2,000 Years of Puzzles,
Three Elegant Rules
February 2, 2026
“The planets wander… sometimes they appeared to move backward.”
For 2,000 years, astronomers asked why.
By the end of today, you can:
The Cosmic Puzzle
Why do planets move backward?
Stars rise and set predictably.
But a handful of bright lights wandered among the fixed stars.
Sometimes they appeared to stop, reverse direction, then resume their journey.
What could cause a celestial body to move backward?
Spoiler: retrograde is an apparent motion — an illusion of perspective.
Like passing a slower car: it can look like it drifts backward against distant mountains.
As Earth passes Mars, Mars appears to drift backward against the stars.
But Mars never reverses its orbital direction.
During Mars retrograde, Mars is actually moving:
The Ancient Answer
Earth at the center
Earth felt obviously central:
Ptolemy proposed epicycles:
Accurate, but complex: by medieval times the model needed dozens of circles.
The geocentric model protected two assumptions:
Earth is stationary and central — we don’t feel motion, so we must not be moving
Circular motion is perfect — orbits must be circles because circles are “perfect”
What happens when we challenge these?
The Revolution Begins
Copernicus, Galileo, Tycho, Kepler
What if the Sun is at the center?
Geocentric (Ptolemy)
Heliocentric (Copernicus)
What was the key advantage of Copernicus’s heliocentric model?
Galileo (1610) used telescopes to test models:
Evidence did not end the controversy.
20+ years of planetary observations
Greatest naked-eye observer in history
Tycho’s data was unprecedented in precision.
When he died in 1601, his observations passed to his young assistant…
Johannes Kepler
Kepler’s Laws
The Main Event
Kepler inherited Tycho’s precision Mars data.
For years he tried circles.
An 8 arcminute mismatch forced a new shape: an ellipse.
Suddenly, the orbit fit.
Planets orbit the Sun in ellipses, with the Sun at one focus.
An ellipse has two special points called foci.
Key property: Sum of distances from any point to both foci is constant.
Eccentricity (\(e\)): How “squashed” the ellipse is
| Term | Symbol | Definition |
|---|---|---|
| Semi-major axis | \(a\) | Half the longest diameter; “average” orbital distance |
| Eccentricity | \(e\) | How “squashed” (0 = circle, near 1 = elongated) |
| Perihelion | \(r_p\) | Closest point to Sun; \(r_p = a(1-e)\) |
| Aphelion | \(r_a\) | Farthest point from Sun; \(r_a = a(1+e)\) |
The Sun sits at one focus, not the center.
| Planet | Eccentricity | Description |
|---|---|---|
| Venus | 0.007 | Nearly circular |
| Earth | 0.017 | Nearly circular |
| Mars | 0.093 | Noticeably elliptical |
| Mercury | 0.206 | Most eccentric planet |
| Halley’s Comet | 0.97 | Extremely elongated |
Most planets have nearly circular orbits — this is why Ptolemy got reasonably good results with circles.
An orbit with eccentricity \(e = 0\) would be:
A line connecting a planet to the Sun sweeps out equal areas in equal times.
Perihelion (close to Sun):
Aphelion (far from Sun):
Bottom line: planets speed up when closer to the Sun.
Think of an ice skater spinning with arms extended.
Arms in: spins faster
Arms out: spins slower
This is conservation of angular momentum.
Same idea: closer to the Sun means faster; farther means slower.
| Position | Date | Distance from Sun | Earth’s Speed |
|---|---|---|---|
| Perihelion | ~Jan 3 | 147.1 million km | 30.3 km/s |
| Aphelion | ~Jul 4 | 152.1 million km | 29.3 km/s |
A planet is moving fastest when it is at:
Before we see the law: if we double the orbital size (\(a \to 2a\)), the orbital period \(P\) becomes:
The square of a planet’s orbital period is proportional to the cube of its semi-major axis.
\[P^2 \propto a^3\]
Let’s unpack the symbols:
In plain English: if you increase orbit size, the period increases more than linearly.
