Site Section Links
The Third Story
The Nature of Time
Nature of Time video
The Nature of Space
The Neutrino Aether
Nature of Force Fields
Origin of Modern
Niagara Falls Issues
Climate Change Model
Climate Change Questions
Modern Mythology Material
1994 Velikovsky Symposium
Pensee Journals TOC
Selected Velikovskian Article
State of Religious Diversity
PDF Download Files
Open letter to science editors
A proposal for an experiment
The Velocity of Light in Relation to Moving Bodies
Copyright 1973 by Immanuel Velikovsky
This paper remains in the
form in which it was written over a decade ago. Ed.
The Michelson-Morley experiment performed in 1886 demonstrated that a beam
of light that issues from a terrestrial source and travels in the direction
of terrestrial motion, East-West-East or West-East-West, needs the same time
to traverse a laboratory distance as a beam that travels at right angles to
that motion (North-South-North or South-North-South). The undulatory theory
of light transmitted by waves in the ether anticipated detection of a
difference in the velocities of the two beams due to the orbital velocity of
the earth through absolute ether-filled space. Half a year later, when the
earth was on the opposite side of its orbit, the same experiment again
disclosed no difference in the velocities of the beams; any possible
compensatory motion on the part of the solar system or the entire galaxy
thus was excluded.
The explanation first offered was the supposition that any material object
(also a measuring rod) traveling through the ether is shortened by a very
small amount; the East-West distance in the laboratory apparatus
(interferometer), being shorter, is traversed by a beam of light traveling a
little slower in the same time that the longer North-South distance is
crossed by a swifter beam (1). Einstein, however, generalized this idea by
assuming that the velocity of light in vacuum is constant in relation to all
bodies, whether in motion or at rest. The ether was discarded in the
Special Theory of Relativity, and Einstein embraced the quanta theory of
light. Both space and time lost the attribute of constancy, and light (its
velocity) acquired it.
Ritz objected to the thesis of constancy, claiming that the velocity of an
illuminating body adds itself to the velocity of propagation of light c, the
resultant being c + v. De Sitter disposed of Ritz' argument by offering to
consider a case of a double star whose two members revolve around the common
center of gravity on orbits situated in the plane of the observer on earth.
If the velocity of light from each of the stars is c - v while receding from
the observer and c + v while approaching, the beams of light, after
traveling at varying speeds for many years to reach the observer, would
present the pair in a most chaotic motion, definitely not in harmony with
the Keplerian laws of motion for celestial bodies (2). The idea developed
by de Sitter originated with D. Comstock (3).
This argument for a constant velocity of light was regarded as
conclusive and the question was not raised again for over three
decades. However, in 1932, Sir Oliver Lodge wrote (4): "'that the
velocity of light in space is really constant, in itself, may be fully
granted ... but that its apparent speed when determined by an observer
moving to meet it, should be unaffected likewise is extraordinary. Yet
this is the postulate of the relativity doctrine propounded in
1905. The doctrine certainly explains the Michelson experiment, and my
experiment; nor has any experiment ever negated it so far; and
yet--well, it hardly seems consistent with common sense. It seems to me
that posterity will formulate the doctrine a little differently. I had
hopes that if Michelson had lived (5) --he might have tackled it by some
hitherto unimagined experiment, and separated the velocity from the
Doppler change of frequency with which it is closely associated.
Something might possibly be done with a pair of rapidly revolving
cameras. Every terrestrial determination of c involves a to and fro
journey of some kind, and therefore is inconclusive. It is now orthodox
to assert that no one will ever measure the velocity of light
differently by moving towards the source. They would certainly not
observe any change in velocity if the source moved toward them; and it
is inconsistent with relativity that there should be any difference
between the two cases. If there were, we should have to take the ether
as our standard of rest (as I want to do), and begin to apprehend
absolute motion through the emptiness of space, as well as the familiar
relative motion with reference to other portions of matter. Let us try
to keep the question open."
Actually the Michelson-Morley experiment is not a direct test; it
depends on assumptions and analogies. The beam of light travels both
directions; whatever might be won in time on approach to a mirror may be
lost in retreat from it; this is also expressed thus: "Because transits
to and fro are involved, the Michelson-Morley experiment does not prove
anything about the velocity of light in one direction" (6).
Therefore, this experiment by itself does not prove the proposition
that the velocity of light is the same in all directions and for all
observers. Besides, in the estimate of the expected results, analogy
with a body propelled upstream and then downstream as compared with one
propelled forth and back across the stream is assumed, and the different
times needed for these two types of travel are computed accordingly.
