77. SOLAR SPECTROSCOPE.

Apparatus. The optical circle, a 60° prism of dense flint glass, and a mirror capable of turning horizontally or vertically, by which a ray of sunlight may be reflected in any desired direction. This is accomplished more perfectly by a heliostat, in which the apparent motion of the sun is corrected by moving the mirror by clockwork. The instrument should be placed near a window into which the sun is shining, or if the day is cloudy, the experiment may first be performed with a Bunsen burner and a wire on which is a little borax, and afterwards concluded by the aid of sunlight.

Experiment. Adjust the optical circle as described in Experiment 72, so that both telescopes shall be focussed for parallel rays, and when placed opposite each other the cross-hairs and slit shall be distinctly visible. Bring them to coincide, and read the vernier. Turn the observing telescope about 45°, and place the prism on the centre plate, so that its back shall be equally inclined to the axes of the telescope and collimator, as at D, Fig. 59. Place the mirror in the sunlight, and turn the collimator towards it, and distant only a few inches. Nearly close the slit, and reflect the light through it by turning the mirror. On moving the telescope to one side or the other, if necessary, a brilliant spectrum will be visible, any part of which may be brought to coincide with the cross-hairs, and its position determined by the vernier.

To obtain the best results, the position of the mirror must be accurately adjusted. This may be done in two ways. Most simply by opening the slit wide, when the position of the beam of sunlight may be seen in its passage through the object-glass of the collimator, forming a bright spot on it. The mirror should then be turned until this spot falls in the centre. Holding a sheet of thin paper against the object-glass renders the spot more visible. The slit must be nearly closed before looking through the telescope, or the eye may be injured by the intense light. A more accurate method of adjustment is to remove the eye-piece and look through the tube, when an image of objects reflected in the mirror will be faintly visible. Turn the mirror until the image of the sun falls in the centre of the object-glass and the light will then pass through the axes of both telescopes. At the same time the prism should be placed so that it shall cover as

much of the object-glass as possible. If no light is seen, even if the slit is opened wide, probably the telescopes are not set at the right angle. The brilliancy of the image depends on the width of the slit, and when the latter is very narrow, the image of the sun will widen out by diffraction. Having set the mirror correctly, it will remain right only for a few minutes, owing to the apparent motion of the sun, and hence must be readjusted every little while. This may commonly be done with sufficient accuracy without removing the eye-piece. Much trouble may be saved by noting the point on the opposite wall where the reflected beam falls, and resetting the mirror by this. Or, a small mirror may be attached, and the direction of its reflected beam noticed. If the shadow of a window-sash falls on the mirror, move the latter across it, so that the further motion of the sun may separate them instead of again bringing them together.

Next bring the prism to the minimum of deviation, that is, so that its back shall be equally inclined to both telescopes. Turn the prism while looking through the telescope, and the spectrum will be seen to move a certain distance toward the red end, and then return. As a considerable motion of the prism corresponds to but a slight motion of spectrum, this point may be found with sufficient accuracy by the hand alone. Now focussing the telescope with care, the spectrum will be seen to be traversed by a multitude of fine vertical lines known as Fraunhofer's lines. Bring one of these to coincide with the crosshairs after setting the prism at the minimum of deviation, and read the vernier. It will be found that the minimum for one line is not the minimum for another. Measure in the same way the position of several of the more prominent lines; which may then be identified by the accompanying table. The first column gives the names, the second their wavelengths, and the third their position on the map of Kirchhoff, which is still much

Name. W.L.

W.L. K.

7605 405 7185 500 6867 594 6562 694 6276 810 5892 1005

A

α

B

C

α

D

E

5269 1528

bi

5183 1634

4861 2080

G

4307 2855

h

4101 3364 H1 3968 3779

used as a standard. The line A is about the extreme limit of the red end of the spectrum, and can only be seen in strong sunlight.

B is therefore often mistaken for it. C is a sharply marked, but fine line in the red, caused by hydrogen. a is due to aqueous vapor in the air, and is most conspicuous about sunset. It then bears a marked resemblance to B. Dis a double line in the yellow, due to soda. The fine lines between its two components were often used as tests, until it was shown by Prof. Cooke that most of them were due to aqueous vapor. E is a close double line in the green in the midst of a group of double lines, some of them very close. b consists of four very strongly marked lines, three of them due to magnesium, of which the least refrangible is b. They contain several fine lines, which form good tests of the power of a spectroscope. F is a strong line in the blue, like C, due to hydrogen. G lies in the midst of a group, among a multitude of lines. h is fine and due to hydrogen, and H consists of two very broad lines, almost at the limit of the visible spectrum.

