AN OBSERVATIONAL TEST OF DOPPLER'S THEORY USING

SOLAR-SYSTEM OBJECTS

J. WILENCHIK

Abstract. This year is the 180th anniversary of Christian Doppler’s hypoth-

esis that colors of light shift due to movement by the source or observer. In
1868, astronomer William Huggins proposed doing an observational test of
Doppler’s theory using planetary velocities. But Huggins calculated that the
Doppler shifts “would not exceed a fraction of the interval between the dou-
ble line D” and that 19th-Century instruments were not up to the task. I
conducted a small observational test of Doppler’s theory, much like Huggins
intended, using planetary and lunar velocities and shifts in Sodium D lines. I
reduced a handful of observations from the data archive for the High Resolu-
tion Echelle Spectrometer (HIRES) without Doppler (heliocentric) corrections,
then compared the shifts to calculated values from the NASA/JPL Horizons
ephemeris system. I found both observations that agreed and disagreed with
Doppler predictions, which is encouraging for future “Huggins Tests.” Per-
forming these tests is critically important, because Doppler’s theory is behind
the theory that our universe is an expanding “primordial atom”–or the “Big
Bang” theory, as it is more commonly known. Doppler’s theory is also based
entirely on another theory that physicists have doubted since Newton. I pro-
pose a simple alternative to the “Big Bang” that is grounded in theories of
Isaac Newton and Lord Kelvin, which I call “atoms all the way down.”

Doppler Shift – Spectroscopy – Astronomical Techniques – History of Astronomy

1. Introduction
Mathematics professor Christian Doppler first hypothesized that colors of light
shift due to movement by the source or observer in a paper published in 1842.
Doppler (1842). Doppler’s original paper was purely theoretical, and it reached
conclusions that were quickly recognized as erroneous even in their own time. For
example, Doppler suggested that the actual color of every star was white or yellow,
and that stars’ apparent colors (red, blue, etc.) were due solely to their radial
velocities with respect to the earth. Nevertheless, the last sentence of his original
paper proved to be prophetic: in “[t]he distant future,” he wrote, his theory would
“offer astronomers a welcome means of determining the motions and distances” of
distant stars and other objects whose velocities are otherwise “immeasurable.”
The instruments of the 19th Century lacked the resolution needed to test Doppler’s
theory on celestial objects with known velocities, such as solar-system planets and
moons. Huggins (1868b). As astronomer William Huggins wrote in 1868: “[t]he
great relative velocity of light to the known planetary velocities, and to the probable
motions of the few stars of which the parallax is known, showed that any alterna-
tions of position which might be expected from [Doppler shift] in the lines of the

E-mail address: jack@atomsallthewaydown.com.

1
stellar spectra would not exceed a fraction of the interval between the double line
D, for that part of the spectrum.” “I have devoted much time,” Huggins continued,
“[and] I hope to accomplish the detection of so small an amount of change. . . [but]
[t]he difficulties of this investigation I have found to be very great. . . ” (Id.) The
first astronomer(s) to use Doppler’s theory therefore focused on targets whose ve-
locities could not be rigorously and independently measured, like distant stars and
nebulae or gases on the solar surface. Huggins (1868a,b).
But a modern spectrograph like the Keck Observatory’s “High Resolution Echelle
Spectrometer” (HIRES) is more than capable of performing the “Huggins Test.” I
report the results of a limited exploratory test of Doppler’s theory on solar-system
planets and moons using the shift in their D lines, much like William Huggins
intended.

2. Methodology
I searched the Keck Observatory Archive (KOA) for solar-system data from the
HIRES, particularly planets and moons with low axial rotation.1 The HIRES has
a precision on the order of meters per second and has been heavily used in searches
for exoplanets; accordingly, its archives contain comparatively few observations of
solar-system objects. A handful of observations were used: two observations of
Venus in 2007 and 2009, one of Ganymede in 2009, one of Europa in 2009, and one
of the dwarf planet Ceres in 2005. The data for various observations of Mercury
were also considered, but the signal-to-noise ratio was deemed to be too low (and
airmass too high) to be included in this exploratory study.
The KOA offers data that has already been reduced and extracted by the Keck
Observatory “MAKEE” pipeline (“MAuna Kea Echelle Ex-traction”). However,
that pipeline normally applies a “heliocentric correction” of up to around ±30 km/s,
which is designed to account for the putative Doppler effect of the movement of the
earth at the time of observation. MAKEE can be run manually with heliocentric
corrections turned “off”; and so I downloaded the same raw science and calibration
data that was used to produce the extracted data in the archives then re-extracted
it using MAKEE without heliocentric corrections. Because I made no effort to
account for the effect of the bodies’ (or the earth’s) axial rotations on Doppler
shift, I treated them as a source of error in the objects’ calculated radial velocity
(see Ecalc in Table 1 on Zenodo2). The speed of axial rotation for each object in this
study was between ±0.01 and 0.15 km/s, and earth’s rotation was estimated at 0.5
km/s, so Ecalc was never greater than ±0.52 km/s. Putative relativistic effects were
calculated to be less than 0.01 km/s and therefore neglected. Finally, the measured
Doppler shift in the D lines was compared to radial velocity as calculated by the
NASA/JPL Horizons ephemeris system. More details on methodology, including
the code and data that was used, are on Zenodo (see Appendix “A”).