Two effects compound:
They have farther to travel (larger orbit circumference)
They move slower (farther from the Sun means weaker gravitational pull)
These combine to give the precise \(P^2 \propto a^3\) relationship.
Form 1: Ratio Method (recommended default)
\[\left(\frac{P_2}{P_1}\right)^2 = \left(\frac{a_2}{a_1}\right)^3\]
Compare two objects orbiting the same central body.
Form 2: Sun-only Scaling
For objects orbiting the Sun, with \(P\) in years and \(a\) in AU:
\[\left(\frac{P}{1\,\text{yr}}\right)^2 = \left(\frac{a}{1\,\text{AU}}\right)^3\]
Problem: A planet orbits the Sun at \(a = 4\) AU. What’s its period?
\[\left(\frac{P}{1\,\text{yr}}\right)^2 = \left(\frac{4\,\text{AU}}{1\,\text{AU}}\right)^3 = 64\]
\[P = \sqrt{64} \times 1\,\text{yr} = \boxed{8 \text{ years}}\]
Problem: A comet has orbital period \(P = 27\) years. What’s its semi-major axis?
\[\left(\frac{27\,\text{yr}}{1\,\text{yr}}\right)^2 = \left(\frac{a}{1\,\text{AU}}\right)^3\]
\[729 = \left(\frac{a}{1\,\text{AU}}\right)^3\]
\[a = \sqrt[3]{729} \times 1\,\text{AU} = \boxed{9 \text{ AU}}\]
An asteroid orbits the Sun at a distance of 4 AU. What is its orbital period?
| Planet | \(a\) (AU) | \(a^3\) | \(P\) (years) | \(P^2\) |
|---|---|---|---|---|
| Earth | 1.00 | 1.00 | 1.00 | 1.00 |
| Mars | 1.52 | 3.51 | 1.88 | 3.53 |
Notice: \(a^3 \approx P^2\).
| Planet | \(a\) (AU) | \(a^3\) | \(P\) (years) | \(P^2\) |
|---|---|---|---|---|
| Jupiter | 5.20 | 141 | 11.86 | 141 |
| Saturn | 9.54 | 868 | 29.46 | 868 |
Same pattern: \(a^3 \approx P^2\).
If a planet’s orbital distance \(a\) increases by a factor of 4, its orbital period \(P\) increases by a factor of:
The Limits of Patterns
What Kepler could — and couldn’t — explain
For the first time, humanity had a precise mathematical description of planetary motion.
Why ellipses? Why not circles?
Why do planets speed up when closer?
Why \(P^2 \propto a^3\)?
Does it work around other stars?
Empirical law: A pattern from data. Describes what happens.
Physical law: Explains why it happens.
Kepler’s laws are empirical.
They describe what planets do, beautifully and precisely.
But they don’t explain the underlying mechanism.
Kepler’s laws are considered “empirical” rather than “physical” because:
Newton (1687) gives the mechanism:
\[F = \frac{Gm_1m_2}{r^2}\]
Symbols:
Key idea: \(F \propto 1/r^2\).
How Science Evolves
From epicycles to ellipses
| Era | Model | Complexity |
|---|---|---|
| Ptolemy (~150 CE) | Geocentric + epicycles | Dozens of circles |
| Copernicus (1543) | Heliocentric + circles | Fewer circles, still complex |
| Kepler (1609-1619) | Heliocentric + ellipses | 3 laws, no epicycles |
Each step toward truth was also a step toward simplicity.
Retrograde is apparent (a geometry effect).
Kepler I: ellipses, Sun at a focus.
Kepler II: equal areas in equal times.
Kepler III: \(P^2 \propto a^3\) (use ratio or Sun-only form).
Kepler describes patterns; Newton explains mechanisms.
Better models are often simpler and more predictive.
Next lecture: Newton transforms patterns into physics.
We’ll see how one equation explains Kepler’s three laws.
Reading: Lecture 6 Reading Companion
Demo: Kepler’s Laws (Newton Mode): ../../../demos/keplers-laws/
Kepler gave us the patterns.
Newton will give us the physics.
