppose that light coming from a star in the direction ao (Fig. 477) enters the telescope at o, and advancing in the direction ob reaches b at the same instant that the eye-piece reaches there as it moves sidewise from e to b. The light will be received by the eye and the telescope will seem to be pointing at the star. To accomplish this the telescope evidently must incline forward so that ob: eb V:v where V is the velocity of light and v is the component, perpendicular to the star's direction, of the velocity with which the telescope is carried along by the earth.

When the earth is moving directly toward or away from a star there is no displacement or aberration, while stars in directions at right angles to that in which the earth is moving have maximum displacement. The apparent position of a star

therefore changes slightly as the earth moves from one part of its orbit to another, so that by careful determinations of its apparent position made during an entire year the maximum displacement or aberration constant may be determined.

Recent observations give as the aberration constant 20.492". Now, the mean velocity of the earth in its orbit is 18.51 miles per second, and we may calculate the velocity of light V from the relation

which gives 186,400 miles or 299,930 kilometers per second.

804. Fizeau's Method.-On account of the enormous velocity of light it was not until 1849 that a method of measuring it was devised which did not involve astronomical measurements. In that year the determination was made by Fizeau by the following method. A telescope and collimator were set up 8.633 kilometers (more than 5 miles) apart. A beam of sunlight L

FIG. 478.-Fizeau's apparatus for measuring the velocity of light.

(Fig. 478) sent through an opening in the side of the telescope was reflected by a small oblique plate of glass G so that it passed directly out through the lens of the telescope to the distant collimator which was provided with a mirror M at its back. The collimator and mirror were so adjusted that the beam of light was reflected directly back into the telescope again, and passing through the plate of glass G was received by the eye at E. Thus light came to the eye at E after traveling to the distant mirror M and back again. At S in the telescope is a small opening which is alternately opened and closed by the teeth of a cogged wheel W which revolves immediately in front of it. If the wheel is

slowly rotated, light from L passing through a gap between two teeth travels to the distant mirror and back again through the same opening to the eye at E. If the notches and teeth in the wheel are of equal width and if the speed of rotation is such that a tooth moves forward just its own width in the time that light requires to go to the distant station and back, light which must have passed out through an opening will on returning find the opening closed by a tooth and will therefore be cut off from the observer at E. If the speed is then doubled, light passing out through one opening will return through the next one; at a still higher speed it will be eclipsed again, etc. It is therefore only necessary to observe the speeds at which the light is completely eclipsed to be able to determine the velocity of light, when the distance between the two stations and the number of teeth in the wheel are known. In Fizeau's apparatus there were 720 teeth in the wheel and the first eclipse was noticed when the wheel made 12.6 revolutions per second. Therefore the time required for light to travel twice the distance between the two stations was only

The stations were 8.633 kilometers apart, making the velocity of light 313,000 kilometers per second. The same method carried out by Cornu in 1874 with improved apparatus gave v = 304,000 kilometers per second.

805. Foucault's Method.-Another method of measuring the velocity of light was devised and carried out by the French physicist Foucault in 1850. The essential features of the apparatus are shown in the diagram, figure 479. A beam of sunlight concentrated on the narrow slit S passes through it and through the inclined plate of glass G and the lens L to a small mirror m, from which it is reflected to a concave mirror M whose center of curvature is exactly at the center of the mirror m. The light is reflected perpendicularly back from M to m and thence back to the glass plate G, which reflects it aside into the eye-piece E. A bright image of the slit S is formed at a by means of the lens L, and this is seen by the observer at E. If the mirror m is now slowly rotated in the direction of the arrow, the image of the slit will be formed at a only when m is in such a position,