3. Results
Figure 1 shows plots of the measured and calculated Doppler shifts. While the
Sodium absorption lines in Venus’ and Ceres’ atmospheres appeared at or near
their Doppler-predicted positions, the lines in Ganymede and Europa did not. The
1
Rates of rotation were calculated from Cox (1999); or in the case of Venus, also from Hori-
nouchi et al. (2020) (indicating that Venus’ atmosphere rotates sixty times faster than its surface).
2doi:10.5281/zenodo.6240436

2
mean absolute difference (weighted by error) in between measured (Doppler) and
calculated (JPL Horizons) velocity for Ganymede and Europa was 9.24±0.72 km/s.
Space-based (Hubble) spectroscopy confirms Na D absorption lines in the atmo-
spheres of both Ganymede and Europa,3 which tends to discount telluric interfer-
ence as a cause for the discrepancy. Its magnitude(9.24 km/s) would also tend to
discount atmospheric winds and other internal dynamics.
The discrepancy is less if the lines are compared to the Doppler-predicted shift
in solar light reflecting from the body, which is given by:
Rhelio Rcalc
(1) Rref lect = Rhelio + Rcalc +
c
(where Rhelio is the object’s heliocentric velocity, Rcalc is its geocentric velocity,
and c is the speed of light in a vacuum). However, the bodies’ spectra do not show
separate lines for reflected light (albedo) and light originating from the object, as
Doppler’s theory would predict. Because this was an exploratory test on a small
sample, a larger Huggins Test should be conducted to determine whether these
results are statistically valid.
3.1. Concerns with the Doppler Modeling of Planetary Spectral Line
Inclinations. The spectra of Jupiter and Saturn are known to be “tilted,” or to
exhibit a linear inclination. (Figure 2.) Historically, the cause of this inclination was
deemed to be Doppler shift due to each planet’s rotation about its own axis. Moore
(1939); Giver (1964). However, the radial velocities of points across a spherical
rotating body should exhibit a curved, sinusoidal pattern. (Figure 3.) The observed
“tilt” is always linear, which suggests a cause other than Doppler shift.

4. Discussion
If Doppler’s theory proves to be incorrect, then it would either undermine or
support various theories in the history of physics. In particular, it would undermine
the standard “Big Bang” cosmological model; but it would support certain theories
by Isaac Newton and Lord Kelvin, as well as an alternative cosmology based on
them, which I call “atoms all the way down.”
4.1. “Some Other Cause”.

When astronomer Vesto Melvin Slipher first discovered the substantial redshifts
in distant objects like galaxies, he immediately applied Doppler’s theory to his
findings–but he harbored serious reservations right from the start. Slipher thought
that the velocities he calculated using Doppler’s theory were too large to make
sense. In his first report on the subject, Slipher wrote: “The magnitude of this
velocity, which is the greatest hitherto observed, raises the question whether the
velocity-like displacement might not be due to some other cause, but I believe we
have at present no other interpretation for it.” Slipher (1913).
Before Slipher made his findings, astronomers Edwin Hubble and Milton Huma-
son had measured the distances to galaxies through methods unrelated to spectral
shift: namely, through measurement of the reduced luminosity in galaxies’ recog-
nized star types. Hubble (1929). Hubble noticed that the redshifts in galaxies

3See Observation ID “o51u02040” (Ganymede) and “od9l140m0” (Europa) in the ESA Hubble
Science Archive.
3
 4
Figure 1. Plots of the D lines (actual and predicted) and Gaussian
fits for the five observations. Error is noted with horizontal bars and
generally thinner than the lines, with the notable exception of the yellow
lines (the predicted shifts in albedo) which had more significant error
due to the calculated rotation of the sun. (See Table 1 on Zenodo.)
 Figure 2A - On the left is an image of the spectrum of Saturn taken on June 25, 2018
(along with the corresponding camera image of Saturn below it, demonstrating place-
ment of the spectroscopic slit). On the right is a spectrum of Jupiter taken on June 25,
Figure 2. At top left is an image of the spectrum of Saturn taken on
2018 (again with a corresponding camera image below it, demonstrating placement of
the slit across the face of the body). The linear inclination in both planets’ spectra is
June 25, 2018. Below it is the corresponding camera image of Saturn,
apparent. (Source: 10.5281/zenodo.3588493, observations nos. 224 and 225.)

which demonstrates the placement of the spectroscopic slit across the

face of the planet. At top right is a spectrum of Jupiter taken on June
25, 2018, and below it is the corresponding camera image, which again
demonstrates placement of the slit. The linear inclination in both plan-
Figure 2A - On the left is an image of the spectrum of Saturn taken on June 25, 2018
ets’ spectra is apparent. (Source: 10.5281/zenodo.3588493, observations
(along with the corresponding camera image of Saturn below it, demonstrating place-

nos. 224 and 225.)ment of the spectroscopic slit). On the right is a spectrum of Jupiter taken on June 25,
2018 (again with a corresponding camera image below it, demonstrating placement of
the slit across the face of the body). The linear inclination in both planets’ spectra is
apparent. (Source: 10.5281/zenodo.3588493, observations nos. 224 and 225.)

Figure 2B – At top (in gray) are illustrations of the expected sinusoidal pattern of spec-
tral lines that are Doppler-shifted by the rotation of a spherical body. At bottom, the
corresponding placement of a theoretical spectroscopic “slit” on the body’s surface is
shown.

Figure 2B – At top (in gray) are illustrations of the expected sinusoidal pattern of spec-
tral lines that are Doppler-shifted by the rotation of a spherical body. At bottom, the
corresponding placement of a theoretical spectroscopic “slit” on the body’s surface is
shown.
Figure 3. At top (in gray) are illustrations of the expected sinusoidal
pattern of spectral lines that are Doppler-shifted by the rotation of a
spherical body. At bottom is shown the corresponding placement of a
theoretical spectroscopic “slit” on the planet’s surface.

corresponded to their distance; but after reviewing Slipher’s data, he also applied
Doppler’s theory to the observed shifts and treated them as “radial velocities”
without question.
If Doppler’s theory is invalid, then what we are left with is the simple fact that
very distant objects exhibit redshift; and the more distant the object, the stronger
the redshift. The only theory necessary to fit these facts is that extreme distance
causes redshift.

