Proof of a conjecture by H. Dullin and R. Montgomery

Imagine you are watching a cosmic dance involving three characters: a tiny, free-floating particle (like a dust mote) and two heavy, stationary stars fixed in space. This is the Euler problem, a classic puzzle in physics that has been around since the time of Euler and Jacobi.

The paper you provided is a mathematical detective story about figuring out exactly how long it takes for that dust mote to complete a specific loop in its dance.

Here is the breakdown of the paper's story, using simple analogies:

1. The Setup: The Cosmic Swing

In this problem, the dust mote is pulled by gravity from two fixed stars. Because the stars are fixed, the particle doesn't just fly away; it gets trapped in a complex, looping orbit.

Mathematicians have known for a long time how to calculate the time it takes to complete one of these loops (called a period). However, there was a catch. The existing math formulas were like a pair of glasses that only worked clearly when you looked at the orbit from one specific angle. If you tried to look at the orbit from the other side (a different range of energy and speed), the formulas became messy, complicated, and hard to use. They hit a "singularity"—a point where the math breaks down or becomes incredibly ugly.

2. The Goal: A New Pair of Glasses

The author, Gabriella Pinzari, wanted to create a new set of formulas that work perfectly on the other side of that singularity.

Think of it like this:

Old Formula: A map that is perfect for the "North" side of a mountain but becomes a jumbled scribble when you cross the peak to the "South" side.
New Formula: A second map that is a bit messy on the North side but gives you a crystal-clear, simple path on the South side.

By combining these two maps, the author creates a complete, simple guide for the entire mountain.

3. The Method: Two Different Tools

To build this new map, the author used two very different tools, corresponding to the two different "sides" of the problem:

The Dynamical Tool (The "Kepler" Trick):
On one side of the mountain, the author used a clever trick involving the Kepler problem (which is just the simpler case of one star and one planet). She realized that if you imagine the second star disappearing, the math becomes much simpler. She used this "limit" to derive a clean, simple formula for the period of the orbit. It's like realizing that if you ignore the wind, the path of a thrown ball is just a simple arc, and using that simple arc to understand the complex path.
The Analytic Tool (The "Complex" Magic):
On the other side, where the dynamical trick didn't quite work, she used Complex Analysis (a branch of math dealing with numbers that have imaginary parts). She treated the orbit as a shape in a complex geometric space. By using a specific type of mathematical "lens" (called an elliptic integral transformation), she proved that the messy old formula is actually mathematically identical to her new, simple formula. It's like proving that a complicated knot is actually just a simple loop if you look at it from the right angle in a higher dimension.

4. The Big Win: Proving the Conjecture

The main reason for doing all this hard math was to prove a guess (a conjecture) made by two other scientists, H. Dullin and R. Montgomery.

The Guess: They suspected that as you change the energy of the system (specifically, a value called the "first integral"), the time it takes to complete a loop changes in a very predictable, smooth way. Specifically, they thought the time would always go up or always go down (monotonicity) without ever zig-zagging back and forth.

The Proof:
By creating these new, simple formulas, the author could easily see the behavior of the orbit.

She showed that the time it takes to orbit is indeed a smooth, predictable function.
She also looked at the rotation number (the ratio of two different periods). This is like checking if the dancer's steps are perfectly synchronized. She proved that this ratio also changes smoothly and predictably as you tweak the energy.

Summary

In short, this paper is about simplifying the complicated.

The Problem: Existing math for calculating orbital periods was too messy on one side of the energy spectrum.
The Solution: The author derived new, simpler formulas for that messy side by borrowing ideas from simpler planetary motion and using advanced geometry.
The Result: With these new tools, she proved that the time it takes for these particles to orbit, and the ratio of their movements, changes in a perfectly smooth, predictable manner. This confirms a long-standing guess by other mathematicians and provides a cleaner way to study these cosmic dances.

The paper does not discuss medical applications or future technologies; it is purely a victory in the world of theoretical mathematics and physics, clearing up a foggy area of a classic problem.

1. Problem Statement

The paper addresses the Planar Euler Problem (also known as the two-fixed-center problem), a classical integrable Hamiltonian system describing the motion of a particle under the gravitational attraction of two fixed masses ( $M$ and $m$ ).

Context: The system is integrable via Jacobi's separation of variables using elliptic coordinates ( $\alpha, \beta$ ). The dynamics are characterized by two periods, $\tau_+$ and $\tau_-$ , corresponding to the motion in the $\alpha$ and $\beta$ directions, respectively.
The Issue: While the periods are well-known in the literature, the existing formulae (expressed as elliptic integrals) are simple and well-behaved only on one side of a specific singularity in the parameter space. On the other side of the singularity, the formulae become complicated (involving variable integration limits and complex root structures), making analytical study difficult.
The Conjecture: H. Dullin and R. Montgomery conjectured that the periods $\tau_+$ , $\tau_-$ , and their ratio (the rotation number $\omega = \tau_-/\tau_+$ ) are monotone functions of the non-trivial first integral (denoted as $f$ ) at any fixed energy level. Proving this requires simple, unified formulae for the periods across the entire domain of quasi-periodic orbits.