To determine the indices of refraction for these lines, subtract from the reading of the vernier in each case the reading when the telescopes were opposite, and the difference D gives the deviation. If i is the angle of incidence, and the angle of refraction for the first surface, since the prism is at the minimum of deviation, the angle of incidence at the second surface will equal r, and the angle of refraction i, as both faces are equally inclined to the light. Again, it may be proved by Geometry, that calling A the angle of the prism, A=2r. The ray is deviated at each surface i hence the total deviation D 2(i r) 2i (A+D). If the index of refraction equals n, sin i

sin (A+D)

or n= sin 4

A, or i

= n sin r

Compute in this way the index of re

fraction n for each of the lines, and see if they satisfy the theoret

В

C
24

ical formula of Cauchy, n = A + B + = A + Bx + Cœ2,

calling x equal to the reciprocal of the square of the wave-length, and A, B and C, constants depending on the particular material of which the prism is composed. By the method of least squares, p. 4, the most probable values of A, B and C, may be found, and compared with observation by a residual curve. To insure accuracy, it is safer to remeasure the indices again, using the other side of the graduated circle and employing the mean. By using a

hollow prism bounded by two plates of glass, the indices of liquids may be measured, and with a prism of quartz the relation of the ordinary and extraordinary indices to the wave-length, established.

To obtain really valuable results in this experiment, great care is necessary, and an instrument of the finest construction. The more powerful spectroscopes contain a number of prisms, thus giving a much longer spectrum. In some the light passes twice through each prism, the collimator being placed immediately over the observing telescope, or better, united with it. With such an instrument a vast number of lines may be seen and identified by comparison with the maps of Angström or Kirchhoff. To measure the exact place of those near together, it is better to determine accurately the position of two or three, measure the rest by a spider-line micrometer, and then reduce to wave-lengths by interpolation.

The distance between the two components of a double line may also be readily determined by the same instrument. It consists of an eye-piece, in which are two vertical spider lines, one fixed, the other movable by a micrometer screw the number of whose turns is commonly measured by notches in the upper part of the field of view, and the fraction of a turn by a circle divided into one hundred parts, attached to the screw-head. Both wires may be moved by a second screw, and illuminated by a light placed opposite a piece of glass inserted in one side. It is used when the field is dark, to render the lines visible. The distance between two lines may be measured as follows. Call the screw with divided head, A, and the other, B. Bring the two cross-hairs to coincide, and read the micrometer-screw, A, repeat several times, and take the mean. Then turn B until the fixed hair coincides with one line, and turn B until the movable hair coincides with the other. The reading of A minus that previously taken, gives their distance apart. After setting both hairs, their position is sometimes reversed, and the distance through which A has been turned equals double the distance between the lines. It is a good exercise to measure all the lines visible in a small portion of the spectrum, and then compare with one of the charts mentioned above.

If the sun is not shining, the Bunsen burner may be employed instead, using the platinum wire with a borax bead at its end. This will give a bright, double line, coinciding exactly with the dark line, D, in the solar spectrum. Its position, and the interval between its components, should be accurately determined. Spectra of great beauty may also be obtained with an induction coil by allowing the spark to pass between terminals of different metals placed in front of the slit. Still finer effects are obtained with the electric light.

78. LAW OF Lenses.

Apparatus. In Fig. 60, A is a fishtail burner, attached to the end of a bar eleven feet long, and divided into tenths of an inch. B is a lens of two feet focus, by which an image of A may be projected on the screen C. Both B and C are movable, and carry pointers to show their distance from A.

Experiment. Place Cat the end of the bar, and B just 100 inches from A. An image of the flame will be formed on C,

B

Fig. 60.

A

which is then moved backwards or forwards until the exact focus is found. When the screen is too near, it will be noticed that owing to chromatic aberration, the edges of the image are red, while if too distant they are blue. The intermediate position may thus be found with great accuracy. Read and record the distance AC, and repeat, making AB successively 95, 90, 85, etc., inches. C will approach A up to a certain point, until AB = BC equals twice the focal length of the lens. Determine this point more exactly by taking a number of readings, moving B an inch at a time. Then continue to diminish AB two inches at a time, until the image falls off the bar.

Write the results in a table in which the first column contains AB, the second AC, and the third their difference, or BC. Now compute the true value of BC in each case, and insert in the fourth column. Calling u and the conjugate foci, or AB and