2 5 6 D O C . 6 8 E X P E R I M E N T O N L I G H T E M I S S I O N

The light source is a narrow beam K of pos-

itive rays that is mapped by the lens into

the plane of the slit S; the latter screens out a

short piece of the image. Lens makes the

light emitted from the image of every ele-

mentary particle parallel; or more precisely,

the surfaces of equal phases are transformed into planes.

According to the undulation theory, light from an elementary process will—due

to Doppler’s principle—have a shorter wave length at the lower rim of the lens than

at the upper rim. The planes of equal phase behind will not be exactly parallel

but rather be slightly inclined to each other like a fan. In a telescope placed behind

and adjusted to infinity, one will see a picture of the slit, exactly in the same

location as if light would be emitted from particles at rest. The mapped points, how-

ever, corresponding to individual surfaces of equal phase from an elementary pro-

cess, will not coincide but they all will fall within the optical image of the slit.

However, the situation changes if one places a layer of a dispersive substance,

e.g., carbon disulfide, between and the telescope. Due to dispersion and due to

the dependence of frequency on position, surfaces of equal phase will propagate

slower at the bottom than at the top; and thus one can expect a deflection of the light

emitted by the moving particles of the positive ray. This deflection must, if it exists,

be easily observable. If the distances and are equal, and the distance

is denoted by , and l is the thickness of the layer of the dispersive medium, then

the angle of deflection is given by the formula

,

(3)

where is the velocity ratio of the particles in the positive ray and that of light, n

the index of refraction of the dispersive substance, the frequency, and dn and

the associated increases of these two quantities. For a -layer of 50 cm in length

with cm, one can expect an angular deflection of more than .

However, if the individual process has a unique frequency, then the frequency of

the individual elementary process is independent of direction; the deflection de-

manded by the undulation theory will not exist. I do not want to discuss the possi-

bility here other than to note that it could very well be brought into agreement with

the existence of the Doppler effect that J. Stark has stated.

Herr Geiger and I decided to look into this question experimentally.

K L1 L2

S [1]

L1

L2

L2

L2

[2]

L2

KL1 L1S SL2

Δ

α

α

1

Δ

---

v

c

-- -

dn

dν⎞

ν

----- -

⎝ ⎠

⎛

----------- -

=

[3]

v

c

-- -

ν dν

CS2

Δ 1 = 2°

[4]

[5]