Numerical Verification
Checking our work with real data
Given data:
Step 1: Convert period to years
\[P = 687 \text{ days} \times \frac{1 \text{ year}}{365.25 \text{ days}} = 1.881 \text{ years}\]
Step 2: Check \(P^2 = a^3\)
\[P^2 = (1.881)^2 = 3.538 \text{ yr}^2\]
\[a^3 = (1.524)^3 = 3.540 \text{ AU}^3\]
Match within rounding: \(3.538 \approx 3.540\) (0.06% difference from rounding)
Why does the shortcut look like time-squared equals distance-cubed?
The full physics statement is:
\[P^2 = \left(\frac{4\pi^2}{GM_\odot}\right) a^3\]
The constant \(\frac{4\pi^2}{GM_\odot}\) is doing the unit conversion.
Given data:
Convert and verify:
\[P = 4333 \text{ days} \times \frac{1 \text{ yr}}{365.25 \text{ days}} = 11.86 \text{ yr}\]
\[P^2 = (11.86)^2 = 140.7 \text{ yr}^2\]
\[a^3 = (5.203)^3 = 140.9 \text{ AU}^3\]
Match within rounding: \(140.7 \approx 140.9\)
Mistake (unit mismatch):
“\(a = 778\) million km, \(P = 12\) years, so \(P^2 = 144\) and \(a^3 = 4.7 \times 10^{26}\)… they don’t match!”
Fix (convert first):
\[a = 778 \times 10^6 \text{ km} \times \frac{1 \text{ AU}}{1.496 \times 10^8 \text{ km}} = 5.20 \text{ AU}\]
Then compare \(P^2\) (in yr\(^2\)) to \(a^3\) (in AU\(^3\)).
Mistake (wrong powers):
“\(P = 12\) years and \(a = 5.2\) AU, so \(12 = 5.2\)… wrong!”
Fix: Kepler III is about \(P^2\) and \(a^3\).
In Sun-only units (years and AU), you can do a quick consistency check:
\[(12)^2 = 144 \quad \text{and} \quad (5.2)^3 \approx 141\]
Deeper Dive
For curious minds
Kepler II says planets move faster when closer to the Sun. But how much faster?
From conservation of angular momentum:
\[v_p \times r_p = v_a \times r_a\]
where \(v_p\) = speed at perihelion, \(v_a\) = speed at aphelion.
What the symbols mean:
Starting from conservation of angular momentum:
\[v_p \times r_p = v_a \times r_a\]
Rearranging:
\[\frac{v_p}{v_a} = \frac{r_a}{r_p} = \frac{a(1+e)}{a(1-e)} = \frac{1+e}{1-e}\]
Sanity check: if \(e=0\) (a circle), then \(r_p=r_a\) and \(\frac{v_p}{v_a}=1\).
Mercury has eccentricity \(e = 0.206\).
\[\frac{v_p}{v_a} = \frac{1 + 0.206}{1 - 0.206} = \frac{1.206}{0.794} = 1.52\]
Mercury moves 52% faster at perihelion than at aphelion!
Compare to Earth (\(e = 0.017\)):
\[\frac{v_p}{v_a} = \frac{1.017}{0.983} = 1.035\]
Earth only varies by 3.5% — nearly circular orbit.
Newton will show us that orbital speed at distance \(r\) is:
\[v = \sqrt{GM\left(\frac{2}{r} - \frac{1}{a}\right)}\]
Assumptions: two-body, Newtonian gravity, spherical bodies.
Sanity check: if the orbit is circular, then \(r=a\) and \(v=\sqrt{GM/a}\) is constant.
At perihelion (\(r = r_p = a(1-e)\)):
\[v_p = \sqrt{\frac{GM}{a} \cdot \frac{1+e}{1-e}}\]
At aphelion (\(r = r_a = a(1+e)\)):
\[v_a = \sqrt{\frac{GM}{a} \cdot \frac{1-e}{1+e}}\]
We’ll derive this properly in L6 with Newton’s laws. For now, notice how \(e\) controls the speed variation.

ASTR 101 - Lecture 5 - Dr. Anna Rosen