4.2. The “Controversy” Behind Doppler’s Theory.

In the introduction to his original paper, Christian Doppler acknowledged that

his theory depended on another theory that was “known to have encountered dif-
ficulties” over the course of its history. Many “highly respectable” physicists, he
wrote, believed that the other theory was “by no means correct and true,” and
5
that it was unworthy of being “defend[ed]” or “cultivat[ed].” But the physics com-
munity nevertheless “tolerate[d] [the theory]. . . for the time being,” Doppler wrote,
“because of its success.”4
The controversial theory that Doppler was referring to was the “wave” theory
of light, and the controversy over it has never quite faded away. The dispute dates
back to at least Isaac Newton, who rejected the theory outright and believed that
different colors of light consisted of very small bodies (“corpuscles”) with different
sizes.5 Physicists’ dissatisfaction with the “light wave” theory seems to come and go
in “waves,” resurfacing again at the beginning of the Twentieth Century in Albert
Einstein’s work and the theory of photons.
In his treatise Opticks, Isaac Newton expressed his own dissatisfaction with the
light-wave theory by pointing to the way that water waves pass around an object.
Waves of water, he wrote, flow around obstructions; but light does not. 6 Ironically,
what excited Doppler about his own theory was mostly the same thing: the way that
water waves pass around a (moving) object. “According to common experience,”
Doppler wrote, “a ship that is even somewhat deep, which is steering straight
against the oncoming waves, has to suffer a greater number and much more violent
wave impacts at the same time as one that is at rest or even moves with them in the
direction of the waves. But. . . why should this not be accepted, with the necessary
modifications, also from the air and ether waves [of light]? It seems as if there is
hardly anything substantial that can be said against it!”7

4.3. A Not-So-Brief History of Time.

Doppler’s enthusiasm for his own theory is admirable, but of course any theory
is just that – a theory. On the first page of his book A Brief History of Time,
Steven Hawking tells the story of a “well-known scientist (some say it was Bertrand
Russell)” who gave a public lecture on astronomy. “At the end of the lecture, a
little old lady at the back of the room got up and said: ‘What you have told us is
rubbish. The world is really a flat plate supported on the back of a giant tortoise.’
The scientist gave a superior smile before replying, ‘What is the tortoise standing
on?’ ‘You’re very clever, young man, very clever,’ said the old lady. ‘But it’s turtles
all the way down!’” Hawking (1988).
According to Hawking, the point of the story was that while “[m]ost people would
find the picture of our universe as an infinite tower of tortoises rather ridiculous,”
“why do we think we know better?” “Where did the universe come from, and where
is it going? Did the universe have a beginning, and if so, what happened before
then?” Someday, our answers to these questions “may seem as obvious to us as the
earth orbiting the sun—or perhaps as ridiculous as a tower of tortoises.” (Id.)

4Doppler (1842) at section 1.

5See Opticks, Question 28: “Are not all Hypotheses erroneous, in which Light is supposed to
consist in Pression or Motion, propagated through a fluid Medium?” Newton (1704)
6From Opticks, Question 28: “The Waves on the Surface of stagnating Water, passing by
the sides of a broad Obstacle which stops part of them, bend afterwards and dilate themselves
gradually into the quiet Water behind the Obstacle. . . [but] Light is never known to follow crooked
Passages nor to bend into the Shadow.” Newton (1704)
7Doppler (1842) at section 2.

6
 The standard model of cosmology strikes many people as non-intuitive—including
Albert Einstein, when he first heard of it.8 Standard cosmology suggests that the
universe began in a singular event for which there was no cause, and therefore no
explanation. But the foundation for all science is that there is a cause for every
effect, and an explanation for everything—which the Big Bang seems to contradict.
So to paraphrase Hawking: is there a better, more obvious answer?

4.4. A New Cosmology: “Atoms All the Way Down”.

At the risk of committing the same “sin” as our friend Christian Doppler – which
is to say, engaging in mind-provoking but ultimately unproven speculation – I pro-
pose a new alternative cosmological model called “atoms all the way down.” To the
mind of a “standard” cosmologist, the model will seem disruptive and radical or
even lacking in scientific basis. But we must keep in mind Hawking’s warning about
the “relativity” of such things: a theory that seems “ridiculous” to one person may
be “obvious” to others.
A good cosmological model should start with obvious and uncontroversial facts.
So we begin with these facts: (1) galaxies are very large; (2) they are very far; and
(3) most of them have whirlpool-like spirals.
The Sloan Digital Sky Survey has uncovered another important fact, which is
not as widely known as it should be, and which poses a very serious challenge to
standard cosmology. By no slight exaggeration, it is also the “largest” fact there
is: the positions of galaxies in the universe do not appear random and reveal larger
structures. Galaxies form superstructures consisting of distinct “walls” surrounding
“voids” – effectively, an enormous “soap-sud like” grid.9 de Lapparent et al. (1986)
As mentioned earlier, there is a long and curious history of scientists like Newton
and Doppler trying to understand important physical phenomena by thinking about
water. In the late Nineteenth Century, William Thomson (a.k.a. Lord Kelvin)
proposed that atoms have the structure of a whirlpool, or vortex. Thomson (1867)
But like most quantum theories, Kelvin’s model seemed insusceptible to any clearly-
visible proof. After all, how do we “see” what an atom really looks like?
But Kelvin proposed his theory before we discovered hundreds of millions of
enormous “whirlpools” in space—or that those whirlpools form an enormous “grid.”
Even Georges Lemaı̂tre, the creator of the “Big Bang” theory, intuited that galaxies
and atoms have some kind of fundamental symmetry; Lemaı̂tre referred to the
“Bang Bang” theory as the theory of a “primeval atom,” which he suggested had
“expanded” to form the galaxies. Kragh (2012)
The search for a new cosmology does not have to go far. If Kelvin was right,
then atoms may have a “whirlpool” structure; and galaxies may show “whirlpools”
because they share this structure. Galaxies may be “atoms,” i.e. they may share
atomic physics. Galaxies may form large-scale “molecular” structures in the same
way that atoms do, which accounts for large-scale galactic structures and groups.