2. Methodology

The author employs a hybrid approach combining dynamical systems theory and complex analysis (elliptic integrals).

A. Parameterization and Domain Decomposition

The problem is reformulated using parameters $d$ (related to energy) and $f$ (related to the first integral). The domain of valid parameters $P$ is decomposed into two regions relative to a "singular line" $f = f_s(d)$ :

$Q_M^\uparrow$ (Upper region): Above the singular line.
$Q_M^\downarrow$ (Lower region): Below the singular line.

The classical formulae for the periods are known to be simple in the upper region but complex in the lower region.

B. Step 1: Dynamical Argument (The Keplerian Limit)

To derive a simple formula for the lower region ( $Q_M^\downarrow$ ), the author utilizes the Keplerian limit ( $m \to 0$ ).

By considering the system as a perturbation of the Kepler problem (one center), the author reconstructs the periods of the two-center problem from the periods of the Kepler problem.
In the Kepler limit, the orbits are periodic, and the periods can be calculated using a specific integral transformation.
The author proves that for the lower region, the period $\tau_M$ can be expressed as a single integral with fixed limits (from 0 to 2), significantly simplifying the expression compared to the classical variable-limit integrals.

C. Step 2: Analytic Argument (Complex Analysis)

To rigorously extend the validity of the simplified formula from the Keplerian limit to the full two-center problem in the lower region, the author uses complex analysis.

Proposition 4.1 (Equivalence of Complex Integrals): The core technical tool is a lemma establishing the equivalence of elliptic integrals under specific Möbius transformations ( $\Gamma_\varrho$ ).
The author defines a transformation that maps the integration path of the complicated classical integral to the path of the simplified integral derived in the dynamical step.
By analyzing the roots of the polynomials under the square roots of the integrands and ensuring no roots lie in specific regions of the complex plane (using Cauchy's theorem and contour deformation), the author proves that the classical integral and the new simplified integral are identical.

3. Key Contributions

New Formulae for Jacobi Periods: The paper provides a new, unified representation for the periods $\tau_+$ $τ_{+}$ and $\tau_-$ $τ_{-}$ (denoted as $\theta_M$ $θ_{M}$ ) that is simple on both sides of the singularity.
- For the lower region ( $Q_M^\downarrow$ ), the period is given by:
  $\tau_M = \sqrt{\frac{2d}{v_0}} \int_0^2 \frac{dz}{\sqrt{z(2-z)(M^2z^2 - 2fz + d^2)}}$
  This contrasts with the classical formula which requires variable upper limits dependent on the roots of the polynomial.
Proof of the Dullin-Montgomery Conjecture: Using the new formulae, the author proves that $\tau_+$ , $\tau_-$ , and the rotation number $\omega$ are differentiable and monotone with respect to the parameter $f$ on every connected component of the parameter space.
Analytic Continuation: The work demonstrates how dynamical insights (Keplerian limits) can be rigorously extended to general integrable systems using complex analytic continuation of elliptic integrals.

4. Main Results

Theorem 1.1: Establishes the identity $\tau_\pm = \theta_{M_\pm}|_P$ , confirming that the new simplified formulae $\theta_M$ correctly represent the periods across the entire domain $P$ , including the region where classical formulae were cumbersome.
Theorem 1.2: Proves the monotonicity of the periods and the rotation number. Specifically:
- $\tau_-$ and $\tau_+$ are monotone functions of $f$ .
- The rotation number $\omega(d, f)$ $ω (d, f)$ exhibits specific monotonic behaviors in different sub-regions:
  - It increases from a finite value to $+\infty$ in the "Small" region ( $S$ ).
  - It decreases from $+\infty$ to $0$ in the "Large" region ( $L$ ).
  - It increases from $0$ to a finite value in the "Periodic" region ( $P$ ).

5. Significance and Applications

Resolution of an Open Problem: The paper definitively settles the conjecture by Dullin and Montgomery regarding the monotonicity of periods, a property that was previously only verified numerically or in specific limiting cases.
Foundation for Perturbation Theory: The monotonicity of the rotation number is a crucial condition for the applicability of KAM theory (Kolmogorov-Arnold-Moser) and Aubry-Mather theory. This result allows these theories to be applied to systems close to the two-center problem, such as the restricted three-body problem or planetary systems.
Nekhoroshev Stability: The monotonicity of periods is linked to convexity properties, which are essential for Nekhoroshev theory regarding the long-term stability of nearly integrable Hamiltonian systems.
Rabinowitz Floer Homology: The results have already been applied in the context of Rabinowitz Floer homology to study periodic orbits in Hamiltonian systems.

In summary, Pinzari's paper bridges the gap between classical mechanics and modern dynamical systems theory by providing elegant, unified formulae for the Euler problem, thereby unlocking the proof of fundamental monotonicity properties essential for understanding the stability and chaos in celestial mechanics.