But motion through a resisting fluid is not necessarily analogous to the
case of light transmitted by waves in the ether.
A direct experiment to measure the velocity of light relative to moving
bodies was devised by me as long ago as 1944, and in December of 1945 it
was submitted to the National Academy of Sciences (7) (also copyrighted
as lecture on February 23, 1945); but at that time and until recently,
such an experiment was considered unnecessary. This I also learned from
a letter written to me by Professor Paul Epstein of Cal. Inst. Tech. on
March 7, 1945. However, I thought that an experiment, if one could be
devised, should also be performed to verify the Michelson-Morley
results. On theoretical grounds, it appeared to me that de Sitter's
argument is not necessarily valid: an illuminating body might not impart
to an erupting photon its own velocity of motion, so instantaneous, most
probably, is the eruption; perhaps the light travels with constant
velocity c in relation to the illuminating body and also with a velocity
c + v in relation to an illuminated body or an observer traveling toward
or away from the source of light. De Sitter's argument may have been
valid against Ritz, but Ritz' conjecture did not encompass the
proposition made here and, in fact, contradicted it.
In recent years (1959-1960), Professor Herbert Dingle raised a question
that lay dormant for decades when he wrote (8):
"In view of recent difficulties that have arisen in connection with the
[Special Relativity] theory, and particularly in view of the fact that
an alternative theory of Ritz's, which was thought to have been
disproved, has now been shown to be a distinct possibility (9) ... it is
claimed that at present there is no experimental evidence for or against
the kinematical requirements of the [Relativity] theory. . ."
"For all practical purposes, then, the choice lies between Einstein and
Ritz, and at present there is not a scrap of experimental evidence even
to make one appear more probable than the other ... We have not
sufficient evidence on observational grounds to give one even a greater
or less probability than the other: we await the verdict of
experiment." Professor Dingle challenged experimental physicists to
find a method for "the comparison of the velocity of light, with respect
to the laboratory, from relatively moving sources," and expressed hope
that "the crucial experiment will be undertaken without delay if it is
An experiment involving measurement of time on moving clocks is also
mentioned in general terms but conceded to be less simple. Professor
Dingle stressed that "all experiments in which an atom is regarded as a
clock are powerless to answer our question" (10).
It follows that Professor Dingle did not consider that in addition to
the hypotheses of Einstein and Ritz a third is possible, namely, that an
observer traveling toward a source may meet its light at velocity c + v,
while the light travels at constant speed c in relation to neither
observer nor source but to the point in space at which it was emitted.
P. Moon and D. Spenser, before Dingle, tried to show that observations
of double stars do not justify de Sitter's argument against Ritz (11).
Dingle's reference to "recent difficulties" that have encountered the
Special Relativity theory has its basis in W. Heisenberg (1955):
"Relativity Theory assumes that in principle no effect can be propagated
faster than the velocity of light. Now this trend in Relativity Theory
leads to difficulties in connection with the uncertainty relations of
quantum theory ... We have found that in quantum theory a clear
determination of position-- in other words, a sharp delimitation of
space--presupposes an infinite uncertainty of velocity and thus also of
momentum and energy" (12).
By 1956, Russian physicists, too, recognized the need for proof of
constant velocity of light; they were not guided by doubts regarding the
truth of the postulate but by the realization that the theory of
relativity lacks experimental verification of this basic postulate. A.
M. Bonch-Bruevich presented in that year a paper entitled: "An
experimental verification that the velocity of light is independent of
the velocity of the source with respect to an observer" (13), in which
the author claimed to have "for the first time performed a direct
The experiment was based on an idea offered by S. I. Vavilov and before
him by G. S. Landsberg. Of the two equatorial points at the limbs of
the rotating sun, one retreats from the observer, the other approaches,
and the velocity difference is almost 4 kilometers per second. The
experiment was performed in Pulkovo on a stretch of two kilometers. The
procedure, though complicated in apparatus, was not sufficiently
sensitive to measure a time differential of less than one ten billionth
of a second, a requirement in the method. Therefore, a statistical
treatment was applied to over 1700 measurements, yielding a result that
"satisfies the hypothesis of relativistic theory."
The experiment and its result were disqualified by another Russian
physicist, A. G. Baranov, who wrote (1961): "On the one hand, the author
begins with some a priori assumptions concerning the speed of light
after reflection from a mirror; on the other hand, owing to the low
accuracy inherent in the method, the observed results have a large
scatter, several times larger than the effects expected from classical
theory" (14). To this, I may add that in the experiment only the
influence of the motion of the illuminating object on the velocity of
light was queried, whereas the problem is also whether the motion of an
illuminated object in relation to a photon can result in c ± v velocity.