8Einstein’s initial reaction to the theory of an expanding universe was to express his disap-
proval; according to the creator of the theory Georges Lemaı̂tre, Einstein commented to him:
“Your calculations are correct, but your physics is atrocious.” Deprit (1984)
9“The distribution of galaxies. . . looks like a slice through the suds in the kitchen sink. It
appears that the galaxies are on the surfaces of bubble-like structures with diameters of 25-50/h-
Mpc.” de Lapparent et al. (1986)
7
 Stars must have a fundamental role to play in this cosmology. Stars are found
throughout a galaxy, and throughout space. In the main sequence of stars, the
mass of a star generally corresponds to its color – with blue stars being larger, and
red stars being smaller.
Remember back to Newton, who was convinced that colors of light are composed
of very small bodies with different sizes.10 Newton suggested that it might even be
possible to “see” these bodies of light, using the optical instruments of his day.11
Stars may be “light.” And in a powerful piece of cosmic symmetry, the color of
a star may correspond to its color of “light.”
This is why Doppler’s theory is wrong. Newton’s intuition was always right:
light is not in “waves.” We can see exactly what “light” looks like, and how it
behaves, by observing and studying the stars. Newton was always correct that we
can see what “bodies” of light look like – just with telescopes, not microscopes.
Atoms emit and absorb light. If stars are “light,” and galaxies behave like atoms,
then galaxies must also emit and absorb stars. This means that galaxies create and
destroy stars continuously. Clearly, this cosmological model is fundamentally differ-
ent from the “Big Bang”– because in this model, matter is created (and destroyed)
continuously, and not just at one singular “primeval” moment in time.
The notion that galaxies “absorb” stars could explain the activity that we observe
in “black holes.” Black holes may be where galaxies “absorb” stars, in the same
way that atoms absorb light.
Many galaxies contain tight concentrations of stars that are nearly all the same
color, called globular clusters. These may be places where stars are actively created
and “emitted” from the galaxy. New stars may be ejected from galaxies and travel
in between them, in the same manner that light travels in between atoms.
When atoms absorb light, they generate electrons/current (a.k.a. electricity, and
this effect is known as the “photoelectric effect”). The converse is also true: when
atoms emit light, they “lose” electrons/current. In other words, atoms convert
light (and other forms of radiation)12 into electricity, and then electricity back into
light/radiation again.
If galaxies are “atoms” and stars are “light,” then what and where are the
galaxy’s “electrons”? To answer this question, we must carry the “atoms” analogy
further. Electrons are made of something much smaller than atoms, and even
smaller than light; and they compose the basic structure of the atom. Like electrons
in the atom, the galaxy’s “electrons” must be made of something much smaller than
galaxies, and even smaller than atoms.
In a universe where galaxies resemble atoms, and atoms resemble galaxies, then
there is an apparent answer to what these even “smaller” things would be. Atoms
10See Opticks, Question 29: “Are not Rays of Light very small Bodies emitted from Shining
Substances?” “Nothing more is requisite for producing all the variety of Colours. . . than that the
Rays of Light be Bodies of different Sizes. . . ” Newton (1704)
11(From Opticks, Proposition VII: “it is not impossible but that Microscopes may at length
be improved to the discovery of the Particles of Bodies [of light] on which their Colours depend, if
they are not already in some measure arrived to that degree of perfection. For if those Instruments
are or can be so far improved as with sufficient distinctness. . . I should hope that we might be able
to discover some of the greatest of those Corpuscles [of light].” Newton (1704))
12If stars are “light,” and the color/mass of a star corresponds to its color/energy of “light,”
then bodies with less mass – like planets and comets – may be the equivalent of “infrared” and
“radio.” Larger bodies, such as more massive stars outside of the main sequence (or even globular
clusters), may constitute “high-energy radiation” like ultraviolet and X-rays.
8
must also contain their own little “atoms.” These “atoms in the atom” – which I
will term “subatoms” – must compose the atom’s little “stars,” i.e. light. In other
words, “subatoms” are the quanta that compose light—also known as photons.
As the next available “smaller” thing than atoms, the “subatoms” (photons) must
compose the galaxy’s “electrons” or “current.”
Let us turn back to Lord Kelvin, who theorized that the atom was a vortex
composed of a fluid. Atoms are composed of electrons, or electricity/current—which
behaves substantially like a fluid. If Kelvin was right, then an atom may be a vortex
composed of electric current. If we apply this to the galaxy, then it means that the
galaxy is a vortex of its own “current”—i.e., a vortex of photons. And if this is
correct, then there should be clear evidence of it: we should observe giant “vortices”
made up of photons, occupying the space in and around galaxies.
Which we do. Another criminally-underpublicized discovery of the last several
decades is the existence of large vortex-like jets, loops, lobes, and corkscrews orig-
inating at the center of galaxies, including our own. These objects often appear
in the invisible spectrum, i.e. as radio or X-rays, and have been variously termed
“jets” or “active galaxy nuclei” See e.g. Dickinson (2018). These massive objects
are evidence of exactly what this theory predicts: giant vortices composed of pho-
tons, occupying the space in and around galaxies.
At the scale of galaxies, photons may behave substantially like a fluid: they
are present nearly everywhere, but with a measurable general direction and mag-
nitude/pressure. Photons may “flow” through the galaxy in the same way that
current flows through an atom—or water flows through a vortex. This “fluid” of
photons constitutes the basic structure or “matter” in a galaxy, while the atoms
and stars constitute its “energy” or “radiation.” This means that when we view an
image of the stars in a galaxy, it is like looking at the light emitting from an atom;
and when we view images of radio jets and loops, it is like looking at the atom’s
underlying electron structure.
Finally, if a galaxy’s “radiation” is atoms, and its “electrons” are photons, then
when the galaxy converts its “radiation” into “current” (the photoelectric effect)
and its “current” back into “radiation,” then the galaxy must be converting atoms
into photons and then photons back into atoms again. This must be how an infi-
nite, recursive universe really works – it creates and destroys things constantly, by
converting things of a larger dimension into things of a smaller dimension and then
back again.13 And all “things” – whether they be galaxies, atoms, ”subatoms”or
beyond – share the same basic repeating vortex structure.
I like to call this alternative cosmological model “atoms all the way down” (in
honor of Steven Hawking and a nameless “little old lady”). The model aims to pro-
vide fundamental explanations for the most obvious phenomena in space like stars
and spiral galaxies, as well as our more recent discoveries like galaxy superstructure
and radio jets. The model also offers potential solutions to a number of otherwise
“unsolvable” mysteries, ranging from exactly why and where galaxies are located
in the “grid”14; to why many galaxies share the same bimodal distribution of star
13This theory also predicts that there are things so incomprehensibly large, that when com-
pared to them galaxies are like the size of atoms. These theoretical structures – which we could
call “super-galaxies”—must also be capable of “converting” (merging?) atoms into galaxies, and
(splitting?) galaxies back down into atoms.
14“Groups” of galaxies may resemble molecules; and the positions of the groups may resemble
a known supramolecular structure. The Milky Way Group, M81 Group, and other small galaxy
9
colors15; to what chemical bonding really “looks” like and exactly how it works;16
to what neutrons “really” are and exactly how they interact with atoms17; to how
black holes work18; to how atoms absorb and emit radiation19; and to how atomic
fusion, fission, and forces like gravity “really” work.20

groups should be studied for their consistency in terms of the number and sizes of their galaxies,
and the relative positions of their galaxies. This can then be compared to the positions and “sizes”
(masses) of atoms in various kinds of molecules; and the arrangement of groups can be compared
to known supramolecular structures.
15If stars behave like “light,” and the color of a star is its color of “light,” then galaxies
may primarily produce only certain colors of stars, just like individual atoms produce primarily
certain colors of light (spectral emission lines). It may be possible to identify the “element” of
an individual galaxy by cataloguing the colors of all stars being “emitted” by the galaxy (i.e.,
its globular clusters). Other means of determining a galaxy’s “element” could include studying
the galaxy’s position and size relative to other nearby galaxies (in its “molecule”/group), and
potentially counting the number of “dwarf galaxies” that are found in and around larger galaxies
(see below).
16Radio imaging of the M81 Group of galaxies (discussed below) shows fluid-like connections
in between the galaxies, which may be evidence of “chemical bonds” between the galaxies. We
can study/reproduce in the lab the process by which fluid vortices form similar connections, which
may help explain how these bonds actually form in atoms/galaxies and thus how chemical bonding
“really” works.
17Many galaxies contain smaller objects that display galaxy-like features (e.g. they show
black holes and/or radio jets or loops) but they are a fraction of the size of a galaxy. These
objects, variously characterized as “dwarf galaxies” and/or “microquasars,” may correspond to
“neutrons.” If this is correct, then these kinds of objects should occur in relatively small, “neutron-
like” numbers near to/inside of a galaxy. (And we may also be able to identify the galaxy’s
“element” by counting them – see above.) The basic features of active galactic nuclei, like the
number and size of their jets and loops, may correspond to basic atomic properties like mass or
the number of electron “shells.” (And in turn: the size of an atom’s “vortex” may correspond to
its “mass,” while the speed that fluid/current passes through it may correspond to voltage/charge.
The periodicity in elemental charges may have something to do with the number of loops/jets in
the atom’s vortex.)
18Fluid vortices dissipate when they run out of energy. It may be that atoms, too, will
“dissipate” when they are in a space that is totally devoid of radiation/energy, like a black hole.
19Because the center of the galaxy is a place with no photons—i.e., a “low pressure” point,
like the center of any vortex – then it may cause the galaxy’s “current” of photons to continuously
flow towards it (thereby forming a vortex). At the center of the galaxy (the black hole), atoms
may dissipate (see above) and then contribute their photons into the galaxy’s “current.” (In other
words, the galaxy “absorbs” them.) This injection of photons into the galaxy helps to sustain its
vortex, and it also increases the speed of the vortex’s flow (i.e. it makes the galaxy’s “charge”
more negative—see footnotes above). But if this were to cease – i.e., if the galaxy were deprived
of atoms/photons to absorb – then its flow of photons would eventually dissipate (just as an atom
deprived of radiation may dissipate in a black hole). With these principles applied to photons in
the atom, it may explain how atomic absorption fundamentally works. “Emission” in the galaxy
may occur when its current of photons encounters intense pressure (due to a faster flow). This
may cause the photons to merge, become atoms, and then start absorbing photons rather than
flowing with them. As matter continues to be generated like this, the “old” matter gets pushed
out, causing emission. Needless to say, understanding exactly how these abstruse mechanisms
work may be one of the final and most important frontiers in all science.
20Atoms may split and merge in essentially the same manner that fluid vortices split and merge;
and they may repulse each other in essentially the same manner that two adjoining vortices repulse
(unless pressured to merge). Gravity in the atom may function in basically the same way that
vortices pull in the fluid surrounding them. Electric fields may be created when current flows
through the atom faster, causing its vortex to “suck” that much harder and attract more slowly-
moving vortices/fluid. (In other words, faster-flowing current creates a “negative” charge; and
slower-flowing current creates a “positive” charge. The reason why more massive elements on
10
 The “atoms all the way down” model also leads to intuitive predictions. For
example, our own group of galaxies is called the Milky Way Group, and the nearest
group is termed the M81 Group. Radio images of the M81 Group reveal fluid-
like “streams” connecting the largest galaxies in the group. This may be evidence
of a “current” of photons flowing in between the galaxies, i.e. of “electron shar-
ing”/“molecular bonding” in between the galaxies of the group. Notably, the M81
Group contains three large (major) galaxies like our own Milky Way Group. The
three major galaxies in the Milky Way Group (the Milky Way, Andromeda and Tri-
angulum galaxies) may also be connected to each other by these kinds of radio-loud
“streams.” This would indicate a similar kind of “molecule”/“molecular bond” to
what is being observed in between the galaxies of the nearby M81 Group. Other
galaxy groups should also contain similar “bonds.” And in general, the relative
sizes and positions of major galaxies in groups should resemble known molecu-
lar/supramolecular structures.