Baranov, in turn, proposed "A Method to verify experimentally that the
speed of light is independent of the velocity of the source" (the title
of the paper), claiming it to be "a scheme by which to verify directly
in the laboratory the postulate of the constancy of the speed of light,"
the two claims being not identical. Also, the method (no experiment was
performed) is not well devised. A beam of light is split in two by a
semi-silvered mirror; the split portions of the beam travel, with the
help of reflecting mirrors, the same route but in opposite directions; a
movable system of two mutually perpendicular mirrors is inserted (closer
to one end of the route) so that its motion lengthens the distance more
for that beam which has the movable system close to the end of its
route. The speed of the mirror is of the order of one meter per
second. The interferometer must show no fringe shift if the speed of
light depends on the velocity of the source (the value c + 2v after
reflection from the movable mirrors would be compensated by the
difference in the lengths of the paths).
However, in this project, the source of light and the observer are
stationary, and only the mirror is moving; also, the problem remains
whether the (hypothetical) c ± v velocity of light traveling toward a
mirror remains the same after being reflected (related to the mirror),
or becomes c ± v. Light loses energy when it exerts pressure on
reflecting surfaces and when it encounters resistance in various
transparent media, like air, water, or glass. In Baranov's scheme one
half of the beam is reflected from four mirrors and passes through two
semi-silvered ones, whereas the other half of the beam is reflected from
six mirrors and passes through none.
In my experiment, as devised in 1944, a beam of morning sunlight is
permitted to travel through a slit provided with a shutter toward a
rotating, multifaced mirror at a distance; through a parallel slit
operated by the same shutter travels light from a source on earth. From
the rotating mirror the beams are reflected to a sensitive plate not far
away. The apparatus travels toward the sun with velocity due to the
earth's rotation amounting to somewhat less than half a kilometer per
second, depending on latitude (40,000 kilometers in ca. 86,000 seconds
at equator). If the solar light approaches the apparatus with velocity c
+ vt, then it will mark its appearance on a sensitive plate
in advance of the mark made by the light from the terrestrial source,
which travels with the apparatus and is stationary with respect to it.
Before sunset the experiment is repeated as the apparatus retreats
before the solar beam. This time the shutter is placed west of the
rotating mirror; precise equality of distances between shutter and
mirror in the morning and afternoon is not of prime importance. The
results of the morning and afternoon experiments are compared.
In this experiment, the light does not travel in two opposite directions
before its velocity is measured by comparison with a parallel beam from
a terrestrial source. The solar light coming from the central portion
of the disc is utilized.
The following quantities may be taken as guides: The distance between
shutter and mirror--one kilometer; the shutter opens for 10-5
of a second; the mirror rotates 5,000 times per second and has the form
of a decagonal prism; each facet replaces the previous in 2 x 10-5
sec.; the distance from the decagon mirror to the detection device is
ca. 2 meters. If there is a difference in velocity between the solar and
terrestrial light signals, it would amount to ca. 10-6 km per
sec.; the difference in time would be 3.3 x 10-12 second; at
a radius of 2m., ten microns subtend nearly 1" of arc; the angular
rotation of a reflected beam of light being twice the angular rotation
of the mirror, at 5,000 rotations per second, the light beam that
precedes by 3.3 x 10-12 sec. would mark itself on a
sensitive plate half a micron in advance of the other beam of light.
Half a micron is at the limit of optical observation, but we are left
with a reserve in the possibilities for increasing mirror speed or
passage and projection distances. An optical dispersion device might be
inserted before the sensitive plate, unless another method
(photoelectric) proves to be superior. Half a micron being the wave
length of green light, an interferometer device could be utilized for
checking the comparative velocities of the two beams.
In this form I communicated my experiment plan to Dr. H. K. Ziegler,
Chief Scientist, U. S. Army Signal Research and Development
Laboratory, Fort Monmouth, N.J., in April-May 1961.
Originally I considered passing the beams of light through a prism
between the rotating mirror and the sensitive plate and comparing the
positions of the spectral lines in the solar light with those from
terrestrial source (neon) in the morning and before evening, but this
arrangement requires a very fast shutter to obtain unsmudged spectral
lines. To distinguish in this arrangement between a shift due to the
changed velocity of light and a shift due to a Doppler effect, the
mirror should be rotated both ways: the Doppler shift would be toward
the violet in the morning and toward the red in the evening, but a
displacement due to the change in velocity of light would follow the
direction of the mirror's rotation.