5. Conclusion
Christian Doppler’s theory that light Doppler-shifts was accepted and widely
applied without an observational test on solar-system objects, due to historical
limitations on spectroscopic instruments. The results of my exploratory “Huggins
Test” raise substantial questions about Doppler’s theory and are encouraging for
other larger tests.
The standard cosmological model is based on Doppler’s theory; so without his
theory, the model lacks a foundation. My alternative model – the “atoms all the
way down” model – is simple: it posits that galaxies are “atoms” and stars are
“light.” According to the theory, the reason for why Doppler’s theory is wrong is
that light is not in “waves.” Rather it is corpuscular, just as Newton always thought.
The “atoms all the way down” model offers immediate explanations for basic
celestial phenomena, like spiral galaxies and stars; and it not only acknowledges
but predicts major discoveries in astronomy over the last century, like large-scale
galactic structure and radio-loud galaxy jets. In contrast, the standard model takes
the existence of such things as stars and spiral galaxies “for granted.” Explaining
them is more of a “distraction” to the “Big Bang” model than a fundamental part
of it.
The standard model basically posits that the universe is an expanding atom,
which is no less plausible on its face than an “atoms all the way down” model.
If embraced, the “atoms” model will lead to new fields of research in astron-
omy/astrophysics, quantum physics, and other scientific disciplines. The most im-
portant questions in astronomy and cosmology can be answered by our knowledge
of quantum physics; and the most important questions in quantum physics can be
answered by our observations of the galaxies and stars.

the periodic table skew electropositive could be that current takes longer to flow through them,
resulting in a slower – and therefore more “positive” – flow/charge.) Magnetic fields may be
created by current passing into and then out of a series of aligned atoms/vortices, before being
pulled back around and into them again. Electromagnetic fields may be a special case in which
the current passing through a series of vortices induces the vortices to align; and their natural
rotation causes current to rotate in the space around them. Again, it should be possible to study
and reproduce these and other atomic phenomena through experimentation on fluid vortices.
11
 6. Acknowledgments
This research made use of the services and data of the European Southern Obser-
vatory (ESO) Science Archive Facility and is based on data products from observa-
tions made with ESO Telescopes at the La Silla Paranal Observatory. This research
also made use of the Keck Observatory Archive (KOA), which is operated by the
W. M. Keck Observatory and the NASA Exoplanet Science Institute (NExScI), un-
der contract with the National Aeronautics and Space Administration. Profound
thanks go to Dr. Zach Cano for his invaluable advice and assistance, as well as to
Dr. Francisco Pina, Dr. Rafael Brahm and Dr. Christina Thöne.

7. Declarations
No outside funding was received to assist with the preparation of this article.
The author has no conflicts of interest to declare that are relevant to the content of
this article. The data reviewed in this article is publicly available from the archives
cited above. Spectral analysis was performed with publicly-available software.

Supplemental Documents
The Appendices and other supporting materials/data referenced in this article
can be found on Zenodo (doi:10.5281/zenodo.6240436), along with a Python script
that will reproduce my data reduction and analysis.