In both versions of the proposed experiment, we have a direct method to
compare the velocity of a beam of light in relation to an observer
moving toward the source of light or away from it with the velocity of
light from a source that is stationary in relation to the observer.
Travel of beams of light "to and fro" is avoided in this arrangement.
Should the experiment result in velocity c for the light arriving from
the terrestrial source and c ± v for the light from the sun, then
Einstein's postulate of constant light velocity would be disproven;
Ritz, however, would not yet be proven because, though we can regard the
sun as moving and the observer as stationary, yet we cannot reverse the
illuminating faculty and consider the observer illuminating and the sun
illuminated. Ritz claimed that the motion of a source is transferred to
the photons and results in c + v where v is the velocity of the source
with respect to the observer. Yet we may, consider that the photons do
not inherit the inertia of motion of the source from which they erupt,
but an observer traveling toward h source of light would meet the
photons at a speed surpassing c. In order to substantiate or eliminate
Ritz, a control experiment should be instituted: beams from both
equatorial edges of the solar disk are compared as to their velocity; an
arrangement similar to that of the original experiment is employed.
In case, however, the original experiment should result in equal
velocities for the solar and terrestrial beams of light in relation to
the observer, Ritz would be disproven, but Einstein not yet proven.
This is seen from the following: If in the experiment performed in the
morning the velocities of the solar and terrestrial beams are c in
relation to the points in space where they were emitted, but c + vt
in respect to the mirror, then the two beams will arrive at the same
speed and there will be no immediate way to determine whether velocity c
± v in relation to a moving body does or does not exist.
To solve this part of the problem, an experiment must be devised in
which light (laser) signals from two terrestrial sources equidistant
from a detector and coming from opposite directions (east and west) are
competing. Also radio signals substituting for fight could be sent to
the sun and registered upon their return, the passage time being
measured and compared in the morning and afternoon. In view of
turbulent conditions above the photosphere and in the corona, such an
experiment needs to be repeated several times in order to obtain mean
values. The moon, too, after rising and before setting may serve as the
reflecting body; it has no magnetosphere worth considering and it may
prove to be a preferable target, though the sun offers an almost 400
times longer route; the time difference for the passage time of morning
and afternoon signals to the sun and back, if not zero, may be close to
one three-hundredth second.
Experiments concerned with the problem of absolute space need to be
repeated at certain time intervals: The orbital and the rotational
speeds of the earth and the motion of the solar system in its entirety
will affect any experiment devised to solve the problem of constancy of
velocity of light in relation to a point in absolute space where the
photon left its source.
The main experiment described in this paper will eliminate either Ritz'
or Einstein's solution of the problem, leaving the winner to face
1. G. F. Fitzgerald (1851-1901), The Scientific
Writings (1902). H. A. Lorentz, Proc. Amsterdam Academy
English edition) 6,809 (1903).
2. W. Ritz, "Das Prinzip der Relativity in der Optik,"
Gesammelte Werke, Paris, 1911. W. de Sitter, Ein astronomischer
Beweis für die Konstanz der Lichtgeschwindigkeit," Physikalische
Zeitschrift, 14 (1913).
3. D. Comstock, Physic. Rev., 80 (1910), 267.
4. Sir Oliver Lodge, Past Years, an Autobiography
(New York, 1932).
5. A. A. Michelson died in 1931 at the age of 79:
Lodge, in 1940, aged 89.
6. R. W. Ditchburn, Light
(1953) p. 331.
7. As a contribution for consideration for the
"Charles L. Mayer Nature of Light awards."
8. H. Dingle, "The Origin and Present Status of the
Special Relativity Theory," Science Progress, Vol. XLVIII # 190
(April 1960), pp. 201-19: idem, Nature, 183 (1959), 1761.
9. Cf. M. N. Royal Astron.
Society, 119 (1959), 67.
10. For more recent discussion of this problem by W. G.
V. Rosser, see Nature (April 15, 1961), 249.
11. P. Moon and D. Spenser, J.
Opt. Soc. Am., 43 (1953), 635.
12. W. Heisenberg, The Physicist's Conception of
Nature (1958), p. 48 (transl. from Das Naturbild der heutigen
13. Doklady, Physical Section, Proceeding of
the Academy of Sciences of the U.S.S.R., 109 Engl. Transl. :
Soviet Physics, Vol 1, #4.
14. J. Experiment. Theoret. Phys. (U.S.S.R.), 40
(March 1961), Soviet Physics, Vol. 13, #3 (September 1961).
A MISSED OPPORTUNITY?
In response to a general request for an article by the editors
of the New York Times, Velikovsky submitted the following
observations on July 25, 1973:
On the day of the sun's total eclipse, the Skylab was without
occupants. The first team departed earlier, the second team did not
arrive until weeks after the eclipse. One of the great opportunities
was lost for science. It appears that not even automatic pictures were
In advance of that day, the last in June, teams of scientists arrived in
Africa to photograph the eclipse. Over 50 years had passed since the
full solar eclipse of 1919 was photographed from the isle Principe on
the western coast of Africa by an expedition headed by Eddington, with
another group going to Brazil. Despite poor visibility over
Principe--the eclipse was partly covered by clouds--Eddington found the
sought-for result and thus confirmed Einstein's General Theory of
Relativity and created a stir in all the civilized world. But soon
questions were asked, the results were placed in doubt and since then
every total solar eclipse was sought out by new expeditions, new
photographs, and new checking. An unequivocal result was eluding the
scientific world. The present expedition to Africa will not supply it
According to the General Theory of Relativity a beam of light from a
fixed star is deflected slightly from its rectilinear path when passing
near a massive body. The massive body, suggested by Einstein, is the
sun. Jupiter's mass, a thousand times smaller than the solar mass,
would produce a deflection too small for precise measurement. During a
total solar eclipse the star light is visible. Photographs are made of
the eclipsed sun together with the optically nearby stars. Then the
photographs are compared with the relative position of the same stars during the
night; sometimes the expedition even prefers to return half a year later to
make night pictures of the same stars.
The ray of a star that grazes the sun on its way to the earth must be
deflected by 1.7 seconds of arc, a very small angle, if the General Theory
of Relativity is true, but only half so much if the Newtonian corpuscular
nature of light is right and the General Theory of Relativity is not true.
At any distance farther away from the rim of the sun, the bending of a ray
of light would be less pronounced. Thus the experimenter works on very
small angles trying to determine whether the star changed its optical
position. One tenth of a second of arc, for instance, is an angle sustained
by a silver dollar placed on its rim on the Empire State Building and looked
at from Princeton, N.J., 50 miles away. Therefore any chance of a mistake
or of interference of natural elements may doom the experiment to failure;
and photography from the ground has several unavoidable interferences that
must leave any obtained result in doubt.
In the first place, in an eclipse, besides the sun the moon is a
participating body; each of them pulls the atmosphere: the created tide must
act as a lens distorting the passage of a ray when approaching the Earth.
For some unexplainable reason this phenomenon, to my knowledge, was never
discussed in the evaluation of the results of the tests.
In the space above the Skylab there is no atmosphere. The column of cold
air over the telescope during an eclipse with a different refraction index
is also missing in space. Actually in a few successive passages of Skylab,
only hours apart, new photographs, if taken, would have revealed whether
Einstein was right. The sky seen from Skylab is black and stars are seen.
If, indeed, the result would confirm Einstein's expectation, then still
another test would be desirable--namely, to see whether the planet Jupiter
which, as already said, is one thousand times less massive than the sun,
performs an unexpectedly strong bending effect on the passing star light.
If this would be the case, it would point not to gravitation but to a
powerful electromagnetic field of this planet
as the cause of bending. It is not known and, to my knowledge it was not
experimented to find out, whether a strong electromagnetic field deflects a
beam of light in a laboratory. In view of the great distance that the light
after grazing Jupiter has to pass to reach the observer (in the order of
five hundred million miles)--and thus a possible bending becoming so much
"enlarged"--a Skylab experiment has all the advantage against a laboratory
In such an observation different effects, if any, should be expected from
the star's light passing close to the magnetic equator of Jupiter or close
to its magnetic poles. This experiment of photographing Jupiter and the
grazing light of a star was devised by me quite a few years ago; it still
can be performed by one of the new Skylab teams. The answer may perchance
make the missing of an experiment with the eclipsed sun less painful.
However, quite possibly it will not be necessary to wait for another fitting
solar total eclipse (which event will not come to pass for quite a long
time) to experiment. With the help of a coronograph--a metal disk covering
in a telescope the bright disk of the sun--the team that now reaches the
Skylab could try to establish whether the sun deflects the star light. The
sun, too, has a magnetic field and the sun spots are strong magnets, but if
it is the mass that deflects light, the sun must produce a much
stronger effect than Jupiter.
I imagine that future scientific expeditions to remote places to observe
solar eclipses will consist mainly of zoologists, anthropologists, and
psychologists to register the behavior of animals and of primitive man, in
order to learn of the vestiges of racial experiences and responses to them
carried over from the remote past. The astronomical observations will be
made not from jungles or deserts, but from space stations.
PENSEE Journal V