References
Cox, A. N. 1999, Allen’s Astrophysical Quantities (Springer, 4th Ed.)
de Lapparent, V., Geller, M., & Hulchra, J. 1986, The Astrophysical Journal, 302,
L1. https://ui.adsabs.harvard.edu/#abs/1986ApJ...302L...1D/abstract
Deprit, A. 1984, in The Big Bang and Georges Lemaı̂tre, ed. A. Barger (Reidel),
370, doi: 10.1007/978-94-009-6487-7_30
Dickinson, C. 2018, Astrophysical Journal, 6, doi: 10.3390/galaxies6020056
Doppler, C. 1842. https://bit.ly/3qpUJpc
Giver, L. 1964, Astrophysical Journal, 139, 727, doi: 10.1086/147798
Hawking, S. 1988, A Brief History of Time (London: Bantam)
Horinouchi, T., Hayashi, Y., Watanabe, S., et al. 2020, Science, 368, 405, doi: 10.
1126/science.aaz4439
Hubble, E. 1929, Proceedings of the National Academy of Sciences, 15, 168, doi: 10.
1073/pnas.15.3.168
Huggins, W. 1868a, Proc.R.Soc.London Ser. I, 382, doi: 10.1098/rspl.1867.0080
—. 1868b, Phil. Trans. R. Soc. London, Ser. I, 529, doi: 10.1098/rstl.1868.0022
Kragh, H. 2012, in Georges Lemaı̂tre: Life, Science and Legacy. Astrophysics and
Space Science Library, ed. R. Holder & S. Mitton, Vol. 395 (Springer), doi: 10.
1007/978-3-642-32254-9_3
Moore, J. 1939, Publications of the Astronomical Society of the Pacific, 51, 274,
doi: 10.1086/125081
Newton, I. 1704, doi: 10.5479/sil.302475.39088000644674
Slipher, V. 1913, Lowell Observatory Bulletin, 1, 56. https://ui.adsabs.
harvard.edu/abs/1913LowOB...2...56S
Thomson, W. 1867, Proceedings of the Royal Society of Edinburgh, 6, 94, doi: 10.
1017/S0370164600045430

12
 Appendix “A”

Version 6.4 of “MAKEE” was used to extract and reduce the data. The version used is dated May 2019 and
available for download from: https://sites.astro.caltech.edu/~tb/makee/makee_6.4-2019.tar.gz. MAKEE was
run in a command terminal using Ubuntu 20.04.3 LTS. The MAKEE pipeline requires at least four “FITS”
(Flexible Image Transport System) images to reduce and calibrate data: an image of the object; an image to
find the “trace” of the echelle orders (which can simply be the image itself, although a star is often used); flat
image(s); and an image of the arc lamp for wavelength calibration. Each image in the Keck Observatory Ar-
chive (KOA) is assigned a unique “KOAID.” The KOAID for each of the raw science and calibration images
used in this paper (as well as the CCD and orders extracted) are listed below:

Obs. Object KOAID Trace (star) KOAID Flat KOAID Arc KOAID CCD Order(s)
1 HI.20091213.17797 HI.20091213.08389 HI.20091213.08988 HI.20091213.10643 3 11
2 HI.20091211.17597 HI.20091211.10571 HI.20091211.12418 HI.20091211.12272 3 11
3 HI.20070606.19972 HI.20070607.01296 HI.20070606.02904 HI.20070606.16831 2 13
4 HI.20090107.16390 HI.20090107.16390 HI.20090107.02523 HI.20090107.01375 2 13
5 HI.20050617.21254 HI.20050616.06005 HI.20050617.09547 HI.20050617.11120 2 6,7

To remove the heliocentric correction, MAKEE was run using the “-nohc” option. The “-koa” option was also
used, which outputs the processed data into “.tbl” files. Finally, in order to run MAKEE, the user must specify
a CCD number to be extracted (using the “ccd=*” argument). The final command for processing each obser-
vation was “makee [Object.fits] [Trace.fits] [Flat.fits] [Arc.fits] ccd = [CCD No.] -nohc -koa.” An optional
“log=*.txt” argument sends the command-line output into a “*.txt” file.

After running MAKEE, the region of the Sodium D lines (5890 – 5900 Å) was identified in the extracted orders.
The wavelength, flux and error spectrum in the region of the D line(s) was then manually extracted into a “.csv”
file (which is contained in the Zenodo depository and named “*_full.csv” for each observation). In Observa-
tions No. 1 and 2 (Europa and Ganymede), the D1 line fell beyond the extracted orders, and so only the D2
lines were used. The D lines in Observation No. 5 fell across two different orders; so the data in Order # 7 was
used for the D1 line and part of the D2 line, with the remaining data for the D2 line from Order # 6. Postscript
images of the orders for all extractions can be found in the “logs” folder on Zenodo, along with the MAKEE
command-line output logs.

To calculate the parameters for a Gaussian fit to each of the Sodium D lines, the “curve_fit” function in Py-
thon’s SciPy package was used (“SciPy: Scientific Library for Python” version 1.7.3). The error spectrum in
the MAKEE-generated data tables (column #7, “Error”) was input as “sigma” in the “curve_fit” routine. This
produced parameters for the best-fit Gaussian function for each D line, as well as an estimated covariance. The
standard deviation in the Gaussian centerline was calculated from the covariance; and this standard deviation
was used for error in the measured Doppler shift of each D line. Finally, for those observations in which both
D lines could be detected, an average of the two shifts was calculated (weighted by error) to reach a final
Doppler shift; and the errors in the shift of each D line were combined in quadrature to reach final error values.
Final shifts were recorded as 𝑅 in Table 1, and final errors were recorded as 𝐸 . The Python code
used for these calculations is included in the “Zenodo” depository, and when run it will reproduce the data
analysis and figures used in this paper. Python version 3.9.7 was used.
 Appendix “B”

The equation for finding the predicted Doppler shifts in solar spectra that are reflected from a target under
observation from the earth (𝑅 ), and expressed in terms of velocity (km/s), is:

𝑅 𝑅
𝑅 =𝑅 +𝑅 +
𝑐

where 𝑅 is the target’s heliocentric velocity, 𝑅 is its geocentric velocity, and c is the speed of light
in a vacuum. To derive this equation, we start with the general Doppler equation for wavelength as a
function of radial velocity; and we apply that equation to find the initial Doppler-shifted wavelength of
solar light reaching the target (𝜆 ):

𝑉
𝜆 = ∙𝜆 + 𝜆
𝑐

where 𝜆 is the target’s wavelength at rest. To determine the final observed wavelength after light reflects
from the target (𝜆 ), we must apply a second Doppler shift due to the target’s geocentric velocity:

𝑅
𝜆 = ∙𝜆 + 𝜆
𝑐
𝑅
𝜆 = 𝜆 ∙ +1
𝑐

Finally, in order to express the observed wavelength as a shift in velocity (𝑅 ), and as a function of
the target’s heliocentric and geocentric velocities, we must again use the Doppler equation (this time
solved for radial velocity) and make the proper substitutions for 𝜆 and 𝜆 :
𝜆 − 𝜆
𝑅 = ∙𝑐
𝜆

𝜆 ∙ +1 − 𝜆
𝑅 = ∙𝑐
𝜆

∙
+ 𝜆 ∙ +1 − 𝜆
𝑅 = ∙𝑐
𝜆

𝑅
𝑅 = + 1 ∙ (𝑅 + 𝑐) − 𝑐
𝑐
𝑅 𝑅
𝑅 =𝑅 +𝑅 +
𝑐
 Table 1 – Summary of spectroscopic data/observations in this paper. All velocities are in km/s.

A. B. C. D. E. F. G. H. I. J. K. L. M. N. O. P. Q.
𝑅 𝐸 𝑅 𝐸 𝛥 𝐸 𝑅 𝐸 𝑅 𝐸 𝛥 𝐸
Epoch and
Obs. Target
Exposure
Image
# Name ID km km km km km km km A Endnote
Length km km km km km
s s s s s s s s s s s s

12/13/09 KOAID#
1 Europa 4:56 UT HI.200912
13.17797
+30.05 ±0.51 +40.34 ±0.03 +10.29 ±0.51 +7.83 ±1.99 +37.82 ±2.05 +2.52 ±2.05 1.52 i
(30 s)
12/11/09 KOAID#
2 Ganymede 4:53 UT HI.200912
11.17597
+14.29 ±0.51 +6.09 ±0.05 -9.20 ±0.51 -10.91 ±1.99 +3.38 ±2.05 +2.71 ±2.05 1.47 ii
(20 s)
6/6/07 KOAID#
3 Venus 5:32 UT HI.200706
06.19972
-14.04 ±0.52 -13.62 ±0.17 +0.42 ±0.55 +0.23 ±2.00 -13.81 ±2.07 +0.19 ±2.07 1.72 iii
(500 s)
1/7/09 KOAID#
4 Venus 4:33 UT HI.200901
07.16390
-12.78 ±0.52 -13.88 ±0.36 -1.10 ±0.63 -0.23 ±2.00 -13.01 ±2.07 -0.87 ±2.10 1.76 iv
(500 s)
6/17/05 KOAID#
5 Ceres 5:54 UT HI.200506
17.21254
+16.89 ±0.51 +18.25 ±0.48 +1.36 ±0.70 +1.43 ±1.99 +18.32 ±2.05 -0.07 ±2.11 1.25 v
(300 s)

D. 𝑅 = the target object’s calculated geocentric velocity at the date and time of observation, from the NASA/JPL Horizons ephemeris system.
E. 𝐸 = uncertainty in the target’s calculated geocentric velocity, due to axial rotation of the earth and of the target body.
F. 𝑅 = the Doppler-measured radial velocity. For more information on the data reduction and extraction process, please see Appendix “A.”
G. 𝐸 = uncertainty in the Doppler-measured radial velocity. For details, please see Appendix “A” or the main paper.
(𝑅 − 𝑅 ): the discrepancy in between Doppler-measured radial velocity (𝑅 , column “F”) and Horizons-calculated radial velocity
H. 𝛥 =
(𝑅 , column “D”).
I. 𝐸 = uncertainty in 𝛥 = (𝐸 ) + 𝐸 .
J. 𝑅 = the target object’s calculated heliocentric velocity, based on the NASA/JPL Horizons ephemeris system.
error in the object’s heliocentric velocity, due to rotation of the sun and target (which were combined in quadrature). Solar rotation was estimated at
K. 𝐸 =
±1.99 km/s.vi
L. 𝑅 = the predicted Doppler shift of solar light reflecting from the target, given by (𝑅 + 𝑅 ). (See Appendix “B.”)
M. 𝐸 = error in 𝑅 = (𝐸 ) + (𝐸 ) .
𝑅 –𝑅 , i.e. the difference in between the Doppler-measured radial velocity (𝑅 , column “F”) and the predicted Doppler shift in solar
N. 𝛥 =
light reflecting from the target (𝑅 , column “L”).
O. 𝐸 = uncertainty in 𝛥 , (𝐸 ) + 𝐸 + (𝐸𝑐𝑎𝑙𝑐 ) .

P. A = averaged airmass (as reported in the FITS header).

i
The D1 line was beyond the extracted orders; only the prominent D2 emission line was used.
ii
The D1 line was beyond the extracted orders, only the prominent D2 emission line was used.
iii
The prominent D1 and D2 emission lines were used.
iv
The prominent D1 and D2 emission lines were used.
v
The prominent D1 and D2 absorption lines were used.
vi
(Cox, A. 1999.) The solar equatorial circumference can be calculated as 2.720984 million miles, divided by the rotation period of 26.24 days to obtain a rotational
velocity at the solar equator of 1992.86 m/s.